Telirati Newsletter #53

In this dispatch, I give a very postive review to .NET, which was, at the time, the best archiecture for multi-tier Web applications, and which, if Microsoft had executed on the potential to use .NET to tie desktops to Web services, could have enabled Microsft to dominate Web applications.

Four years later, Microsoft still can't seem to unholster the .NET gun, which could make Google's JavaScript hacks look pathetic compared to the level of desktop integration that mail, calendars, and search that employed .NET could provide.

Telirati Newsletter #53: What C#, .Net, SOAP, and NGWS Really Means

Microsoft holds some kind of world record for inept naming schemes for what is really pretty simple stuff. The whole COM/ActiveX nomenclature swamp, for example. It all refers to various levels of interface conventions in the COM distributed component system. (And you thought DCOM was the distributed version, eh?) That’s it. Now that you understand that, be assured the Microsoft nomenclatura have been working overtime to bring you the next deluge of impenetrable labels. Which is why you should turn to your trusted chronicler for the Magic Decoder Ring:

C#: A computer language. Who needs a new one? You do, trust me. And C# is a good one. As good as Visual J++, but without the lawsuits. Better than Visual C++, which is encrusted with macros and declaration obligatos that impede the coder and obscure the result from others understanding it. Better than Visual Basic, as much as Java is better, adding strong typing and removing the legacy compatibility that invites mistakes through lack of historical awareness.

.Net: It means many things (and that’s Microsoft for ‘ya). But the one we’ll focus on is that it means a common set of class libraries for all languages that enable creation of a desktop user interface, access to databases, access to other important Windows APIs, as well as a common set of class libraries and architecture for creating Web server applications. It may also mean support for Windows code associated with Web pages that enables code to “animate” Web pages behind the scenes. This aspect of .Net means Microsoft has finally unified the class libraries, and, implicitly (when you consider support for Java-style reflection and COM in the class libraries), the execution environment, for all languages for Windows programming. Halleluiah! But this is more than convenience for programmers: it represents the recognition that the class library, and not the API, is the platform. Pop-quiz for would be Microsoft-busters: Is .Net part of the OS, or a common library used by applications? Who maintains it? And who directs the architectural direction? Bzzzt, time’s up, thank you for playing, Judge Jackson.

NGWS: Next Generation Windows Services. This is stuff built using .Net that makes building Web sites easy. One of the most important aspects of it is that it enables easy outsourcing of Web site functions, like credit card processing, user identity management, chat support, call-center integration, etc. It represents the productization of distributed component interfaces.

SOAP: A common underlying protocol for distributed component (object) systems. Whew! And you thought this was going to be a simple newsletter. Again, trust me, you need it. The old way of making COM talk to CORBA or other distributed component systems involved gateways that were supposed to translate interactions. These presented scalability bottlenecks. They were always implemented by COM bigots, or CORBA bigots, or some other bigot who thought the other guys’ architectures were secondary. And, though I’m sure those who toiled in the gateway vineyards might differ, they never worked as well as one would want. SOAP puts interoperation where it belongs: in a common, standardized, Web-oriented protocol based on XML. It makes finding out about distributed component interfaces an XML application. And it makes software that does not employ distributed component systems, such as legacy database and transaction processing systems, able to play in a distributed object world through the use of XML.

What does it all add up to? It adds up to the fact that Microsoft gets it. It will be easier to make architecturally sophisticated and interdependent Web sites with Windows programming tools, Windows 2000 servers, and services from Windows-oriented B2B sites than with any other family of tools, OSs, and Web services.

Yikes! Does this mean Microsoft has a good shot at being the dominant force in Web-based systems? Does this mean that the loosely integrated Oracle/Solaris/Java way of making big Web systems might be overwhelmed? It sure does. Has Microsoft behaved virtuously? It has: SOAP is a benefit to all, though it does have the effect of bringing into relief the fact that COM is by far the most widely deployed distributed component system (marketing studies indicating CORBA and COM are in some kind of parity only take into account large IT projects, and ignore the fact that COM is vital to the operation of every Windows desktop).

Does it mean Windows CE will topple Palm with a startling burst of Web-oriented distributed applications that take advantage of the multi-platform execution environment in .Net? Ah… no. Not real soon anyway. That will have to wait for the deployment of real mobile wireless Web access at attractive prices and I don’t mean WAP.

One thing to understand is that while Microsoft has shipped some dodgy stuff, in the form of an over ambitious windowing environment built on top of a creaky DOS foundation, Microsoft has generally prevailed on the merits. Even Windows 95 was better than Apple’s development torpor at the time, not to mention Sun’s desktop user interface strategy. Oh, sorry, they didn’t have one. Allegations that Microsoft out-markets its competitors will have to come up with an explanation for the goofy naming conventions: perhaps a New World Order conspiracy to lull the masses into paying no attention to Microsoft’s plans for world domination. Well the Orbiting Mind Control Satellites are powering-up again, this time to turn all those Visual Basic zombies into tools of Web domination.

More seriously, those who fail to notice that Windows 2000 is very nice product indeed, and who fail to see the parallels between Microsoft having gobbled up the desktop, from school-child’s learning toy to engineer’s CAD station, based on products that are cheaper and easier to buy and use, will fail to see that complex multi-tier interconnected Web sites would benefit from the same formula. Sun won’t vanish tomorrow, but if Sun does not come up with a direct response to the PC server challenge, Sun will soon enough be marginalized, just as minicomputers and mainframes have been. And those who think this is a bad thing, and who would step up their efforts to crush Microsoft with the tools of state, will only deny the economy of the benefit of a significant productivity gain. If Microsoft’s potential to make Web development more productive is considered in the same light as a new vaccine or a more-efficient electric generator, keeping Microsoft down qualifies as equivalent to cutting off our nose to spite our face.

Copyright 2000 Zigurd Mednieks.

Telirati Newsletter #52

In this Telirati newsletter the question of when to follow and when to lead is addressed.

Telirati Newsletter #52: Imitation, Flattery, Competition

The Telirati do not do product reviews. In this case, examining some of Microsoft’s new products indicates company direction. Mostly, it is the right direction.

Microsoft is revamping MSN. You may not like the result, which is a customized browser with plenty of Internet dumbing-down, like big colorful buttons you can hit two drinks after your friends hide the car keys. But this is what was needed long ago as AOL was carpeting the U.S. with disks. The new MSN is exactly what one expects from Microsoft: aggressive pursuit of a market leader with a high-quality implementation of a flatteringly similar idea.

The main lesson here is that marketing strategy is mainly a question of when to follow and when to lead. If you follow, your results will be predictable. In a growing market, achieving results similar to those of a market leader is a good thing. So, while pabulum content and icons-for-morons might annoy you, you don’t have to read it and you can still use your unmodified browser at MSN sites. Embracing AOL-level customer intelligence and extending the idea to a very well-done Windows/IE-oriented implementation will, no doubt, be to Microsoft’s benefit.

Microsoft has another large initiative in the works: enabling digital publishing of music and books. This encompasses a range of products from the new Pocket PCs to the new Media Player software to a digital document reader called, obscurely enough, “Reader.” Here the mix of imitation an innovation is more problematic. Like many electronic book technology providers, Microsoft has included “digital rights management” (DRM) technology in its software. That’s fine. Publishers can choose to use it, or not. Microsoft’s Media Player incorporates similar capability for music.

The problems with DRM are twofold: For Microsoft, the problem is that the goal of popularizing Reader, their new Media Player, and Pocket PC is in conflict with the goal of getting content publishers comfortable with electronic publishing. Here, imitation flatters only past failures. Past electronic publishing initiatives foundered on over-protective content protection, and so, it seems will current efforts. Without an aggressive move to break out of the copy-paranoid mode of previous electronic media publishing attempts, aggressive following will only lead down the path of similarly tepid results.

Digital Rights Management and other content schemes are inconvenient, brittle, and contrary to the notion of being secure in one’s own computer. DRM enables bits, including executable bits, to be hidden from a computer’s owner. This should be alarming to anyone who takes the security of electronic documents seriously. On the one hand, Windows 2000 offers excellent security features like an encrypted file system that can probably put documents out of the reach of anyone except the most elite code-cracking resources, and the fact that the extent of national security code-cracking must be kept secret effectively prevents the abuse this capability in garden variety corruption. This is good. But then Digital Rights Management creates an infrastructure for snooping. This is very very bad. A trusted system cannot serve two masters. Your computer can be trusted by you, or it can be trusted by content publishers.

Content publishers are deathly afraid of rampant illegal copying. This has driven them not only to demand dumb and dangerous content protection technology but also to supporting dubious laws like the Digital Millennium Copyright Act (DMCA). Invoking the DMCA, which was supported by Microsoft, has put Microsoft in such worthy company as the Church of Scientology, and it has pitched Microsoft against worthy outfits like Slashdot in a public relations fiasco over what was supposed to be an act of magnanimity on Microsoft’s part in releasing a specification for Kerberos extensions. One could, one has, and one no doubt will go on at length about the evils of the DMCA, but the down-side for Microsoft is that sticking to the same old recipe will yield the same old results, and, on top of it, cause much embarrassment and loathing.

In telephony, (once again, that pesky topic sneaks in) the question is in how to crack the stagnant CPE market. Small and medium business CPE is a large but torpid trade where market-share shifts with the speed of molasses in January. Here, using the same recipe, which is to say traditional channels and promotional tools, not only limits your market to current established boundaries, it limits your ability to make headway in that market to what appears to be a speed limit of about 0.5% per year. At that rate, most startups will starve before they reach a viable sustainable market share. Aggressively following established leaders into the conventional telephony channel has proved to be a quagmire for telephony servers, convergence products, and other novel CPE.

The answers have to be comprehensive and bold. In order to get Pocket PCs and digital media distribution off the ground, two approaches can break the logjam: One is to make large amounts of public domain and freely distributed content available, and the other is to lead the publishers by distributing paid content with less-intrusive security measures like watermarks that would still enable detection of industrial-scale piracy without inconveniencing, or reducing the privacy of, the consumer. This may not seem particularly bold, but it does mean scaring rather than seducing content publishers. Unless publishers fear commercial irrelevance more than they fear digital distribution, digital distribution will remain stifled by heavy-handed content protection. But this is the only way forward. Unless Pocket PCs break away from the past, they will follow a predictable path into obscurity.

In telephony, breaking out means developing products that are consumer friendly, have features that truly deliver value – screening nuisance calls, for example, because telephone companies will never respond with an equivalent subscription service (Caller-ID helps, but it is not a cure), and, especially, it means finding and exploiting channels that are different from established telephony channels. Where are the established player not? They are not selling direct. They are not advertising on the Internet. They are not selling superior customer relationship management capabilities to e-commerce startups. They are not linking their products to B2B portals. They are not creating features that antagonize the telephone companies by serving the individual rather than the telemarketer. These are the places telephony ventures must go in order to crack open a valuable but heavily encrusted prize. In telephony, especially in CPE for small and medium business, aggressive following, much less tepid and timid following, will not get you far.

Copyright 2000 Zigurd Mednieks.

Telirati Newsletter #50

In this number, I kick sand in the face of Y2K alarmists, having successfully predicted nothing would happen. The meanstream media was wrong about Y2K, and blogs like Slashdot got it right, presaging the current "discovery" of the blogoshpere by the mainstream media by nealry 5 years.

Telirati Newsletter #50: Why Too 'K?

No accidental nuclear war. No food riots prompting the imposition of global government implemented by U.N. zombie soldiers in black helicopters. Did you even get a fax dated 1900? Y2K went more than just OK, it went swimmingly OK. But then, this was predictable (and predicted in the newsletter entitled "Why Tu Que?") Before returning to rubbing it in, let's look at why this was so predictable.

First, and most important to those of you in the computing and telecom industry: the number of mission critical systems is vastly overestimated. Most systems hardened against failure and operated by a team of operators that wear their pagers proudly to display their importance are not actually mission critical outside of the mission of enhancing the significance of the people operating them. By contrast, the small number of actually critical systems were long since well taken care of, if there had been any Y2K concern over them in the first place.

Corollary to this rule are the fact that Russia, Nigeria, Mexico, and other chaotic places are chaotic for reasons like cleptocratic government, organized crime, and other forces far more life-threatening than any computer bug. It is unlikely that the balance of chaos in these societies would be upset by anything less substantial than a few planeloads of AK-47s and a coven of spies. Computers, as powerful as they are, are not up to fomenting revolution, and their relative lack of importance is viscerally known to every banana republic strongman with the wits to live through next week.

Second, expectations of Y2K were driven by a singularly unreliable transmitter: the "mainstream media." If there is one lesson to learn from Y2K it is that independent, non-traditional, and largely Internet-based information sources are at least as good as the so-called "mainstream press." Yes, kooks use the Internet as their trumpet, but they are easy to spot. Instead, the calm that pervaded independent Internet news sources like NewsMax, StratFor, the undeservedly maligned Drudge Report, and the non-traditional computer-oriented sites like Slashdot indicated that there would be no significant trouble. This illustrates that the "mainstream" press has been disintermediated as an effective way to get the most important news fastest. It now picks up and retransmits things we already know to the unplugged and generally apathetic, while retaining its other function as spin amplifier for views sympathetic to the journalistic subculture of people who have never made an actual functional work product and sold it. Hence we see the spectacle of formerly respectable institutions such as the New York Times and BBC proclaiming the non-millennium as if it were the millennium if only to avoid contradicting the political objects of their toadyism who have stood for dumbing down the calendar.

It is far too much to expect the traditional press to comprehend and accurately portray facts such as that computer systems of any non-trivial complexity have dozens to hundreds of bugs that can cause them to stop functioning. Increased taxes would not fix this. Regulations would not fix this. More social workers would not fix this. A lawsuit would not fix this. One could blame it on white men, but those were antediluvian times when white men with short hair wrote code. Nobody would get the connection, since everyone who relies on the mainstream media knows that most programmers look and act like Jaron Lanier (who invented virtual reality in the lab just down the hall from where Al Gore, who must have at the time sported a similarly Rastafarian haircut, was inventing the Internet). It is therefore impossible for most young reporters who have been to journalism school in the last 20 years to write a concluding sentence to a news item accurately describing the danger, or lack of it, from the Y2K bug.

Then there is the minor point that it wasn't actually the millennium. The new millennium, the next thousand years after the first and second thousand in the Western calendar's numbering scheme, begins at the end of this year. The first year of the new millennium is 2001. If one is really serious about immanentizing the eschaton, which is not a task to be taken lightly, one would make sure not to martyr oneself and appear before the higher authority a year early. Oh faux pas! Which only goes to show that the average fanatical zealot can get the calendar right when a certain prodigal Rhodes scholar cannot, or chooses a path of ignorance on purpose. Should one be more worried about the end of this year? Probably. But even if you think the world is done on the granularity of millennia, it is a bit arrogant to suggest that you will be walking the earth when the End of Days does arrive. Even the fieriest millennialist may have to concede that Bill Clinton makes a pretty lame antichrist. He's smart enough, but the ease with which he is distracted from his job does not exactly fit the profile of "working like the devil."

There are however, some serious pitfalls to avoid. While there is no orderly or competent New World Order, there are thousands of little Pol Pots in bureaucracies all over the world that would take advantage of any opportunity to regulate your life in accordance with their pet theory of society. Some boondoggle monger got the U.S. President to stand up and spend a considerable amount of credibility on asking for government money for an anti-cyber-terror squad just as the FBI was admitting there was literally not a single incident of Y2K cyber-terror that stood out from the normal level of cracker jiggery-pokery of port scans and spammers shanghaiing mail servers. Resisting such cyber-pork swindles will be a tiresome but constant and important chore for cyber citizens that want to keep the golden goose out of the kettle of the government chef.

Watch your wallet around government men. Don't trust the big media companies. Do listen to your own judgment based on data you collect yourself. Go forth and prosper in the New Year.

Copyright 2000 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #49

Four years ago, the specter of content protection stalked the PC. Four year on, it's still out there, threatening to turn the magic of a general machine into something more like television.

Telirati Newsletter #49: Content protection: diabolical, or just evil?

First, we start with the good things about content protection: (let me know if you come up with any).

OK, now let’s get into what is wrong with it:

Content protection is the end of personal computing. This may sound a bit apocalyptic, but it really only means that the reasons the personal computer is popular could all be extinguished through content protection. The personal computer became popular because it was completely under the control of the individual who owns it. Subsequently, in the name of orderly corporate computing environments and other supposedly worthy causes that control is eroded. But, still, you can purchase a computer, a completely general machine, capable of any function, programmable to any purpose, and control it completely and utterly.

The attractions of this total control range from those based in the philosophical freedom it gives to create forms of information processing that authorities cannot control and that can embody seditious intent, to the rather more mundane blessing of getting one’s work done no matter how inefficient one’s IT department is. Everything from the free choice of digital music formats to the world-shaking possibilities of anonymous digital cash depend on the individual being able to completely control and trust a computing machine.

What happens when you destroy this trust? The PC industry, as it is currently constituted, dies. Forget, if it makes it easier, the moral dimensions of content protection. Content protection is directly inimical to the foundation of the industry and the attraction the customer has to its product.

Content protection is also equivalent to key escrow. Remember key escrow? That cold, clammy feeling of the spook’s hand on your shoulder all the time? Content protection, which must enable your computer to keep secrets from you and rat you out is literally identical to key escrow in its ability to enable Big Brother to take up residence in your computer. Perhaps even worse. The presence of hidden software, undetectable execution and storage, and use of network connections without computer users’ knowledge of such communications can turn the PC into an instrument of surveillance. Content protection is, in fact evil.

But is it diabolical? Yes, if the criterion of diabolism is that it can make intelligent people act against their self-interest. Microsoft, for example. Microsoft is in the process of flushing tens to hundreds of millions of dollars down the toilet in the name of content protection. How? By failing to see that Windows handhelds need a killer app. Well maybe they see it, but they don’t have the will to pursue it.

The killer app, of course, is digital media delivery. Windows handhelds, with the high resolution text and powerful media player, are potentially a very attractive platform for delivery of digital audio and books. So far so good: an application has been connected to product features. Somebody read up on how the Mac’s bacon was saved with desktop publishing. But Microsoft has left out a key ingredient: creative destruction.

If you are old enough, you might remember that desktop publishing destroyed several industries and professions: Page composition workstations, layout service bureaus, and several other crafts and their tools went extinct, or were decimated, by desktop publishing. Almost all publishing today is desktop publishing. Apple was big enough, and the conventional page composition industry and service infrastructure was small enough and invisible to the wider public, enough so that it could be wiped out without much outcry. Microsoft has a bigger challenge.

If Microsoft really wants Windows handhelds to take off it will do this: Instead of playing footsy with the content publishers, seed the content directly. For music, this means allying with a Napster-like service or adding unattended file transfer capability to Windows Messenger, NetMeeting, or some other utility and feeding it with an alliance with MP3.com or the like. For books, it means directly funding one of the sleepy and underfed public-domain electronic library projects and making sure they can deliver in a Windows handheld-friendly format that takes advantage of the new high-resolution text display technology. Or an e-learning initiative like the one in Cambridge Massachusetts that aims to teach an MIT-level computer science education in one year – free or low-cost college courses in the palm of your hand. In all these cases, however, someone’s ox ends up as the main course of the luau.

The music industry has demonstrated that they are bastards of the first water. Book publishers, while more gentlemanly, nonetheless will not be cheered by a flood of public domain reading material packaged in the hottest new format and delivered by a super-slick Web site. And universities in the U.S. have spent the last two decades raising tuition far faster than costs in general. But the fact is, if nobody gets pissed off at you, it is a sign you are not delivering value. What do you expect? To be able to create sufficient value out of thin air to motivate people to spend $500 on a new and untried tool that already has a history of dubious utility? No, unless take the value out of some other industry, you don’t have a killer app. If you don’t have the will to destroy, you will never create anything big.

The question is: What is it worth? Does Microsoft really want to succeed with handhelds, or are they a hobby? Does Microsoft remember what made PCs an unstoppable force in the first place? Or do they really think they cannot be replaced by a system that respects the customer more than it respects slimy payola-dealing record companies? Is content protection a concept so diabolical it can cause the downfall of an enterprise more illustrious than any that came before, and do it far more surely than a bunch of DoJ pinkos?

The other side of the same coin is that actors, musicians, and authors and others sometimes act as irrationally as makers of digital media hardware: Before there were record labels, almost all musicians had day jobs as music teachers. A small number of elite musicians were subsidized by sponsors. Now, a small fraction of a percent of musicians make a living by selling records, supplemented by commercial sponsorship of concerts, and the concept of “elite” extents to The Backstreet Boys. Even if the “worst” were to happen and record companies went out of business, the effect on musicians would, in the context of their total population, be almost undetectable. As horrible as it is to contemplate that the world might never have heard of Brittany Spears, it is not outlandish to suggest that the diffusion of commercial activity now concentrated on both the deserving elite and the manufactured pop icons out onto the whole real world of music-making might be a net benefit.

Then there is the danger is that the idea of intellectual property could be fundamentally damaged. Laws granting limited protection to intellectual property have fostered an era of wealth creation that was impossible in a time when land and minerals were the only defensible form of wealth and value. Some people believe that intellectual property is so artificial as to be immoral. Some even believe all property ownership is immoral. The record, however, shows that people take care of what they own far better than any form of collective stewardship has ever done. So when abominations like the Digital Millennium Copyright Act do the equivalent of loosing vicious dogs in the neighborhood in order to protect a single lot from trespass, the reaction is to shoot the dogs and hang the landlord. Some time ago we reached a sensible and fair balance between absolute and eternal control over a work of art or literature and the public good of eventual free dissemination. Modern efforts to enlist government in oppressive and unfair schemes that directly contravene established doctrines of fair use only strengthen the hand of radicals who would overthrow property in general. These bad laws also create needless conflict between the nature of the Internet and the personal computer on the one hand, and intrusive laws that create thought-crimes and sneaky and brittle protection technologies on the other hand. This conflict is bad for commerce as a whole.

Far better to let change happen. Far better not to subject the legal traditions of patent and copyright to artificially created stresses since they face sufficient natural challenges. And far better for the shareholder, the musician, and the author for Microsoft, or Palm if they are unwilling, to take up the challenge of digital media distribution in a post-Time-Warner era. Such are the risks and rewards that will create the next great wave of wealth.

Copyright 2000 Zigurd Mednieks.

Telirati Newsletter #48

This newsletter, written 4 years ago, describes a lesson still valid today: New technologies have to be an improvement over what they replace - a lesson VoIP has only now applied with the superior sound quality available with Skype.

Telirati Newsletter #48: What Roger Ebert Can Teach Us About Telephony

What can a movie critic, a fat man in a tweed jacket, teach us about telephony? A lot, as you will see. Telephony, because it is so widely used, is about what people want, much the same as making and showing movies. In a recent column, Roger Ebert, a noted movie critic, takes his gaze away from the screen and looks into the projection booth. There he finds digital cinema technology trying to make an unwise leap and landing under the wheels of the juggernaut of refined existing systems:

I have seen the future of the cinema, and it is not digital. No matter what you've read, the movie theater of the future will not use digital video projectors, and it will not beam the signal down from satellites. It will use film, and the film will be right there in the theater with you.

Then there is the hype:

But how good is digital projection? I saw it demonstrated in May at the Cannes Film Festival, and have read reports of those who've attended the custom "Phantom Menace" installations. A system offered by Hughes is not very persuasive, the witnesses say, but the Texas Instruments system is better; reviews range from "85 percent as good as a real movie" to "about as good." The special effects in "Phantom Menace" looked especially sharp, viewers said, and a reason: They were computer-generated in the first place, so they arrived at the screen without stepping down a generation to film. And because they depicted imaginary places, it was impossible to judge them on the basis of how we know the real world looks.

For all the hype, "about as good" is about as good as reviews get. Remind you of anything? Like how H.323 and most other forms of IP telephony is "about as good" as normal switched-circuit telephony? "About as good" should ring alarm bells to management and investors. Nobody wants to get a new technology that is about as good as the old one. In one case, we have a digital cinema technology that delivers less resolution than film, in the case of telephony, a technology that could deliver better phone calls, but generally doesn't, because of poor choices in product formulation. In the case of cinema, what is going to keep digital projection at bay?

"Dijection" offers a wonderful new prospect, if it's for real. But it's not the only possible future. Far from the boardrooms of Texas Instruments, which has unlimited financial resources and wants to grab the world movie distribution market, there is an alternative film-based projection system that is much cheaper than digital, uses existing technology and (hold onto your hats) is not "about as good" as existing film, but, its inventors claim, 500 percent better. That is not a misprint. This system is called MaxiVision48. I have seen it demonstrated. It produces a picture so breathtakingly clear it is like 3-D in reverse: like looking through an open window into the real world. Motion is shown without the jumpiness and blurring of existing film projection, details are sharper, and our eyes are bathed in visual persuasion.

Is Ebert describing something like IMAX? No. IMAX has its own problems. The system he describes is a relatively simple and quite inexpensive optimization of 35mm film projection:

Without getting into labyrinthine technical explanations, here is how MaxiVision48 works: It can project film at 48 frames per second, twice the existing 24-fps rate. That provides a picture of startling clarity. At 48 frames, it uses 50 percent more film than at present. But MV48 also has an "economy mode"... MV48 uses a new system to pull the film past the projector bulb without any jitter or bounce... MV48 completely eliminates the jiggle that all current films experience as they dance past the projector bulb. Watching it, I was startled to see how rock solid the picture was, and how that added to clarity... The result: "We figure it's 500 percent better than existing film or the Texas Instruments video projection system, take your choice," Goodhill (the inventor) told me.

Unfortunately for the digital cinema equipment providers, this refinement is startlingly inexpensive:

It is also a lot cheaper, because it retrofits existing projectors, uses the original lamp housings and doesn't involve installing high-tech computer equipment. MaxiVision's business plan calls for leasing the projectors at $280 a month, but if you wanted to buy one, it would cost you about $10,000. Estimates for the Texas Instruments digital projector range from $110,000 to $150,000 per screen.

In this, we see reflected other aspects of how the new telephony is actually unfolding, in contrast with some predictions. Instead of the telephone infrastructure (and incumbent operators) being replaced, they are being refined. Fully IP-based infrastructures may be on the way, but solutions that focus on true network operator needs are likely to do quite well until IP calls are not just "nearly as good" or even "just as good," but will likely be dominant until anyone offering an alternative technology delivers along with it phone calls that are really better.

Well, more expensive and not better seldom beats cheaper and better, but this is not the most interesting part of the comparison. The really interesting part is the fact that the refinement trumps the new technology by offering a higher quality expeience to the viewer. The system's inventors seem to grasp the needs of the people making the pictures, which is how they arrived at a system that delivers something superior to every point in the value chain: a superior tool to the cinematographer, an economical system to the cimema operator, and a visibly better result to the customer buying the ticket:

One advantage of a film print is that the director and cinematographer can "time" the print to be sure the colors and visual elements are right. In a digital theater, the projectionist would be free to adjust the color, tint and contrast according to his whims. Since many projectionists do not even know how to properly frame a picture or set the correct lamp brightness, this is a frightening prospect... We saw a scene that had been shot for Goodhill by another cameraman who likes the system, Steven Poster, vice president of the American Society of Cinematographers. Poster deliberately assembled a scene filled with technical pitfalls for traditional film and video systems:

We see actor Peter Billingsley walking toward the camera, wearing a patterned shirt. He is passed by another guy, wearing a T-shirt with something written on it. The camera tilts down as Billingsley picks up a hose to water a lawn. The camera continues to move past a white picket fence. In the background, a truck drives out of a parking lot. Not great art, but great headaches for cinematographers, who know that picket fences will seem to "flutter" if panned too quickly, that water droplets will blur, and that the sign on the side of a moving truck cannot be read. All true in the old systems. With MV48, we could read the writing on the shirt, see every picket in the fence, see the drops of water as if in real life and read the side of the truck. Case closed.

This is a devastating case. Simpler, better, cheaper, more compatible. By now you may be thinking "In film, I can see the advantages, literally. But what do you do once you have delivered a full-rate voice telephone call?" And this question is at the crux of why alternative telephone technologies have had such little impact on sales of telephone equipment. Has nobody asked if you (or, more importantly, the customer) would want a better phone call? The very term "full rate" inspires a lack of inspiration. Why go to all the trouble of digital phones, network interfaces, complicated protocol stacks, powerful servers, abundant bandwidth, and the imposition of QoS on a network that mostly spells c-h-a-o-s when all that is delivered at the end is a full-rate phone call, perhaps not even, and almost always with marginal to poor latency?

Can we do better? Why yes. And this answer is not even an outre suggestion, it is right under our noses, at least if we are familiar with ISDN or ATM. "High quality voice" is an ISDN service. It bears reminding that ISDN means Integrated Services Digital Network, integrating the definition of services into the definition of the network. We only need support a long-defined service in order to provide to users voice quality that is more like broadcast-quality audio than a phone call. ATM is even more flexible, allowing us to trade off network bandwidth for signal processing horsepower in situations where that is appropriate. H.323 and IP telephony on 100Mb networks would easily afford the bandwidth. Yet we are stuck with systems that deliver a product that is "almost as good" as the decades-old systems they want to replace. Unlikely.

But can such attention to a quality experience by the user find powerful allies? In film, it can:

The big film companies such as Kodak and Fuji should like the system, since it will help them sell more film. The directors who love celluloid, like Steven Spielberg and Martin Scorsese, should know about MV48. And there are other applications. Retail outlets use "video walls" to create atmosphere. Rainforest Cafes could put you in the jungle. NikeTown could put you on the court with Michael Jordan. No more million-dollar walls of video screens, but a $10,000 projector and a wall-sized picture.

What about telephony? Both IP telephony and voice-over-ATM systems have the prospect of delivering a truly better product, in audio quality, in call information, in lightning-quick connection, in integration with PC and Palm-based information, in the richness and friendliness of voice interface. Why use bells and tones and buzzes when you can explain things to the user and offer alternatives? Still, makers of new-style systems seem to think their work is done when it is "almost as good as" phone calls delivered over the very first, and now decades-old electronic exchanges.

Are the developers of new telephony systems stupid? No, they are merely tired. By the time they have gotten their products to work, with their investors breathing down their necks, they are happy to have something that will satisfy a specification and a business plan, rather than make a customer say "Wow!" But if you really want success, you won't stop until you have something that is really remarkable, not just respectable.

Moreover, the best way to come to an understanding of how to assemble the right set of capabilities into a system that delights not only the end-customer, but every hand along the value chain to that end-customer is to develop as integrated a view of your pursuit as Roger Ebert and the inventors of the MV48 projector have of cinema. Without the ability to translate artistic need, viewing delight, and cameraman's craft, all the way down into the engineering detail of how film is pulled through a projector, they would have produced something only prosaic, incremental, and only a half-step improvement in one place in the value chain. Similarly, without the ability to integrate knowledge of users, network operators, and all levels of system design and implementation, just having the ambition to produce a great telephony product won't get you to actually have one in the end.

Copyright 2000 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #47

In this newsletter, written in 2004, I made a number of predictions. My scorceard looks like this:

Apple did make a UNIX my mother could use. But now they face the task of taking advantage of, rather than being run over by, the coming desktop Linux wave.

Microsoft, after the huge technology achievement of Windows 2000, did face the challenge of remaining relevant, and largely failed. Microsoft isn't moving fast enough, hasn't had a break-out into new product categories - with the possible exception of XBox, and hasn't created any new product categories to dominate.

Until recently I was right about VoIP being irrelevant because incumbent networks could comfortably drop prices. Cisco is finally getting some traction (by default) in VoIP CPE, and Skype has discovered pricing and a business model that (probably) works.

I was thuddingly wrong about VoATM. Never heard of VoATM? Yeah. Anyway, DSL and VoATM are wrong and deserve to lose. VoIP will set you free to buy service from any telco. VoIP wins.

In 2000, I predicted content protection is futile, and it remains so. Public domain literature still isn't in mass-market circulation. Nobody has, for example, brought Soviet-era prices to iTMS for back catalogs of Soviet-era classical recordings. eBooks are still a total commercial failure. Bah! How many years of clue-stick bashing will it take?

9/11 has forestalled both shrinking government and making it more transparent. The population is the target of tag-and-bag technologies, not the government itself. It will be a long road to turn this one around, and this is Not a Good Thing.

Oh, and other failed prediction: Windows CE, Web terminals, and other non-PC products went nowhere for Microsoft, and continue to go nowhere. Maybe CE will make a comeback in Xbox II. Maybe.

Telirati Newsletter #47: At the Crossroads

A number of technologies, economies, companies, and policies are at a crossroads now. They are worth marking.

Apple is flush and healthy. They have had time to really finish OS X. When Steve Jobs started Next, I could have told him it was a dim idea to use a Unix variant as the basis for his operating system. Some ten years later, he might be done taming Unix, and in time to catch the Internet wave. Apple will thrive or die based on whether OS X really is a Unix my mother could use.

In newsletter #46 we looked at Microsoft?s operating systems design leadership with and found it a mixed bag when it comes to being equipped for the next fight. Speed has become the issue. Microsoft has many of the right answers, but can it put them into products, have the products accepted, and have applications that take advantage of the capabilities quickly enough that those right answers turn into value, instead of just white-paper fodder? Focus, simplicity, and clear-headed assessments of whether leapfrog moves have to be made when incrementalism gets bogged down will be the key. The positive signs for Microsoft come from the fact that Windows 2000 is an excellent product. It really does deliver the ease of use that only Windows95/98 had before, and then some. Go ahead! Windows 2000 is an NT that doesn?t hurt to install anymore. Your mother could set up Internet connection sharing. A wonderful product. But is a wonderful desktop OS enough? Will COM+ win the middleware war? Will Microsoft break out into post-PC devices? They are fighting a war on many fronts, and victory is not at hand.

The technical battle of packet-switched voice vs. circuit-switched voice spirals into irrelevance. Much as Intel showed the world that manufacturing prowess can overcome the purported advantages of RISC, circuit switched communications carriers are showing that bits are bits and that the incremental advantage of packet switched voice communication can be countered by incumbency, scale, and experience in network operations. In the middle term, ATM will emerge as the currency for apportioning DSL bandwidth between voice and data. ATM is a connection-oriented virtual-circuit technology that will prove to be just flexible enough to enable DSL to use its bandwidth very efficiently. DSL is not as pure as IP over a cable modem, especially for supporting household servers visible to the Internet, but clever cooperation between CPE devices and carrier equipment will make even this distinction irrelevant.

In telephony, CPE vs. network is a battle that morphs into an Internet incarnation as Cisco acquires a lot of carrier-oriented technologies like big-iron unified messaging technology. They are forgetting that the network is dumb and the edges are smart. While the network is very good at being dumb, reliable, and transparent, it is as hard as ever for attempts to add value to the network to keep up with the cleverness that springs up at the edges. They will find customers who are happy to buy more efficient ways to deliver capacity, but dubious about providing more than an e-mail address over and above the raw bits.

Intelligence at the edge, plus voice-over-ATM, will produce a new class of mass-market telephony applications for making individuals? telephony experience smarter and better. These applications will live in integrated access devices and will arbitrage the boundary between ATM WAN connections on one side and IP SOHO LANs and wireline/wireless telephone handsets on the other side. The most imaginative applications that break down the boundaries between business and residential needs, and that take best advantage of the WAN architecture will form the basis of a new and vibrant class of intelligence at the edge. Voice mail will morph into a more user-friendly blurring of the real-time/store-and-forward boundary, much as instant messaging is blurring this boundary in the text medium.

Music and books are at a crossroads. Publishers and creators can either face up to the fact that content protection is futile and change their business model to suit, or they will break their pick on the matter. Good guys in governments around the world can give this a heavy shove in the right direction by publishing all material under government ownership, the public domain, and material in their possession that is not protected by intellectual property law in open and unprotected formats. The result would be an efflorescence of cultural wealth, and a crushing blow to both piracy and harebrained schemes to infringe on the doctrine of fair use with fragile software locks. Government, museum, and foundation archives are brimming with recordings, images, texts, movies, and video. A large amount of it belongs to the people, or to institutions chartered to enhance the people?s lives through culture. It is time to use the Internet as a giant amplifier of the cultural legacy of mankind. The alternative is quite dark. The Digital Millennium Copyright Act tramples well-established doctrines of fair use, and creates novel thought crimes, like knowing how to break cheesy ?copy protection? schemes that at times take the form of a bit in a binary file, a setting in the Windows registry, or a line of text in a configuration file. It is not exaggeration that attempts to move past business models into an unaccommodating future will, in the short run, make the government into publicly funded goons for music and movie publishers.

The above comments might make your chronicler out as some kind of anti-capitalist flower child. Ha! The next victim standing at the crossroads not minding the traffic to be run over by the V-8 powered luxury sport utility vehicle of progress is government bloat and waste. Here the people are the ones faced with a decision: Do we let the fat times continue to blind us to the fact that government is at least twice as large as it needs to be? Or will the ethic of cheaper faster better begin to gain traction when people contrast the performance of government against the performance of technology. Will we accept more intrusive surveillance, or turn the camera the other way so we can see in real-time how many road crews are hanging out at Dunkin Donuts? We have ?sunshine? laws and the Freedom of Information Act. Government, in theory, operates under the scrutiny of the people. Why not make it a matter of course that all government meetings are recorded and the recordings posted on the Internet, along with all documents? In every department outside the FBI and DoD this should be the norm, and even in these departments, 80% of matters are mundane and would benefit from the sanitizing benefits of sunshine. The rest can be recorded and electronically numbered and signed in a way that would absolutely prevent destruction of evidence should they be unsealed in case they become the subject of an inquiry. Or, if these arguments do not sway you, think of it this way: Where do you think the productivity gains that will take your online stock portfolio to new heights will come from? Look at that nice fat government over there, like a great fatty slab of bacon. Wouldn?t privatizing another 20% of the GDP look nice on that mutual fund statement? Flower child indeed.

And finally, back around to Microsoft: What a simple idea, that Windows CE should be the foundation of simple, inexpensive machines that surf the Web. Not exactly a crossroads, it is more like ?Well, duh!? An idea whose time came about a year ago but nobody took the call. This product, together with Microsoft?s game console architecture again puts Microsoft at the crossroads of non-PC mass-market products. Microsoft tried, in the distant past, to promulgate a game console architecture. This time around, the x86 instruction set looks like part of the mix, which raises the likelihood of success considerably. But is it enough to withstand the rock of falling PC prices, while resting against the hard place of game console competition?

We will be keeping an eye on these crossroads.

Telirati Newsletter #46

Every now and then I would reach pretty far in my newletters, and this is a case in point: I analyize David Cutler's influence on Microsoft. Ambitious, but perhaps also useful in that the influence of people like Cutler is one reason why Windows seems to move so slowly to respond to the threat of Linux. He shaped Microsft, and his influence, and the momentum of what he set in motion, is still strong.

Telirati Newsletter #46: Cutler on the Couch

One thing that makes Microsoft great is that titanic personalities can still mold large parts of the company in their image. David Cutler, the father of NT, is one such titan, and his worldview is about to start shaping your desktop computing experience. What a strange world it will be. In a recent interview Cutler gave in a Microsoft newsletter distributed to the Windows 2000 beta test community, he made some telling remarks. The interview itself was a puff piece, and the writer was in awe of her subject. But Cutler reveals more than he or the writer might know.

One striking thing about the interview is the reverence over the original NT specification document. Your's truly enjoys regaling his co-workers with stories of how he had to walk five miles through the snows of Illinois to a log cabin where he toggled in the boot load code on the front panel of a PDP-8 until their appreciative snoring reaches its crescendo as much as the next old fart. But it does not exactly exude an aura of Internet speed to blow the dust off the original NT spec. It more reminds one that the ghost of VMS lives on in many of the clunky aspects of NT.

Everyone has to deal with the legacy of their previous efforts. This is nothing new. The more important thing here is the legacy of one's management philosophy and leadership. When Cutler was hired, the move was widely praised as signaling the maturation of Microsoft. Here was someone who could bring order to the chaos of microcomputer software, which had hitherto (and long after) been the province of cowboy engineers who had little patience for the style of software development practiced in corporate software engineering projects.

Microsoft no longer had to scrimp and improvise. It could for the first time afford to do things the right way, and Bill Gates thought there wasn't enough "right way" thinking inside Microsoft. It would be better to score a system architect from an established firm that did things the right way. At the time, the rightest company on the planet was Digital Equipment Corporation. Microsoft harbored dreams of PCs being able to reshape computing to the extent that the kind of rigorous design and development of an operating system that was a hallmark of DEC would become a requirement. So they went right to the source: David Cutler, father of VMS, would lead Microsoft to the future of microcomputer-based systems used alongside, and perhaps even replacing minicomputers.

Cutler is, as already stated, a titanic figure. He had the overwhelming force of personality to move a company like Microsoft, already filled with strong personalities, to a direction of his choosing. He would not shy away from spending whatever it takes to accomplish his goals, and, in general, has self doubt in amounts that, if it exists, must be hunted down as assiduously as a neutrino. In some ways, the perfect man for the job. In some ways, not.

One way to measure the level of self assurance is in the way the discussion of NT's goals are framed. Cutler refers to the original spec (which will be donated to the Smithsonian!) and an early book on NT:

We had five or six major goals: Portability, reliability, extensibility, compatibility, performance. I think that's right. Let me see.

He was then asked if the goals had been achieved. The response is intriguingly self-contained, internally consistent, but not even coming close to examining the ways in which reality has impinged on the relevance of the goals and the way in which they were met:

We tried to create a system that had a good, solid design, as opposed to one that would run optimally on hardware of the time. When we started, we were working on 386/20s. At the time that was a big, honking machine. Since our design had to be portable, we didn't allow people to optimize code in assembly language, which is hardware specific. This was hard for the Microsoft mentality at the time. Everyone wanted to optimize code in assembler. We didn't embed operating-system semantics into the kernel, so when we switched from OS/2 to Windows, we didn't take a major hit. If we had built OS/2 threading or signals into the kernel, we would have been in trouble. Instead we built the OS in layers.

Later he adds: "The basic, internal architecture has not changed, except for Plug and Play?"

Anyone who has ever complained to a Microsoft engineer that a product or programming tool might be more than necessarily complex or difficult to apply is familiar with the form of Cutler's response. Correct, but it leaves one with a sense of unreality. Two facts jar the mind here: Windows NT was forced to adopt an architecture where notoriously unreliable video drivers were allowed into the kernel for the sake of performance, and Linux achieves a high level of reliability with a low-tech monolithic kernel architecture. Multiple user interfaces, multi-protocol-networking, and other NT architectural keystones have become cybernetic teats on a boar. Shockingly, the museum-piece spec dates back only to 1989. NT feels much older.

Napoleonic siege development processes and forced-march debugging cycles do not feel right in Internet time. It is as if a bunch of splendidly dressed German officers on horseback showed up to fight the narco-terrorists in Colombia. What saves Microsoft is the fact that Windows 2000 development is not a single, monolithic army. It is a multi-service-branch task force with a variety of leadership styles and operating procedures. Good thing, for as Cutler says:

What I think is cool, is that the system doesn't crash, and it doesn't lose my work, and it has functionality. I could care less that the visuals are flashy if my 32-gig hard drive goes away.

Never mind the fact that Apple has resurrected itself on the restored ability to deliver the flashy visuals on a kernel that was iffy in 1984, and little better today.

Cutler's remark shows he is more brethren to the Linux people than one might think: architecting a system to make user interface systems changeable and disposable indicates a similar contempt for what we feed on through our eyes all day long. The kitchen may be a gleaming temple of stainless steel, impervious to infection, equipped with the fastest-cooking ovens, but if the food is unmemorable, never mind its objective quality, it will go uneaten in favor of the kebab of dubious street corner origin.

To this point Microsoft was up against companies that were either far more stultified and bureaucratic, or that had inadequate resources for a credible challenge. With Linux as the main competitor, the old measures of relative strength are mostly useless. Any inefficiency could be the cause of Microsoft's decline. They can no longer count on IBM to botch OS/2, or Apple to become a seething pit of office politics, or BeOS to be the consumptive Bohemian in the garret. Linux is immune to these diseases. Not to say Linux is destined to succeed. Only that the old paradigm that sucking less is enough may no longer hold. Can Microsoft make NT development really conform to what people want? Can they do this before Linux oozes into the gaps NT has left open, and metastasizes?

Telirati Newsletter #45

Overall, I find it eerie to reread what I woute five years ago. For an industry that supposedly moves at lightning speed, the predictions presented below are remarkable mainly for the fact that their outcomes are still unfolding.

Still, it is possible to tell that most of these predictions have come true. Moreover, it is also possible to say quite clearly that Microosft has failed to use the previous five years to take actions that would have been effective in minimizing the competitive threat from Linux and other open source software.

One can also see that the importance assigned to these threats has held up: Price hasn't dislodged Windows from the desktop, where Windows pricing is low enough to fail to deter any customers. It is the other attactions of open source software that matter more.

Telirati Newsletter #45: Categorizing the Linux Threat (Opportunity)

How much of a threat is Linux to Microsoft? What are the main opportunities for Linux? In answering these questions, it is best to keep in mind that there are layers to the answer, and some layers matter more than others even though the theories behind each layer are supportable. Here, we attempt to order the layers according to importance:

Big threat: Linux is good enough. Linux is good enough for a lot of purposes. Serving simple Web content over moderate-speed links is simple, and Linux is more than good enough for this purpose. Inexpensive desktop PCs, or used PCs, are good enough for this purpose. If you need to put content on the Web, Linux is good enough. As broadband access explodes, a lot more people will find low-end PCs in combination with Linux ideal for this purpose. The vast majority of computers and the OSs they run are not applied to anything remotely approaching their highest and best use nor to the limits of their capacity. It should worry Microsoft that good enough might be good enough for most people.

Big threat: It's better than stealing. A lot of people got into computer software for the same reason people become drug pushers: they don't want to pay for their own habit. But not everyone can cadge free goodies. Linux, in addition to being good enough is free or near to it. As makers of commercial software squeeze out freeloaders and thieves, these people will have to make a decision on how to go legit. In this, Linux has the distinct advantage of not having a commercial entity associated with it that can be embarrassed by this market.

Big threat: It's better than cadavers. Students can cut Linux open and start hacking away without having to be part of the select few organizations with source access to Windows OSs. If students come to be familiar with the anatomy of Linux more than Windows, Windows will lose a key slice of mind share among the most fertile minds.

Moderate threat: IBM likes it. And so do all the other also-ran software makers. The problem is that there are a lot of these guys out there and they might get it together enough to add significant momentum to Linux as a platform that supports a full suite of software. And don't get all worked up about how IBM has higher software revenues than Microsoft if that is still true. Computer Associates is a very large software company, number two or three by most measures, and I defy anyone to tell me how that might be relevant to this discussion.

Moderate threat: Responsiveness. One of the most visible fruits of community development is multi-platform portability for Linux. It runs on the x86 platform, of course. And Alpha, so that it covers all the ground NT covers. Now add PowerPC, MIPS, and a number of miscellaneous platforms. The reason this isn't a large advantage is that the x86 architecture is so dominant. But it does draw in all the makers of platforms Microsoft doesn't have time for. To counter this threat, Microsoft has to reduce the cost of maintaining NT and it's successors on a variety of platforms, which is right now in the tens of millions of dollars per platform per version.

Moderate threat: The vast hordes of Linux coders. Only a small minority of these coders matter. If Microsoft can reform its culture, and there are some indications it will try, Microsoft can more than meet the challenge of open source development.

Moderate threat: Open source. Important, but vitally important to only a small minority of users. If you are running a defense organization outside the U.S., you would want an OS you can inspect for the presence of exploitable features. This is no joke. It has been reported that the Chinese government will not use Pentium IIIs because of the PSN and related features. Microsoft can blow it here by incorporating content protection features into Windows that will create the same kinds of security concerns that Intel inflicted on itself in the Pentium III. If this does happen, more people will justifiably ask: is my computer capable of betraying me?

Moderate threat: Stability. Only a small minority of people leave their computers on all the time. Microsoft has an opportunity, with Windows 2000, to meet the threat of Linux stability, which, in no small part comes from the fact that most applications of Linux are exceedingly simple. Still, it is not outrageous to suggest that a Unix-like OS would be more stable than Windows 98, and Microsoft will have to show it can make systems that inspire confidence before it becomes commonplace to leave one's computer in charge of one's house.

Not much of a threat: User interface. The Linux community is only now discussing user interface in something other than dismissive tones. Even now, various attempts at a user interface are regarded as interchangeable, and the discussions in most Linux fora still treat the topic as if it were a check-off item, like USB support. Wrong. The OS exists mostly to support the user interface, and the theory and architecture of the user interface are the most difficult aspects of OS design. Windows 98 is manifest proof of this. Apple understands this (and may produce the first Unix-derived OS that embodies such an understanding), and so does Be. Unless Microsoft goes on the long march to Bob II, Linux won't catch up. More likely, Microsoft will come up with at least a few interesting UI innovations while Linux is still learning the basics.

Not much of a threat: Price. Other than the aforementioned freeloaders and brigands, the price of a Windows OS has never been a barrier for any important customer that can move a significant number of machines. PC makers have common cause with Microsoft in maintaining price levels and margins, and, when they shift, to adapt to new levels by segmenting the market and pricing appropriately. Consumer preference is far more important than the difference between the cost of a Windows OS license and zero.

Copyright 1999 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #44

Five years ago I wrote this newsletter, and I came not to praise speech recognition, but to bury it. In the subsequent five years, speech recognition lives on as a novelty in mobile phone handsets and cards. One of my colleagues recently demonstrated his new Honda's abilities. "Take me to the nearest whorehouse!" he exclaimed. And, dutifully, the navigation system displayed nearby hospitals, in anticipation, no doubt of him catching some STD.

My prediction here is that speech technology providers would continue to bark up the wrong tree, and they have maintained that course with stubborn steadfastness. Interdisciplinary approaches are still left uninvestigated, and products are only slightly less of a laughingstock than they used to be.

Telirati Newsletter #44: The Shock of Recognition

Speech recognition is a curious beast: Sometimes it appears to have been tamed. It jumps through hoops on the trade show stage and once again we are drawn to believe. Perennially, Bill Gates makes the assertion that, really truly, we will have a "natural" interface to our computing environment. But sometimes the full horror is revealed. It twists our words into parody or simply refuses to behave. It cruelly mocks our ambition to move beyond the keyboard. For more than a dozen years, just on the PC platform, it has been "real soon now." This darker side comes into view only occasionally. One such occasion was a recent article by BT's chief technologist. Writing in his regular column the London Telegraph, he described his encounter with dictation software he bought because of a wrist injury. This is a singular event because this emperor of R&D has some 30 boffins working on nothing but speech recognition technology and applications. He had every reason to assert that he emerged from the dictation software fitting room fully clothed. Instead his candid account of the state of speech recognition serves notice to the believers that they are nakedly overoptimistic:

Like a dummy, I believed what was on the box and purchased a speech-to-text program. After five hours of continuous training (of the computer, not me), this expensive package was capable of creating complete gibberish. A quick search on the Net found many alternatives, one of which was a tenth of the price. Having purchased this cheaper package, I found it equally qualified in the gibberish department. Why is it taking speech-to-text such a long time to become a practical reality? And how come some people seem to be able to get them to work... or do they?

Well, what to make of this? Dragon Systems, IBM, and Lernout & Hauspie have been making and selling dictation and voice control software for almost the entire existence of personal computers, and this category of software has been available on mainstream retail store shelves since the days of Windows 3.1. People buy this stuff. They use it. They haven't all returned it (though, if you consider the amount of shelfware on your own shelf you might find the last part not too surprising). Is Mr. Cochrane too picky? Does he not get it? As it turns out, the problems with speech recognition software now are the same ones that existed a decade ago. Not only are the problems discouraging, the lack of progress in solving them is a strong indictment of a lack of creativity among the purveyors of speech technologies:

With military systems, banking and telephone operations, it is now possible to embark on an adventure of human-machine voice interaction. Even cars and television sets can be reliably controlled and commanded by speech. So why can't my laptop understand what I say? Well, there seem to be a number of key problems with all generalised speech-to-text technologies. First, the acoustic environment, noise and echoes in the room play a critical part in disrupting performance. It also seems to be vital to get the microphone precisely positioned and the computer set up just so. It is also imperative to be clear in your diction and dictation, leaving adequate spaces between words, which means adopting the monotonic regularity of a robot. It also pays not to have a cold or to be thirsty. Even worse, the problems associated with our wide vocabulary and use of several words for one meaning or different contexts seem to kill developers' efforts to create a generalised environment.

Cochrane knows it can be done, in some environments, and for some applications. It makes the lack of progress in addressing the environment of general purpose dictation that much more frustrating. BT, as a user of recognition technologies, has figured out where to apply them, what conditions are suitable for speech recognition, and what special measures have to be taken to make recognition work as well as possible. With dictation on personal computers, almost none of this kind of applications sense has been applied to the problem. The underlying recognition technology gets better each year and with each faster CPU, but that only affects the margins. Cochrane's overall experience is little different than when dictation software was a novelty.

Cochrane goes on to suggest a few things that could help: more computing power, and more attention to the context in which words are used. But this does not go nearly far enough, and, with a more open consideration of possible remedies it should be possible to make dictation work significantly better without a radical increase in computing power: Lip reading, using an inexpensive camera. Detection of the position of the user's head. Where is the user looking? With USB, a speech recognition product could be packaged with a microphone that gives uniform and predictable results, unlike the present situation with a motley assortment of cheap-as-possible mics and sound cards. What about a mic specialized to speech recognition, worn directly against the throat, to exclude environmental noise? And these brief musings are only the surface of what can be done to address an obvious and longstanding problem.

Creativity, cross-functional integration, and sophistication in integrating speech technology with the computing environment are the elements of success. But, instead of addressing these, speech recognition technology providers still put almost all their resources into the core recognizer technology. The folly of this should be easy to see: the level of sophistication of recognizers varies widely, but their performance in demos is uniformly wonderful. So instead of mocking the crudeness of a competitor's recognizer, some speech recognition company out there should take this as a hint that other avenues of competition are open, ready to be exploited, and may yield differences that customers will find very easy to see and appreciate.

Copyright 1999 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #43:

In this newsletter I take a shot at the server craze of the time. Over the years, server software has remained a very active area of the software industry, but Linux and other free and open source software has cooled the gold rush mentality.

Recently, tools like distcc have brought distributed omcpilation to Linux, confirming my prediction that software development would be one of the first targets for workgroup clustering.

But the real importance of this is in striking a blow against the TV-izing of the Internet. The less distinction between a user's node and a server, the better. An interesting example of the virtualization of server software in use today is Skype's distributed directory. A contining trend in this direction is all part of the return of the Internet to its network-of-servers roots, and it is a Very Good Thing.

Telirati Newsletter #43: Servers, Re-centralization, and the Promise of the Personal Computer

A wave of servers is sweeping over the face of computing. If you want an Internet presence, you need a server. For many prospective customers, this fact finally answers the question: "Why do I need a server?" For some people, mainly those who were never wholly comfortable with personal computers, servers feel like home, and the term "re-centralization" has come into currency, meaning the return to importance of a central computing resource.

I don't like it! No, I am not suggesting that the server market is over-hyped. It is as real as can be, and will make everyone serving this market a lot of money. However, it is easy to take enthusiasm for servers, server technology, and server-based software too far. One reason why it is so easy is that one can charge thousands of dollars for server software, and the vision of every-business-an-e-business with a server makes people who sell server software positively tumid with anticipation. On the other hand, the personal computer hardware business is now an exercise in extreme brutality of competition. Software, outside of Microsoft's dominant spheres of operating systems and desktop productivity applications, and a few high-end specialties, is also suffering from reduced price expectations, and Web-based replacement technology. Servers and server software offer a comfortable refuge. So it is no surprise that servers have sucked up all the available mind share.

The problem is that, while the server business is booming, servers have a long way to go to catch up with the computing power being deployed on desktops. Not the computing power of individual desktops, of course, but the aggregate computing power of the millions of desktop PCs built every single month of every year, with unit volume that continues explosive growth. This even as the revenue and, worse still, the profit picture for makers of PCs looks bleaker by the day. Nevertheless, desktop computing power dominates the computing landscape as never before. You can cluster servers, of course, and if you get several dozen of them together this way the results are very impressive, but it is still but a tiny fraction of what could be accomplished by turning large groups of personal computers into a kind of "hive mind."

Before this develops into a speculative discussion, here is an application of workstation-level clustering, as you might call such a technology: Software development. Compile-time is a critical element of coding productivity. Typically, coders are given fast machines for software development, so that they spend less time waiting for a compiler to complete its work. The process of "doing a build" - compiling all the code in a system and linking it, is also a key bottleneck in software development. On large projects, doing builds is a monumental effort. Still, I have not heard of one instance where the aggregate power of the idle or mostly-idle workstations in a software development organization are brought to bear on the task of compiling an individual programmer's modifications to a module or to the task of building a complete software system.

Compilation is an ideal task for exploring the possibilities of this type of distributed computing: It can be implemented at the level of the software development applications. It can be implemented with existing distributed component technologies. It can be administered using the same group administration user interface and database as used for source code version control. Compilation is almost entirely risk-free. It does not impinge on the state of the PC except to use a modest amount of temporary storage. The results would be valuable. Even a two-fold improvement in compile time would be a boon to everyone developing applications software. If it proves useful, such facilities can be migrated into the OS and made more user-friendly so that they can be applied to all applications.

There are some research implementations of this idea extant. Legion is one such system. The ideas in Legion are not too distant from what can be extrapolated from Microsoft?s directions in the development of COM, intentional programming, and next-generation component software systems. Jini incorporates some of these ideas, too. Market drivers for this technology could come from such end-user categories as game software.

But I want it now! This opportunity, and others, such as a successful implementation of personal communication in the form of IP telephony and conferencing, true integration with consumer electronics, the lack of any credible commercial attempts at an interface beyond the desktop metaphor, the lack of commercial attempts to replace the notion of data stored in files, the lack of breakthroughs in speech processing, etc. all indicated that anyone whining about a lack of opportunity to compete with Microsoft in operating systems is just not looking to see where the holes are.

Diversity in personal computer operating systems appears to have died with OS/2's larger ambitions. This need not be so. If there is a competitor out there that can put eight or nine zeroes in front of the decimal point of an operating system development budget, there are ample domains of product definition where superiority can be established. Compared with having to compete with Intel?s ability to spend a few billion on each new generation of a fab, going up against Microsoft is less daunting. Would it be a thankless task? Only if you think half a trillion dollars in market value is not worth the effort.

What of current challengers? Linux is a worthy opponent. Linux could acquire capabilities like ad-hoc workstation clustering sooner than Windows. But bringing a real challenge to Windows' superiority means surpassing Windows in several areas of product formulation. Linux, with all its advantages, doesn't have product formulation. It is a remarkable phenomenon, but it cannot go in an inspired, risky, and uncharted direction, which is exactly what is needed in order to compete with Windows. BeOS is still out there, waiting to get enough zeroes behind its budget, and BeOS does have developers of a particular vision. Lucent's Inferno and Plan 9 have, at least potentially, more zeroes than Croesus could bring to bear on a problem, plus the singularly fine mind of Dennis M. Ritchie. But Ritchie has recently given interviews that indicate he has given up on challenging Microsoft.

The likely outcome will be that this concept will be pioneered in Linux, achieving some popularity in, for example, cracking a crypto key or analyzing vast amount of planetary imaging data. The trouble is, to be well implemented, the concept has to incorporate the use of a distributed component technology, and CORBA is just not that intimately a part of Unix-like OSs, the way COM is part of Windows. So when Microsoft addresses this competitive threat, it will do so in a way that is more comprehensive and more interwoven with the way Windows software works. Unless someone else decides it's worth taking a run at Microsoft.

Copyright 1999 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #42

I predcited in 1999 that Microsoft would start to sort out the winners and losers from their Internet ventures. As with many other things that look like an action plan for a year, maybe two, at most, this process continues to drag on, and Microsoft is, still, part owner of an also-ran cable TV channel portal hybrid, among other oddments.

Telirati Newsletter #42: Species of Internet Ventures

There is no one kind of Internet startup, and it is a multidimensional space in which they exist. But at least one axis of this space can be mapped out, and it is to our benefit to do so since we can then place Internet ventures along this axis to see what type they are. And, more usefully, if they appear not to be in the right place, it can tell us which way they ought to be going.

At one end of this axis are companies that have business models enabled by the Internet. These include Ebay, PriceLine, and every company taking a new whack at direct sales based on the prospect of the Internet changing the rules of direct selling. Some of these companies will find that novel business models, like PriceLine's, will, in fact, change the world and build for themselves a defensible position over the long term. For these companies, the Internet truly is an enabling medium, before which their ideas were not possible or practical. Some, like many of the direct marketers, will find that the Internet did not change the rules so completely that products that had been far outside the scope of direct marketing before the Internet are now viable businesses on the Internet.

It is the companies with the novel business models that have attracted most of the attention. This is in part deserved, because some of these companies will be new giants. An idea, a new medium, and relatively little technology, are all it takes to change the world in a big way. Investing early in a winner of this type is what every investor dreams of. But happy endings to these dreams will be inevitably rare, and difficult to discern at a distance. More often the result will be that the Internet did not work magic, or that the idea itself was not a winner.

Below is a diagram of the spectrum under discussion:

--technologies enabled by the Internet-------------business plans enabled by the Internet--

At the other end of this spectrum: Businesses that hinge on technologies enabled by the Internet. Like most geek pursuits, these kinds of Internet ventures get short shrift in the coverage of the Internet economy. Indeed, the coverage this kind of venture gets has also been marred by the recent fashion of Microsoft bashing: trade publication coverage on the trend toward plug-ins that support enhanced interactivity have been slanted toward pointing out that Windows is a beneficiary of finally waking up to the fact that the Internet is a network of computers, not of browsers. But it is only a fact of life and of market share that when servers and clients on the Internet start to talk in any language more expressive than HTML it is likely to be a Windows-specific from of communication.

There are good reasons to pay attention to pay attention to those geekier companies at the more technology intensive end of the spectrum, and for companies at the other end of the spectrum to begin to move toward the middle.

Among the reasons to look at technologies enabled by the Internet is that Microsoft is waking up to these possibilities. Microsoft spent several years now trying to be a PC software maker and a constellation of Internet ventures. Some of these Internet ventures have been quite successful. Others, however, are also-rans. Is there a way for Microsoft to succeed more reliably in Internet opportunities. The answer is yes, if they pick opportunities that are linked to the needs of Windows users.

It seems simple, but the reason Microsoft is involved in, among the odder pursuits, a hybrid of a cable TV news channel and Internet portal, is that when the Internet hit big, it made people panic and worry that computers, operating systems, application software, and other mundanes of that nature would no longer excite. Now, with the confidence that Microsoft can make and sell world-beating Internet software and incorporate it into the development plan for Windows in a meaningful way, Microsoft has a collectively clearer vision of what the Internet is for: It is for making Microsoft customers happy.

Sit down you Linux users! Put away the pitchfork and torch. What's the matter? Birkenstocks too tight? Beard itchy today? The Internet is a computer network, not a hierarchy of browsers and HTTP servers. And this fact applies to all pursuits on the Internet, Windows-related or otherwise. Examples of Internet ventures based on a technology enabled by the Internet include streaming media technology providers, Internet-based off-site backup services, and some really innovative ventures like FusionOne, a service that enables synchronization of files, databases, and other aspects of one's computing environment.

It isn't just Microsoft that will benefit from the realization that the Internet does not begin and end with a small set of standard protocols and applications. Tight integration with the user's environment, the ability to reach into contact management databases, messaging interfaces, and user interface mechanisms in general will expand the horizons of the Internet. Which is all well and good, but how does this expansion of horizons affect which ventures will succeed or fail?

Here is why this matters: New business structures are exceedingly rare. Existing business structures translated onto the Internet may or may not benefit from the characteristics of the Internet. But if you have a useful software capability enabled by, or that has capabilities that are amplified by the Internet, you have something of value. These types of Internet ventures carry a far lower risk than attempts to create a novel entity in the economy as a whole. Moreover, there is a shortage of creative thinking in the area of how to amplify the power of individual computers by connecting them. The answer does not lie in the direction of technologies that reduce the user experience to the lowest common denominator. Rather, Windows PCs will find parts of the Internet serving the unique needs and capabilities of such systems, while other parts will be inaccessible to anyone not playing a game on a Playstation.

There are, of course, ways of going about the technology-intensive end of the Web business spectrum the wrong way. The Palm VII is likely to become an example of this. Instead of projecting the capabilities of the Palm OS out over the Internet, one is expected to find value in a constriction and limitation of Internet-based information for the sole purpose of making it possible to implement a wireless connectivity scheme with limited and immature technology with the wrong pricing model.

In general, however, one is more likely to succeed by treating the whole computer as the medium, not just the browser. This holds true even for business models enabled by the Internet. As competition moves in where it can (and often these business model-centric ventures have little or no barrier to entry) even the most innovative basic thinkers in business models will find they must enhance the sites that express their lovely ideas with more and more technology. Auction sites will become the sources of rich media content to make up for the lack of touch, taste, and smell. Internet direct marketers will apply community communications technology to make their constituency feel as if they have a common cause. Meanwhile, the entire idea of software will morph into something where the extent to which the Internet is brought to bear on users' needs will be the single most important distinguishing characteristic.

Copyright 1999 Zigurd Mednieks.

Telirati Newsletter #40

Here in the archives of the Telirati we have a classic tragedy of bad management. It is as current as ever in informing the process of management selection in small ventures.

Telirati Newsletter #40: Care and Feeding

In the ecosphere of the enterprise there are two species that require special care and feeding: engineers and salespeople. Salespeople are, to me, a delightful mystery. They live eat and breathe interaction with customers. Yet, the last time I had the pleasure of sharing a car ride with three of them, they were volubly discussing the role of Altoids in White House affairs while I was on the phone with an important investor and potential customer. Engineers, on the other hand, are more my cuppa.

Companies that keep their engineers happy reap worthwhile rewards. They skate over the perils of product transitions, they build high walls around their markets, erecting a thicket of features that are hard to duplicate. They acquire a halo of high touch that can sustain them when they stumble.

Much has been written about keeping engineers happy, and most of it is useless. "Don't step in the leadership" just about sums up that body of work. In keeping with the general approach of these newsletters we will pursue more modest goals here: First, do no harm.

This is evidently harder than it sounds. I recently had a chance to see, at close range, how hard it is. First, be assured that the example I am giving here is taken from real life. It is not a composite or a cautionary fiction. Second, if you think at times I have gone over the top in describing managerial incompetence, I assure you the same situation can be mined to much greater depths and would yield up even worse. For example, the manager I will describe in this document includes in his technique being overtly nasty to a woman seven and a half months pregnant, who put in 80 hour weeks, while pregnant, to get a product out the door. So it isn?t that the villains here deserve protection from universal opprobrium. They go unnamed just because it isn't strictly necessary.

This story begins before your humble chronicler arrived on the scene, in a startup where a charismatic founder and leader has assembled a team of engineers that would be the envy of anyone. The company, having spent many months banging around between unsatisfactory product formulations, sought the stability and direction that some management structure might confer on the product development process. This founder, so brilliant at attracting and retaining top-flight engineers, has only a spotty record of hiring good managers. As a further examination of this example will show, this was the downfall of the company as well as of the founder. Nevertheless, the founder hired an experienced and credentialed manager, ex-of IBM's Lotus division (some of you can at this point see where this is headed, but stay put, it will be worse than your expectations). I should add he was ex-of the non-Notes part of Lotus that had managed to fritter away a very strong franchise in desktop productivity software. But the resume looked good, and he seemed nice enough.

It didn't turn out that way. He was out of touch. He did not get involved in implementation decisions, not out of confidence in his crew, but because of a something like a character defect. He required detailed reports, but did not ever get deeper than this needless paper to find out what was going on. He was what Machiavelli would have called an indolent prince. He arrived early and left rather early. Decisions were made around him, among the engineers and the company founder, late at night when real work was happening. Eventually this manager was, in effect, fired by the people working for him.

The story should have ended here, but it didn't. Lesson #1: Anyone can make a hiring mistake. But to really screw up, you have to botch the firing. The more-serious error here was in not getting rid of this guy the first time (oh yeah, there is a second time).

Engineering resources are too valuable to waste. This is true, but the wrong conclusion was drawn here. It is the people writing the code that are the resource that matters, especially if they are proven to be highly productive. But here, the failed manager was installed in another role: running the company's IT and e-commerce infrastructure.

OK, so you can guess the result. But it bears examination just for the lessons to be learned. An insecure and incompetent manager, rejected by the team he was hired to lead, will make some predictable hiring decisions: Web site development at this company was staffed by third-tier plodders that posed no threat to their second-rate manager (this, after an outsourcing misadventure). The result, for a company with ambitions of becoming the Dell of small business phone systems, was disastrous. A hideously expensive effort yielded up an e-commerce Web site so undistinguished that you can buy aftermarket Beanie Babies from slicker sites that were homemade by housewives. The champagne-grade enterprise software systems do not mesh well. Telesales is sluggish because customers call, intrigued by the product and price, but puzzled over what it does exactly, the Web site being almost opaque on the topic.

Can it get worse? As promised, it does. Through a series of connivances and errors, the board of this company dismissed its founder. As you may have guessed, the ex-VP of engineering, a failed, embittered, and vindictive bureaucrat, was part of the plot to oust the founder from his company. As a reward for supporting this move, he got his old job back, heading R&D as well as the firm's gold-plated yet leaden IT infrastructure. Before going on to describe just how well this is working out, let us recall that the lesson here is "First, do no harm." You are reading the story of how a mistake by a well-intentioned and decent founder, for whom the company's engineering team would walk through fire, got amplified into a full on catastrophe. It happened under the direction of managers who, though they might not be the most benevolent people on Earth, do not have the excuse of lack of experience. It happened under the chairmanship of a long-time pillar of the Boston venture capital community. It happened in a small company where the issues are all easy to see and take little time to grasp. It happened where the stakes are petty. It happened. And if you don't pay close attention to your engineers, it can happen to you.

So there it stands: A manager who failed so thoroughly once, and twice, is back to wreak revenge on the competent and hard-working people who gave him the boot. The environment cannot be described in terms of worker morale, which has evaporated completely, leaving only distilled venom. The engineers are openly contemptuous of their boss, who, in return, is boorish as he can be. He dare not fire them because he hasn't got a clue about how the product works. Headhunters swoop and hover, some of whom placed engineers at this company and were stiffed on their fees. The table has been set for the engineers who will quit right away: Internet startups with great pay and entrepreneurial upside have made their offers. It is tragic for this group of engineers to have to abandon their baby. But their current management has reached a level of awfulness that is comic. Flipping the boss the bird and not getting fired for it is entertaining enough to make some of the group stay on just to see how things turn out. The possibility of a recovery, for the engineering team and for the company in general, still exists, but it would be a reclamation, not a save. You see the picture as if on the moment of impact, like an Edgerton strobe photo of a hammer striking a porcelain vessel.

Copyright 1999 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #38

This isn't one for the ages, but it is an example of giving kudos where due. Microsoft often set the standard for handling problems. It will be interesting to see if content protection and other priorities can be accomodated without negatively affecting that customer focus.

Telirati Newsletter #38: How it’s done

How it’s done: When Microsoft got caught in a privacy problem, using unique ID numbers generated on the fly (these are the backbone of every software service that has to identify things across networks) to identify Windows customers across sessions with Microsoft Web sites, and for other purposes, Microsoft immediately issued a statement to the effect: “We were wrong, and here’s a fix to get rid of it.” Immediately, as in hours after the first press reports. On the weekend.

How it’s not done: Intel could have just stopped programming unique IDs into some Pentuim III CPU chips and allowed customers to make up their minds whether to buy chips with or without the numbers. Instead, Intel shot another hole in their foot claiming that ID numbers found in some Pentium II chips were a production error. That this is a breathtakingly unbelievable excuse is the kindest thing I can think to say about it.

How it’s done: Microsoft has released an internal memo, widely leaked to the press, explaining in clear terms that having one’s witnesses savaged by an extremely talented litigator is not the same thing as being on the wrong side of the law. While a strategy of being obstinate bastards is probably the right one for Microsoft, it does open witnesses to ridicule. Information explaining why such cross examination virtuoso turns don’t amount to much is the right antidote.

How it’s not done: While Microsoft, in it’s memo, points out that cases are decided on law, not PR, Microsoft have not done well against the government which wants the case decided on PR, and not law, or at least not a conventional reading of the law. By fumbling the ball in the creation of illustrative videos on the one hand, and by not consistently painting the government’s case as novel, radical, and anti-capitalist on the other hand, Microsoft has allowed the press to lazily focus on it’s executives getting reamed over relatively minor points, peripheral to the core issues.

How it’s done: A clock, built in 1746, housed in the museum at Versailles, that indicates the day, month, year, phase of the moon, and position of the planets, will tick over into the year 2000 without fault. Perhaps the maker knew that the new millennium begins in 2001, a number that should cause no difficulty to a competent clockmaker. Perhaps people thought differently before they were so artificially influenced by digital computers to think that first things are numbered zero. Is your firstborn child number 0 in your family? When you count a dozen eggs, do you start at zero?

How it’s not done: Intel, in containing the damage from the public reaction to processor IDs, has described the public as confused over the difference between privacy and anonymity. This and other attempts to “educate” the public in this matter are doomed to fail. There is nothing to educate about: either you want a processor serial number, or you don’t. All this, however, does not get at the underlying question: Do you want stuff in your computer that you don’t know about, and that cannot be subjected to independent review to determine if it causes security or privacy problems? It’s a simple question.

How it’s done: Sun, SGI, and now Apple are trying, with varying degrees of success at this point, to incorporate the “open software” model into their licensing regimes. This is more than just a fad. It is obviously useful to ISVs who have to struggle with incomplete documentation, and it is critical to creating and maintaining confidence in precisely the place that Intel has squandered confidence. Customers want to know what they are getting, at least to the extent that they have confidence that others who may be more interested in security will review open source systems and publicly report problems. It is a great shame that the continuing DoJ litigation against Microsft has effectively blocked Microsoft from making dramatic changes to its licensing models, since drama must be reserved for the settlement.

How it’s not done: They’re from the government, and they’re here to help. The new Cyber Citizen initiative of the US Department of Justice includes “personnel exchanges” between government and private industry to help thwart cyber-crime. Ick. How do you spell “warrant-less search?” If you want a heads up on the next pool of toxic public relations poison, here it is. Stay far, far away.

How it’s done: Microsoft has partially admitted to itself that there are more market segments than there are Windows operating systems to fill them, and has backed off a probably doomed effort at pounding a big, sharp-edged square peg called NT into the round consumer hole.

How it’s not done: Public statements by Jim Allchin and Bill Gates indicate that they are still true believers in NT for everything. Make no mistake, NT is a great product, and the next version will be an enormous step forward. Further, the NT development organization has successfully organized development efforts in several significant areas, among which networking and browser development, and forged them into a compartmentalized, yet coherent whole. No small accomplishment. But this accomplishment should not obscure the fact that it appears that Microsoft fails to grasp its own success with Windows 95. It was a magical thing to create a product so thoroughly accepted by individual consumers making individual choices. NT does not capture that magic. And, now that Microsoft wants new versions of Windows 98, there is, in effect, no Windows 98 team in place. The way way OS development is organized will serve Microsoft well in that large modules, like the browser, Universal Plug & Play, TAPI, and Multimedia enhancements can be rolled together into a major update, but the opportunity for a major kernel upgrade will probably have to be let go. A shame, since focusing on the x86 platform, and on a personal OS, to the explicit exclusion of NT’s portability and security, could yield significant performance advantages. The question: What will Microsoft learn from this?

Copyright 1999 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #37

For four years now our privacy has been hanging in the balance, and there it remains. The next major developments of PC technology could turn personal computers into tools of mass surveillance. Microsoft could, still, do for Linux what Linux has failed to do so far: provide a compelling reason to use Linux on the desktop.

Telirati Newsletter #37: How not to win friends and influence people

Intel recently attracted a lot of bad publicity for the Pentium III processor. They managed to take a fairly innocuous feature, a processor serial number, and turn it into a scary public relations monster that will be very hard to kill. A boycott was called by several privacy advocate groups, which Intel’s waffling response has so far failed to lift. Intel’s mistakes stem from a recent trend in the industry to forget what the letters “PC” stand for. They stand for “personal computer,” of course. And that implies a number of things that, if not properly attended to, can lead to ruptures in public trust deep enough to jeopardize a serious percentage of market share for the Pentium III.

A personal computer is personal, more personal than a car – more like one’s toothbrush, or underwear. Let’s see how our personal effects can be improved through technology: One could create a product that would detect and report body odor, tag the data with a unique ID, and ship it out over the Internet, whereon your Internet service provider could do you a substantial favor by sending you an e-mail suggesting a fresh change of knickers, with suggestions based on the weather report, perhaps. With on-chip sensors and a bit of radio technology, such a product could be built cheaply and pay for itself by optimizing the use of deodorant and laundry detergent. And we have all known cases in the high tech industry where better underwear management is needed. The low-cost technology both Intel and Microsoft have developed for interactive toys could be applied here.

Expressed in these terms it should be easy to see why technology that invites the wider world to pry into our personal communications and works, and in our choices of entertainment and diversion are soundly and broadly rejected. Yet, for some reason, people who should know better cave in to government spooks and entertainment industry rent-seekers and end up kissing the hot stove of visceral public rejection.

As computing technology creeps ever further into consumer electronics, similar faux pas follow, and with just as much success. Remember the V-chip? The TV makers are trucking their V-chip equipped sets to the landfills after they blighted store shelves for longer than retailers could stand. Space in the dump has also been reserved for DIVX, a video disk technology that reports your viewing habits to a license authority, via modem, while you are sleeping. Too few customers see the charm in this.

Intel’s blunder does not end with serial numbers, unfortunately. The same entertainment industry wizards that spawned the V-chip and DIVX have put some truly awful stuff into Intel’s hardware plans. According to Intel’s Dan Russell, as reported in IT Week, "Other security primitives that will be built into future PCs will support random number generation, digital signatures, secure storage, hidden execution, smartcard access and biometrics.” Very nice: your computer can execute code you cannot detect, it can store stuff you cannot get at, and if you get busted for making an illegal copy of Zardoz, it will pick you out of a police lineup. This is supposedly for the purpose of “content protection” but there is nothing to prevent really nasty invasions of privacy, much worse than could be caused by the processor serial number, through the same facilities. Personally, I’d rather have the underwear monitor.

Microsoft has so far avoided this mess. But perhaps not for long. Microsoft is attempting to create operating system facilities for content protection. One wonders how this is to be accomplished. Will it become illegal or against license provisions to write multimedia device drivers that spoof the system into thinking it is playing to a device, when actually the software is recording an unprotected copy of a work? Preventing this from happening implies mechanisms that would assault the freedom of peripherals makers, ISVs, and customers so deeply as to invite a stampede for the exits. Microsoft has a knee-jerk reaction to Intel’s serial adventures in promulgating standards. “Leave it to us. We do the software.” This is the canned response Microsoft puts up in every case. For content protection, however, Microsoft would do well to leave this public relations Afghanistan to the hardware guys.

Many of the most tantalizing possibilities for computer and communications technology are in areas that can make customers uncomfortable if handled incorrectly. Speaker recognition, transcription of casual speech, intelligent video and machine vision, even voice stress analysis, all hold opportunities to create systems that can amplify our productivity. Systems that know what we said, who we said it to, on which topics, and that can automatically link this information with documents, schedules, and Web searches can actually make us seem smarter than we are. Unless customers are certain these technologies will not be used to amplify the power of surveillance as well, they will go unused. If you develop these technologies, make certain you use strong cryptography in conjunction with them. The result could be systems that enhance human powers in unimaginable ways, and, simultaneously, protect us from unwelcome prying, and make us more secure in our documents and effects. This is a future people will readily pay you to provide.

Copyright 1999 Zigurd Mednieks.

Telirati Newsletter #36

Here is a little-known "success:" Apple's line of servers. The thing I missed was that unless Apple servers are as friendly as Apple's desktops, the purpose of having an Apple server is... what? It would be nice if Apple made servers my mother could configure and operate, but that is not likely to happen without years of effort.

Telirati Newsletter #36: Can everyone be right? Can everyone be wrong?

Can everyone be right? Yes.

Can Steve Jobs be doing a wonderful job of reviving Apple, and can Microsoft safely view Apple as irrelevant? Yes to both. Even if, as a worst case (from Microsoft’s point of view), Jobs has successfully managed Microsoft’s expectations w.r.t. OS X and Apple successfully rolls out a competent server product without drawing Microsoft’s ire in the form of letting Office for the Mac platform wither, Apple cannot divert the freight train momentum of NT servers reshaping corporate and small-business computing. Apple can, however, carve out a slice of the server business at least as large as its desktop market share, which would be very nice indeed.

Is Microsoft moving too slowly? Yes. Why doesn’t Windows have built-in virus protection? I wanted Universal Plug & Play yesterday, not later this year. Is Microsoft moving fast enough? Yes. The incrementalism of technological improvements to the Windows platform is often maddening. One can wait for years while technologies and capabilities percolate up from the OS to full exploitation in applications. But without a carefully crafted and executed plan to migrate Windows toward the future, large dislocations would constantly call the value of improvements into question. As it is, Windows moves about as fast as Microsoft can convey the value of improvements to ISVs and enterprise customers. Which, of course, brings us full circle: Is Windows moving too fast? Yes, if you are one of those enterprise customers that is as yet unreconciled to the fundamental nature of PCs. Hint: they are called “personal computers” for a reason.

Can everyone be wrong? Yes.

The easiest way to show that “everyone” can be wrong is the millenium. The vast majority of the planet will celebrate the “millennium” one year early, at the end of this year. This is the result of a curious combination of innumeracy, the spinelessness of everyone in a position of authority to correct the situation, and the odd influence of the Y2K bug. There was, of course, no year “0.” So the first thousand years A.D. ended at the end of the year 1000. The second millenium ends at the end of 2000, and so on. The end of this year has no greater significance than the fact it is the last year to begin with the numbers “19.” It is enough to make one into a millennialist nut case just in order to wreak some real havoc on the real turn of the millenium. Instead we shall have to settle of a simulacrum of millennial suspense as we wonder how many computers will crash.

But the serious point of this is that popular assent to an erroneous result can drown out science, religion, and things as objective as the calendar itself in propagating an incorrect statement. No conspiracy required, no intent needed. Surely nobody gains by making a counting error on a global, indeed millennial, scale. But once the publicity machine gets going, inconvenient facts are easily ignored. Now consider the kinds of assertions that are pushed in the popular press behind which there is money, influence, or power at stake. If the press is so weak-willed and weak-minded as to go along with an incorrect date for the end of the millenium, what tendentious and harmful lies could they be convinced to transmit? Virtually any well-organized effort to promote a point of view, properly executed, can succeed in using the press to trumpet that point of view and imprint on it the stamp of a press that never hesitates to promote itself as professional and truth-seeking.

The good news is that you no longer need the conventional press. I get my press releases directly from PR newswires, and filter them by keyword. This way, I get what a company intended to say (to the best of their marketing department’s ability to say it) without the intervention of an “editor” that is usually a poorly paid junior employee of a machine for selling advertising to the trade, and without any question as to the point of view of the author. I also use sites like Slashdot.org where dedicated amateurs find interesting articles in the trade press and general press and post them with commentary. This type of Web site, where amateurs provide analysis, counterspin, and content that springs from genuine personal interest are almost always more informative than the piles of trade and general magazines and newspapers that used to clog my desk. The downside is that I have to split more wood for the wood-stove where a substantial portion of my winter fuel needs once were delivered to my mailbox.

The lesson: trust your judgement and your own calculations.

People will lazily accept astoundingly flawed business plans, marketing plans, scientific assertions, etc. simply because they are cowed by the credentials of the people submitting the result, or are stultified by the process. Most of the math in business planning, and even in the analysis of market studies is very simple. If the numbers don’t make sense to you, odds are something is wrong. And in marketing studies, if the experiment design looks flawed, it probably is. High-tech marketing is a young and underdeveloped field that has yet to find a high overlap of both technical and traditional marketing knowledge and experience. The only person you can rely on to check the correctness of plans, budgets, projections, and analyses is you.

Copyright 1998 Zigurd Mednieks.

Telirati Newsletter #35

Here I make a singularly safe prediction: Unless AOL/Netscape plays to win, they should just give up. Subsequently, the collapse of that entire "threat" to Microsoft was more complete than Bill Gates's fondest hopes. Even if the illustration turns out to lack suspense, the lesson remains: Don't take on large problems with small plans.

Telirati Newsletter #35: New horizons in competition

There is a theory I developed as I plied my craft early in my career: Most management failures are a result of insufficient radicalism. I was a hothead, a trouble-maker. I was always being told that, in the “real world” you had to be patient and that people could not digest too much new technology at once. You can see this mind set writ large if you pick up any trade publication oriented to corporate data processing: “Why do they keep revising software?” is a persistent theme of columnists in those pages. Those bad, bad, software people just do it to make problems for the decent folk that help keep computers running. What is really wrong with the world is that software changes too slowly.

Consider hardware: having crashed through the $1000 barrier, computers plummet toward the frightening figure of $500 per. Hardware doesn’t have to change, it just has to scale. (The fact that all the pieces, from CPU, to ram, to rotating memory, to CRTs, to printers have all kept pace with this same plunging price curve is another marvel which will have to get it’s own dedicated newsletter topic.) So hardware can change the economics of operating computers, change the kinds of applications everyone can use, change everything without themselves changing in any fundamental, architectural way.

No so with software. Make a word processor four times as fast and lo, I still type like leper that dropped a few digits. In order to take advantage of the new hardware, and the new economics of hardware, software has to change, deep down, architecturally, fundamentally. This happens, but it happens slowly compared with the changes in hardware price and performance. There are effects such as enabling nonlinear video editing for the masses that cause software to be pulled into the mainstream as hardware becomes bigger cheaper faster, but, in general, the software that runs adequately on today’s machine will not change your life on a new machine. Software, as an industry does not scale, it deforms.

This means that the nature of competition in software does not scale, it shifts. Which brings us to our example: AOL/Netscape is a potential competitor to Microsoft not just in competition with MSN, but also in competition with the way software is delivered. Even more fundamentally, AOL/Netscape can change what people expect to pay for software and how they expect to pay it. Mark this then: Microsoft’s most important software product is not Windows 2000, or Office 2000. It is Windows Update.

Given this potential for tectonic shifts, how can the DoJ build a case against Microsoft based on an economic theory that says Microsoft can build on its advantages? Every company can build on its advantages. If AOL/Netscape fail to change the software landscape it will be due to failure to build on the advantages of realizing early on that any level of investment is justified in pursuing dominance in Internet access. Every shift in the industry places new players on the field with interesting advantages. Execution, and not the advantages themselves is what makes the difference.

And if AOL/Netscape succeed what might be different? First ask, what is a client environment worth? Why is Windows worth its present price? What if the valuation of a company is based on something other than the price of a desktop OS. Why not give away a client environment? Radical? Someone was sufficiently radical to ask: “Why not give away a Web browsers? Why not integrate it tightly with the OS?” Any response would have to be sufficiently radical as well, and open source isn’t enough.

What are the tests for plausibility here? Size: A browser, a full suite of plug-ins, and a Java environment (especially a full Java 2.0 environment) is about the size of the rest of Windows98. Applications: Is the Internet not application enough? How many people who buy computers are retail stores need Office? Installed base: 15 million AOL users is larger than the Macintosh installed base.

All the ingredients are in place for making life interesting. The only question is whether AOL/Netscape has sufficient audacity. Will they see that, like the early stages of AOL, any level of investment is worth the prize? Can they get a hardware maker to market computers that don’t run Windows or find some other path to a native platform? Can they get their arms around all the potential changes to their business model such as challenge would entail? A winning formulation here could take a variety of forms. But it is certain that the cautious forms, like those pursued by Sun, cannot win.

One last question: Are we not glossing over the matter of an operating system here? A Web browser is fine, Java provides an API, AOL is a place to go and a means of connection, but what takes care of the hardware? The answer is that AOL/Netscape are spoiled for choice: Be, Linux, perhaps the carcass of OS/2 could be disinterred, QNX, etc., could all serve as an underlying layer.

If they don’t rise to the occasion, you heard it here first: It was not Microsoft, not the consumers, not the channel, not the Asian crisis, Y2K, or bad karma that prevent AOL/Netscape from mounting a credible challenge to Microsoft. If you are not radical enough, don’t take on big problems.

Copyright 1999 Zigurd Mednieks.

Telirati Newsletter #34

Four years ago I warned that computers would have to simultaneously continue to acquire the capabilities of large-scale computing systems, and they would simultaneously have to get easier to operate, maintain, and defend against cracking. There has been precious little progress since then: PCs can't be easily clustered, despite the potential .NET has to enable pervasive distributed processing. Security from attacks is still elusive. Encypted file systems are still an obscure feature. An Internet of servers is as far away as ever. And the prospects are still dim. Microsoft has distracted itself with DRM while Linux prepares to make they jump to desktops.

Telirati Newsletter #34: A long way to go

According to a recent report in The Register, a researcher at Sony devised a novel way of designating files to be copied from one computer to another: He used a pointing device, a pen in this case, to indicate which file he wanted to transfer on the monitor attached to the computer on which the file resides, and then using the same pen, to indicate which folder the file should be copied to on a destination computer. This very simple improvement illustrates just how far we have yet to go:

How will multiple computing devices be made workable for individual users, or informal, unmanaged groups of computer users such as families sharing two or three (or more, as in my case) computers at home?

What is the essence of “personal” in personal computers? Wholly possessing your computer has enormous benefits. So does merging its environment with that of other computers on a network. While it is a simple, almost trivial, idea to enable one pointing device to operate on more than one computer, consider the implementation details – how is this to be made easy and intuitive? (Light-pens have certain properties that may make it relatively easy to implement such shared use, but that does not make pens the best choice for users.) Currently, notions of network, computer, state, environment, etc. are far too crude to form a virtual environment as fluid and intuitive as our physical environment.

Sharing a computer is needlessly complex because current concepts of security and resource sharing developed in an era where a person made it his profession to manage the arrangements for sharing. This is no longer economically sound in an era of $1000 servers.

An individual with multiple computers is in even worse shape. Some efforts to manage portable environments has been made for networks of computers, but, as with servers, if computers cost $500, why not have three of them? Why not, but for the fact it is a pain in the behind to keep three computers in order. Document synchronization software is a venerable genre of utility, but a built-in system-wide solution is the only really satisfactory way forward. I should be able to say that one PC is a slave to, or subset of, another, and specify the character and behavior of subset in a simple fashion.

These laments are not mere carping. I get paid to mess around with computers. For most people, the messing around is a dire cost. What does this mean? It means we are not nearly done with the evolution of personal computing environments.

Another measure of the extent to which we are not done is the extent to which concepts from large-scale computing have been absorbed by personal computers. A networking software company recently tried to bring the concept of clustering to PCs. Not servers, but desktop PCs. Why not? If clustering is good for servers, if it makes servers more reliable and more “available” it should do the same for a network of workstations, as they are dynamically added or taken away from a network, data should not become inaccessible, and aggregate computing power should expand as more computers are networked. Unfortunately they only got as far as clustering storage, and had to try to sell the idea as an interesting type of peer-to-peer NOS.

PCs will have to simultaneously continue absorbing the stuff of “serious” computing, and keep getting simpler to operate. Increasingly, there isn’t any computing more “serious” than what can be accomplished with a cluster of PCs. Increasingly, the only administrator is you. Security, privacy protection, redundancy, automated vigilance against cracking are all going to have to become simple, possibly invisible, parts of personal computing in order to enable PCs to unlock the power of pervasive networking.

With wireless networking, you may find yourself in the presence of networks all the time. Some might be your neighbors’ networks! If only to prevent, on a large and horrifying scale, the equivalent embarrassment of overhearing one’s neighbors on the nursery intercom, managing what is personal and what is connected will have to advance with more than incremental steps. Or, to put it in context, it would be embarrassing if “Furby” toys were more adept at self-configuration when networked than a PC is.

Copyright 1998 Zigurd Mednieks.

Telirati Newsletter #33

Here is a blast from the pre-9/11 past. In this newsletter I proposed that secure communication could be the killer app for IP telephony gear. Now that we have entered an era of surveillance society, such an idea seems impossible, even if it isn't yet illegal. Still there is plenty of reason telecom gear should enable secure communication. Many parts of the world are unfree and suffer from official corruption. Businesses and individuals should still be interested in securing their communications from these threats.

Telirati Newsletter #33: Privacy as the norm

Recently, the President of the United States was reported to have mused to one Monica Lewinsky that a foreign government might be listening to their calls.

This has all kinds of interesting implications, but the one I would like to focus on is that is the guy who can command billion-dollar spy satellites cannot routinely secure his less momentous communications.

There are serious national security implications to this, but my point is more modest: If the President cannot be sure of privacy in his use of the public switched telephone network, who can be?

Recently, an assistant to an editor at a publishing house asked for my social security number in order to pay me for the review of some book proposals. The total amount at issue is less than a few hundred dollars. In order to get paid, without making a big deal of it, I had to send my social security number in an unsecured e-mail. I know some people who are paranoid enough not to have done so. I did, but it made me feel decidedly vulnerable, the way I feel turning over my car to some guy in a red vest standing outside a restaurant.

Why should anyone feel so exposed? With the end of the Cold War, the reasons for keeping citizens away from a completely inoffensive means of securing their privacy has diminished. And the tools for implementing security and privacy have become inexpensive enough that businesses have very few excuses not to use them. It is as if otherwise sensible businessmen were leaving sensitive documents lying around in the airport, or leaving the doors of their businesses unlocked at night.

We may be comfortable with the laws and systems in place to prevent corrupt officials from tampering with communications in industrialized countries, but who in this age limits their business to industrialized countries? All the big growth potential is out at the boundaries, where there is what is politely referred to as “insufficient institutional development.” Which is to say that the Prime Minister’s son-in-law runs the local outfit competing with you, and can access the mechanisms of security and intelligence to his advantage. It is naïve and irresponsible not to take some simple measures to minimize the damage that can be done to your business in these circumstances.

Which brings us to a solution to another problem. Over recent years, the margins on business telephone equipment have been eroding. Fancy speakerphones, LCD displays, arrays of buttons, headset ports, etc. have faded in their ability to impress customers who are jaded by the spectacular price erosion of computer power, and expect the same from other electronics on their desks. What can CPE vendors do about this?

Sell security. Security is a high value item. The cost of providing security has dropped to the point where you don’t have to gouge the customer to make a buck. Which opens an opportunity: security has to be present at both ends of the conversation. If it isn’t cheap enough to make it standard, it won’t fly. Security, therefore, should be used system-wide, across the product line, and be the default condition under which a call takes place. Price-wise, security should be used to bolster a market position above commodity vendors, and to differentiate products for “serious” customers from those low priced competitors only uninformed dilettantes would buy. Security can replace the old tools of differentiation that have succumbed to price erosion.

Who can benefit most? Makers of telephony servers. To this point, this is the typical position of a telephony server vendor: NT-based product, integrates IP telephony and wireline telephony, integrates unified messaging, costs about twice as much as a conventional PBX and conventional voice mail. The retort to the price issue is that the telephony server provides high-end features at a lower cost. This is correct: the telephony server provides IP telephony and unified messaging for about half what it would cost to provide the same with a conventional PBX plus IP telephony gateway, plus unified messaging. The trouble is that the market penetration of unified messaging is stubbornly low. IP telephony is a better star to which to hitch one’s wagon, but it too is in market penetration infancy.

Telephony server makers need to identify a feature, identify a market segment that will pay for that feature, and focus their efforts in order to win that first slice of the market. Currently the typical business plan of telephony server makers is fairly modest: a few hundred to a few thousand units, a few million dollars, and based on a modest market penetration rate for new systems in segments ranging from middle to large size key systems to small and medium size PBXs. Some server makers are targeting vertical markets. Good, but so have been vendors of unified messaging systems, with little effect.

The question is: what will connect these products to end-user purchases? What will transform the business of telephony servers from guesses at plausible numbers to a steady stream of purchases from identifiable types of customers? There is no guarantee that security and privacy are the killer apps, but there are proxies that indicate this might be the case: Customers pay for Internet security. Customers pay for call logging systems for sensitive operations. Customers pay for fraud detection systems. These proxies provide a basis for market research, and a source of leads among existing customers for these other products. This is what real money-making businesses are built on.

Furthermore, there is a fundamental idea on which this all rests: Privacy as the normal condition under which people communicate. You should not have to worry about the sneaky competitor, the corrupt banana republic, or the bureaucrat with an axe to grind. And you should not expose your business to these risks. So privacy and security are not just a wedge to crack open the market. Privacy and security can drive turnover in CPE and market penetration for vendors that provide them as a standard feature.

Copyright 1998 Zigurd Mednieks.

Telirati Newsletter #32

Many people have forgotten the Microsoft Cordless Phone. Some, especially those invovled in developing the product, may still be in therapy.

Telirati Newsletter #32: It’s only a phone, how hard could it be?

I don’t do product reviews, and this isn’t a product review. So what am I doing writing about the Microsoft Cordless Phone? Yes, if you haven’t yet heard, Microsoft now makes cordless phones. Naturally, the phone connects to your computer, and thereon hangs the lesson in this newsletter. More often than otherwise, software people come to think “It’s only a phone, how hard could it be?” This, usually just before wading into the quagmire of computer telephony never to be seen again. The Microsoft Cordless Phone is less a product than a touchstone. It characterizes Microsoft’s development approach to a deceptively difficult problem. Let’s see how well they did…

First, what do you get? You get a phone with a charger-base, and a separate radio base-station. This radio base-station is what plugs into the phone line and into the computer’s serial port. (I thought the phone was going to be a USB device. Maybe they just sent a serial version to maximize compatibility?) Both parts are black with gray rubber antennae. The phone is larger than most cordless phones. The buttons are ovoid and bean-shaped, with the feature buttons arranged in rounded groupings that remind me of the Ford Taurus radio and climate control buttons. The shape of the phone is hand-friendly, much like the Microsoft “Dove Bar” mouse. The charger-base does not wall-mount, which is fine with me since anything with a power-cube plug is hard to satisfactorily wall-mount. Installation is a breeze. The drivers work marvelously, playing and recording clear audio in real time over the serial link. The drivers and phone firmware run diagnostics and keep-alive checks that make sure the phone is in touch with the base station, and you are alerted to any errors with informative messages that make it easy to fix problems. Curiously, this phone, which works very well as an audio device, requires a separate audio device in the PC.

You can voice control the phone. I won’t get into the details of this feature, since doing so would occupy the entire newsletter, but the significant aspect of voice controlling the Microsoft Cordless Phone is that this represents getting voice control right. Voice control is used appropriately because when you are holding the phone, you can’t type. This changes the productivity equation. Voice control is, in this context, not competing against the 100% accurate keyboard and mouse.

Voice control human factors are implemented correctly: There is a voice command button on the side of the phone (right handers have their forefinger on the button, lefties their thumb). You tell the phone when you want to issue a voice command with a button press. You get a start beep, and a confirmation beep (or a less pleased sounding un-confirmation beep). There is no mistaking what is a voice command and what isn’t. Since the button is positioned under where your finger rests, there is no hand movement, or anything else that takes you away from your task, associated with issuing a voice command. The feedback is audible, so you never have to look at anything to see if the command was correctly interpreted. And, since the command is spoken into a telephone handset, and all the results are audible through the handset, voice control is not any more intrusive to the people around you than normal phone use.

Microsoft’s hardware division is a curiosity: It is non-strategic. It is a relatively low-margin business (compared to software, any hardware is low-margin). And it makes very nice products. I am typing this on a Microsoft “earth shoes” keyboard and am mousing with a Microsoft mouse. Nice stuff. Always high touch. The phone is no exception. Microsoft must have spent a bundle developing it, and it shows.

Well, it shows in the industrial design, in the product formulation, in the quality of the device drivers, and in the attention that went into conceiving a really useful and usable application of voice control. But what about the application software that comes with the phone? Here things were less well executed: The phone comes with a dialer/answering machine program that doesn’t merit long discussion because it is no better than any above-average prior attempt. The flaws come in with poor integration into existing messaging client environments, in which contact data is shared, but messages end up in their own separate message store and must be accessed through their own UI. A hole in the head, or another place to put messages – which is needed least? Capping the mediocrity of the dialer/answering machine is the UI itself, which is a “consumer-look” UI that, while I have no fundamental objections to friendlier-looking UIs, fails to justify its differences. Indeed it points up the gratuitous-ness by implementing the typical tree view of multiple mailboxes, only with bigger and clunkier graphics than the usual Windows look.

To be fair, designing a multi-user messaging system for Windows PCs is not child’s play. You have to make such a system a true-store-and-forward system in order to integrate properly with messaging clients and Windows’ multi user capabilities and security. A lot of work. But, in addition to illustrating how hard it is to manage design quality across the large spectrum of technologies that go into making a computer-connected phone, the failure of the application software to measure up to the quality level of the rest of the product shows a general disconnection between an otherwise brilliant hardware division at Microsoft and the kind of technology inter-locking that makes Microsoft software hang together so well.

There are three key areas where opportunities went unexploited, either to the immediate detriment of the product, or resulting in missed chances to reinforce Microsoft software technology initiatives: a) Messaging architecture: The messaging software could have shown the way in applying Internet and Windows messaging standards and capabilities. Instead, it settles for using a little of it and mucking up some of it. b) Audio: A telephone handset represents a unique opportunity to create a business audio tool. It enables audio to be used in a cubicle setting without disturbing co-workers. It provides a microphone that won’t get misplaced when most PCs do not have one connected. And the human factors issues of muting other audio while the phone is in use for calls could also have been handled. Perhaps with the USB version. c) It is remarkable that this phone is not well integrated into NetMeeting. For the same reasons that a phone is the right audio device in an office setting, it is also just what is needed for NetMeeting.

It’s only a phone. And this is how hard it can be to get right. With what is probably ten times the money of any other computer connected phone in the development budget, and a result that is head-and-shoulders above anything that came before, there is still plenty to do for rev. 2. While that sounds like a lot of work just to do a good job with a hardware peripheral that probably will not do nearly as well this Christmas shopping season as the Microsoft “Arthur the Aardvark” dolls (I’m not kidding!) there are, in fact, profound, strategic, and highly lucrative opportunities for getting audio, telephony, and the link to conferencing tools like NetMeeting right.

Microsoft has a history of getting it right by continuing to try. The Microsoft Cordless Phone is definitely not “Bob,” and deserves a continued high level of investment. The future of computer telephony could be getting a lot more interesting.


Copyright 1998 Zigurd Mednieks.

Telirati Newsletter #31

In 1998 the world was only a few years into the widespread commercializiation of the Internet, and the Internet bubble was only beginning to form. Now, after it has burst, we still have a "TV" model of Internet service for most customers - no servers, limited ports, all kinds of restrictions on what, still, should be a nework where every node is potentially a server.

Telirati Newsletter #31: Who will pave my information super-driveway?

I live at the end of a nine-hundred foot long dirt driveway through the woods of rural Massachusetts. My local CO is not a 5ESS nor a DMS100. It is an earlier model, and Bell Atlantic is moving no faster than Nynex was in upgrading. No chance for ISDN here. My cable company, which had been a sleepy little outfit that purposely made itself hard to find in the phone directory, was recently bought by Paul Allen, and has now vowed high-speed Internet service on an HFC network by next year. But that might be by the end of the next calendar year. I am on the fringe area of a radio-outbound telephone-return ISP, and the equipment costs too much. Who will pave my information super-driveway?

The two main contenders are cable-based services and digital subscriber loop (DSL). The phone companies and the cable companies are realizing that the Internet matters, a lot. Some very intelligent people, not to mention the investment bankers raising the money for these build-outs, are backing one side or the other in this race. It is worth taking a look at why either side might win or lose.

This isn’t to discuss the state of the race, rather to lay out some of the ways in which these new Internet access providers will succeed or fail. I will set out the criteria in the form of questions:

Stingy, or generous? This is likely to be a key question in the acceptance of Internet connection technology. Customers don’t know, except in the broadest terms, that streaming video is several orders of magnitude more bits than the largest document they would ever care to read. Customers do know they don’t like nasty surprises on their bills. This is why I don’t use satellite Internet service. The rate plan is too complicated, and who wants to count bits? Currently cable ISPs discourage multiple computers on a single connection, they throttle bandwidth, and in other ways build up bad karma among early adopters that tell the rest of us what things will be like. Initial DSL trials are overpriced and artificially under-perform. This is meant to protect T-1 revenues. Forget it.

What about the track record? Here, telcos are in a hole. ISDN was bollixed up by telco inability to absorb a new product into their service procedures. ISDN was just enough more-difficult to install that the service training mechanisms telcos have in place could not carry the burden. The result was a shortage of technicians able to install ISDN. Cable companies do not have comparable comprehensive disasters on their record in absorbing new set-top box technology.

Open, or closed? Do you have to buy, or rent the equipment from the provider? Is it standardized and licensed so that third party suppliers can play? Here the telcos have an advantage. They have now gotten used to customer-owned equipment, and if they can suppress greed for just this one decision, they will benefit. Open set top boxes probably scare cable companies. Open systems will have more choices: USB or Ethernet? Built-in router and address translation, or vanilla bridging? Choice, along with channel power, will help decide the winner.

Service with a smile, or a scowl? Telcos and cable companies are in a race for the bottom here. I had to do some serious detective work to find out who my local cable company is. No listing in any local phone directory! Nynex (now part of Bell Atlantic) was legendary as the place where Soviet bureaucrat and management efficiency lived on. Who will be smart enough to change or work with a partner that can deliver real service?

Do they get telephony, or not? How could telcos not get telephony? Easy: they could try to change for IP telephony separately from the rest of Internet service, or force customers to use their gateways. Cable companies will have to resist some of the same temptations. Some of the pioneers in cable Internet have already invested in their own carrier class switches. Will they force customers to use their non-IP telephony over cable, or to use their gateways? Or will ISPs tempt customers into the fold with integrated unified messaging and other Internet-based services for telephony? Who will understand that offering separate voice mail and e-mail (and possibly separate fax services) is a losing idea?

With this many variables of the competitive equation still unconstrained by experience, it is impossible to say who will win. Personally, I hope wireless technology is able to add a competitive dimension that is certain to be to everyone’s benefit.

Apart from my selfish interests, Internet numbers remain quite exciting. A tenfold increase in Internet subscribers and a twenty-fold increase in speed will make new economies and new product classes possible. The stepwise speed increases in the U.S. and Europe presage worldwide penetration. Much has been written about the Internet’s impact of society, but we have not seen a fraction of what will soon come. India, for example, has only 50,000 Internet subscribers in a country of a billion people. Once that fraction, in India, China, the Arab world, and Africa reach the same ratio as existed between literate people and the population as a whole when printed books became widely available, interesting social changes will probably take place. Computers prices overlap televisions substantially. Windows CE-based systems, and other lightweight OSs, will bring Internet access devices into parity with television in cost. These global numbers leave open several generations of access technology to sweep over various regions of the world and provide the connections that will put Web access in every tea room, and the private homes of the middle class, in the developing world. At that point the border-less nature of information flow in the Internet will have a profound influence.

Copyright 1998 Zigurd Mednieks.

Telirati Newsletter #30

He shoots! He scores! This is one of the best things I ever wrote.

Telirati Newsletter #30: Why tu que?

We are shaped in our outlook by our geography. The north woods of Maine offer a desolation unimaginable to a Western European city dweller. The American west is even more impressive, and the emptiness of some quarters of it gives rise to the romantic notion of escape and refuge. The latest group to pack their pickup trucks with guns and mean dogs and cast their gaze toward the distant hills is year-2000 alarmists.

Positing a frightening cascade of system failures, the alarmists predict global meltdown of economic activity and social order. Most of these alarmists are American. Not coincidentally, I think.

The possibility of running away from a collapsing society and the prediction of such collapse are linked. Whereas a client of mine from the Netherlands pointed out that there they have a) Not very many guns; b) No hills; c) Dogs that are mostly the terrier rug-rat sort; d) No place that is very far away. The Dutch aren’t going anywhere when the Y2K bogeyman comes, except maybe down to the pub for a beer until the computers are rebooted.

The cascade of failures is one problem with Y2K alarmism. Another is that most industrial uses of computers are often no less useless than most desktop deployments. They are just more expensive. What we will find when we reach the millenium and a number of elderly systems crash is that many computers are employed foolishly, on tasks that barely require automation, much less the maintenance of costly legacy systems.

Another thing we will find is that the same methods that work when computers fail now will work when computers fail due to the Y2k bug. A lot of the U.S. air traffic control system runs on alarmingly antique systems. It breaks often. There is a manual backup method or controlling air traffic when the automated system breaks. It is only slightly less alarming than an antique computer, and involves slips of paper on a note-board. But it works when the antique computers break.

It drives my wife nuts when it takes me days to set the date and time on the oven when we have a power failure, but I find clocks that read wrong not very disturbing. Ever hear of the Y2K fax machine problem? Yes, lots of fax machines will start screwing up the dates on faxes. Got to trash the lot of them. The fact is, most of these buggy fax machines will continue to be used, wrong dates and all, until they die a natural death. The “service” light on your dashboard? A square of black electrician’s tape will take care of that more efficiently than replacing the car’s computer.

The enterprises that derive a real competitive advantage from IT have solved their Y2K problems and will be quick to mop up the unforeseen ones when the time comes. Which is why Wal-Mart will be up and running, and their cheesy competitors still won’t have that plastic sandwich container I was looking for, only they will blame it on the Y2K bug and suggest I should have hoarded them when I had the chance.

I was once asked by the proprietor of a bartending school how powerful a PC he should buy to automate his database of alumni. I asked him how many database records he needed to keep track of. He pulled out a stack of index cards. There were about 250 of them. I nodded and decided that I had to ask some more questions to really understand his computing needs, like: “Do you like those 3-D shoot-em-up games?” His academy is in more danger from too much lab work than the Y2K bug.

Even systems that make good use of computers, like just-in-time (JIT) manufacturing supply chains, generally have non-automated precedents. Some Japanese implementations of JIT rely on color-coded parts bins, blackboards, and slips of paper. When the parts bin is brought to the supplier, he fills it. Relatively few industrial processes involve so many operations that must be tracked or calculated in detail as to actually depend on computer automation for their existence.

There is one effect that is underestimated: A number of hardware and software companies are either blaming a current downturn in sales on companies applying their resources to the Y2K bug rather than to deploying newly purchased systems. Other companies, notably voice mail companies, are benefiting from customers deciding to finally scrap antiquated systems in their Y2K update process. This effect will not decrease after January 1, 2000. It will increase as Y2K bugs that are not fatal, but are deeply annoying, drive an intensified Y2K bug hunt. This means continued hard times for big, expensive, hard-to-deploy systems, another nail in the coffin of mainframes that can be replaced by commodity servers, and a continued up-tick for purveyors of moderns, standards-based systems that can be quickly deployed.

To sum it up, we will learn that many computer systems are now misapplied, and could be done entirely without. We will learn that some computer systems are needed, and used, and well taken care of, and that there is a real benefit derived from these systems. And some situations where ancient systems have been kept on in fear of change will find the Y2K bug driving modernization. Finally there will be genuine botches, where some businesses will fail due to the failure of their badly run IT. But it isn’t the end of the world.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #29

In 1998, the idea of Java becoming a threat to the Windows desktop environment was, already, dead. Now Java is the dominant langauge for creating mobile applications for mass-market handsets. Back then, desktop Java died, in part, for lack of a "home" platform. Now the question is whether the 450 million telephone handsets shipped each year provide such a home. J2ME - the mobile variant of Java - has been used mainly for games, and Java is not tightly integrated into any mobile handset OS, with the exception of SavaJe. This is surely a topic for blog posts to come!

Telirati Newsletter #29: Who killed Cock Robin?

No! Not the ice cream parlor! (Though that would make another study in bad management.) I mean why is Java as a serious threat to the Windows platform all but dead?

It was a simple and beautiful idea: Start from scratch. Build a language. Build a computing environment with system services and user interface services. Without all the baggage of history, how could Java fail to be more elegant and wonderful than Windows?

When I first saw Java, this idea was tremendously appealing. I felt almost as I felt on seeing the first Macintosh. I saw not an incremental improvement, but a leap big enough to make a difference. Here was a credible attempt, sufficiently bold, to unseat Microsoft from dominance.

Java was also linked to the Internet through applets, which offered a glimmer of a software future where software arrives when needed, serves the user, and disappears before making a maintenance nuisance of itself. This future will arrive, but not in the form Sun intended.

Why has Java fallen so far short of challenging Windows in any serious way? It has fallen much shorter than the Mac OS, and Mac OS has the burden of past failure on its shoulders. Why is Java relegated to the uninteresting role of providing a plausible way to modernize software development on a range of medieval IBM systems, plus NetWare, for which software development is only incrementally less painful than the Spanish Inquisition.

The answer is: Java lacks an iMac. The iMac is a pleasure to contemplate. It captures what you can do when you control both hardware and system software (like actually make USB popular, something Intel and Microsoft have labored long and hard at without conclusive results). The iMac is wonderful even without the next generation of Apple OSs, and it will fuel upgrades to the new OS when it arrives.

Java does not have an iMac. Java does not even have a Cobalt (a nifty Linux-based server box). Java has got squat. Why is this lack of a compelling native platform fatal to a language who’s fundamental dogma is non-nativeness in the form of “Write once, run anywhere?”

The trouble is that the dogma is defective. Not enough people care about writing once, running anywhere. Everyone who has succeeded in selling desktop software sticks to the more practical belief system that can be summed up thusly: “Write it for the platform.”

Without a native platform this is what cannot be said to potential Java software developers (imagine, for a moment, there is a really neat Java computer): “The reason you should write Java programs, and stick to Sun’s view of the correct set of user interface classes, is that we have this nifty Java computer here. On it, your programs will run marvelously well, and look stunning. We guarantee to keep it that way by continuing to develop new and better Java computers. So while we have only 5% market share now, we will be doing much better in the future. And, to top it all off, your code will run on all those Windows computers, too.” Instead, all you get is a compromise position: Windows compatibility.

Sun could not even muster the sense of direction to revamp application development and user interface on Sun workstations running Solaris to use Java. Likely NetWare will make better use of Java in managing NetWare than Sun will in managing Solaris. While I do not know if any serious discussion took place on the matter, Java could have become a supercharged form of AppleScript, able to script other applications and create new ones. But this would mean adopting Apple’s user interface classes and becoming platform-specific in such an implementation.

The irony is that becoming a platform-specific language is Java’s fate: A better Delphi than Delphi, Windows-specific, and the optimal middle road between Visual C++ and Visual Basic. Microsoft will end up successfully assimilating Java into the MicroBorg, where tendrils of COM technology will ensure Java is inextricably intertwined.

Why did this happen? Was it timidity in taking on ever-cheaper PCs? Was it development projects gone bad? I do not know. But the result is that Java has no home base, no native country. And this prevents Java from ever having a chance, as a language and environment together, of changing personal computing. Too bad.


Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #28

Looking back at 1998, before the launch of OS X, Apple was, once again, caught trying to reshape strategy on the fly. But, despite all the last-minute course corrections, and as I predicted, Steve Jobs would re-emerge as a force in computing.

Now, however, Linux has become an even more prominent force in computing. If Jobs can position Apple relative to Linux and Open Source software in general, he has an opening for a true breakout in the market.

Telirati Newsletter #28: Steve Jobs, The Most Dangerous Man Alive

A wise man once said “Never say ‘Watch this!’ before throwing a Frisbee.” Steve Jobs evidently listened to this sage, and is reaping the rewards with the iMac. A craze of comparable magnitude to the Volkswagen Beetle revival has greeted the iMac. (Even though the iMac bears a disturbing resemblance to a deservedly forgotten vehicle, the BMW Isetta.)

Had Jobs explained what the iMac was intended to be, it would have been counted a failure, or at least a bunt, when it was delivered. The iMac was supposed to be the first of a breed of machines running a simplified Mac OS, so that Mac OS could live on as an OS for PDAs after Rhapsody took over as a desktop and server OS. Instead, the iMac is heralded as, and deservedly so, a brilliant repackaging of the original Macintosh idea.

In Newsletter #10 I posited that turning Mac OS into an NC or PDA OS would fail, because it is devilishly hard to drop baggage once it has been acquired. So it was. But the iMac is a success anyway. And so likely will be the revised Rhapsody strategy. Instead of relegating the Mac OS to PDA and NC duty, and attempting to convert ISVs to a new religion, Mac OS X will seduce rather than command converts with a high degree of compatibility with current applications and a powerful new set of technologies for creating new applications.

It is Mac OS X that makes Steve Jobs the Most Dangerous Man. With OS X, Jobs is the only competitor to Microsoft with a full suite of technologies, across the full span of market segments to be able to go to a corporate IT user and say “I have a better way, on the desktop, in the servers, over the network, for the Internet, than Microsoft.” Sun doesn’t have it. Novell doesn’t have it. IBM doesn’t have it. And Linux, while it has many advantages in creating cheap, simple Internet servers, isn’t even close to having it.

What has Steve Jobs got? He has got:

An installed base: Millions of happy Mac users now have a reason not to give up and go Windows. The iMac and succeeding machines will keep these users in place and grow their numbers to give Apple plenty of time to release OS X.

Momentum: The success of iMac and the likely success of follow-ons means that Apple will launch OS X in an environment of expansion and not crisis. Contrast this with Novell: No matter how good the next Novell OS release is, Novell is, at best, fighting a rearguard.

An OS that can compete with NT: Remember what I said in Newsletter #10 about the difficulty of dropping baggage? How many years ago did Steve Jobs figure that using UNIX as a starting point would lead more quickly to a capable system that would be better and cleaner than the Mac? Well, he’s finally near to done: Mac OS X is comparable to NT in almost every way that matters. And it is also the only UNIX that has successfully morphed into something my mother could use.

A user interface technology: Apple’s Mac user interface is the only alternative to Windows that has attracted a significant number of ISVs. This fact alone means Apple stands alone as a possible effective direct competitor to Microsoft. The OpenStep-based UI technology in OS X is a credible claimant to call itself the most advanced UI technology available.

An object technology: Open Step, or whatever they are calling it these days, is an object-oriented application framework. Plus it has distributed and component capabilities that make it, and the Objective C language system, roughly (though certainly not completely) equivalent to COM on the NT platform. Like COM, the object technology in Mac OS X is pervasive (unlike CORBA) and used throughout the system. This is currently the weakest point of Apple’s appeal to corporate customers, but it is a beginning.

I am not ignoring the fact that this is the same Steve Jobs who offed his OS licensees. Apple may find it is still too little, too late, and there are too few allies available. Implementing a real threat to Microsoft might involve merging with one of more companies that have a corporate installed base of UNIX systems, and migrating those customers to OS X servers. Do you see anyone else in this position?


Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #27

It wouldn't be a retrospective without a few thudding failures in prediction, like this one. A complex and only tenuously relevant analysis leads to an utterly wrong conclusion.

Telirati Newsletter #27: Picking the hits.

It has been an abiding interest of mine to figure out when technologies will become widely accepted and used. Recently it was brought to my attention that my interest in determining when, for example, real customers will buy unified messaging systems parallels a fashionable trend in the analysis of the history of technology.

Writers like Jared Diamond, in seminal books like Guns, Germs, and Steel have brought an interesting new infusion of rigor to the history of technology and even to more-general topics in history. This new wave of historical analysis finds the underpinnings of historical outcomes in geography, climate, minerals, plants, and disease conditions of the regions of the earth.

This is an attractive analysis. Genetics tells us that man is far more alike in all the ways that matter than superficial racial differences suggest. Finding alternative explanations to why dead white men, mostly English, dominate the formation of the modern world lets the air out of any racist analyses from all sides of historical debate over why this is so.

Diamond’s analyses, and those of historians with similar approaches, show that Greece, Rome, the Christian church, the Renaissance, the American Revolution, the industrial Revolution, and everything else leading up to modern liberal capitalism may have happened where and when they happened because of the guns, germs, steel, flora, fauna, rocks, and water of the vessels that contained those events.

The applicability of this type of analysis to technology product acceptance was brought home to me by a recent article in Feed, an online publication that surveys this new wave in historical analysis. The article describes the approach used by Jared Diamond, who places great stock in geography, Paul Levinson, who takes a Darwinian approach to the history of ideas, and Brian Winston, who hitches a school of linguistics to his plow.

Winston theorizes a multi-phase process through which ideas are accepted: ideation, prototype, social necessity, invention, suppression, and adoption. These steps are inspired by Saussurian linguistics, which, argues that speech is expressed through a process of social acceptance in which ideas are integrated into society before the structures needed to express those ideas become a part of language.

All of which sounds suspect to this one-time linguistics and cognitive science major. But the links, however tenuous, to some claim to a connection to the way that the mind works makes Winston’s analogous application of these linguistic ideas to history relatively plausible. After all, the geography of high tech seems to consist uniformly of cubicles (though you should recall that I place some credit for Microsoft’s success on putting their coders in offices). To be a bit more serious, the Internet makes Singapore and Stockholm equally likely candidates for the location of the next breakthrough.

The Web itself is an example of a technology that traversed Winston’s six steps: Early hypertext systems were heralded as a fundamentally new and better way to organize and communicate knowledge, with numerous benefits. But those early products languished. I cannot exactly map the social necessity of hypertext, but those who proclaimed hypertext in the first place were solidly convinced it was a step forward for mankind, even as their idea lay fallow. In the invention phase, an interesting thing happened that may well be a significant elaboration of a theory of product acceptance: Hypertext stopped being a product as the World Wide Web escaped the halls of academe, where it was invented and lay in suppression for a number of years. The Web just is and does, and few users think: “That hypertext stuff is a great idea!” In fact, the Web is deficient compared with earlier hypertext systems in that it does not manage broken links other than to report errors when such links are encountered. The Web has become a widely accepted idea at the same time the users of the Web forgot why it was created in the first place and forgot about the product category, hypertext, that it is part of.

My theory is that a similar thing will happen to unified messaging. Unified messaging has passed through the early stages, through prototyping in which proprietary technologies were employed, and now is poised to break out. This will happen, however, once people stop thinking of unified messaging as a separate, high-end product. Unified messaging will suddenly and without much heralding replace voice mail as we know it. The key is to find the inflection point from the present state of suppression to the future state of ubiquity. I think that point comes with you stop hearing the words “unified messaging” and begin to hear telephone companies talking of making their business model more Internet oriented. One obvious expression of this is that my cell phone’s address book will be automatically updated from an LDAP directory, and that one access option for my voice mail will be IMAP. This will happen without hoopla, and, very importantly, it will happen without price increases. The capability, like HTTP serving today, will simply be the price of staying in the game. Acceptance is, after all, ordinariness.


Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #26

IP telephony used to have no "killer app." Really, it still doesn't. Which may have something to do with the fact that IP telephony is only now taking off, now that mobile telephony has eclipsed land lines and buying a conventional PBX seems a bit like buying a horse cart. IP telephony still does not, in general, offer better privacy, or even better voice quality than conventional circuit-switched telephony.

Telirati Newsletter #26: Is privacy the killer app?

One major reason why IP telephony makes sense for replacing PBXs is manageability. Data network managers have efficient, powerful, and reasonably priced management tools available to them. Tools for managing voice networks are more expensive due to the fact that the market for these tools is confined to the high end.

A close relative of management is logging. Implementing voice logging in IP telephony systems can be a matter of configuring the software, which makes it cheap, easy, and invisible. In conventional systems, voice logging involves a voice processing port for every trunk line on the system. Combined with the easy and powerful management available in IP telephony, logging systems can tag logged calls with all kinds of call data, for more powerful cataloging and readier retrieval.

This combination of effective, efficient, easy, and inexpensive administration and logging can be a blessing. In environments like retail brokerage operations, voice logging is a necessity. But there are cases where this ease of surveillance can be abused. One might argue that instances where the greatest potential is for abuse rather than benefit are vastly in the majority.

This is no fault of IP telephony. Rather it is the difference between the closed and opaque world of telephone systems and the open, transparent, programmable, and examinable world of data networks.

Privacy and encryption are central to implementing IP telephony in a way that users find trustworthy. Both the cause and cure related to easier snooping in IP telephony rest in another characteristic of IP telephony: the smart endpoint.

If the increased intelligence of the IP telephony system, and its programmability and flexibility make it ideal for (and what a quaint word it now sounds) “wiretapping” the increased intelligence at the endpoints that makes IP telephony possible also makes it possible to guarantee privacy.

This means that to be trustworthy, IP telephony networks do not need to be trustworthy in themselves, only that the endpoints provide trust through strong encryption. This is the requirement: the call has to be secured inside what the user can see, control, even own. The PC, or the dedicated IP telephony handset, has to contain software for encrypting the call so that it can thereafter traverse all manner of exposure on its way to its destination. Without being certain his handset is able to secure the call, users of IP telephony will reasonably conclude that their calls are exposed to the increased capability to perform monitoring and logging in IP telephony systems as well as the increased hazard of hacking in any system as open and diverse in application as the Internet.

So people should want encryption. But people don’t always want what they should want, as many companies selling backup and anti-virus software (and dental floss) will tell you. Is there then, a desire for privacy and security driven by more than innate virtue? If there is it probably lives near the reason people use IP telephony these days: to bypass international long distance charges when calling obscure corners of the globe. This type of call, more than any other, is susceptible to official and semi-official snooping. And often this snooping is more for the purpose of finding financial advantage than for the cause of state security.

Are there enough people exposed to real hazard and with awareness of that hazard to popularize strong encryption for voice telephony? This can only be determined from primary source research into the market: How many people use IP telephony in situations where they have a real reason for concern about snooping? How many people would use IP telephony to secure sensitive calls if security and privacy were available? My impression is that this second group would be larger than the first, and that privacy constitutes something of a killer app, the kind of capability that justifies the whole system.

The downside is that the profile of the early adopter resembles that of the early videotape and Internet enthusiast: he may be more interested in keeping his communications with his Monica-equivalent private than he may be with the general security and trustworthiness of the communications system. So if this is the privacy customer, it would not be the first time a worthwhile technology was paid for by not the visionaries, but rather those blindly driven by their glands. Farfetched? Recently Salon magazine covered the chagrin arising from the fact that Microsoft’s directory servers for users of NetMeeting have become hotbeds, so to speak, for people who want to share more than their pie charts in multimedia conversations.

The best outcome would be like that of other technologies pioneered in their use by less than platonically motivated users: the technology of privacy becomes widespread enough that the default behavior of people is to choose privacy.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #25

In this rant, I let rip on what was, in 1998, Microsoft's inability to make the NT elephant dance. The fact that now we all use NT - in the form of Windows XP - is a huge credit to Microsoft's abilty to execute.

Telirati Newsletter #25: Microsoft’s soft underbelly

What’s wrong with Microsoft? Not much. Expect NT to continue to consolidate the victory it has already scored over Unix and NetWare as a server platform for mainstream business server needs. Expect Windows 98 to be a winner in the upgrade market despite the usual grumbling gurus that say “don’t tamper with a working Windows 95 system.” Expect Windows CE to start taking off.

But don’t forget that not everything made in Redmond turns to gold. Bob and At Work clunk the loudest when you shake the box, but there are incremental losses as well. Predicting such troubles is an extremely dicey business. To give you an idea just how dicey, I not long ago said that a particularly likely peril of perdition lay with Java, and Microsoft then corks it with Visual J++ 6.0. But it can still be instructive to look at the places where Microsoft skates on thinnest ice.

Compaq now owns Alpha, Digital’s CPU architecture. Alpha languished at Digital because it is a 64 bit chip running 32 bit NT. Yes, it runs 64-bit Unix, but that had little impact (which says something about Unix, but that’s not today’s topic). Merced, Intel’s new architecture, is also a 64-bit chip, not due until 1999. I could hear something that sounded like a huge sigh of relief emanating from Microsoft when Intel announced Merced would be late. The brave announcements that Microsoft would have a 64-bit NT ready for Merced, right on the heels of NT 5.0, never sounded very convincing, and look less so as NT 5.0 slips, and other aspects of NT 5.0, like Wolfpack, get scaled back in ambition.

If the lack of 64-bit NT hurt Digital so deeply, what would the presence of Merced do to Microsoft if they are not ready with 64-bit NT? This question assumes Merced can drive OS decisions, an assumption unlikely to be universally true, but serious enough to be worth asking. Speculation that this will hurt Microsft is already n the trade press.

It is also worth asking why is NT so late. It is because NT is not like most Microsoft software. What? Isn’t NT the very definition of Microsoft software? Isn’t NT the destiny of all PC software at Microsoft? No, and not necessarily. NT was for many years the most alien presence inside Microsoft, and only recently has NT and the rest of Microsoft begun to converge on a single philosophy. NT is big, unfriendly, and hard to deal with. It is less unfriendly than and less hard to deal with than NetWare, which is why it is sweeping NetWare from the playing field. But taken on its own, NT is no joy.

Contrast NT with Windows 95, which was a ringing success in finally delivering on the promise that a Macintosh-like GUI environment could be created on PCs. The easy part was making a windowing environment about as good as the Mac’s. The hard part was making installation of software and peripherals as easy on PCs as it is on the Mac. This latter success was won on the basis of brute-force hard work, and deserves a newsletter of its own to tell the tale. Windows 95, for all the compromises and architectural awkwardness of a system that has to carry the considerable deadweight of a DOS legacy, went beyond Macintosh is delivering a consistent, object-oriented, 32-bit programming environment that spans client and server systems to PC. No small achievement. Where was NT when Windows 95 changed the PC world? To be fair, without the development of the Win32 APIs and COM on NT, Windows 95 would be rather less revolutionary. But NT did not pay much attention to making the customers life easy. The Windows 95 user interface was adapted to NT with NT 4.0. The miraculous level of hardware compatibility found in Windows 95 has not come to NT yet.

What marker can one seek to measure the distance between Microsoft’s consumer OSs and NT? Windows 95, and now 98, are sold to any Tom Dick and Harry with 90 bucks and PC. This customer is likely never to open the documentation. NT is served by a priesthood of MCSEs, geeks who have taken courses to master its complexity. So, while superficially, one might get sucked into thinking that, after this product generation, NT could entirely replace Windows 95/98, the divergent cultures of these two systems will have to be reconciled through a tremendous amount of work before the blessed event of one PC OS will ever happen.

Let us take a moment to consider this MCSE thing. Certified geeks to operate computer systems is an idea Microsoft copped from Novell, which had long made scandalous sums of money training and certifying people so that they could successfully install and operate their product. This might make some sense in a world that thought of computers, at least of servers, as some type of industrial equipment. You would send your machinist to a training course to learn about his new lathe. But computers have advanced beyond this stage. In the previous newsletter I covered small servers. If nothing else, small servers indicate that, in order to sell lots of servers, they have to be at least as easy to buy, install, and use as desktop PCs. Whereas a culture that encourages semiprofessional certification as an indication that a person can use a product means that product is not ready to be used by the general public. Microsoft product managers could respond to this with reams of white papers and presentations that would tell you that the complexity of Windows NT can be either hidden or encapsulated, and that not everyone would need to learn the complexity of the system or to access every feature. Don’t believe it. Dry feature lists and usability studies do not capture the essence of the problem. Not until the culture of trained system operators is not only obsoleted, but actively denounced and overturned will NT be ready to replace Windows 98 as an OS for everyone.

Back, then, to the root of trouble: It is the culture within Microsoft that produced the culture among users that requires trained operators that is the greatest risk to Microsoft’s future. This is for Microsoft, what comes after their successful response to the Internet. The next task is to internalize and weave into its institutions the fact that sauce for the user goose is sauce for the administrator gander. The people running systems deserve at least the same level of effort that goes into making a writer’s, or a Web designer’s, life easy in outstanding products like Office and FrontPage. When there are lots of servers, there won’t be enough administrators. Users, customers, will be administrators. It is therefore useless to think in terms of better administrative tools if there is no one to learn those tools. This is the question: Can Microsoft make servers customer-friendly? Not administrator-friendly, but customer-friendly. Customer-who-runs-a-business-not-servers-friendly.

Big OS upgrades are a dangerous undertaking. Failure of big OS development projects caused the fall of Novell and Apple. For Microsoft as well, there is a simpler danger than the complexity faced by customers, which can be papered over in the short run. NT 5.0 is late. And while Microsoft, not less, and often more than some other software organizations trumpets their methods and controls that make it possible to undertake huge software projects, the reality is that isf these methods worked as advertised, projects would not go late. Is adding pressure to simplify the user experience just going to make OS projects even more complex, and therefore more late?

While it is too late for NT 5.0 to benefit from a re-think, it is worthwhile to examine whether user-hostile administrative interfaces and late projects are somehow linked. How could these two factors interact? Take, for example, configuring NT. In theory, NT server and NT workstation are one OS configured only slightly differently. In practice, the set of NT server facilities that enable networks of workstations to operate securely and in an orderly manner are what make NT server different from NT workstation, and it is impossible for the average computer user to understand these services and, especially, how these services are intertwined as they share information about users, servers, security, etc. If there were a visual tool, not an ugly pile of dialog boxes that only serve to format obscure textual information rather than interpret and represent it in graphical form (they do call it a graphical user interface, which should be a broad hint to interface designers), to manage configuration, that would be progress. In order to make such a configuration and management tool safe for users, modularity, consistency-checking, and error handling would have to be raised significantly above the level it is today in NT management tools.

Take this simple datum as an indictor: in most administrative tools, there is no “undo” operation. Start with that, and then implement every ease-of-use enhancement since 1984 and you will have raised administrative tools to something like what the average marketing drone gets with PowerPoint. Then go a few steps beyond that, and the average PowerPoint user might then be able to install and operate a server. This is one way to express the goal of simplification, and not a fatuous way either. That guy with the PowerPoint slides is your server customer. And that server customer won’t be able to hire a technician for his server. And that technician makes ever less economic sense as server prices fall.

But back to the synergy with server development. For all the gasconading about rigorous development methods, which do you think might impart more discipline on the process: More management controls over the development process with more bug-counting and code metrics, or the discipline of having to make NT installable, manageable, and configurable by the average PowerPoint user. If Microsoft made NT the way it makes products for real human beings, it would not only benefit the customer, it would make NT development more connected to disciplines that matter in making quality software in ways that software project management methods seldom adequately capture.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #24

Here is an idea whose time didn't come, but is ripe for a second look soon. The success of Linux in embedded and consumer electronics applications means that many of these appliances will grow to become the personal servers described here. The general stability of file sharing protocols has made them possible to reverse engineer, and Macs, Windows PCs, and Linux nodes will like happily together.

Telirati Newsletter #24: The next battle: Personal networks, tiny servers

A lot of the smoke and noise coming from the computer battlefield over the next three years will be over servers and networks. For most, the perception will be that this is all about servers for the un-exploited small and medium sized business market, and about the further downsizing of formerly mainframe-based systems. These are, certainly, important developments. But like a really big rock rolling downhill, there are not going to be many surprises about where all this is going.

There is, however, a market in a wonderful state of ferment: personal networks and really small servers. Big servers are capable of a wide range of purposes: file sharing, backup, faxing, telephony, etc., even if the each server specializes on one such task. Small servers will have a fixed purpose. They are built for a single task, or related set of tasks. Their hardware is designed for those tasks and no others.

Who will buy these small servers? What person needs a network? Good questions and ones that do not necessarily have answers that are optimal for the interested parties. So we start with those asking the questions: PC makers who have seen the price of a typical PC erode. Previously, PCs got vastly more powerful without dropping from the $2000 to $3000 price range. But the next level of PC penetration and the popularity of the Internet ended this comfortable situation. PCs have fallen to, and through, the $1000 level. Which has led to a hope, a scheme, a theory, under which the size, power, and total price of an individual PC customer’s purchase can be lifted to higher, more profit-yielding levels. This theory: the personal network.

This is a term you probably have not heard. Personal networks sound a bit self-contradictory. They are not fully personal, but they cannot require the level of management networks require now.

Some of the problems that have to be solved before personal networks will work well:

User management. I don’t manage how my wife and daughter use their computers. Their Macintoshes and my PCs do not inter-operate to any extent except to share a LaserWriter, which happily operates on an AppleTalk network and on a parallel port connection to a PC, no further configuration required than to plug both connections in. We don’t share an Internet connection, and the effort of acquiring a high-speed connection and sharing it would be too high, although it is possible to do. Personal networks must accommodate the style and practice of personal computer use in order to become manageable and applicable to commonplace problems. Currently, PC users accommodate networks in order to use them. Personal computers are just getting over the inherent complexity of being completely general-purpose machines. While one can expect the first personal networks to be made in the image of existing data networks, this will have to change, otherwise personal networks will remain geekware.

Simplification of communications. Shared Internet access is the driving force behind personal networks. It is all people really care about. Everything else is in a distant second place. Personal networks and small servers must provide a solution to this problem in order to be viable. At first, the mainstream solution will rely on multiple modem connections. Then personal networks will have to transition to enabling sharing of high speed connections, as these finally become widely available.

Current market segmentation. Even if networks can be made simple, there are forces operating in the market to prevent this happening. The tide sweeping over the industry is one in which small businesses are being networked. This is a hugely profitable trend. It is driving the near and mid-term profit expectations of Microsoft, and of the server operations of PC makers. Altering the price and value composition of servers and networks, and tampering with the channel for delivering these products is considered dangerous. The computer industry will have to learn a new type of market segmentation, and accept some cannibalization of its small business market in order to open a new mass market among individuals and the smallest businesses.

If these difficulties (and other outside the scope of this missive) are overcome, we will see a world where all devices, such as printers, become print servers, and where a set-top box or “home router� also becomes a communications server and the convergence point of home computing. The personal network may have one server, say something like a multifunction printer/fax/scanner with a brain, or there may be several specialized servers. These servers will have to learn to cooperate and stay out of each others’ way, to figure out which function (DHCP, for example) is the responsibility of which server, and enable management that does become a geometrically growing burden as the number of small servers increases. In this world, the concept of a home computer will be replaced by multiple computers that truly are personal, a concept worth keeping in the front of one’s mind: PCs are more like underwear and toothbrushes than they are like the silverware.

So if this is the future, where is telephony in this picture? This is what must be considered: if the conventional small business key system looks out of place next to today’s PCs, how dislocated must it seem next to a future of pervasive networks. A fundamental task for the telephony industry will be to come up with telephony CPE that fits, exploits, extends, and simplifies this coming reality.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #23

Here we looked at the ultimate ridiculousness of the government's pursuit of Microsoft. As the software business gets more mature and as dubious technologies like DRM make regulation possible, in addition to keeping an eye out for the irony of Microsoft forging its own chains, we must continue to look out for government folly in the area of software regulation.

Newsletter #23: Mr. Pot, I’d like to introduce you to Mr. Kettle

Political matters are a touchy topic in business. All the more so in technology which has been innocent of politics for so long. Those of you reading this that come to computer telephony from the telephone operating companies may chuckle a bit at the innocence of computer people. In this light, Microsoft appears naive, and the U.S. Department of Justice involvement in the question of browsers integrated with operating systems signals the maturity of the software business.

There are two reasons to think differently: First, computer software and the Internet share few characteristics of industries, including telecommunications, where anti-trust litigation has been successful (at least by some internally consistent definition of success). Second, the whole idea of thinking Microsoft naïve and incapable just because they are dealing with a new problem is dubious. The argument I put forth here has nothing to say about whether Microsoft is good or bad. All the ill effects I document are purely the creation of the action against Microsoft. All of which is to say that we, the industry, and the high tech economy, are about to shoot ourselves in our collective foot, never mind what we think we are aiming at.

First, the fundamental difficulty of regulating the content of software products:

Software changes faster than anything else. Steel girders are still steel girders. Oil is very much the same oil that was ever pumped from the ground. Even frivolous products, like Coke and Pepsi (which have, nevertheless, attracted the attention of “trust busters�) are eternally just brown, fizzy, sugar water. Even biotechnology, which deals in the business of life, moves slower. Software moves more swiftly than any human endeavor that came before. Software concepts become obsolete before the ink is dry on laws.

In the midst of this tumultuous and rapid change, everybody in the computer industry has the same ambitions: to dominate some portion of the computing ecology and become irreplaceably embedded in desktop computing or the Internet. Calling Microsoft a gang of monopolists is no more sensible than the pot calling the kettle black. So far, except for the fact that Microsoft won against Unix, OS/2, and Mac OS, etc., I have not heard anything different about what Microsoft intended than what the losing players themselves had in their plans. Apple and IBM wanted to retain and regain, respectively, complete control over both hardware and system software. Sun, despite protestations that theirs is an “open� system, finds no significant takers for Solaris licenses, and Sun has control over hardware, software, and the CPU chips in their systems.

To illustrate the futility of the best laid and most viciously executed monopolist strategies, look at video games: Atari, Nintendo, Sega, Sony, and next Nintendo, no Sega, or someone else entirely. Each of these companies build intentionally closed systems, with closed markets for software, and surround these strongholds with platoons of nasty lawyers to sue anyone with reverse engineering on their minds. Game consoles are a nakedly monopolistic pursuit. Nevertheless, no game console maker has remained at the top of the heap for more than a few years.

Set aside the practicality of applying anti-trust laws to software and ask what if it comes to enforcement? Software is the ultimate contraband. Software has no mass, and travels at the speed of light. Better than electricity, you don’t need a billion dollar power plant to make lots of it, and you don’t need a well-regulated power grid to transmit it. Better than oil, there are no fixed wells, pipelines, or gas stations. Better than railroads, there is no track, no locomotives too heavy to move around the world easily. Encrypted, it can hide in plain sight. It has no radar signature, no DNA test can track it, no spectrometer can sniff it out. Making a desirable software product illegal to distribute in the U.S. is probably the purest folly that could be pursued by any body that makes and enforces laws.

These are all reasons to avoid trying to regulate the content of software systems for anti-trust and other reasons. What are the potential consequences of ignoring these indications?

Really Bad Outcome #1, The cancer spreads: Didn’t Apple strong-arm Power Computing, Motorola, and Umax out of the clone business? Sure. Completely intentionally, Apple drove legitimate licensees out of business. They did it to implement a strategy that does not even hold any guarantee of saving Apple. Has Microsoft done anything comparable? I don’t recall anything like it. But it is the prerogative of platform vendors to behave this way. Consequently, there are no bounds to the underlying theories that the DoJ is bringing to bear against Microsoft that prevent Apple and other companies from becoming targets. The back-up tapes of every company with an ounce of ambition are chock full of scheming to crush the competition. And if the Microsoft suit looks like a losing proposition, the government may turn to a secondary target just in order to score a win and bolster the theories under which they pursued Microsoft.

Really Bad Outcome #2, The tables turn: I recall this story starting with some company thinking “Gee, they’ll never figure out how to make a good browser!� Oops. Many critics of Microsoft succumb to believing in the insults they hurl at Microsoft. Now it’s “Oh they’ll never figure out how to play the government regulation and lobbying game.� Imagine Microsoft becoming as canny at enlisting the government’s help in maintaining a monopoly as their Seattle neighbor, Boeing. Yikes! Or a darker outcome would be that Bill Gates turns out not to be the John D. Rockefeller of the 90’s, but rather the Ted Vail of the next century. AT&T used the government to create regulatory barriers to competition and created a truly oppressive monopoly. Is this the lesson we want to teach to Microsoft?

Really Bad Outcome #3 They go nuclear: Factories move to escape direct costs and regulatory costs. Software licenses can be sold from any spot on the planet. The DoJ has not touched he cutthroat game console industry which makes feudal vassals of the software makers not because they see how quickly leadership in this business changes, but because all the game console companies are based in Japan and it would be futile to try. Because of the portability of intellectual property, software can be developed anyplace (much game hardware and software is developed in the U.S.), and sold from anyplace. If software companies change their legal location it will have only minimal impact on development projects under way, but it may have a long term negative impact on the U.S. as the center of the software world.

All this without even discussing the merits of the DoJ suit. Which indicates that if the computer industry is following telecommunications and other industries into the regulatory corral, it is out of a false sophistication, and the results will ill suit the cowboy capitalists of Silicon Valley.


Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #22

The decadence of the desktop metaphor, and the lack of a repalcement gestalt, still draws too little attention. The phone UI is an example of the cost of ignoring this deficiency. Phones continue to lack a consistent, easy, and powerful user interface, which is one big reason why all the features of switches and networks don't get accessed.

Ironically, it will be trivialities like ring-back tones and "video ringtones" that bring people to look at call state and how it is presented at the handset. Hopefully something more useful than annoying MIDI melodies will come of it.

Newsletter #22: Everything you know about user interface is wrong

Software user interfaces are about to change, a lot, and for the better. This change represents the next stage in the evolution of the desktop metaphor, and it opens the door to evolving the desktop metaphor to encompass a hypertext world where most documents will never be printed on paper.

This change centers on Dynamic HTML or DHTML. In DHTML-based user interfaces, hypertext documents form the outer layer of the user interface experience. Instead of the current limited types of interactivity, where forms and simple scripts scarcely extend the Web user interface experience beyond basic hypertext interaction, future interaction and richness of presentation will be more like what you get in Windows applications like Office. Direct manipulation, complex typography and graphics, context-sensitive commands, color, animation, and multimedia will be the hallmarks of good Web interfaces.

Stylistically, this change means a merging of user interface with the way information is presented. Technologically, this means that merging programs with hypertext is no longer limited to the awkward embedding of something obviously a program, and obviously designed by a programmer, inside of a rectangular boundary, inside something obviously not a program, and obviously not designed by a programmer.

Dynamic documents have been possible for some years. Microsoft’s Office suite long had a programmer’s interface that enables the minute manipulation of word processing documents, spreadsheets, and other documents editable through the suite of Office applications. If you had wanted to, you could have implemented an entire active, dynamic user interface by directly manipulating items in an Office document. Click on a spreadsheet cell, for example, and run any arbitrary piece of code. This code, activated by interaction with an element of a document could create, modify, or delete other elements of the same document. There has never been any a priori reason to confine the tools for manipulating documents to dialog boxes and the peripheral toolbars and menus.

There are other examples of entire user interface systems based on a display language behind which programs operate: OpenStep, the NeXT user interface environment, uses Display Postscript for purposes roughly equivalent to DHTML in an architecture that is broadly similar to DHTML-based user interface systems.

So why are Dynamic HTML-based user interfaces more interesting than these previous attempts at luring us into a world of active documents? DHTML is more interesting because HTML is successful and pervasive to an extent far beyond that of previous tries at hypertext. Further, it looks as if document formats are converging on XML, which provides a meta-level framework for document formats based on HTML-style tags. This is leading to a world where Web browsers will be universal document viewers, and where document creation applications will shake out to the number actually needed in order to provide specialized capabilities.

This huge success of a hypertext Web is driving other developments in software. Where software once came in boxes you bought at the store and was installed, software increasingly appears as needed, is updated without user intervention, and is paid for through a business model linked to presenting some type of content. Now this new species of software could be created in the image of the previous generation of software and simply use the Web as a distribution mechanism. But that is unlikely to be the end result of the influence of the Web on software. Instead, software will have to become more like Web content.

At first, it was thought that Web content would become more like desktop software, where the anarchic structure and interface of web pages would be shaped into an equivalent of the tidy, gray toolbar and dialog box world of desktop productivity software. It didn’t happen that way. Web sites kept their unique navigation tools. Users didn’t mind. None of the interactions that had migrated from toolbars into Web pages migrated back into toolbars as the Web matured. Instead the interaction on web pages has become steadily richer.

So if the Web’s HTML hypertext outer layer and the future of software are inextricably bound, where does this leave Java, and especially, where does it leave Java’s user interface packages, AWT and Swing? Java’s philosophy of cross-platform user interfaces goes along with the model of the Internet as software distribution mechanism, but ignores the importance of the Web as user interface. Which all leads to a truly strange outcome: Microsoft may well have perceived the future of Internet software more keenly than the inventors of Web browsing and the inventors of Java. (And there is some evidence that Corel, once burned by Java in creating an office suite may turn to DHTML, not Swing, for building the user interface in their next attempt.)

Take a look at the white papers on Microsoft’s next Java product: Visual J++ 6.0. You will find an interesting view into the next generation of Windows programming technology. You can find the documents, along with a beta of the product you can download, here: http://www.microsoft.com/java

Forget the Java religious wars for a bit. If Microsoft intended only to produce a travesty that would sink Java, Microsoft’s Java efforts would quickly fade from relevance. Instead Microsoft has a Java virtual machine that performs better than any other on Windows, and Visual J++ 6.0 looks to become the most powerful and productive Java development environment. Take them at their word at least as far as the objective measure of their products. Microsoft’s goals are to produce a Java that stands on par with, and in many ways leads the evolution of their entire family of development tools. The result is a Java that delivers excellent performance for interactive applications, a Java that is not weighed down by an enormous user interface package, and a Java in tune to the next wave of user interface technology.

Even if it had not been the explicit goal of Microsoft to divert Java, as a movement, from direct competition with Windows as a computing platform, it would have been impossible to do without presenting a superior alternative in the form of Java as a Windows programming language. If the Windows Foundation Classes did not offer a sublimely flexible DHTML architecture that spans client and server-based execution, and what is probably the easiest way to create Windows applications, then there would be nothing to weigh against Sun’s approach, and nothing to support Microsoft’s position that Java is a fine language, but a second-rate platform. Instead, Microsoft has created an alternative view of the future of Java that is both interesting in its claim to a fundamentally superior approach to user interface building, and compelling in that applications writing using Microsoft’s approach to Java will likely perform better in all practical situations. Add to this the fact that the “practical situations” I’m referring to will likely be built out of NT servers, Windows 98 desktops, Windows CE set-top boxes and game consoles, and, possibly, Macintoshes that run a Microsoft-influenced Java VM, (N.B. the lack of Java OS-based systems in this picture) and you see that Microsoft’s position on how Java should be done is not just a defense.

The Department of Justice’s lawsuit against Microsoft has put the Java battle into the popular arena. There are large elements of strategy in this directed at preventing Sun from building Java into a rival platform. But Microsoft’s response stands on its own merits. If it didn’t, it wouldn’t have a significant impact on whether Java is viewed as a platform defined by Sun. Jurgen Schrempp, chairman of Daimler Benz, in talking about the merger with Chrysler, said “It all comes down to product.” Which says a great deal about the gulf between the perception of how business is conducted, and the reality.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #21

Well, now! Here we have strong evidence of excellent prognostication. All this has come to pass, and we even find Linux in Motorola mobile phones.

Newsletter #21: Wham! Another surprising result of the Internet culture

No, not George Michael's old band. I mean another bolt out of the blue is in the process of striking the software industry. This time it is open source software.

It used to be the case, back in the late seventies and early eighties, that every university that used UNIX had a source license. That meant anyone could make alterations to suit needs like real-time data collection, add network protocols, or even port the whole operating system to another CPU architecture. Access to OS source code was a necessity when adding a device driver meant re-linking, if not recompiling, the OS. Many sites that did not pay for source licenses "borrowed" source access from nearby universities. Other OSs widely used in universities were written there, like ITS. Commercial OSs like VMS and IBM mainframe OSs were less popular at universities, since you couldn't rummage around inside them.

During this period in the history of computers, universities played a leading role in seeding the industry with innovations. Sun workstations, Lisp Machines, and the Connection Machine, among many others, were spin-offs of university-based projects. UNIX was also the workhorse of university life. UNIX processed documents, using emacs or vi editors, and the troff formatting language. This was crude by comparison with today's visual tools for document editing, but it worked.

As commercial software, from Lotus and Microsoft, began to make inroads in universities, the old UNIX tools were pushed aside, and rightly so. Users didn't care about source access. The Macintosh and even Windows 3.1 were easier to use for document creation than UNIX and emacs and troff. PCs replaced terminals and UNIX workstations outside of labs, and, eventually, in many of the labs. Commercial software and PC hardware got cheap, and the process accelerated. In the end, almost without anyone raising a voice of protest, the computing infrastructure of universities, which had been home-brewed, was replaced by store bought stuff.

So? Well it did have a bad result for universities. They became largely irrelevant to commercial software development. Indeed, to people like me who were observers of the dynamic from close-up, Lotus's tendency to hire academics to lead important development projects was, in part, their downfall. Software had moved on to an era where blending entrepreneurialism and a feel for the market with the ability to architect software was more important than what the most fashionable professor from MIT or Stanford was thinking. Was the end of home-brew computing a cause, or a symptom? I can't say for sure, but it all happened suspiciously around the same time. And the challenge remains: how can universities maintain relevance, especially at the postgraduate level, to the software industry when most of the commercially relevant cutting edge stuff is in commercial ventures? Has commercially relevant software become so complex and expensive to develop that universities can no longer undertake projects of that scope?

There may be an answer to that challenge in open source software. Open source software is a mechanism where software is either created in the public domain, or put in the public domain. (Or some variation of public domain. This article is not meant as a full explanation of, much less a detailed comparison of the licensing arrangements for open source software, the intent of which are to maintain the openness of the source without enabling a commercial vendor to “hijack� the product by coming up with a more-attractive variant, among other objectives of such licenses.) This type of software is typically available in source form for free downloading from the Internet. There are also, typically, one or more commercial distributions of the software as well. These commercial distributions are meant to provide the comfort-level and convenience of buying such software in a box, with a manual, and a tech support phone number, plus additional services. VARs have also picked up the gauntlet on open source software by providing systems integration and support services for systems using open source software. Three examples of open source software are Linux, a flavor of UNIX, the Netscape browser, and the Apache HTTP server.

All of this must sound like a really weird idea to telephony people who spend more time than most figuring out ways to erect patent protection barriers around their products, much less ever consider releasing source code for free on the Internet. What makes open source software so interesting is that the Internet seems to have made it possible. It has been proven that dozens to hundreds of people can collaborate on a software system, often without detailed central control over who is doing what, and useful results are produced. One effect is that universities now have the ability to teach students about operating systems by having them use and modify an open source system like Linux. But this is more than a free and limitless source of specimens for dissection in the university software lab. Linux is commercially relevant, and is making serious market headway in server systems, especially Web servers, so students can experiment in making improvements to something that is used in the real world. The choice these days seems to be NT, Solaris, or Linux. For simple, budget-constrained Web sites, the answer is surprisingly often Linux.

This is possible because Web serving is not a trip to the moon. A Web server serves up pieces of hypertext documents when asked, which makes it a fancy type of file server. Some basic performance requirements need to be met, but most servers communicate with relatively low-speed connections to the Internet, or are not engaged in applications where absolute performance is required. If you are not shopping for the fastest server, you are probably not going to stretch the capabilities of the OS. If you are not putting hundreds of thousands of dollars into developing an interactive system that uses OS features for linking Web pages to software components, you will probably be happy with a Linux-based solution running Apache. That is to say, in almost all the cases where a hand-me-down or brand-X server can be used for simple web serving, so can Linux.

But what about everything else? What about user interface systems, which can take the efforts of dozens of coders years to create? What about the thousands of Win32 calls that form the programmer's interface to Windows? Surely Linux can't duplicate them. No, Linux can't. But those thousands of system calls and calls into other subsystems can be divided into three groups: Grotty old calls that need to be there so programmers that can't be bothered to update their software don't complain; calls that do really useful things; and calls that never should have seen the light of day. Linux can do with fewer of the useless calls, in part because it is a younger system, and in part because there is no product manager at Linux, Inc. (there being no Linux, Inc.) that can be cowed into keeping old baggage around, nor stampeded into the latest API fad. Aesthetically, the difference between Linux and NT (as it was between UNIX and VMS) is exemplified in the way Linux takes a more uniform view of the world, where all drives on the system fit into one file hierarchy, where i/o devices and files have a uniform interface that makes it easy to “pour� data from one container into another.

What will happen as systems mature and are transformed by future technologies? This is where the comparison gets really interesting. A proprietary system, so called because its creator owns all rights to it, should be flexible, and responsive to the market. This is illustrated by how quickly Microsoft climbed to a parity position and then some relative to Netscape, and then went on to begin migrating the whole world of Windows user interface to a Web paradigm.

As it grows, open source software faces the following challenges:

Can open source software compete? An interesting question. Some software, like Netscape's browser, have a marketing department and engineers to steer the product in response to the market and implement high-priority features. Linux depends on a community effort, and yet it is doing quite well competing against other operating systems.

Can open source software find a channel? The open source software that started out commercial already has a channel, and will continue to for supported releases. The cost of some types of software has fallen so low, that supported releases of open source software will not fall below the threshold for retailers to carry it. But open source software was enabled by the Internet, and the Internet itself has provided an adequate channel. If systems integrators take up using open source software in serious volume, that will show conclusively that open source software can thrive without traditional marketing programs and channels.

Can open source software innovate? Can open source software make big technological transitions? Windows, for all the criticisms leveled against it, especially by open source partisans, has incorporated innovations at a rapid pace: messaging, telephony, and now Internet tools. Can Linux grow a Web-oriented user interface? Can an unmanaged collection of coders make the right design decisions? This may become the test of renewed relevance of academic research in software.

There has been a lot of chest thumping among open source enthusiasts about open source software being the cure for the Microsoft monoculture in PC OSs. I can see great potential for government sponsored communitarianism in cajoling commercial software developers to donate resources for Linux and other open source development. Al Gore could do a lot worse than to tout open source software in his efforts to bring the Internet to schools and libraries. Demonstration projects in which schools are automated with donated PCs and Linux servers, and government-funded development of Internet-based training courses in Linux internals are two examples of simple, inexpensive things that are helpful and that do not inject the government into high tech economics or policy. Done right, such government leadership is an expression of the positive role of government in technology, and rather less dubious than treating Windows 98 as the moral equivalent of passing out smokes in the schoolyard.

Is open source software a threat to Microsoft? It is not a realistic threat to Microsoft's business in any way that would change the value of Microsoft shares. The only thing that could change that would be if systems integrators took up Linux enthusiastically in creating large-scale server environments. The odds of this happening quickly are low, since many systems integrators simply do not have the right attitude to succeed with open source software and be accepted by the open source software community. But much to Microsoft's credit, they are able to feel the heat from even small challenges. So far, they have no reason to respond, since Netscape's move to open source was only a last ditch defense to ensure survivability among a community that would respond with fierce loyalty to an open source browser. But if the advantages of open source software accumulate, one thing that open source enthusiasts should never count out is an open source response from Microsoft.

And in telephony? Let's say you are a second-tier voice processing card maker. Let's say you are bummed-out about the fact that TAPI has made nary a dent in Dialogic's dominance in voice processing APIs and appears set to make their S.100 suite the next generation standard in telephony server software. Let's say you have your own APIs and middleware, equivalent in function to many aspects of TAPI and S.100. Let's say you make little or no money licensing this software. What would you stand to lose by making your telephony middleware the open source telephony subsystem for Linux? And what might you stand to gain?


Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #20

In this and the previous newsletter I prescribe how to get CPE telephony products out of the dark ages. This advice is no less heed-able today as when I first gave it. So there you have the benefits of glacial progress: stuff you thought of years ago can be restated as completely up-to-date advice. The bursting of the Internet bubble and the general weariness of the CPE telephony business ensure that outside of a few small corners of the industry, it won't be galloping away from this position any time soon, either.

Newsletter #20: Envisioning the Telephony Future, the Enterprise Edition

In newsletter #19 we visualized IP telephony for the small business customer. Now we’ll try on the large size enterprise suit. I had promised to discuss moving IP telephony products through the small business channel, but we’ll take a side trip here, for the reason that IP telephony in the enterprise has different and in many ways simpler requirements for formulating the product to the correct channel.

This is very unlike small business: the large enterprise has money and people to apply to the matter of why to do IP telephony. The large enterprise has proven needs, and an institutional memory. The large enterprise, if well run, measures what it does, and so can understand benefits that look alienatingly theoretical to the small businessman.

In many ways, the large enterprise is an ideal place to start selling IP telephony.

But there is a catch: Where visualizing the problems with selling to small business clearly illustrates the problem with a product that is in many instances unfinished, there is a temptation to use the enterprise market as a gradual on-ramp to a complete product vision. This is a false hope. The properties that make the enterprise market more receptive to the technical sell of IP telephony make it correspondingly more demanding that the technology deliver measurable benefits. On top of this, the enterprise customer demands sophistication that many IP telephony systems lack at this time.

Where the small business customer has simple, but demanding needs, no time for chit-chat, and less patience for products that are not full replacements for conventional technology, the enterprise has complex needs that go beyond what the conventional technology can deliver. Each enterprise comes into these needs through its business functions, but we can obtain a good approximation of what the enterprise customer wants by conducting a thought experiment into what a telephone system is.

A telephone system transports audio from station to station, it arranges and manages the connections that do this, and it provides other services, such as call answering, conferencing, logging, accounting, directory services, and a management console. A large part of the impetus toward IP telephony comes from the fact that tools for managing networks have become more evolved in data networks than they are in voice networks, and the unification of management tools centers on standards bred in data networks. With the increasing multimedia content of the Internet, objections to putting voice on data networks have become mostly obsolete. So the functions of a telephone system can be implemented at least as well on a data network as they can on a voice network.

As with our small business customer, you still can’t expect an enterprise customer to forgive lousy station sets, kludgy interfaces, and difficult system configuration. These are the telltales of an incomplete product, and if the basics are not complete, the customer has every reason to expect the nifty stuff to be substandard as well. Risking antagonizing his users by installing dicey new technology is also something the enterprise customer has little or no incentive to hazard.

On the plus side of the ledger, there are things the enterprise customer pays attention to that the small business customer does not. But these are not the features typically touted in the collateral for IP telephony products. One wire to the desktop has no attraction for a company that is plenty well wired. Instead, look inside the company’s IT organization. What does a modern IT department do? It creates solutions based partly on technology they buy, and partly on technology they build. These solutions are distributed systems, and communicate via DCOM, CORBA, or Java RMI technologies. The bought technology must provide interfaces compatible with these distributed software interfaces, and must provide functions that are valuable in creating solutions. If you want your IP telephony product to be part of this picture, it must integrate with solutions created by a company’s IT department.

So far, this picture is much simplified. Reality is much grittier. Lots of enterprises are nowhere near the level of sophistication described above. Their needs are much more like the small business customer, who must be shown a simple benefit that goes straight to the bottom line. But we will continue to assume for our purposes that enterprise customers can be quite sophisticated so as to illustrate how such customers are different, and represent a different opportunity. One simplification made by omission here is that the classic problem of selling IP telephony, the split between data and voice, no longer exists. This is a simplification, but it is also correct for the type of customer we are painting a picture of: one with enough IT savvy to make use of the distributed software technologies he finds in other products.

What would an IT department want to do with a phone system? One place we can look for ideas is the penetration of modern enterprise planning and fiscal management software from SAP, PeopleSoft, Baan, and others. These software systems enable the enterprise to link, via databases, all the important activity in the firm. When orders are taken, new materials are ordered, transportation arranged, etc. This revolution in enterprise fiscal management and planning turns what had been lumpy, un-integrated processes into a continuous, always-up-to-date, whole. This improves efficiency because the management’s view of corporate activity is instantaneous. The CEO doesn’t have to wait until the end of the quarter to discover that sales turned good or bad. He doesn’t even have to rely on projections, or “feel.” He’s got the numbers from the last day, or last hour if he chooses.

Linking telephone activity into such a system would enable lead tracking, funnel reports, and information about when, exactly, a customer ordered a product to be used in the planning and management processes. Some of this is as simple as integrating call accounting so that up-to-the-minute data on telecom costs flows into the system, other uses are more imaginative, such as measuring the sales response to out-calls and fax broadcasts. It is in this type of application of IT to business processes that the value of IP telephony will be realized in the enterprise. This is potentially very important. If enterprise fiscal management software enables an enterprise to track business processes from the moment a product is ordered to the time it ships, linking telecom information into such a system could make it not only perfectly responsive to requirements for materials, etc., but also anticipative of coming requirements. Different patterns of telecom activity can be linked to sales results, and used to detect problems with selling a product, channel stuffing to cover for problems, as well as anticipating a better than expected response and ramping up manufacturing to meet it.

What about scalability? Isn’t that an important part of selling something to the enterprise customer? Yes, but this is one place where enterprise requirements are not absolute. Departmental systems that meet requirements for integration with enterprise IT initiatives are valid products in the enterprise space. Scaling down is important, since the life cycle of an enterprise purchase is likely to include a technology demonstration phase where a departmental deployment is made. If this requires all of Hannibal’s elephants to accomplish, it won’t look good. And the key to scaling up is modularity. IP telephony does not have some of the physical limitations of distributed switched circuit systems, and in practical terms, integration with a corporate data network make being distributed much easier. As long as the system is designed to work as a manageable whole even if it is divided among several servers, you have a good scalability story until you start to tun into large scale network design issues. At that point there is no substitute for having built and tested multiple large configurations.

Since the requirements, like integration into IT strategies and initiatives, and advantages, like the ease with which IP telephony blends in with network management, closely resemble the positive characteristics of other enterprise IT systems, this broadly hints at how IP telephony should be sold into the enterprise: In the small business example of the previous newsletter we ran into the problem of the reseller channel. How do you get the interconnects and small business VARs to understand IP telephony. By and large, you don’t, and you must be able to sell IP telephony as if it were conventional telephony. But in the enterprise, you have a natural ally in the systems integrator. If your system is complete, and if your system is responsive to the IT activities systems integrators support, then you have a path to building a channel for IP telephony in the enterprise market that understands IP telephony and understands its advantages.

To sum up:

The enterprise has different requirements, but can’t be considered a shortcut to marketability for incomplete products.

The enterprise is going through changes in the way software is used to control the fisc and business processes. This is what IP telephony and telephony servers have to link in to in order to get the attention of the enterprise customer.

Not all enterprises are progressive and managed using modern tools. Target the ones that are.

The systems integrators are your natural ally, but you have to have the end user’s requirements in place before attempting to build this channel.

Go get ‘em!

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #19

Here again we provided a healthy slap on the fanny to a CPE telephony industry that not only failed to heed the call, remains as much asleep today as it was then. Maybe when I go shopping for that Cisco or 3Com IP PBX I will see some inkling of these developments.

Telirati Newsletter #19: Envisioning the Telephony Future, the Small Business Edition

I have, in previous newsletters, commended the virtues of visualization, of picturing an outcome in your mind. In telephony, sitting down and doing some serious visualization is in order, because the industry seems blocked from advancing by some unknown force. Visualization makes such impediments visible.

Visualization is also useful because it makes one see a situation in sequence, and from a particular point of view. In this case, we will start from the point of view of a small business proprietor buying communications capabilities for his business. From his point of view: “I gotta buy a phone system.”

Get inside his mind, shove your way past the sports scores and the concern over his daughter’s preferences in boyfriends, and you’ll find a small slice of mind share available to deal with how his business communicates.

Does he see the world the way we do? No: Does he see that computers and telephony are converging? No. He mainly sees his phone bill. If it’s too high, he knocks heads in the office and figures out why. If we are lucky, his phone bill is high for some specific reason, like he has a customer or supplier in some faraway place with high telecom costs. Does he fax a lot? Some, but not enough to bother with a fax server. Does he have a call center? No. At best he has a salesman motivated enough to be on the phone all day. Does he have any clearly defined application of telephony in mind? No. He just needs to call people and have people call him.

OK, we’ll do a focus group. We’ll ask the customer! Customer, what do you want? “I want youse not to ask me so damn many questions about phones, got it?” Mostly, the customer wants answers, and not to be troubled. This is an important difference between selling a product to Fortune 500 corporation and selling a product to your local small businessman. The fortune 500 corporation has a person whose job it is to make an intelligent and informed decision. This person finds satisfaction in complex comparisons. You can build a great relationship with people buying your products in large companies. They will practically design your product for you. Not so the small businessman.

Look in your local yellow pages, especially if you live in a small town: Insurance agents, real estate agents, loam and gravel, septic tank pumping, propane dealers, driveway pavers, exterminators… This is not a think tank waiting to be discovered. All of your neighbors who are actually interested in the technology behind your product work for big high tech firms, or they make incredibly expensive handcrafted flutes, and take 20 orders per year on an ancient answering machine. This is not a market that will design your product for you.

Back to visualization, back to inhabiting your customer’s mind, which, if not introspective, is at least unguarded from this type of examination. So far we find a mind that responds well to the way telephone systems used to be designed: over-engineered, bullet-proof, trailing-edge, mysterious. Any un-PBX that wants to make headway will literally have to respond to a competitive analysis versus a conventional PBX where all the strengths of the conventional PBX are prioritized at the top of the list. Unfair! But required. Look at the requirement this way: Unless the system can be sold without telling the customer it is an un-PBX, it is market-limited.

What about the way phone systems are sold? How does this mind respond to the telephony channel? Mostly, he’s annoyed: He doesn’t want to understand his telephones, so the un-PBX people have no chance with him yet, but he feels taken advantage of by the mystery and obscurity of telephone systems. Where to look for an opening? Computers, natch.

If the customer has no computers, or an old hodgepodge of systems with a NetWare server that has not been upgraded in five years, he is going to be a tough sell. Unless your system really truly can sneak in without being outed as un-PBX, skip this prospect. If, on the other hand, the customer has a server, there is hope. If, furthermore, the customer is connected to the Internet, there is quite a lot of hope. Now why, having painted a portrait of the customer as troglodyte, do I now posit he has a server on the Internet? Everybody has a phone system, but what is the market penetration of Internet connections? Am I mad? Am I suggesting an approach that will drive a telephone system, that should be a mass-market item, irretrievably into a niche?

The answer is in the fact that now the Internet is more important than telephones. The Internet has a deep hold on people not normally motivated to hassle with computers. I could see years ago that something was up: people who could not be bothered to upgrade their systems or applications software, were voluntarily wrestling with add-in TCP/IP software for Windows 3.1 and the early primitive Web browsers. The same kind of people that don’t care about un-PBXs, and who don’t want to be bothered about whether their phones use the same wires as their network, care deeply about the Internet. I recently read an article about the high market valuation Yahoo has, in which this line stuck in my head: “The Internet is like television in the early fifties.”

Think about that. That’s an idea the guy with the Sans-A-Belt slacks can get his head around. It’s like the early days of TV. Who would not see the importance of that? TV defined consumer electronics for two generations. Losing our TV manufacturing was a national crisis worthy of intervention by the federal government. The Internet will be way bigger than TV. Next to the importance of the Internet, marketing messages about the cost of wiring to desktops and the benefits of having your phone system run NT pale to insignificance. You will not move the customer off the mark without hitching your message to the Internet.

Hitch your message, and your product, to the Internet. Visualize how that might look: The user sees he telephony server as a Web server. “Installation” of client software consists of giving users a URL to visit with a Web browser. There they find a Web site they log into to place calls, use directory services inked to the Internet, and get their voice mail. Logging in to this site from home produces the effect of turning the user’s home phone into an off-premises extension, or, if he has no second line, using an H.323 client for the OPX.

Visualize what’s here and what isn’t here: There isn’t any client software or hardware installation. This implies the system has to have handsets that can, at least, go off hook in speakerphone mode on command from the un-PBX, so that on-screen commands don’t have to be coordinated with lifting a receiver on cue. This is an off the cuff example, so OPXs might not be the killer app, but this example does integrate the Internet in a way almost all users can grasp and use. Is there anything us telephony geeks like about un-PBXs visible to the user? No. Everything we think is cool about using IP for telephony is boring to users of phone systems. Importantly, this system could be installed in an office without a computer network, and the users would be unaware they were using an un-PBX.

These are the conclusions we can draw from this visualization session:

Follow the Internet. Small businesses with servers and, at least, ISDN or fractional T-1 Internet connections have already paved the way for computer telephony by modernizing the computers and network they have. The Internet is more important than telephones.

Don’t sell integration. Put the customer’s phone system in a separate server. Provision it for high reliability, with an UPS, remote rebooting, etc. When he’s happy with it, sell more integration, with e-mail, system management, and applications. Be patient. The follow-up sales opportunities will come.

Sell comfort. Make the system simple, reliable, un-intrusive, and undemanding. Brand it. If your brand isn’t strong, find one that is and swallow your pride. Or spend the money it takes to identify your brand with a comfortable purchase.

The use sees the phone, not the box. So far, un-PBXs have lame handset support, with kludgy PC-based interfaces or analog handsets that lack too many features compared with proprietary handsets on conventional PBXs. The un-PBX needs an un-handset.

In the next newsletter we will tackle how to package and move a product like the one we visualized in this newsletter through the channel.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #18

Now here is a truly interesting bit of Windows and Java history: Visual J++. Visual J++ could have been a lot of things. It could have been .NET delivered two years earlier. It could have made Java relevant to Windows desktop applications. It could have made .NET more widely used by giving it a head start.

Instead, Sun succeeded is suing Visual J++ out of existence. This delayed the use of the Visual J++ class libraries - which are where most .NET ideas were first hatched. This prompted Microsoft to redouble their efforts and, as a result, .NET is much more than the Visual J++ class libraries. But now Microsoft's impact on server applications and an active and interconnected set of Internet-distributed systems has been blunted by Linux.

What did Sun get out of this? Not much.

Newsletter #18 Can Microsoft redefine Java?

In previous newsletters I have mentioned Java because it an important development:

Java is the most interesting challenge to Microsoft’s hegemony since the Macintosh.

Sun has, so far, not committed as many errors as Apple did when Apple frittered away the Mac’s advantages.

Java is likely to be important in creating and delivering a lot of software in the future.

Java can solve some sticky multi-platform problems, and it has some advantages in creating computer telephony systems.

Java is a delight to code in. It is C++ designed by an individual, not a standards committee.

So the question of whether Microsoft can redefine Java is a key question: If Microsoft cannot assimilate Java into the set of tools used by Windows programmers to create Windows software, the only other choice is to stand astride progress in Java - a much more brittle and hard to defend position. But doing this is no walk in the park. What does it mean to assimilate Java? How to overcome the inevitable negative response of people who find the simplicity of Java a refreshing break from the mounting complexity of COM and the Windows APIs? How do you, in practical terms, hijack the product direction of someone else’s set of APIs? There have been several attempts to provide alternative or cross-platform object libraries for Windows programming and all failed. Apple’s developer tools group at one point thought they could take their development tools cross-platform and become the development environment of choice for Windows – no need to ask what they were smoking. Why should Microsoft expect to succeed in steering some other direction for Java than that chosen by Sun?

There are other reason to think it unlikely that Microsoft could succeed in creating an alternative development path for Java. The class hierarchies for Java were created to be part and parcel of Java. There is no non-object-oriented interface to anything in Java, so there seems to be no practical way to create alternatives to what Sun, Java’s creator, sees fit to be the complement of classes underlying Java software. On top of this, there is some distance between the concepts of multi-threading in Java and Windows, among several other architectural divergences. The stage was set for any attempt at diverting Java onto some other track to turn into a first rate train wreck. On top of all this there is a need for Microsoft to make their class libraries simpler and more productive.

Surprise! Microsoft may pull it off. The Windows Foundation Classes and Visual Java 6.0 are an astoundingly credible attempt at what seemed like a doomed mission. Not only is it very possible that Microsoft will succeed in creating a branch of Java oriented toward Windows, Microsoft may well have come up with solutions to a number of other problems as well. Like the serpent uttering the promise “et eritis sicut dii, scientes bonum et malum� Microsoft offers to expand the minds of Java programmers in interesting and productive new directions, among which temptations:

A clear shot at producing applications in Java that will look good, perform well, and have all the capabilities expected of modern desktop productivity applications. Programmers who use Microsoft’s Java will not end up like Corel, Oracle, and Lotus, with Java-based products that are meritorious but not complete or competitive.

An answer to the problem: Visual C++ is too complex and Visual Basic is too wimpy. Java is now a full member of Microsoft’s language pantheon. And it has important characteristics, like pseudocode, JIT compilers, garbage collection, and a relative lack of legacy baggage, that make it an ideal choice for lots of developers now using VC++ and VB.

The first manifestation of Microsoft’s ability to simplify programming with a better class hierarchy. The Windows Foundation Classes make Windows programming easier than the MFCs or AFCs. Programmers willing to accept Microsoft’s vision for Java get these goodies first.

The best tool for exploring new user interface models. The WFCs are the best tool now available for realizing the future of user interfaces. By enabling the manipulation of DHTML, the WFCs provide the first object model, the first real architectural skeleton, for the creation of applications with a Web user interface. The importance of this cannot be understated. It is well and good to say, as I have in these newsletters, that the desktop metaphor is being replaced by Web-oriented user interface philosophy. But without an underlying software architecture for programs that work by manipulating components of hypertext documents, it all remains speculation and experimentation that is destined to be discarded. Microsoft has now shown the way.

In addition to what VJ and the WFCs have accomplished already, they indicate some interesting future directions: imagine applications delivered from servers that run on any Windows computer, be it a PC, or a CE machine with one of a handful of RISC CPUs. Write once, run on Windows may sound facetious, but right now, where else do you plan on running your software? It is quite likely that the future will find Windows everywhere – in your car, in your stereo, your TV, your wallet, your telephone handset – are you IP telephony people listening? Imagine an IP telephony handset that really is plug and play, which really requires no setup and administration to deliver a rich applications environment. Microsoft may end up not only assimilating Java, but relying on it to implement the only practical way to deliver software to networked Windows CE platforms, based on a variety of microprocessors, scattered throughout our consumer products.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #17

While some of the past I commented on recedes quickly - like the memory of Digital Equipment Corporation - other things, like the matter of when IP telephony will expose some function or value directly to the user, remain fresh. This is of course another illustration of the glacial speed of telephony, especially wireline CPE telephony. My company's next PBX will probably be an IP PBX, but we will still qualify as an "early adopter," and, when asked, we will be hard pressed to answer what specific benefit we derive from owning an IP PBX.

Newsletter #17: How to Lie With IP Telephony Statistics

In earlier newsletters I discussed some of the technical challenges to IP telephony, particularly IP telephony to the desktop. But when you look at IP telephony in the real world, many of the challenges do not come from the technical aspects of implementing IP telephony or formulating IP telephony products. Instead, the economics of IP telephony and the process of creating IP telephony companies often means more to the success, and more often the failure, of IP telephony ventures. Treating these aspects of IP telephony in light of their relatively greater importance also brings into relief the technical aspects that can, in fact, mean the difference between success and failure.

One of my favorite books on statistics is called “How to Lie With Statistics, “ a book that humorously (as humorously as math gets) looks at ways to distort reality with numbers, charts, and graphs. Let us examine the reality distortion field surrounding IP telephony.

Lie #1: IP telephony is cheaper to develop.

Look along the roadside at the twisted wreckage of un-PBX startups. Their mangled business plans are a gruesome sight. Try as you might to avert your eyes, the venture capital oozing onto the pavement only draws your attention. What was in those business plans that caused the drunken swerve into the ditch? What was the fatal flaw? It was the proposition that IP telephony, or LAN telephony of any type, represents a uniquely efficient opportunity to break into the PBX business. When telephony is reduced to software, software people are tempted to propose, and investors are tempted to buy the proposition, that a telephony startup is no more costly than a software startup. Temperance!

In fact, IP LAN telephony is no less costly to develop than conventional telephony. And here is an argument that it is more costly: If you set out to develop a PBX for small and middle-sized businesses, you would find vendors of ICs for your project ready with application notes to assist your design. You would find experienced consultants who have designed telephone switches. You would find manufacturers in Asia experienced in building cheap telephone sets. You would find the fundamental design choices, such as whether to have pure digital or digital/analog hybrid station sets, to be well-worn paths. You would find among your competitors dozens of examples of how to configure and manage a telephone switch more or less well. You would find correspondingly as many examples of how to document those systems and train operators. If you are building an un-PBX, you are bereft of all this experience and communal expertise.

When you add it up, the startup capital requirements of an un-PBX company look like this:
The cost of developing a conventional PBX (Including the handsets!), plus…
The cost of unexpected technical challenges associated with a novel approach, plus…
The cost of developing new, novel, yet valuable features that take advantage of IP telephony, plus…
The cost of breaking in to an established market, plus…
The cost of convincing customers your novel approach doesn’t add risk to their purchase decision.

So why invest in Un-PBX companies? Why pull the wrecks from the ditch and put them back on the road? The reason is that, one day, all PBXs will be un-PBXs. Not only that, but a large number of second-tier PBX makers will not make it past that change, and the slice of the market they own will be up for grabs. Don’t ask me when that will happen. The reason we are discussing the wreckage in the un-PBX sector is that a number of venture capitalists who thought they knew were wrong. I’m no smarter. But I can say that if you run an un-PBX company as if it were a conventional PBX company, you stand a better chance of living to see that day in a condition that lets you take advantage of it.

Lie #2: IP telephony long distance networks are cheaper to build and operate.

People are building and operating IP-based long distance networks. This is an important development. It adds experience to an industry that is adolescent at this point. It paves the way for the convergence of communications networks. And it further establishes IP as the basis of advanced telephony services. You won’t however, read about these prosaic virtues in press releases about IP-based long distance networks. You will read they have cheaper rates than the established large players do in long distance.

Well hooray! IP telephony has bottom line impact! Wait, wait, maybe not… There are some other factors to look at. New telephone networks are cheaper to build and operate than old telephone networks, no matter the underlying technology. The fiber is cheaper and carries more calls. The transmission equipment is cheaper and has a higher capacity. The computers, software, and other IT infrastructure for operating a telephone network are vastly cheaper. All the technology, hardware, and software of a brand new long distance network can be operated by a far smaller staff than it takes to operate an older network. Anyone who has experience building and operating a long distance network, and many IP long distance companies are an encore performance, carries in his head dozens of ideas on how to do everything better faster cheaper, few of which have anything to do with IP.

So why are they building IP telephony networks? Why take the risk? The answer lies in the maxim that if you use the same recipe, you get the same bread. The only chance these new long distance networks have to be anything other than another cog in the economic machine that will continue to grind down profit margin in long distance is to be ready to offer telephony convergence. And to properly take advantage of telephony convergence, the un-PBXs have to be ready to supply the converged customer premises systems that bring the advantages of IP telephony to the users’ desktops. Unlike the un-PBX startups, however, the IP long distance startups have their intermediate strategy in place: offer cheaper rates, build out as quickly and efficiently as possible, and keep an ear to the ground for the change that surely is coming.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #16

Remember Clipper? Thankfully, in the intervening years strong cryptography has come so far out of the toothpaste tuble it is unlikely ever to return. But in the era of "Patriot" Acts the general warning to be on guard for similar attempts at expanding the surveillance state are valid as ever.

Newsletter #16: Bit crime

E-commerce and IP telephony require data security that has to withstand challenges that range from a curious colleague in the next cubicle to corrupt foreign government officials out to commit multi-million dollar larceny. In other words, security needs to be secure globally, and cannot rely on the good graces of the laws of developed nations.

Now imagine being able to identify and access any telephone conversation from any past time period, between any two parties, at the touch of a button. You would know with complete certainty who the parties were, when they spoke, from which locations, etc. Sound outrageous? This would be the consequence of requiring key recovery in secure communications systems.

You may have thought that key recovery only meant that law enforcement agencies could access encrypted communications when authorized to do so. That is only part of the story. In order to use key recovery, you have to be able to identify which communications can be recovered with the escrowed key. That means everything you ever said or otherwise transmitted over a system employing key recovery can be traced back to you. This record is absolutely reliable, eternal, and easy to manage.

In fact, key recovery makes the task of managing and categorizing intercepted communications much easier than it is today when unencrypted traffic is recorded. Speaker identification technology needs to be applied to figure out who is talking (or the people recording the conversation must know the identity of the speakers). Each communications system has its own way of identifying the terminal ends of a call. Some PBXs do not identify inside stations at all. Key recovery would sort this tangle out in a tidy, system-independent way. And the trace-ability of key recovery systems still works even if users take the trouble to encrypt their message using a truly secure encryption system.

To a secret policeman, this is better than tattooing a bar code on everyone’s forehead.

Americans tend to believe that we washed enemies lists and the use of law enforcement to exact revenge on political opponents out of our system with Watergate. All that nasty stuff happens only in other places. Our politics is as sanitary as Disneyland.

Unfortunately, we have more the illusion of clean government than we have it in fact. You don’t need a massive conspiracy to do you in, just a corrupt cop, judge, IRS agent, or politician. There are plenty of them. The expanding complexity and increasing penalties in regulatory law mean there are new flavors of cops, with new opportunities for corruption, being invented every year. On top of all that, treaty obligations can open your communication to inspection by foreign law enforcement officials who may be under corrupt influences entirely outside the power of our own laws to curb. Mexico, which is now tied in to all kinds of economic, environmental, and drug interdiction laws with the U.S. has been cited by our Department of State as a cleptocracy second only to Nigeria. Making surveillance orders of magnitude easier makes abuse of power easier by the same amount.

The other side of the coin is that strong cryptography is one more tool legitimate business can use to protect their operations in such hostile environments. The expansion of legitimate economic activity is among the best hopes for reforming corrupt economic environments. Key recovery systems give the nativists and isolationists one more way to scare businesses away from dealing in difficult business climates.

Making strong encryption illegal poses another threat: it expands the realm of thought crime. It makes the possession of bits illegal. This has at least two detrimental effects. First, it exposes professionals with a legitimate need to protect systems from hacking to denunciation and arrest. It in effect lumps these people in with bomb makers and other terrorists. Just because you and I have grown up with the ethos that bits are harmless, and that computing represents a uniquely safe environment in which to build things, watch them fail, and learn from those failures, does not mean we won’t end up in the same public relations pit with the other “mad scientists” of popular culture.

Don’t believe it? This was recently said, on the record, at a public meeting, by the U.S. Attorney General: "There are now new criminals out there that don't have guns. They have computers and many have other weapons of mass destruction." But don’t think that the U.S. Attorney General is stupid enough to think that computers are as dangerous as nerve gas. She is instead incompetent and dissembling enough to lump computers in with nerve gas in order to cover her ass. She can, by demonizing computers, blunder about heavy-handedly. People will forgive you a lot of collateral damage if you are going after a crazy man with anthrax. And if you can make computers seem just as dangerous you can get away with shooting a few innocent bystanders in this domain as well. This kind of ass coverage is politically neutral and is as likely to be practiced by the next administration as this one.

Am I for going easy on malicious hackers? Heck no! I’ve had my cell phone cloned and it’s damn annoying. Cell phone cloning operations in New York City are organized criminal enterprises with a strong profit motive that feeds those profits into other organized crime activities. These are bad people that will kill if they feel the need. It is fully appropriate for the FBI to bust these guys with all the same precautions they would take with drug gangs. But larding the law with all manner of software crimes makes it ever harder for law enforcement officials to distinguish between the evil-doers and merely criminally stupid kids with a modem and time on their hands. By blurring these lines, and applying law in quantity and not with quality, we won’t be ready to distinguish between a really malicious attack that might bring down an e-commerce site with an economic impact in the hundreds of thousands of dollars and your run of the mill Web site defacing artist.

Strong cryptography, reliable digital signatures, and respect for privacy will do a lot more to secure e-commerce and the new IP-based telephony systems than new laws. Having these systems in the hands of citizens will make it more difficult for the corrupt and criminal to harm us. Key recovery, on the other hand, opens new avenues for abuse.

Can law enforcement continue to be effective in a future where people have, in effect, more privacy than in the past? The answer lies in the fact that while strong cryptography might close some avenues of evidence collection, technology produces countervailing improvements in other ways to catch crooks. Listening devices get better and smaller. Cameras are easier than ever to hide. Chemistry and biology provide new tools as well. But what distinguishes these tools from mandated key recovery is that they require somebody to go to the location of the crime and use the tool. They are not sweepingly Orwellian in scope. They are not tools of mass surveillance. This means we would have an easy to understand practical limit on the application of police power.

So when you have a choice between a key escrow system and a strong encryption system in a new product, do the right thing. By letting the horses as much out of the barn as possible now, we all can do something to make the future better, to prevent future gulags and police states, as well as to keep the systems we will increasingly rely on for our economic well-being secure.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #15

Among the insights here are the comparison of the scale of telephony and computing. Although the Internet has grown faster than wireline telephony, the large-scale comparison is still valid today, especially in measuring mobile communication against computers: More than 400 million new mobile handsets each year, and billions of mobile subscribers versus 100 million PCs and hundreds of millions of INternet user. And, since I now work on mobile entertainment products, the roughly hundred-to-one ratio of mobile handsets to game consoles is an even more startling comparison, espcially when considering that connectedness among game consoles, even if it quickly reaches 100% of all game consoles sold, will always remain a tiny fraction of the globally available, mobile, and globally connected numbers of mobile handsets.

Newsletter #15: Feedback works

Some of the feedback I have enjoyed the most over the course of creating this newsletter has been with a friend who works for Microsoft in the role of (I wish I thought of this description) the “Ollie North of the anti-Java effort.” From him I have received a reliable stream of sometimes outré but often prescient views on the course of the conflict between COM hegemony and Java incursion. It has been well worth the effort to field the occasional over the top claim that no one would want a layer over Win32 (What, then, is COM and the MCFs or AFCs?) in order to obtain a clearer understanding of just how completely the future of server environments, and server software development, is dominated by NT.

As a result a balance has been reached between my attraction to newness (e.g.: I wrote a book on Macintosh programming when the Mac first came out) and his partisan position on the subject. This, from his point of view is no doubt made easier by the fact that Microsoft has really achieved every realistic goal in containing Java. They have watched Java proponents wheel from “Java everywhere” to “Java on heterogeneous servers” (What heterogeneous servers?) to some undefined position now as Java has yet to gel with any ISV products. Java will be the pixie dust sprinkled on Web pages to make them more interactive, and this is something Microsoft can live with. Although Microsoft’s spirit of winning every battle is well illustrated by the doomed attempt to compete with Java in this role with ActiveX.

The future will then look like Web pages made active through Java applets, forming the interface to server systems built largely out of distributed COM objects on NT servers. The typical call-center console, for example, will be a collection of Web pages. Microsoft will be pleased that server software is safely their domain to dominate as it grows into the most economically important software of the next several years. One does not know yet how completely Microsoft accepts this outcome: How well will Microsoft support creation of Java enabled Web interfaces? Another way to look at this question is: Is Microsoft succeeding with Visual J++, or have they lost Java developers’ confidence by flirting with non-standard Java? The answer to this question reveals the degree to which Microsoft has assimilated the result of the Java struggles. The answer right now is: not completely.

The other most fertile area of feedback has been the overall state of computer telephony. When I first set out to create a computer telephony product, the promise of the future was one where standards would make the de-coupling of hardware and software possible. In this future, VARs and integrators would sell mix-and-match combinations of telephony software and voice processing hardware. PBX-integration would have been driven below the hardware API layer, and voice-processing cards would compete on performance. The biggest disappointment of computer telephony is that this future did not arrive. The greatest question of computer telephony is what will come along to take another shot at expanding the market.

Let’s take a step back and recall why it is important computer telephony succeeds: 65 million people in the U.S. are on the Internet. Most of them use it for e-mail. I don’t know the Internet numbers worldwide but 100 million might be a good guess. Now consider that 2.5 billion people have reasonably good access to telephones. If we guess that of the 100 million Internet connected people 25 million surf the Web to any significant extent, it means that the population able to interact with computer telephony systems is 100 times larger, that is two orders of magnitude larger, than the population able to interact with Web-based systems. Two orders of magnitude is a big enough multiple to spend some effort reaching. As we create more and more server-based, Web-delivered applications, the question of bridging off-Web participants into these applications becomes a serious matter of total market size. In the race to dominate a segment and scale up faster, those off-Web consumers, even if they are not as rich or hip as your digerati friends, still have, on aggregate, a heck of a lot of buying power. Call centers, IVR, and fax are the tools for reaching out beyond the Web. Now computer telephony has to come up with some answers about integrating server-based Web applications with the telephony-based tools we have been reaching customers with since the first salesman picked up a telephone.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Teirati Newsletter #14

Just five years ago we buried Digital Equipment Corporation. This post-mortem looks at the incidents that led to Digital's demise, and also briefly looks at how voice processing finds it self in a similar bind. The intervening years have, of course, revealed that when Compaq ate digital, it ate poison, and Compaq itself never recovered from the downward momentum of all the lines of business it acquired with Digital.

Newsletter #14: Strategies work, until they stop working

Recent events in Washington illustrate a point: strategies appear to work until they stop working. While it did not result in a complete catastrophe, it is difficult to see boinking interns as a sustainable practice in the executive branch. Similarly, or perhaps just contemporaneously, the purchase of Digital by Compaq brings to mind once again the long and painful history of how Digital came to decline from the second largest computer company on the planet. Despite the dominance of the PC, it is startling to see Compaq, a company which “owns” very little of the technology on which it has built its tremendous prosperity buy Digital. Digital developed the computers that enabled my generation to see computers as a personal tool, cheap enough to deploy in a lab and manage with student labor. I was that student labor.

But enough nostalgia. The point is that you need to be able to tell when a strategy stops working in time to do something about it. It took Digital many years to go from the most vibrant force in computing, to a troubled giant, to a diminished once-great, to a potential alternative power in microprocessors, to their current end-game. The complex trajectory of Digital’s decline illustrates several ways in which strategies are evaluated more or less successfully. To take just two: How could Digital have prevented its greatest and ultimately fatal early mistakes, and did Digital take its best shot at renewal with Alpha and Windows NT?

Much apocrypha and irrelevant information, like Ken Olsen’s statements about personal computers, has clouded evaluation of Digital’s early errors. This is the bottom line: First, early in the game, when it would have been easy, Digital did not learn how to sell computers, PDP-11s for example, the way PCs are sold. Second, when the openness of the PDP-11, and the number of independently developed peripherals for it far outstripped what was available for the IBM PC, Digital did not enhance this openness and instead created a new, closed, patent-protected bus for its VAX line of computers, driving many potential allies in the market out of business. Finally, the failure to aggressively capitalize on and dominate networking technology, especially Internet routers, foreclosed the most likely path to renewal at Digital. It is particularly astonishing to recall that, at one time, the Internet consisted almost entirely of PDP-10s, PDP-11s, and VAXes.

It took three separate and massive errors to sink a company as big, powerful, and loaded with top talent as Digital. Two of these errors illustrate why evaluation of strategy is often difficult: Why not sell PDP-11s like PCs? When I first saw a PC, it was, by comparison with PDP-11s, a joke. Cheaply made and shoddily engineered. The motherboard was laden with “ECOs” – hand soldered connections correcting errors in the printed circuit board. Such things were rarely seen on anything Digital shipped. The 8086 processor was a caricature of the PDP-11 architecture. My first thoughts were “If DEC (that how we all though of “Digital” at the time) could make a small, cheap PDP-11, they’ll kill this PC thing.”

How naïve of me. Every mature man in a suit at Digital looked at the PC and thought “Not worth responding to.” If the IBM PC was so awful, why cut the margins made on PDP-11s? Here is the first reason why evaluating strategies in time to make a difference is so hard: every timely change means sacrificing safety. And big corporations are made up of herds of safety-seeking guys in suits who are, generally, in over their heads when it comes to seeing strategy. Not the best environment for making timely changes. The second reason it is difficult is the time-scale. All the people at Digital who made that fateful decision were probably doing something else by the time the consequences began to bite. Those consequences did not even begin to appear by the time the VAX was introduced, along with the Unibus successor, the SBI bus.

Here was a second chance: If Digital had failed to grab the emerging PC market, they could move their successful formula of open architectures up-market. Instead, they saw a chance to increase their profits and hold on to margins in the peripherals and enhancements business. Instead of opening the VAX’s SBI bus, Digital kept it for themselves, and left to independent peripheral makers only the slower Unibus. And it worked: Digital prospered while the companies that made the PDP-11 such an attractive computer for its flexibility and range of choices in peripherals died off. Nobody, I can confidently say without rummaging through Digital’s personnel records, got fired for keeping SBI bus proprietary.

And what to make of the muff in networking? Just as a cigar is sometimes just a cigar, in this case an error is an error. What can you say about blowing a chance where no safety is sacrificed, and nobody need risk their career? Digital could have been Cisco.

What can we learn from these errors? The first thing to learn is that the same sort of “situational ethics” that enable stuffed shirts to make bad decisions and then move before they are called to account can be turned against these poltroons. Use what you have at hand: Internet time-scales for example. Light a fire under your sluggish colleagues, and stoke it with fear that they will be caught out and embarrassed. Create an environment, through the use of press accounts of competitors’ announcements, where strategies are evaluated sooner rather than later. If the culture in your company is intractable, change horses or become one of the poltroons yourself. If you are in a leadership position, you have fewer excuses. The environment is what you have created, and if you have not created an environment where processes are transparent, you stand to lose exactly the sort of people who will be your ablest help in making important transitions. If your people don’t know what you are thinking, you are in trouble.

In our next example, things get more complex: Can good strategy fail? Yes, especially if starting from a hole that is too deep, into which rocks and debris are falling. So it was at Digital. Alpha is a better chip than the Pentium, and NT is destined to dominate enterprise computing. Trouble is, Microsoft did not get around to taking advantage of the 64 bit Alpha in NT, and Alpha itself could not overcome the space-warp economics of modern chip fabs. There Intel sets the pace, and sets it high enough to overcome any architectural advantage its competitors have.

What can one learn from an untenable situation? First, just as one needs to learn that apparent success may be the illusory, “locally optimal” result of a shortsighted decision, a losing position is not necessarily the result of bad decisions. One need ask what else could Digital have done? If there are no better options, they may have pursued the only reasonable strategy. Nine billion and change from Compaq may have been the best possible result. N.B.: this applies only to the question of strategy. Details of implementation, like the question of whether Digital cut back deeply enough quickly enough are outside the scope of the discussion. But to me it was evident: Driving on I-495 one could only ask “What the hell are all those people in all those Digital buildings doing? How many people does it take to make VAXes and VMS?” Now there are many fewer buildings so labeled. Now the signs say 3Com and Cisco and a large variety of smaller, nimbler companies.

I follow once again the pattern of developing a general idea and then mapping it onto telephony:. This time in focusing on the makers of voice processing cards. They face a decision point when it comes to strategy. They will feel interesting macroeconomic inputs that will warp the time-scale of decisions and results. The outcomes of decisions taken today will not be known for three years or more. Yet the fate of the voice processing card makers hangs on decisions taken today.

The first challenge voice processing card makers face is the temptation to do nothing. If nothing else happens this much is sure: the rest of the world will fill up with call centers and voice mail just as the U.S. has. This creates a huge temptation to do nothing. Do nothing and you will surely grow. Do nothing and you risk nothing. This temptation is reinforced by the face that efforts to break out of the core markets and proprietary habits of voice processing card makers have in relative terms not amounted to much. Technologies like TAPI face a credibility gap when it comes to delivering money into the pockets of voice processing card makers.

To overcome this challenge, management in voice processing companies must make the strategic process open and collaborative. Otherwise even the best lieutenant will just keep his head down and ride the “rest of world effect” to success. Only if made clear at a leadership level that the inevitable expansion of current markets reached by conventional means is not enough will a better result be reached. And only if resources are ventured on obtaining a better result will it be realized.

The second challenge voice processing card makers will face is that time frames for results from alternative strategies are largely unknowable. The right strategy may fail once or twice before it succeeds. Only by making clear at the leadership level that pursuit of the right strategy is the right thing to do even in light of uncertainty over the time frame in which it will succeed will you get anyone but the most naïve and foolhardy to implement that strategy. You can easily see the potential for failure to spiral out of control in that picture. Conversely, if a company’s leadership recognizes good strategy and good implementation, and is able to sustain efforts while making such wise evaluations without the external verification of immediate market gratification, then leadership will draw the following needed to make such strong implementations happen.

Forgive me if I do not go into the specific strategic choices facing voice processing card makers. Time and space have grown short. But I will make the following observation: it will be fascinating to watch the result. It is likely that one of the top tier companies will lose their position due to bad strategic decisions made at this point. There is also the opportunity for a second-tier company to vault up in the next few years based on the right decision. So watch the initiatives these companies make this year. This year’s decisions will determine the outcome of a shake out that will happen as telephony changes due to the impact of the Internet.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #13

While this collection of predictions holds up pretty well, it also shows how little consideration I gave to mobile telephony at the time I wrote it.

Newsletter #13: Perception vs. Reality

Perception: Microsoft is a boorish, arrogant company that harms your well-being by driving alternative browsers out of the market. It also squeezes PC makers by unfairly requiring them to bundle Internet Explorer with every copy of Windows.

Reality: The browser controversy is unimportant. Microsoft’s real weakness, and how they ill-serve their customers the most, is that Microsoft is not set up to be responsive to its customers. The customers I am talking about are the computer manufacturers. They are the ones paying Microsoft for Windows. Retail sales of upgrades, while nice, are not the driving force in Microsoft’s OS revenues. Where Microsoft’s would be competitors should have tried, and either missed the opportunity or failed, is in better serving OEM customers. Little things, like not bothering to update faxing for Win98, or leaving call control out of the TAPI 3.0 object model, stick in the craw. Bigger things, like being too slow to embrace 64 bit addressing in Alpha CPUs, can break a company. Compared with responsiveness to customers, a bit of hardball when it comes to contract negotiations is insignificant. But, with competitors that, for example, failed utterly in an attempt to port OS/2 to Power PC, are other companies even able to take up the challenge of customer-responsiveness? The evidence, so far, is against it, without even mentioning Apple’s willful destruction of their licensing business. There remains an interesting battlefield, however: Windows CE vs. Palm Pilot OS. Perhaps a Java OS will rise to challenge in set-top boxes. So far, Microsoft has been able to have it both ways. TCI will integrate Windows CE and Java in their set-top boxes, which won’t be at all difficult. But other customers will not find the DIY approach acceptable. In the future, Microsoft will have to decide whether to lead or follow in Java support on the CE platform.

Perception: IP telephony will eat the PSTN.

Reality: IP telephony will have to start by taking smaller bites. One of two victories IP telephony should accrue before taking a bigger bite are in corporate infrastructure, replacing the need for two networks to each desktop, and replacing a telephone switch with none-to-node communication internally and a network-to-PSTN gateway for external communication. The other is in cable telephony, and here there looks to be direct competition with switched-circuit telephony, not only from the installed based of wire-line telephone systems, but also from the various DSL technologies, all of which treat telephone calls separately from Internet traffic. The question for cable TV providers will be whether they can exploit potential advantages of IP telephony. The possibilities include PC-based telephony applications, in-home IP-based distribution systems, and synergies with set-top-box applications. If IP telephony cannot win in the last mile, it is unlikely to win in the network. These milestones, along with the projected change in the relative size of the Internet and the PSTN will mark the point at which switched circuit telephony will become the legacy subset of a new kind of telecommunication.

Perception: It’s about Java vs. the Windows API

Reality: TCI got it right. It’s about Java and the Windows API. Microsoft got all exercised about Java because it is the most viable alternative to Windows as an application platform. Who won? Well, nobody lost, so, in a sense, Microsoft did not win utterly and completely. On the other hand, Microsoft did successfully execute a strategy that contained Java hype to the point there is really no danger that Java threatens the relevance of Win32 API and COM+. Sun, on the other hand, succeeded where IBM, Apple, and countless less significant contestants failed, and created a new platform for widely distributed applications. The other side of this happy-happy everybody wins coin is that Microsoft will end up ceding a significant segment of applications to a platform other than Windows, and Sun has succeeded in creating that platform without making any money off it. The results are sure to be galling to both parties: Java applications running on a Microsoft-developed Java VM in Internet Explorer, served from an NT servers running IIS and Exchange Server, using Active Server Pages to bridge the server end of the application into the NT environment.

Perception: Computer telephony is inexorably drawing telephony software development to a small and well-understood family of standardized interfaces.

Reality: Don’t hold your breath. In fact, there has been considerable backsliding, with Active Voice finding growth in small, unsophisticated, non-networked, non-unified messaging voice mail, and Microsoft distracted by Internet conferencing. S.100 received an explicit un-endorsement from Bill Gates, and Q.931, a potential unifying factor in telephony signaling, remains sub rosa. The reality is that computer telephony will have to undergo another generation of development, beyond even that now under way, in order to achieve its promise. Anyone interested in making a lot of money in computer telephony will have to solve problems across several technological categories. Most big companies are uninterested in doing so, and most startups, even if they see a multifaceted solution to the problem of product creation won’t get funded because they appear too unfocused.

Perception: Conventional telephony’s advantage is reliability.

Reality: Conventional telephone networks are run by the same sort of people that make urban subways a pleasure and a joy. Just yesterday I found myself unable to dial 10 digit numbers inside my area code. Evidently my CO was incorrectly programmed as part of an area code changeover, and I ended up with no combination of area code, new or old, and number that would enable me to dial a local consulting client. I was also unable to dial the telco’s service number, or even reach an operator! No, conventional telephone networks’ advantage is that, in my neighborhood, anyway, the cable operator has even fewer clues. Which leads to the following gap between…

Perception: The fight is between the people with the wires, and they are fighting over voice telephone calls, because that is where the money is.

Reality: Internet access is more important than voice. One consequence of which is that wireless technologies, even ones which cannot support voice with wireline reliability and quality, are going to compete for Internet access in some places before the cable operator can afford to upgrade to HFC infrastructure, or the telco figures out how not to make DSL marketing just a replay of their huge ISDN success. It is also no sure thing that voice is where the money is. Even before local loop competition becomes commonplace, the diminishing importance of voice telephony will put downward pressure on prices. So all the alternative access companies that viewed Internet access as a dress rehearsal now find themselves on stage in the big show. It is this flip in the relative importance of voice telephony and Internet access that will also cause the change in viewpoint that will drive IP telephony into the mainstream: voice is just another medium for the Internet. We may not come to view this as the dawn of IP telephony, but rather the rise of voice as an integrated medium in computer-based communication, which is to say the end of telephony as a distinct discipline.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #12

Five years aso I pointed out the lack of a "killer app" in the area of computing and telecom convergence. The intervening five years have not brought about that killer app, but they have brought about a transition in my career from mainly CPE wireline telephony to mobile applications - specifically mobile games. And that leads me to two observations: First, important applications need not be related to productivity, and, second, that there still is not "killer app" that makes the increasing amount of intellingence in mobile handsets truly worthwhile.

Newsletter #12: The Personal Communicator Revolution

The personal computer revolution brought control over computing to the people who demanded that control: capable individuals who were fed up with unresponsive central control over computing.

So why did we have a PC revolution, and no “personal communicator” revolution? Telephones never oppressed the masses. You picked up the phone, you got dial tone. You dialed the phone, you got a connected call or voice mail. What’s not to like? No waiting for reports, or new applications to be installed, or more disk space. A phone was, and is, a reliable and useful tool that is ready when needed and undemanding otherwise. The telephone priesthood was never challenged like the minicomputer and mainframe clerical order, and never fell out of favor.

Let us look at the now-decadent revolution in personal computers: PCs proliferated, and the people who feared them, loathed them, and resented the adept, were equipped with PCs. The MIS department began to regulate the use and content of PCs. Approved vendors, control over requisitions. These are the tools of counter-revolution. So instead of armies of drudges sitting in front of terminals, we got armies of drudges sitting at PCs that seldom were on the leading-edge of anything, running dull applications, connected by duller networks.

What is the lesson here for us toiling in the vineyards of computer telephony? Is “computer telephony” anything more than the side effect of the hegemony of PC technology? Would we be thinking of computer telephony if PCs had not swept minicomputers off the table as a leading edge technology? In other words, is there anything to computer telephony beyond the fact that PC-based servers are cheap, and so make good application platforms?

A valid question, since, thus far, there is no personal communicator revolution that parallels the personal computer revolution. The reason for this is not just as simple as the fact that revolutions happen for a reason and there has not been reason enough in telephony. Revolutions are also made by people. Would the PC have reached so far if Andy Grove and Bill Gates did not push their parts of that revolution so far beyond the limits one would have expected in the normal course of events? What if the Macintosh, along with Apple’s management of Macintosh OS development, marked the high water mark of personal computer development so far? Would we be trying to build telephony applications on an OS that supported but one address space and no preemptive scheduling? What if OS/2 were the best 32-bit OS for the x86 architecture? Again not an entirely rhetorical question since there are still telephony applications that have not migrated from OS/2 to NT, having initially chosen OS/2 because it was the only reasonable choice at the time.

This looks bleak. If there is no reason to revolt against the telephone, and no Andy Grove to weave an architectural sow’s ear into global dominance, what hope is there? Is there the opportunity for a personal communicator revolution? Yes. A seminal article on why this is so is called “RISE OF THE STUPID NETWORK.” You can find it there: http://www.isen.com/stupid.html. In fact it has long been a given, in some quarters, that unless personal computers became personal communicators, their market would be stunted. Intel invented and developed TAPI for this reason. Neither thing happened in anything like the way that was foreseen. Personal telephony software for computers does not deliver added value outside a relatively tiny user community. PCs have thrived as never before. And market transitions like the sub-$1000 PC were far more important than any form of telephony or other person-to-person communication enabled by the PC. But the fact of the relative enormous size of the telephony industry compared with the computer industry is still a fact. Computing must become communications in order to grow into its next larger form.

This all is not to deny certain trends: Chat rooms propelled AOL in its infancy, and it is no surprise that Internet chat technologies that bring various combinations of rich content, speed, convenience, ease-of-use, and etc. have each found their niche among Internet users. That is personal communication. But chat may prove to be a product category with an intense but circumscribed audience. Unless chat technology takes on and beats H.323 in developing the business-oriented use of conferencing, chat will become an unimportant technology serving an odd user community that finds chat an interesting diversion. Chat is not the “killer app” of personal communication.

“Killer app” is itself a phrase that gets thrown around without much thought. Before it got diluted to the point of losing all its flavor, “killer app” meant an application that could make an industry. VisiCalc was a killer app. PageMaker was a killer app. As nice as it is, Visual Voice is not a killer app, and Artisoft, which bought the product in hopes of transitioning to higher growth in computer telephony, is finding that something as simple as sharing an Internet connection is, in the short run anyway, more profitable. PIMs have not been the killer app, either. The rise of the dumb network has not yet been balanced by the rise of any “killer app” to make use of the increasing computing power at the terminal ends of the ever-increasing capacity of the dumb network. This is something that should worry everyone.

Without a compelling application for the individual end user, computer telephony adds up to no more than a nice growth of call centers as the developing world catches up to the U.S. in over-the-phone selling. What makes this a particularly thorny problem is that the impediments to a killer telephony application cross several boundaries and disciplines. Computers, still, make lousy phones. Fixes to this problem, still, require adding a specialized audio card, and so the penetration rate of any good solution will be very low (compared with, for example, the QuickCam, or the Zip drive, both of which very much enhanced their market penetration through easy hook-up). Phones, still, make lousy computers. Uniden’s e-mail phone, and the IntelliPhone are likely flops, and while CIDCO may have something closer to the successful formula Web TV found, they will find their success limited by low-speed Internet connectivity and other infrastructure gaps in connecting devices within the home.

Where do we find the killer app? One thing is certain, it’s someplace “out of the box” we’ve been rummaging around in. Finding it will take integration across hardware and software, something rarely achieved in computer technology. Making that integration possible will take innovation in building industry relationships, because one company, even if it can conceive of the right combination of hardware and software, will not be able to execute it all. It may even be the case that the right application may be installed on your system already, but it lacks peripherals or network support to really make it sing.

When this killer app is found, it will change the industry. And such changes are poised to happen: IP telephony, real-time group interaction, unified messaging, are all on the boundary of change. Like a knife-point in superchilled water, it will instantly crystallize the industry, and what had been invisible will become solid. But until all the pieces are in place, how to evaluate investments in telephony innovations? If you can’t put all the pieces into place, can you at least have reasonable success by providing a subset of those pieces. The evidence says you can: video telephony shows no sign of becoming universal, or of driving IP telephony into the mainstream, but Connectix’s success with QuickCam is sure better than a poke in the eye with a sharp stick.

When looking at an opportunity, look at it in terms of the confluence of successful concepts around it. Look in the overlap of communication, thin servers, networks easy enough for any small business or home, common communication needs, familiarity, harnessing RoW (Rest of World) effects like artificially expensive conventional telephony. These are the external effects that a product formulation can harness to reach a really large target market.

Is there a way to get all the way to a complete success? I think there is, but it requires a comprehensive system approach that only a handful of companies can contemplate putting into place. The reward for doing this, however, can be large. If one’s entrepreneurial competitors can see only part of the solution, there is real value in grasping the whole solution.

No matter your position in this competition to find the real telephony killer app, happy hunting in this New Year.

Copyright 1998 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #11

This is one of the best essays I have ever written. In it, I reject the idea that the complexity of PCs can be managed away. Just read it. True as ever.

Newsletter #11: Live ammo

The armed forces go to great expense to train people in the use of deadly weapons. Misused, mismanaged, misapplied, or poorly understood, weapons kill more of your own than of your adversary. After all, your troops are at closer range than your target. The army takes recruits who are generally teenagers, and, through enough training, make it possible for them to carry infantry weapons and operate heavy weapons without causing carnage on the training field. Military training stems from the nature of the job and, pivotally, from the nature of the tools of the job.

In my two previous newsletters, I touched on the subject of PCs having overrun their natural constituency. PCs are more successful than they should be in an ideal world. PCs are general-purpose computers built in the image of their minicomputer ancestors. They give the user unlimited flexibility and unlimited ability to make mistakes, far outside the scope of any other kind of system people use. The PC, when deployed as a standard-issue office accouterment, the way a rifle is issued to every infantryman, is like passing out an automatic weapon with live ammo. PC users get nothing like the level of training that the teenagers with the real live ammo in their clips get.

The result is carnage: People make mistakes. They blame their tools -- the PCs. They blame the people charged with keeping their tools useful – the MIS department. They come to fear or loathe this very useful tool. Or they worship it and supplicate before its priests like some cargo cult. To add to the generally unsatisfactory situation, the relationship between PCs, the users of PCs, and MIS departments is very unnatural.

Before there were PCs, there were mainframes and minicomputers with terminal attached to them. The computers were operated and maintained by a cadre of trained technicians. In those historic times, computers and telephone systems were remarkably alike in the way they were installed and operated. Dumb terminals (“very thin clients,” if you are looking for capital to build them today) connected to the computer and phones connected to the PBX. This diverged when some users of the computer – the “power users” that mattered most in justifying the cost of the computer – became dissatisfied with the lack of responsiveness from the MIS technicians to their needs. Telephone users, by contrast, never had such a beef.

The power users were, generally, influential in the organization, effective, and intelligent. They got what they wanted: computers on their desks entirely under their control. They thrived. They analyzed hundreds of financial outcomes while the PC-less drones waited for a single report. They created documents online. They stored their own databases. They became even more powerful, effective, and insufferable. Everybody wanted to be like them. Everybody wanted (or at least made as if they wanted) a PC.

And the MIS department still wanted their jobs. So the old guard made a appalling deal with the masses: They would all get PCs, just like the revolutionaries who stirred up all this trouble to begin with, but MIS would undertake responsibility for making PCs safe for non-revolutionary people.

It is as if the army, having seen the success of rifles in a pilot program in the Special Forces said “Let’s give everybody a rifle and ammo” and omitted to train the soldiers except to tell them to “point and click.” Instead they would use “rifle management systems” to try to maintain central control over rifles, and might commission arms makers to create “zero administration rifles,” or perhaps “network rifles” in which only officers control the triggers. It is pretty clear this won’t work with rifles, and it works not much better with PCs.

How does a teenage soldier manage his rifle? How is it that his buddies aren’t constantly being treated for accidental wounds? Why is there no need for a “rifle help desk?” The answer is that the soldier is intimately familiar with his rifle. The soldier knows his rifle, literally, inside and out. He could take a pile of rifle parts, rusted, pitted, wet, and clean, repair, and assemble them into a perfectly working rifle and do it in the dark with a battle raging around his ears. The rifle, like the PC, is not an appliance. It is a general-purpose tool, and a very powerful one.

In a business culture that is overall very fond of military analogies, in which concepts of unit cohesion, tactics, and strategy are borrowed from modern and ancient military thinkers, the PC, the business infantryman’s rifle, has thus far gotten short shift. This is due to the fact that senior management is just now getting past the stage where a computer is either forthrightly banished from the CEO’s office, or sits on his credenza as a decorative item. A senior military officer that is so out of touch with the “grunts” is a clear sign of trouble, but in business, this is the rule, not the exception.

There is also the fact that, for all the talk of “empowerment” in business, the typical PC user is fairly powerless and confined. Open book management, flattened organizations, and internal entrepreneurship are going to have to take deeper root before many PC users can make the most of the power of the systems at their fingertips. But even those that have the level of access and autonomy to make full use of a PC are hampered by training that, if present at all, is limited to pointing and clicking. What if the typical PC user had the same level of familiarity  the almost matrimonial familiarity  that the soldier has with his rifle? You would have an employee as deadly to his adversaries as a well trained soldier is to his.

You, in your organization, have the opportunity to make your colleagues and employees as familiar with their PCs as infantrymen are with their rifles. If you are typical, you are awash in old PCs  486 PCs with too little memory and disk to be useful. Take a handful of these old clunkers. If you have to buy them, they should cost about $300 apiece. Dismantle them. Buy some copies of a book a PCs, of which there are several, that shows how to build up a PC from parts. Give a handful of your people copies of the book, an up-to-date Windows installation disk, and a dismantled PC. The course should be largely self-directed and self-paced, but should be guided by a person who has already built up a PC from components and can steer the participants to the sources of information they will need to answer questions.

The result should be a group of people for whom the PC holds little mystery. And there is every indication this should be broadly applicable. Building a PC is easier than overhauling a car engine, plowing and planting a field, building a shed, or fixing a leaky pipe. It is easier than learning to play a pennywhistle. It requires no talent or dexterity or practiced moves. The downside risk, which is a “smoked” 486 motherboard, is easily absorbed. What is the advantage of knowing one’s PC from the inside out? The advantage is that, even though Microsoft has done a brilliant job of hiding the complexity and the quirks of the PC architecture form users, the basic nature of the PC still comes through, especially when something goes wrong. With a couple hundred dollars more in hardware, these training PCs could be networked by the participants in the training program, and even more mystery would fall away from the PC.

This all started with the thought that the PC is not for everyone, and it has become something everyone has, and that this imbalance is the root of a lot of evil, especially in the way larger organizations manage their PCs. One way to solve the problem is to look to simpler computers. The larger Windows CE systems will address this need. But we will be cranking out tens of millions of new PCs every year as far as they eye can see. And they will have almost all the legacy baggage of today’s PC (some will begin sloughing it off). The best solution for the world as it stands is to try something radically new to raise the level of confidence of the average PC user. A lot of people arrive at this level of competency and confidence on their own. The sales of books on how PCs work attests to this. As long as PCs remain, inherently, the tool of the power user, it looks like a better bet to upgrade the power of the user than to try, through management mechanisms that were never designed for the PC in the first place, to make the PC “safe.”

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #10

This newsletter contains much that is still valid on the topic of legacy baggage, plus a real howler of a bad prediction: That Windows CE (a.k.a. Pocket PC) would be successful. I should have listened more closely to my own words on how the crushing weight of PC clone numbers keeps any other computing platform from being competitive in price and performance.

Newsletter #10: Check your baggage?

On the underside of success you find the encrustation caused by years of neglecting to scrape old features off a product as it sails toward its future. These barnacles can slow a product down until it stops progressing.

It is remarkable that under the weight of an antiquated peripheral bus, a motley assortment of ports of various speeds and interface peculiarities, power and cooling from the dark ages, floppy drives with ridiculously small capacity, a tightly limited number of interrupts devices can use, and an instruction set evolved from an 8-bit processor that would scarcely be considered fit to run your toaster today, the PC marches on. Some of this remarkable buoyancy is evidence that volume conquers all. It doesn’t matter that a stamped sheet metal chassis full of screws and brackets is less efficient than a molded box with snap-in fixtures. If you make tens of millions of the inefficient metal ones they will be cheaper than the cleverest design produced “only” in the hundreds of thousands. More interestingly, it doesn’t even matter if your processor is burdened by the most inelegant instruction set and registers of any processor save, perhaps, the 6502. If you build fabs that cost more, produce more, and drive the state of the art in manufacturing technology and fab construction, you can bury any alternative, like RISC, under unconquerable numbers. Numbers also float the other hardware crud in the PC that would weigh down anything not made in such quantity.

Baggage, it’s real weight, and externalities that make it heavier or lighter control whether products and companies win or lose. It's time to check the weight of your baggage, and your competitors’ baggage.

Look at that keyboard in front of you. Function keys, arrow keys, a numeric keypad, more function keys. The thing looks like the deck of an aircraft carrier parked on your desk. Is there any harm in this excess considering how cheap keyboards have become? Well, yes. How far is your mouse from where your hands are while typing? A good 8" farther than it needs to be. Why? Because at some point in the past, spreadsheets were the reason people bought computers. If it were not easy to enter lots of numbers, computer sales would suffer. Even Apple, which banged it's head against the wall of corporate computer sales without substantial effect for however long gradually polluted the original, well-designed Macintosh keyboard with all the same crud adorning PC keyboards.

Perhaps USB keyboards will save us. With USB, you can daisy-chain devices, like keyboards for all us mouse users that don't carry excess baggage, plus, for those of you that need it, the extra numeric pad, pad-o-buttons, joystick, etc. Apple, of course, had this very solution in the form of Apple Desktop Bus for years already. Did that stop them from crudding up their keyboards? No.

How about those windows? You may recall from newsletter #7 I explained how the desktop metaphor is here to stay until someone comes up with a better metaphor, or framework, for architecting a user interface. But we could have some non-rectilinear shapes by now, for heaven's sake. Gray backgrounds to buttons and beveled edges are the state of the art in user interface. Little color, less motion. We need some ovoids, some anthropomorphic shapes, for relief from the on-screen monotony for rectangular paper shapes. And not be just for the sake of whimsy: A tab-shaped title bar to a window would hide less of what is behind it. And the tabs might move to reveal as much of the tabs on title bars behind the front window as possible -- kind of a self-organizing set of file folder tabs. They would also be easier to see: our eyes process edges. If all the edges on a screen are the same, they fade together and camouflage one another.
Controls that pop out from the edges of a document as needed would also focus the screen display on content and not user interface overhead. Smooth shapes would help distinguish controls from content. If PCs are to converge with TVs, you'll have to clear the screen of non-content material until it's needed. The pop-up task bar in Win95 is a start. Now let's have more of this good thing. And if you are starting from scratch, leave the baggage of old gray toolbars behind.

Icons: Icons are icons of user interface. Only they have silly restrictions that limit them from being pictures, and that makes them baggage. As capacity increases, the number of different data types that have their origins in efficiency or expediency should decrease. A picture is a picture. Size, shape, and such mundanes and bit-depth should cease to be issues when labeling things in the computing environment pictorially.

Despite this user interface baggage, and a good deal more that we don’t have time or space to examine here, Windows looks to make good on the boast of “Windows everywhere.” Why hasn’t baggage slowed down Windows? Unlike Intel, Microsoft cannot shut out competition by redefining industry economics. Making software doesn’t require 3 billion dollar factories. And some of Microsoft’s would-be competitors have spent as much, and sometimes much more, on OS development only to come up short.

Microsoft actively sheds baggage. One recent example is COM+, which effectively shed the ungainly baggage COM attached to C++, and which makes coding COM+ apps in C++ nearly as convenient as Java coding. Windows 95, and even more so Windows 98, go to great lengths not to be weighed down by hardware baggage the PC has accumulated. Imagine the effort it must have taken to treat almost all non-plug-and-play peripherals as if they were plug-and-play by discovering them and how they are configured, and automatically installing the correct software. This works remarkably well, and has taken what had been a nasty, error prone, and time-consuming task, and turned it into the near-equal of plugging peripherals into the elegant self-configuring bus Apple introduced in the Mac II. All this through brute effort at making the PC pig fly. The lesson here is: if you are stuck with baggage, work like the devil to make it lighter.

Other OSs, notably OS/2, did not learn this lesson. OS/2 relied on plug and play support from the BIOS vendors while Microsoft was busy obsoleting separate plug and play (and PCMCIA card and socket services, too). You can’t rely on others to implement something as strategic as making your OS installable on nearly any grungy patched-together PC. Or perhaps it was just Microsoft’s cussed resistance to allowing anyone other than Microsoft to have a franchise on a required bit of software going into a PC. In any case, the result is that you take a card, stick it in the ISA, VESA, PCI, or PCMCIA bus and it just works. N.B., makers of voice processing cards and high-density fax cards, this is the difference between selling a few and selling a lot.

Speaking of OS/2… Two desktop operating systems have recently attempted to change direction. Both Mac OS and OS/2 have branched off into variants that are supposed to operate as NC OSs. There’s a problem with this, however. Desktop operating systems are kind of like very sophisticated Tamogatchis: if you don’t feed them, water them, and play with them, they clog with crud like temp files, configuration files, settings and preferences in system databases, etc. Desktop OSs are by nature the software needed to operate a complete computer system. So, while the NC idea, that the PC has hugely overrun its natural constituency of people who actually want and need a complete computer system on their desks, is a good idea, it won’t get to Jerusalem running Mac OS or OS/2. This is an example of fatally heavy baggage.

Some people get lucky about their baggage. Sun had done a poor job of managing the baggage of Unix. For years and years, Unix stayed as antediluvian in configuration, installation, and operation as it was when I was knee high to a PDP-11. In many ways, it still is just as bad. (Which is not to say it was bad 17 years ago. Unix was cool 17 years ago.) As a result, Sun, which has shone as a beacon of relative stability in the balkanization of Unix, began to lose workstation customers to fast PCs and Windows NT (which is no configuration picnic itself, but at least gives you stultifying and arcane dialog boxes rather than stultifying and arcane text files). Then came the Internet, and some of the things intrinsic to Unix became suddenly all the rage. For a brief window, you could look at NT and Unix and say "Unix would make it easier for me to get on the Internet.” Sun is still riding the momentum of this huge bit of good fortune. Trouble is, Microsoft has long since slammed this opening closed, and is rapidly making NT simple to configure and connect to the Internet so that any small business that wants a presence on the Internet will choose NT. So here is a prediction: If Sun cannot produce a Java OS that sheds the baggage of Unix, and cannot lighten the load on Unix as well, Sun’s momentum gained from the Internet explosion will run out in two years. In five years it’ll be “Sun who?” and everyone will think Bill Gates invented the Internet.

Much of the entire attraction of Java is the lack of baggage. One person had tight control over what was in, and what was not in. Trouble is, you can’t be a fashionable tourist without accumulating souvenirs, and even a retinue. The Java Telephony API is huge. And I recently saw announcements that Sun and IBM are working together on a future user interface standard. Now if that isn’t the crack of doom. There’s probably even some Taligent office space still available.

There is one very interesting lightweight traveler on the scene: Windows CE. While Windows CE cannot be said to be a huge success yet, here is a bold prediction: CE will be very, very important, and will rival the unit volume of Windows 95 sooner than almost everyone outside Microsoft expects. This is because CE is the answer to one of the most fundamental baggage problems: Many people who use a PC don’t need one, and many people who don’t use a PC do need some of the capabilities of a PC. Why is this so? This is because a PC, no matter how well each component has overcome its historical baggage in terms of driving down the price of a PC, it is still a complete, whole, general purpose, super-flexible computer. A PC is made in the image of the minicomputer. Everything is there: you can add any software, remove any software, including the OS, and and remove hardware components -- you can, in short, royally screw up if you don’t know what you are doing. Look to your left, look to your right. Do most people know what they are doing? So to compensate between the competency mismatch between computing and computer users, all manner of kludges and patches, like DMI and Zero Administration, have cropped up to keep Dilbert’s boss from blowing away his already meager productivity and then blaming it on the computer and the people in his company charged with keeping Dilbert’s boss’s PC useful. PCs were made for revolutionaries -- hotheads who can’t wait for the MIS department to get it right. CE is more like what many people really need. If I could screw up my car (which, the manual says, has a 32 bit CPU) by turning it off at an inopportune time, I would not consider it acceptable. Windows CE is more like my car, you turn it off, it stops computing, you turn it on, and it starts again.

In telephony (see, I did work telephony into this newsletter), PBX makers are unloading their hardware baggage and moving the soul of the PBX, the control program, to PC servers. These servers will either control slimmed-down “dumb” switches, or perform PBX-like services for users of IP telephony on intranets, or both. The question is, however: do the functions of a high-end PBX constitute value, or baggage? Do customers really need to minutely control dialing privileges, or are these features just the answer to the question “What do we do now that the switch connects calls?”

How do you integrate user management when moving the software infrastructure of a PBX into a network? Or is that even the right question? Does it make sense to have restrictions on individuals on a data network, beyond restricting them from the data they can access? Do the complex relationships of covering extensions, break-in capability, etc. make sense in an environment where the number of channels open to a desk is effectively unlimited?

What kind of telephony should be ported to data networks? The whole enchilada? Or something more like a simple switch that enables calls on trunks to reach inside extensions? How much of the idea of a PBX can really be ported, and how much must be invented afresh? And what does that imply for which market segment makes the right first target for network telephony?

One approach to an answer is to trace the evolutionary path of telephone switches, and strip the functions back to an earlier stage in development. This approach implies that complex PBX functions, especially the more esoteric user management functions, are not good candidates for implementation in IP telephony servers. This is because they overlap the later, more sophisticated development of similarly intended functions in data networks, and they are farther away from the basics of why we built telephone switches in the first place: to share resources.

So far so good, but we are left with a dilemma: telephony servers are, nowadays, expensive. They won’t ship in large volumes. They are, mostly, direct analogs of high-end PBXs. Big organizations that would gain the most in cost savings by consolidating their wiring plant and using the Internet to economize on calls to odd and expensive locations are now the target market for IP telephony. But is that really the right market given where telephony servers should begin their evolution? The challenge may be less to implement a comprehensive reinterpretation of what a complex PBX does, than to find a way to sell IP telephony to smaller organizations, and then to drive up-market as the product evolves.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #9

A bit of a slog for those seeking instant condensed wisdom, this old nugget on the fax server industry does maintain some relevance: One of the more innovative applications of camera phones is in document facsimile and other document-like imaging functions, like capturing images on a whiteboard.

Developers of these fax-like products will have to contend with the fact that the fax machine is still a standard of usability: paper is fed into one machine, and an identical paper comes out the other. Calling this process "imaging" is a cop out. If it isn't as easy as teleporting a document, mobile camera document imaging applications will find only a minority of their potential user base.

Newsletter #9 The future of faxing:

The future of faxing? How about the future of buggy whips, eh? Not quite the same. Typewriters did not exterminate pencils. Neither did computers. PowerPoint-obsessed managers and interminable meetings viewing "presentations" only sharpen the appetite for doodling. Until man stops marking paper with pen or pencil, we will need to transmit those images. To boil it down to the essentials, fax machines look to be about as permanent as phones. Not to say "eternal" but functionally close to it for the sake of this discussion. And this discussion is about how faxing will evolve in the near term.

Faxing stands on two pillars: The fax machine, and enterprise fax management. The fax machine has embedded itself into the office environment as much as any other office machine, like the copier, the phones, and the postage meter. It is part of how large numbers of businesses view the typical complement of an office, no matter the function of that office. Fax servers that implement enterprise fax management come in when the deployment of large numbers of fax machines adds up to a cost that is higher than it ought to be. Fax servers can also serve particular business processes or applications such as broadcasting information, or order processing resulting in documents sent to large numbers of customers, for example.

So, while fax servers are sold, in part, to reduce the need for fax machines, cultural and human-factors issues prevent the fax machine from being eliminated entirely. It is a vain hope that the need to put a sheet of paper in a fax machine, send it immediately to a recipient, and confirm that the recipient has the document facsimile in hand, will go away. Fax server vendors now know they cannot promise their customers that every fax machine, and every phone line serving those machines, can be consolidated into a server.

Fax servers are the easier of the two pillars of faxing to treat: fax serving has a bright future. An increasing number of businesses will see the scale of their operations grow to the size where buying a fax server makes economic sense. Microsoft is including a simple outbound fax resource sharing system in their NT operating system, which brings the incremental cost of the simplest form of fax serving down to zero. If Microsoft increases the functions of their fax sharing tool in NT 5.0, it may move the fax server business exclusively into the high end, but from the user's point of view, and from the hardware-maker's point of view, all the news is good.

The greatest threat to fax servers on the customers' premises is Internet faxing. Internet faxing was originally conceived as a cost reduction tool. The Internet would be used for long distance bypass. That model flopped for this reason: any business that does so much faxing that long distance savings matter can already seek competitive bids from long-distance providers. The Internet fax service providers never achieved the scale that enabled them to reduce their billing overhead so that the long-distance bypass savings show up in the customer's pocket. That, and the fact that the "last mile" to the recipient's fax machine constitutes the bulk of the cost in both Internet and PSTN-based faxing, made Internet faxing a dud in transmission cost savings. So how is Internet faxing a threat to fax servers?

The Internet faxing business model needs to be revised to an outsourcing business model. This is a model already understood and implemented by Internet backup services. "Don't buy that tape drive. Don't incur the administrative overhead of making backups. Don't hassle with keeping the tapes in a safe deposit box or secure off-site storage. Let us do it all for you." Similarly, the cost of phone lines for a fax server, server maintenance and administration, etc. could all be moved off-site. The more administrative features, like logging and archiving fax traffic, the user requires, the more compelling the off-site, outsourced model becomes. This will also have the effect of moving the fax server business to the high end, since the service providers will need large scale, multi-locations fax server systems. (One wonders why MSN, which has pioneered all manner of Internet-based, revenue-generating services such as airline ticket sales and real estate sales, has not taken on serving the direct needs of Microsoft customers in backup, virtual fax numbers, voice mail, fax serving, etc. An odd blind spot for so big an investment.)

The fax server business is divided into layers, and the upper strata are further divided by the way fax serving is used. At the low end, fax serving is mostly resource sharing: take the individual fax modems off the users' desks, and pool them into a server. Most such resource sharing is used for outbound traffic only. Most inbound traffic in low-end installations comes into fax machines. Computer Associates is a leading player at the low end, dominating low-end fax serving on the NetWare platform. ("CA?" you say. Well they got the business when they bought Cheyenne -- kind of like a whistle in a Cracker Jack box. It is in fact, just as you may think, unclear how fax serving fits CA's business model. Yet it fits no worse than much of the ragbag of hundreds of products CA sells.) OmTool has made hay in the low and low-to-middle market on the NT platform. But it is a vague business at the low end. High return rates and a large percentage of "shelfware" characterize the low-end market. Many customers of low-end fax serving find the fax server software cheap enough to try, but they don't have the scale, or administrative discipline, or, frankly, the need to make the move away from their fax machines. But, as I have heard many times in this business "You can sell anything on price."

The middle of the fax server business has lately been the playground of RightFax. RightFax is now part of AVT, a voice mail vendor. AVT is known for products that always look clunkier than those of its Seattle neighbor Active Voice, but that always seem to sell better. AVT is the only second-tier voice mail company to really break open the unified messaging market and derive a serious fraction of its income from that highly sophisticated part of the voice mail business. AVT does it by hard work: building a dealer channel that consists of dealers that have the scale and sophistication to sell products that are difficult to install, as all communications products are, relative to their data networking cousins. This must be rubbing off on RightFax, which has done well in the mid range fax server business through the same attention to what works in the channel. Success has given AVT an appetite: they recently bought AIFP, an application-oriented enterprise fax server company.

Mid-range fax servers require attention to the dealer channel, because the same fate of high return rates could befall these fax servers if not for a dealer channel that can install them. Typically, the installation process consists of installing a fax board in a server PC of the customer's choice. Most fax server software supports more than one brand of multi-port fax card. The fax cards are usually no pleasure to install, with peculiar IRQ and IO port requirements and no support for plug-and-play detection of these resource needs. And, since there are no software abstractions in Windows for either the fax boards, nor is there a fax API, there is consequently no concept of standard drivers for fax boards. Microsoft tried and failed several times to arrive at fax software interface standards, and now seems to treat the subject with the same enthusiasm the Russians have for Afghanistan.

High end fax servers are, of course, generally bigger. They are also, generally, turn-key systems -- built by the vendor. When they are sold, they stay sold. Customer returns of these systems are rare and if they happen, they signal some significant fault in the vendor's product. These systems are deployed two ways: First, general purpose enterprise fax management, in situations where numerous different documents are sent to and from desktops throughout an enterprise. Second, in application-specific settings, where application-generated traffic dominates: travel itineraries, purchase orders, invoices, etc. High-end fax servers are well placed to avoid challenge from outsourcing. High end servers will become the tools with which companies manage the use of in-house and out-of-house telecommunications resources for faxing, blending Internet fax services with least-cost routing and load balancing.

What of the fax machine itself? Remember that there are hundreds of millions of people in this world who consider a fax machine easy enough to use, almost as easy as a telephone, while they find computers impenetrable. So, while the world's 100 million fax machines do not form as formidable a host as the number of telephone handsets, these are enough to rival the number of computers in use. And, since fax machines usually serve a group of people in a singular, specialized purpose, the number of people who regularly send faxes far outnumbers e-mail connected computer users. These great numbers, combined with the cultural barriers that keeps a large segment of fax users from converting to computer-based communications, and added to the basic need to transmit images on paper, means fax machines will not be displaced by e-mail soon. It also means that new standards like G5 faxing are not exactly responding to a real need. You will likely see all the need for a combination of e-mail and faxing met by a combination of Super G3 and existing e-mail standards.

Do these product categories overlap? Sure, especially if you look only at product features, and not how the products are delivered or how the customers use them. For examples, small fax servers that resemble inexpensive print servers bring the turn-key model to mid range fax servers. Castelle and JetFax are examples of this type of fax server. The convergence of the Internet networking standards and all-in-one peripherals will beget innovations in the turn-key fax server device category. These "thin servers" represent the future of the fax machine: an intelligent terminal device for the intranet as well as a server. If such devices can be made inexpensively enough, and if they remain as easy to use as a fax machine, they will actually penetrate the market for fax machines, which is far, far bigger than that for the existing market JetFax plays in. The barest surface of benefits derivable from a combination of the Internet and faxing has just been scratched, and these benefits are far more concrete than those offered so far in the various Internet-enabled telephone handsets.

Faxing, therefore, in every product segment from the lowliest fax machine in an un-automated office to the most monstrous server bristling with T-1 lines, has a good long way to go before it has exhausted either the potential for greater productivity, or the potential for innovation that can be directly translated to user benefit. There are at this moment two pens and a colored pencil on my desk. It looks as if I will remain close kin to my stone age ancestors (and not just for reason of what my wife refers to as my "jutting brow") in marking flat surfaces with a stylus. And the need to transmit those markings will remain as well. There are, in faxing, more segments than any vendor in any subset of these segments will see or admit to, though almost all are likely to grow.

The future belongs to those vendors that will meet this continuing need with undiminished simplicity while delivering more value (probably via the Internet) to that simple task.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #8

This is another that has stood the test of time. Now that I am involved in mobile game development, rereading this newsletter put into focus many of the project management tactical issues and product strategies I now must use to succeed.

Newsletter #8: You will make mistakes

You will make mistakes. So will your competitors. On average, their engineers are not smarter than your engineers. On average, their code has just as many bugs as your code. On average, their projects are as late as your projects. If you have respect for your competitors, you will soon realize they are not significantly better than you are, and you can't count on them being that much worse. So how do you beat your competitors?

You beat your competitors by managing better. In this newsletter I will discuss, specifically, managing your product development better. Managing your product development better will not change your schedules, bug-counts, or the make up of your engineering staff enough to beat your competitors, but it will prevent you from defeating yourself. Then you just have to keep that up long enough for your competitors to defeat themselves. So you don't actually beat your competitors, you just make sure you're there, in good position, when they blow it, so that you end up so far ahead they can't catch up to you.

You might think you can do this by avoiding mistakes. Not so. You will make mistakes.

Make your schedules so that they are not doomed to fail. And then, when they do fail, learn from them. How do you make a software project schedule? You make it simple. Schedules that have lots of parallelism and few major milestones that tie parallel efforts together are doomed to fail. Take a look at each of those milestones. If you have not met an early milestone, count on this: there is no such thing as making up the time. Push your whole schedule back by as much time as you missed that first milestone, and do not factor in partially completed tasks on the other side of that milestone.

Don't shoot your own officers. When a project slips, is it the impulse of your senior management to replace the project leader? Start a new, parallel project? Cover their asses? Flail around without diagnosing why a project got into trouble? If so, go find a new job. You are doomed. You might as well work for the IRS. If your company is incapable of learning, and reacting to problems in a way that extracts knowledge, it can't produce software. If it ever did, it was an accident unlikely to be repeated. A cleverer man than I called Apple, while Apple was still fat and happy, "the Cultural Revolution run by men in suits." What he meant was that Apple was run by a bunch of managers who were disconnected from the idea of making great products, who, indeed, were quite cynical about the whole "insanely great" thing now that Steve Jobs was gone. These stuffed shirts would cycle Apple's money through a lot of big, expensive, safe, and mediocre development projects. They avoided mistakes (they still made a bunch). They avoided risk (and still made mistakes). And, in avoiding risk, they avoided any possibility of Apple breaking out of its niche. And then, when things went badly wrong anyway, they purged each other like a Party Central Committee after a bad harvest. If you punish risk-taking, if you fail to learn from errors, these are the results you will get.

Communicate clearly. A specification running hundreds of pages, showing stultifying screen by screen depictions of a desired result is not clear communication. Typically, managers with a poor grasp of implementation technology generate specifications of that nature, and their only result is to alienate the design engineers into a stupor. Instead, make sure every engineer has a clear picture in his mind what the result will be, and what guidelines will be followed to achieve that result. Take one aspect of the project and have a visualization session that encourages the implementers to describe the ideal result in detail so that the engineers know how the specs should be visualized. Keep the specs short and sweet.

There is no substitute for strong leadership. Every product needs one person responsible for the result. The difference between your typical GM dullmobile and a BMW is that some individual cares what the BMW looks like. While there are no absolutes, and every product with any economic impact is invariably produced in a structured corporate setting, it is clear enough from the results when an individual has triumphed in expressing his will in a product, compared with the mediocrity produced by committee. This may sound like a rather poetic take on management. Ignore it at your own cost, however. Software is a thing that costs nothing to make, and can cost anything to buy. So if BMW can charge twice as much as GM for the same lump of steel fashioned into a car, just think what the difference could mean when what you sell is bits on a disk.

Pick a few clear themes and repeat the message often. Every major product should have one or two key goals. It should be clear that these goals are to be met at all costs, and that every other goal is secondary. This is the way to keep things on schedule. This is the way to make sure the key design elements receive the most implementation attention. This is the way to make sure that, if you hit rough water, every oarsman in the boat knows which way to pull. You won't have time to fix the clarity of communication about priorities in the middle of a crisis.

Count on losing a few. The way to attack a market segment is to find a set of approaches to that segment and prioritize them according to factors of development risk and likely reward, and when you think the market windows open and close for each opportunity. Then, count on the fact that you might not have a hit on your first try. Know, beforehand, then what your second, and third, try should be.

Compare and contrast results: Microsoft's Internet Explorer 3 is competitive, and version 4 is perhaps dominant. Windows, on the fourth try, dominates. Most of their competitors have quit already by the second round. Microsoft's approach bears all the hallmarks of: 1. Managing projects carefully. 2. Responding to delays intelligently. 3. Learning from errors. 4. Keeping at it long enough that, on the one hand, the things that have been learned can be applied, and on the other hand, the dim-bulb cover-your-ass managers at the competitors firms have long since failed. Simple enough. Why isn't everyone doing it?

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #7

Of my old newsletters, this one is still almost fully relevant today. Telephone user interfaces are as awful today as the first horrid attempts at GUIs for placing phone calls. Even the extra features on mobile phones don't integrate well with the fundamentals of dialing a call. Progress? It can be summed up in one phrase: on-hook dialing.

Address books, calendars, call logs, and messaging are as un-integrated as ever, in both the widely used mobile user interfaces and in the finally-becoming-mainstream desktop IP telephony interfaces. Smartphone UI, with a completely flexible touchscreen interface never delivered progess, either.

What does this tell us? For now, it means that, even with 400 million handsets delivered each year, nobody is putting a serious effort into fundamentally improving the way existing mobile phone features integrate with one another. It means that SMS and MMS are no closer to integrating with voicemail than they were in 1997. It means that there isn't even one application that uses call control in an innovative way on the desktop or in mobile phones. Bah! And people wonder where the bad karma of the telecom crash came from?

Newsletter #7: User interface

I like nice artifacts. A car that looks right and drives right. A kitchen implement that adds to the pleasure of making and eating a meal. A house that speaks to you about the life inside it. We spend so much time in front of our PCs that user interface is a matter of the quality of one's life these days.

No wonder then, that some people look at the Web as the work of the devil. In newsletter #6 I spoke of the apparent end of history when Windows 95 was announced, and how the Internet unraveled that end. One reason why Windows 95 was such a milestone was that it marked the point where Microsoft had effectively eclipsed the Macintosh in every important way, especially in user interface.

Up to that point, it was the Macintosh that lead the way to a future where strict rules would ensure that every application in the desktop metaphor was similar enough that, just like the papers and files in our offices, every document and tool for manipulating documents was obvious. Operations would be standardized -- made generic. Pick an object, perform an action. That's it. No wrong choices are possible because once the object is selected, only the valid actions are highlighted.

The world on screen was headed for a kind of super-reality where direct manipulation of on-screen objects made us unerring draftsmen, arithmetic geniuses, and creative typographers. History would be a smooth continuum where the remaining ugly "dialog box" inheritances of our menu-based cave-man life would be gradually replaced by ever more powerful direct manipulation of documents on screen.

Sure, there was some backsliding. "wizards" are no more than a hideous rebirth of those lame-brain suggestions we've all encountered early on in GUI evolution: "Gee if you just made a sequence of dialog boxes appear, we could port our menu-screen software to the Mac!" But all in all, the virtual-desktop super-office world was coming to life on our screens. Homogenous office documents, all happy at home framed by powerful, yet familiar and uniform tool-bars. Then wham! The Web.

In contrast to the uniform, business-like world of the desktop metaphor, the Web is more like the pop-music magazine metaphor. No standards. Intentional violations of convention. Loud, clashing graphics. When you hear Microsoft's principal software architects dissing Java, the subtext is that the whole Internet has put a wrench into the works. Object models, messaging architectures, document formats, the architecture of multi-protocol networking, and, not least, the user interface are all brought into question by the Internet.

Why is the question of user interface interesting to telephony software developers? Part of our industry's problem is that, so far, user interface is a stepchild issue among the big issues of computer telephony. To take one example, the state of the art in interface to the on-screen dialer is generally shameful. A picture of a telephone on a computer screen? Is that the best we can do? Fake LCD displays? Heck, I've seen fake wood-grain on PC phone interfaces. Can an industry that created an icon of modern life, an artifact that is as emblematic a part of civilization as the water faucet or the automobile steering wheel, be so ignominious in its failure to translate the basic ideas of telephony into the computer user interface milieu?

The dawn of the Web as central focus of user interface ideas has brought us to a crossroads: one way lies Gomorrah, using the Web as an excuse to return to the bad old days of pictures of telephones on the screen. The other road leads to a shining city on a hill where for the first time in a century, we may have a fundamentally better user interface to the telephone. Viewed in the context of the number of telephones and their far greater historical span, the telephone user interface looms large among computer user interface problems.

Before we delve deeper into the telephone user interface problem, lets see where battle lines are currently drawn. Up to this point, the infrastructure of our desktop super-reality has been an increasingly powerful set of generic operations accessed by selecting menu items or pressing toolbar buttons. These toolbars became attached to the standard architectures of the underlying programs. In Windows COM applications, embedded objects could take along pieces of their "home" menus and toolbars and cunningly and complexly merge them into the menus and toolbars of the applications editing the documents in which these objects are embedded. This in contrast to the Web world, where designers slang about "blowing away the chrome," their term for the gray menus and toolbars that encrust documents like user interface barnacles.

Trouble is, so much of what passes for user interface design discussion these days is just so much "anti." Anti "chrome," anti desktop metaphor, etc. Add up all the anti and no wonder flitting from Web page to Web page seems at times regressive. Back to the bad old days of one screen of menus leading to another screen of menus. Some contemporary commentators on user interface go on about speech recognition, or 3-D, without realizing that it's not 2-D that makes interfaces the way they are. It's not even bitmap displays. The desktop metaphor is rooted in virtual coordinate systems first invented for CAD applications so that objects could be moved without disturbing their internal data about their dimensions. Until the would-be Web-based alternatives offer some real alternative to the logical structures underlying the desktop metaphor, it cannot replace it with something more powerful.

The hope of replacing the architectural underpinnings of the desktop metaphor might well be in vain. Let's look at the way humans have consumed information for millennia: marks on flat surfaces. Paper and ink, the objects modeled in the desktop metaphor are but a recent adaptation of a fundamental human communication form. So while modeling the office work environment might well be on the way out, it would be premature to think that marks on a flat surface, whether smudged by the thumb of a cave man or traced by electrons on a screen, will cease being our main non-verbal communication.

Post-literacy has been cited by some as the downfall of desktop metaphor. We will return to the tribal village of emotion and physicality, conveyed across space by our video images. Faugh! Even McLuhan who first elucidated the relationship among types of media did not think that new media and their evident enabling of the return of tribal structures and interactions was a good thing -- it was just a thing, a fact, and one he could explain uniquely well.

So stuck between the imperfect implementation of the desktop metaphor, and the ill-understood changes driven by the Web, the would-be designer of the next telephone user interface stands, far worse-off than before. What can he hold on to? What should he let go of? Where will the new world of user interface design lead?

Modeling reality vs. the document metaphor: This is first hurdle of user interface design in computer telephony, and one that few contestants overcome. Should a telephone user interface model the look of a telephone? The answer is yes and no, but clearly and resoundingly very much mostly "No!" (N.B. we cannot in this space answer the deeper question of whether a telephone user interface can be entirely put on a computer screen -- that is, whether the receiver should be as fundamental a part of the physical complement of a computer's user interface as the mouse and the keyboard.) But we can say a hearty "no" to pictures of phones on the computer screen. The telephone operates a network, creates a paper trail in the accounting software at the CO, and takes input from our knowledge of who we want to call, when, and at what number. These are the deeper fundamentals of telephony, not the keypad and especially not the feature-buttons of a telephone. Translate these deeper fundamentals into operations -- document oriented operations -- on the computer screen, and you are on the right track to a telephone user interface.

Windows look vs. Web look: Well, if we've cleared the hurdle of fake LCD displays and flagrantly breaking the generic-operation model of the computer interface by putting a telephone's button's on the screen, we now need to check which path to take. Do we translate the phone into a Windows COM program, with multi-tab dialog boxes and neat rows of toolbar buttons? Or do we step into the intimidating open spaces of Web design? The latter looks more risky, but consider how many things Microsoft has thrown overboard in their pursuit of dominating the Internet scene. Will Windows go "chrome free?" Will the Web kill dialog boxes that remained impervious to the desktop metaphor drive toward direct manipulation? My bet would be with the Web. Kill your toolbars. Slaughter your dialog boxes (and good riddance). Annihilate your gray beveled buttons.

In other words, do things the Web way for the right reasons. We should have the freedom to shape our user interfaces distinctively. But don't abandon the elements of the desktop metaphor that are rooted too deeply to be moved by the Web: The act of phoning people produces a document, a record of who you have phoned, a plan of who you will phone, and notes on what was discussed. Phone calls have relationships to other documents -- mail messages, address book entries, etc. A picture of a telephone on the screen captures none of these relationships. Do this right, and you have a chance at making a genuine improvement in the way people work.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #6

In this sixth newsletter, I can claim a reasonably prescient view of COM as being insufficient for competing against Internet standards. Surprisingly, .NET is not really pulling people into a Microsoft/Internet world, either. .NET really does deliver a suite of technologies for making the Internet an active network of connected software, rather than a very large hypertext web.

But .NET has not set the world afire. Perhaps it took too long. Perhaps it has too few widely used applications with high value. Whatever the case, nobody is saying "Wow, .NET changed the Internet."

Perhaps there really is nothing Microsoft can do to retain a position as utterly vital in the software ecology.

Newsletter #6: Diversity vs. compatibility

Some people think I'm a Young Turk. But compared with the engineers I hired at my last start-up, I'm a geezer. A running gag we had was that "When I was young I walked three miles through the snow to toggle in the boot load code on the front panel of a PDP-8." Back then, we had diversity -- lots of different hardware and software. Back then, end users often enough wrote their own operating systems. Back then, a VAX with 16MB was a honkin' machine. Back then, it was not unusual for a new computer company to create a CPU, a bus architecture, an OS, a language, and applications -- all on the R&D budget of a start-up.

Gradually, all that came to an end. The first beta release of Windows 95 signaled to me the end of history: We would all write COM objects, for Windows OSs, in C++, and hundreds of millions of x86-based computers all over the world would run this software. In a way, this was exciting. Adding tens of millions of customers each year to a uniform, standardized market for software is exciting by any measure. But while this looked like an ever-expanding opportunity, it also represented the end of a certain class of opportunity. In almost all the major categories of software used on business desktops, Microsoft stood astride the market. Sure, PeopleSoft, Baan, and SAP have the enterprise fiscal management software realm to fight over, but the world plainly is not waiting for the next great word processor, spreadsheet, business graphics program, page layout program, or other mass-market desktop software.

Then came the Internet. Now, before you say you've heard that one before, recall the suddenness with which things were changed. I would have told you, back when Microsoft first announced MSN, that AOL was doomed because the AOL technology was clunky and sure to be overwhelmed by the elegant, well-integrated MSN software. And I would have dismissed the Internet entirely. Who, after all, would take the trouble to learn all that Unix-oriented arcana? Who would tolerate managing IP addresses, telnet, ftp, and all that nonsense? And I stood in good company.

Now things are damn weird. Take COM. COM and DCOM have clearly trounced CORBA as a distributed object technology for Windows software. Writing COM software is not simple, but it is nothing like all the hoops you have to jump through for CORBA, and COM has the distinct advantage that Microsoft creates highly productive application frameworks for COM, and Microsoft's software development tools are ideal for writing this kind of software. And this situation continues to improve. COM+ is better, more powerful, and in many ways simpler. But do you need COM?

If you are writing Windows applications, yes you should use COM. And if you don't use it, you'll suffer. Your applications won't share information and capabilities with other Windows applications. You'll be an outsider wearing a loud tie and a bad suit in a very conformist club. But do you need COM to make distributed applications? That depends on who you are. Who endorsed COM+ immediately? PeopleSoft, Baan, and SAP. These are exactly the companies out there replacing the mainframe-based enterprise fisc with modern distributed systems running on NT. Good for them, and good for their customers. Good for all of us now liberated from ever having to consider buying a mainframe no matter how big our business grows. But what does it mean to the company writing an Internet-based personal financial management package? What does it mean to the company writing multi-player simulation games? What does it mean to anyone with fewer than 100 software engineers not writing a product which their customers will spend millions of dollars buying and installing -- that's millions each, not in aggregate?

Let's say you were designing a unified messaging system today. How are things different from the time of the Windows 95 announcement? How are things different After the Internet?

Before Internet: The customer would install your product from a CD disk, first at the server, then at the clients. A more-advanced system might make use of software management features of SMS.

After Internet: The customer installs the product on the server. Client installations are made by accessing Web pages on the server that contain a client software download, interactive instructions for installation, and a tutorial on using the product.

Before Internet: The best way to make a distributed system ready for big companies is to make it operate across multiple protocols -- IPX/SPX, NetBEUI, and TCP/IP. This can be accomplished by using RPCs or distributed COM objects. Both these technologies make it possible to write multi-protocol distributed software while containing the nasty details of networking in one place.

After Internet: Use TCP/IP only. Even if you are using DCOM or RPCs, you can assume the network is an IP network. Remember, we're talking about a new-generation unified messaging system here, which means the network will be chock full of H.323 clients and Internet telephony gateways, which means IP. Yes, you really can assume IP on everything, especially in telephony applications.

Before Internet: Document-level compatibility meant Microsoft Word. Or, if you wanted to play in the multi-application document integration world, you had to implement OLE linking and embedding. Compound document architectures like OLE assume that there will never be a single document format, so they must enable disparately formatted document objects, and the code that edits those objects, to be embedded in foreign documents.

After Internet: Just like there is one protocol, IP, there is one document format: HTML. HTML together with XML mean your documents and data can be read and modified anywhere. The result will crack the hegemony of Word in document editing and viewing.

Before Internet: Your messaging system should be a MAPI service, either directly connected to your messaging server, or integrated with Exchange Server.

After Internet: Your messaging system should still be a MAPI service, either directly connected to your messaging server, or integrated with Exchange Server. This is because Exchange Server will be Microsoft's Internet messaging solution, and will be very successful. But you should also include IMAP support in your server so Internet mail clients without MAPI support can access your server and present a unified messaging environment to users. And you should have a Web access option as well, where clients can access messages using only a browser. Did your job just get three times bigger implementing this part of unified messaging? Yes.

Before Internet: You had your choice of proprietary management systems -- SMS, Solstice, etc. -- with their own integration interfaces, plus some open systems that use SNMP.

After Internet: SNMP. None to their credit, Sun seems to be trying to foist Solstice's interface on Java, instead of providing the best possible open, SNMP-oriented management tools.

What can Microsoft do to continue to dominate the scene? One answer is "Nothing." Not everyone needs a distributed object and compound document architecture, and less so with Internet standards widely adopted. The other answer is that Microsoft could do a better job of marketing their distributed object technology: First it was OLE, then ActiveX (which gave us the ActiveBuzzword nomenclature), and now DNA, which sounds like a throwback to some DEC announcementware related to networks of VAXes.

One thing Microsoft can do -- and is doing -- is a good job adopting Internet standards. Microsoft is doing is a good job making Exchange Server a good Internet mail server, and Microsoft's Outlook Express is a good Internet mail client (though they did screw up their upgrade path by omitting MAPI support in Outlook Express). And not least, networks of Windows PC and NT servers support TCP/IP very well. To say that Microsoft is doing well adopting Internet standards may sound odd in the face of the tiff over Java, but Microsoft's general advantage in the industry is that they execute better than their would-be competitors.

Execute better than your competitors. Microsoft, which has often enough been wrong about the first release of a product, doggedly pursues top-notch execution. And they do it even if it did not work out right the first time, instead, as is often the case in the industry, a weak start dooms a product to insufficient follow-on investment. What the Internet has done is open the door to new opportunities, new companies. But Microsoft's formula for success still holds true: execute well, and then, when you are done with the first release do not -- do not even consider -- allowing the product to become stale before you aggressively upgrade it in anticipation of, not in reaction to, feedback about improvement.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.
.

Telirati Newsletter #5

In this, my fifth newsletter, I take aim at a fat target: The New PC. In this case it was the "Simply Interactive PC." Remember the "Simply Interactive PC?" Nah, didn't think so. But it did suck a lot of mind-share and ink in it's day. Many lessons for today, as well...

Telirati Newsletter #5: Playing chess

I can swing a golf club. I don't know how to play golf. Just because you know the rules of chess, does not mean you can play chess. Newsletter #4 was about visualizing, and how you can use visualization to understand the distance between the present and the desired future. This newsletter is about playing chess, and a different kind of vision - chess vision.

Chess vision means you can see what will happen. You can see moves ahead. You can imagine not only a sequence of moves, but also how the board looks when those moves are made, so you know the outcome and can play for the best outcome. To be an adequate chess player, you need to see four moves ahead. Rated chess players see far deeper.

There are leaders in this industry making decisions without looking moves ahead. Here is an example: Microsoft and Intel think of a great idea (actually, Steve Jobs though of this great idea around 1982, but hey) called the Simply Interactive PC. It is a sealed box. It has two external busses, one fast, and one very very fast, that eliminate the need for almost all internal expansion capability in desktop PCs. It is an information appliance that can be packaged attractively for people like my mother, who is currently better off with her old Macintosh. In fact, the Simply Interactive PC is a Macintosh - the original Macintosh, before Apple lost its way and made the keyboard too big, the boxes ugly sheet-metal contraptions, and the OS over-complex (but that is a much longer and more woeful tale of lack of vision).

The Simply Interactive PC would revolutionize the peripherals and enhancements business: instead of having to open your PC, you would put the simple, medium-performance devices like modems on one bus - the USB bus - and fast peripherals like DVD drives and digital camcorders on the other - the 1394 or FireWire bus. And you can "hot" plug and unplug to your heart's desire. This would greatly expand the market for enhancements and enable innovative new uses for PCs, along with the hardware and software enhancements that go along with these new uses. Home video and music editing and other creative applications were high on the list of possible new high-growth areas. More important, another tier of customers would be added to the market, those that find current PCs a noisy intrusion of a complex office machine into their homes.

What happened to the Simple Interactive PC? It turned into the flop du jour: the NetPC. With a crude false attack, Sun lured the PC world off its plan to fundamentally expand the PC market. Sun's JavaStation and the even more unicorn-like NC from Oracle (a hardware maker?) stampeded the PC industry into defending its business desktops from this unlikely incursion. The Simply Interactive PC, the PC for my mother, became the defender of the business desktop.

That's what comes of thinking one move ahead: Sun and Oracle posit a device, and the PC industry moves, with a heavy hand, to squash it. Trouble is, like chess pieces, technology strategies unfold in several moves. Microsoft and the PC makers, seeing an attack, failed to see the intermediate moves, failed to see the weak ending position the attack would have, and in the process incurred an opportunity cost: all the millions of people who would have bought a Simply Interactive PC, and all the nifty toys to attach to it, have none to buy. Apple was left breathing room. My mom is left with her old Mac. I don't have a good excuse to buy a DVT camcorder ("Honey, we'll want to edit down those tapes of the new baby for great-grandma."). And we're all waiting for smart set-top boxes when the network interface and decoder hardware for all that fancy digital cable should all just be attached to the 1394 bus of one of these Simply Interactive PCs, and that the set top box should not be smart at all.

Meanwhile, on the other side of the chessboard, Microsoft is trying to stuff the Java genie back into the bottle. Unlikely. Every muffed opportunity has a cost that spans the chessboard: What if Simply Interactive PCs were out in the market now, connecting to set-top boxes, providing Internet connectivity in homes, and, probably, spilling into the parts of the business market where they were really needed? How much better positioned would Microsoft be to promote ActiveX as the power behind making the Internet more interactive?

Chess, and business competition, is all about delayed gratification. If you have no patience, if you are prone to first-order analyses, you'll never see even four moves ahead. But even if you do have the discipline to see four moves ahead, will the big companies who can hire firms like Bain and McKinsey to do the looking ahead for them still beat you? Don't worry. Big companies are run by gangs of vice presidents waiting for other vice presidents to make a mistake and back-stab them. It's hard to look ahead and watch your back at the same time.

And now, it's time to play computer telephony chess.

Move one: Create the UnPBX. Put a switch fabric on an add-in card and you have a telephone switch inside a PC server. Sell it to a customer.

Move two: two years from the time the customer installs the server, the server is looking long in the tooth. In three years, the server needs replacing. The new server has only one ISA slot, the rest are PCI, and that UnPBX is implemented on four (?!) ISA cards. Ow! Later that day, while grousing to a friend over a beer, customer learns his friend has a 15 year old PBX, and no plans to replace it.

This is a bind I've mentioned to a lot of my colleagues, and the best answer I've gotten so far is "Yeah, well, uh, people will start replacing their PBXs the way they replace their servers." Right. PBXs are more like plumbing than computers, despite the superficial resemblance. Imagine your plumbing incorporated a PC.

Look at this, instead:

Move one: Customer with conventional PBX, and a network of PCs, wants to use Internet telephony.

Move two: vendor of PBX enables customer to remove the control processor of the PBX and replace it with a link to an external server over - pick one, depending on your preference and the system size - Ethernet, ATM, 1394, or USB. Now the external server runs the PBX, the control software of the PBX having been ported from whatever grotty old OS the PBX maker had been using to NT. The server also has access to the PBX's internal bus, and can "export" and "import" a large number of slots in the switch fabric (total depending on link technology, but could be as high as the total bus bandwidth of the PBX) to the server from the network and vica versa. So the server is also a gateway from the phones on the customer's desktops to the Internet.

Move three: Three years later, customer upgrades server. Now the server can do more: gatewaying voice calls from telephones into online collaborative conferences, simulating a conference bridge, fax in software, etc. Customer cuts over to new server with a less than one-minute outage - the time it takes to unplug and plug in the link cable.

Move four: Two years later, customer wants still more: adds a DSP resource inside the server for more faxing, more conference bridging, etc. Does he need a separate bus inside the PC? No, this new server has 64 bit fast PCI, and dozens of voice channels are a drop in the bandwidth bucket. Later that day, over a beer, customer makes his friend with the 15 year old PBX insanely envious of his operating efficiency.

Am I Altigen bashing here? No. I want an Altigen. It's cool. I like it. I'd use one knowing full well that I'd toss it in three years. But Altigen themselves have to consider what happens to the bulk of their customers without my sense of adventure (and I still wouldn't put a PC in my plumbing).

PBX makers have to realize that their product formulation depends on them having the right chess vision. Misadventures like in-the-skin PCs, PBXs in PC chassis, etc. are wild guesses, seeing no moves ahead. Seeing the product lifetime differential between PC servers and PBXs is the first step to seeing far enough ahead to deliver the right product. What if the Lucent Partner small business switch came with a USB interface, a built-in voice processing resource, and a Unified Messaging software package you could install on any PC that you would like to attach to this system? Such a configuration would put that hoary old epitome of conventional small business switches right on top of real customer needs.

Some other examples of not seeing moves ahead:

Move one: Apple offs their clone vendors. They might be right that they are subsidizing those vendors, but the numbers are open to other interpretations as well.

Move two: Loyal customers, fanatically loyal customers, get really pissed-off. Did anyone calculate the value of all those Mac fanatics - people who had been actively urging people to buy and not pirate the most recent software update? You cannot buy brand loyalty like that. You can only spend it. Wasted.

Move three: Motorola, Apple's CPU vendor says: "We're probably out of the desktop CPU business." Holy supplier risk! Apple better be very sure about the timing of the transition of their customers to Next technology. They may face a very bleak period where they bleed market share due to un-competitive CPUs.

Move four: Apple launches Rhapsody, and licenses it to… Yeah, right. Want to see a venture capitalist die laughing? Hand him a business plan for a Rhapsody-based desktop computer product.

What to make of this? Steve Jobs is very intelligent. So it is possible that he has a very obscure plan to win. Trouble is, plans for winning are usually not obscure. They are usually simple, like Wal-Mart using computers to keep useful things in stock. Or Ford using TQM to reduce defects in cars. These are very simple strategies. The chess vision involved in executing these strategies is in the fact they take multiple moves to implement: Wal-Mart had to get their suppliers to change the way they do business, and not all were ready to do so at once. Ford had to take it on faith that Americans, who, the numbers plainly showed, bought lots of crappy American cars in the 70's, would respond better to quality. What Apple should have done, and may not have been able to, under any circumstances do, is face the fact that they would have to ride out their current licensing regime until they could replace Mac OS with Rhapsody.

What is the value of honing your chess vision? It gives you the ability to put in place and defend a strategy that takes multiple steps to implement. Remember that the strategy itself is usually very simple. The multi-step execution is where people get lost. Especially when put under pressure: "People love vinyl tops, 'opera windows' and fake wire wheels. Look at all those cars we sell!"

Look ahead and prosper.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #4

In the following, the fouth of my Telirati newsletters, I rake speech technology and IP telephony people over the coals for their lack of visualization of how these technologies would actually be used by customers. That was in 1997. The situation today is scarcely better: Speech recognition is, as predicted, better than ever. It is, however, just as distant from being a viable tool for interacting with a computer, or even with limited computing devices like handhelds or smartphones. The reason it has not progressed is as it ever was: lack of a gestalt for a voice user interface that is comparable to the desktop metaphor.

Even IP telephony is nearly as badly off: Cisco sells a decent IP phone, as does 3Com. And after that, well... I'd have to Google it up. I have no idea who else sells IP phones. One would have hoped by now VTech or Siemens would sell a cordless phone that plugs into Ethernet, and sells with a software PBX. Still science fiction.

No doubt this pattern of an inability of the telecom industry to overcome the obvious will play out over the rest of the archive of Telirati Newsletters as I dig them up and dust them off. And that is one of the big reasons I enjoy the game business so much. In wireless mobile games, especially, the industry changes, in big ways, from day to day. Some changes are good, some less so, but the field is so large and dynamic that something will happen in a commercially interesting way, somewhere in the world.


Newsletter #4: One of these days…

I'm a firm believer in the power of visualization, or at least the discipline of it. I used to work in speech recognition. There is a field full of vision, but little discipline of visualization. On the BBC today, there was a story on IBM's new Chinese dictation software, and a nice woman from IBM was enthusing on how this was going to be revolutionary, because typing in Chinese is such a bother. First thing I noticed, the woman was French. The BBC reporter, probably having heard "Speech recognition, this time for sure!" before, asked the woman if any Chinese users had been consulted as to the usefulness of the system and she assured him IBM has a lab in China dedicated to all Chinese-specific aspects of the project. Now I neither speak Chinese nor have I used any of the keyboarding tools available for typing in Chinese, but I do know some things from my experience in speech recognition:

First, you can't talk all day. My voice is a bit hoarse right now from reading to my daughter for an hour and a half because she wanted to plow through all the way to the end of C. S. Lewis's The Voyage of the Dawn Treader tonight. Talking to my computer all day would be more tiring and, in addition, tiresome. As unnatural as a keyboard is, a speech-based user interface just trades one perversion of nature for another.

Second, assume you have perfect speech recognition - how much of the speech user interface problem have you solved? Not much. It's a bit like saying the fellow who invented the mouse invented the modern graphical user interface. Or the inventor of bit-mapped display. The graphical user interface was an interdisciplinary solution that wove technologies and disciplines as disparate as CAD drawing systems, typography, modeling, and visual language together into a solution.

The development of the GUI must have been guided by visualization that knowledge workers manipulate documents directly - by typing them, writing on them, moving them physically in ways that organize them. This immensely successful and fundamental development in user interfaces has become so ingrained that we do not think about it. In fact, there has been little intellectualizing about what the Web means to user interface design when it has, in effect, liberated users from button down (no pun intended) models of physical papers on desks in offices to a world that resembles colorful magazines. In this new world the previously uniform user interface language of gray buttons of a certain exact size, all a certain exact distance form one another, etc., etc., is replaced by a world of few if any conventions. If the cursor turns into a finger that seems ready to click, go ahead and click, and see where it takes you.

So if visualization has not saved speech recognition from the perennial proclamation of Real Soon Now, how will it help telephony developers? In my pervious newsletter I described a problem with Internet telephony - lack of terminal devices. No lack of computers, just that computers make lousy phones. Let's visualize:

I have a line I use for a modem connection, and a line for voice calls. The Lucent speakerphone next to my PC can access both. I would prefer to have neither, or at least be able to scale back to the one voice line. My computer should be on a high-speed Internet access device. So far so good, MediaOne will probably be in my town soon enough. But there's a problem with that: Current MediaOne consumer-grade access allows for only one device, with one IP address. That, as we shall see, is a problem.

But back to the phone set for a moment: I like my Lucent speakerphone, but it would have to have better audio in order to be my PC audio device as well. A headset jack would be nice, too, not just for phone calls, but for PC audio use as well - so make it a stereo headset, with a separate mic jack. Now to connect the "telephone set" to the PC - USB, please. No phone line to the phone? Nope. But that means the PC has to a) have a two-line "modem" device, or b) make the phone itself an Internet phone, and put routing software in the PC it is attached to (and routing in the set-top box to get to the cable-based internet access).

My house has wiring that is grotty old phone wiring, installed by the pot-addled hippies who build the place, who economized by using ancient four-prong sockets. If I want to upgrade, I'll have to saw bigger holes in the walls to start with. Ick! Think wireless: to begin with, make the receiver on the telephone set on my desk a wireless handset instead.

Now visualize this: I get a phone call, I pick up call on the speakerphone at my desk. My wife wants to make another call, in another room, because all this consultant talk gets on her nerves. She takes the wireless handset, flips it on, and dials her call. Underneath this simple activity, this is what is happening: The phone's base station starts a second Internet phone call, which the PC routes onto the household LAN, and which the set-top box routes onto the cable company's line.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #3

In my third newsletter I had I grown to take a skeptical position as the core of the piece, providing the reader with a gauge for measuring the viability of a technology that has long held promise, and still does. Again, however, we see that hype of a 1997 vintage is still being poured today from new bottles.

As with my previous retrospective, it's still hard to say what to buy and for what reason, in CPE or in infrastructure. The vitality of VoIP all comes from alternative applications like voice chat and the implementations of push-to-talk in non-iDEN mobile networks (which is an intriguing example of how an application like voice chat that was deemed too low-quality to warrant notice by telecom people comes back around to influence a significant telecom technology).


Newsletter #3: IP telephony is here when…

IP telephony is coming, no question. But the real question is "When is it close enough to my products that I better take it into account in my next product generation?" Not everyone is making IP telephony products - products that enable IP telephony directly. But every telephony product will be affected by IP telephony. So how do you get a weather report? What do you watch to tell how close it is to your market and your product? Here are some things to look for on the horizon:

IP telephony is here when…

A major wireless service provider builds an infrastructure based on IP telephony.

Wireless service is at the wave front of providing telephony where is has been non-existent or sub-par. The newest infrastructure, in the most challenging environments, is wireless. By a "challenging environment," I don't just mean a jungle or a desert, I mean countries where the per-capita GDP is less than your cellular bill for a month. Cost and price are as big a challenge as snakes and alligators. If aggressive use of new technology cuts costs, wireless providers will try it. A key indicator of the readiness of IP telephony will be the use of IP telephony to reduce the cost of a wireless telephone infrastructure. Alternatively, you might see IP telephony in a wireless infrastructure geared toward providing Internet access as a key selling point.

A significant long distance carrier runs a major part of their network packet-switched.

IP telephony is the new frontier of the brigands of telephony - callback operators. They have decided to hoist the Jolly Roger and sail out in pursuit of the domestic telcos' galleons, having brought home enough booty from overseas PTTs to make them bold (and hungry for the next big kill). Their entry into this market will test the technology of IP telephony. It will also test a number of other things - all of them important markers that show how real IP telephony is: Can they runs these networks with acceptable reliability? What will billing look like in the Internet telephony domain? Will corporate telecomm buyers use these services as readily as they use callback overseas? Can these buccaneers with their risky technology learn how to market themselves as respectable?

You can buy a $200 telephone that runs IP telephony.

The IP telephone is really a set of solutions to a very infrequently mentioned problem in IP telephony: the terminal. Telephones are good voice terminals. Computers are not. Telephones are an accepted and well-tuned part of how people in a business setting communicate. Computer multimedia devices are not. My mother can use a telephone. I won't hold my breath for her to use any current IP telephony technology. What the world needs is both a stand-alone device that connects to an Ethernet jack and provides what looks like normal telephony to anyone who lifts the receiver, and a multimedia i/o device for computers that is as nice as a good speakerphone handset from an established CPE maker. Some makers of network telephony solutions enable users to plug a phone into their expansion board, which is installed in a PC. Windows can manage multiple audio devices reasonably well, and this is about as well as can be done at this point, but it is a far cry from being as good as those really nice speakerphones Nortel makes. You can't make a 2500 set hooked up to your computer auto-answer, you can't use the programmable buttons or the "hold" button as you would normally - it just isn't like using a phone. And then there is the problem that opening that many PCs and installing that many boards is a non-starter. Ordinary modems, though almost all have both a phone and a line jack, cannot be an Internet telephony i/o device - that jack exists only to make the phone usable on the same line as the modem. This is not a minor quibble: I can get PC multimedia devices and telephones at every office supply computer store, and consumer electronics store in the country. I cannot get a device that serves my PC multimedia and telephone needs in one package anywhere. The solution to this problem is interdisciplinary and difficult. But lacking a solution to this problem means IP telephony is not ready for prime time.

A call center, for both Internet and telephone-based callers is built in the IP telephony domain.

Here is a nicely constrained problem: You are building a call center. Everyone has a PC on their desk. Everyone wears a headset all day. Since it is a new call center, you are quite free to specify audio cards for the PCs that are well suited to Internet telephony, as well as every other attribute of those PCs. You have control over the network. You know exactly what will be moving on that network - you can calculate worst-case traffic down to the bit. You choose the NICs, the hubs, the switches, the routers. Why then should you not do away with the cost of a call center switch, switch-to-host integration, etc. and go straight for IP telephony for handling all calls, whether they come from the Internet or the PSTN? How convenient for your server software to talk directly to the IP telephony software, inside the same server, or at least inside the same OS on the same LAN, with all the OS's inter-process communications capabilities easing the way. One OS, one wire to each desktop, one network, one management system, one software technology for implementing every component - all very attractive, and all likely to save you a bundle. Of course, if you were really doing this, there would be hell to pay if it did not work. So it makes a nice barometer: when you see someone stick his neck out and build a call center this way, IP telephony is on your doorstep.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #2

So here comes the second Telirati newsletter, and it is further evidence that skepticism seldom goes far enough. In this case, I predicted the decline of conventional telephony CPE, and its replacement with telephony servers. However deep that pessimism was, and the decline of that industry is as deep as I could have imagined, the full measure of pessimism would have also encompassed the possibility that no replacement would emerge.

What do we have today? Lucent and Ericsson shed their dowdy CPE businesses on the way to bubble-economy doom. Nortel's CPE hobbles along, and none of the big iron CPE-makers have made a dent with VoIP systems that don't change end-user costs significantly. The answer to "Who will replace legacy PBX-makers?" appears to be "Nobody."

I don't even know what I would buy if I were shopping for a PBX today. Cisco's VoIP CPE? It's mature enough, but no bargain. If cost matters, I suppose a quality used system would be the first thing that comes to mind. All the horrors of CPE costs - voice mail storage rip-offs, add/move/change fees to retailers, etc. are still there. The replacement of business CPE with mobile wireless service is only in its infancy in Europe, and the sight of an Ericsson receptionist using her mobile handset to handle calls to an attendant station confirmed my suspicions that more work is needed.

So here it is, the second Telirati newsletter, and another lesson that pessimism and skepticism can hardly ever be too deep:

Newsletter #2: Telephony Servers

Two words: segments and alliances. Without thinking about the telephony server business in terms of segments and alliances, it is overwhelming. With this approach, the problem of creating a viable business in telephony servers becomes manageable.

First, let's think about it the wrong way: I'm going to build a telephony server. S.100 seems like a good place to start. I'll just hire a bunch of engineers to fill in all the blank spaces in S.100 hardware and software implementations. OK, but my telephony server will probably run on a Windows NT machine, which doesn't use S.100 for installing, identifying, controlling, and interfacing software to telephony systems - it uses TAPI. More engineers and I'll have a TAPI-to-S.100 layer. Now for the applications. I'll need unified messaging, an attendant position, call queuing, ACD, IVR, fax, pager notification, call accounting and nice Internet telephony gateway to top it all off. That'll be 200 man-years altogether, give or take. Heck that's way smaller than a good-sized switch project used to be - why are you all looking at me like I'm crazy? I'll see you in three to five calendar years with a fully formed product. Good day.

Some people think that way. Some other people want to make money this year. "Internet time" has replaced the glacial product time-scale in the telephony industry. Segments and alliances help get things moving. By identifying segments, you can identify the best approach to creating the product, to selling it, and to finding the right partners.

Let's start at the low end:

Hardware: It'll be new, high volume, and cheap. Don't worry about the installed-base. Don't worry about standards other than TAPI. To be cheap, the hardware is probably a single board, probably made by the company integrating the product. But single boards inside PCs are not the be-all and end-all. Every variant from radical new interpretations of KSU-less key-systems to mutations of key systems that enable easy connection to PCs will flower in the low-end. There are lots of ways to cost-optimize.

Software: TAPI is the standard to obey. Here's why: the hardware at the low end is going to be a single-board, no separate voice processing resource or network interface - all in one. Software makers can't count on the Dialogic or Rhetorex proprietary APIs to be there, and unless they want to go nuts interfacing their products to a bunch of different hardware from a bunch of companies attacking the low end, most of which will be out of business in a year, the software and hardware people will have to agree on TAPI. Unified messaging is the most important software to bundle, with Internet telephony gateways close behind.

Channel: VARs and OEMs. OEMs are finicky - use them to figure out if your product is good enough. Does it install and uninstall cleanly. Does it look right next to Microsoft Office, or is it a specimen from the User Interface Museum of Horrors?

Partnering: Nobody can do this alone, but it's a balancing act. At the low end, the product must be inexpensive. Altigen , for example, focused on creating a software bundle encompassing the basic requirements of unified messaging so they would not have a royalty burden there. Altigen probably shouldn't roll their own Internet telephony gateway. Partnering with VocalTec might make sense there. Low-end fax sharing from Microsoft is good enough, and the price is right. Would an OEM take this product? Picture this: major clone maker ships telephony servers that include everything a small business needs for Internet connectivity, e-mail, voice mail, fax, and you hook your phones up to it. Easy enough to sell mail order. Not quite there yet, but tantalizingly close.

That's the low end. Low cost, high volumes characterize the low end. A lot of the customers are new or growing businesses, and their capital investment in a small business phone system is small enough to be replaced. Their installed base of computers is also small enough to be easily upgraded. Sounds like nirvana except that these are picky customers, too. They have no time to mess around with these systems, and nobody on the premises dedicated to taking care of them. These customers will measure products against off the shelf, mass market software and peripherals. Does the telephony server install as easily as a Zip drive? Does the user interface look as good as Outlook? Now the mid-range segment: The key differences here are that the customer won't throw away what he has, but he'll pay enough for the system to allow for more partnering, a more component-oriented design, and more support.

Another characteristic of the mid-range is diversity. Flexibility in product formulation in the mid-range breeds diversity. S.100-based servers, and servers that ignore S.100. Servers based on NT, servers based on Unix. Servers that use Exchange Server for unified messaging, servers that use Internet mail standards. Servers that put switching inside the PC, and servers that put the PC inside the switch. ISA bus, PCI bus, Compact PCI, and none-of-the-above, in combination with SCSA or MVIP.

Here are the hardware, software, channel and partnering possibilities in the mid-range:

Hardware: Modular. At the low end of the mid range, you will find multiple ISA-card products, consisting typically of a network interface, a switching fabric and station interface card, and a voice processing resource, or, more generally, a DSP resource that does voice processing, fax, and data modem. SCSA or MVIP bus connects the cards. All this running under Windows NT. Moving up, you'll find Compact PCI for the modular systems, and for the PBX-centric telephony servers you'll see "in-the-skin" deployment of pre-configured computers, mostly PCs, mostly NT, but more and more UNIX as you go up-market.

Software: More layers, more standards, more choices. S.100 will gain traction in the mid-range, but so will servers based on TAPI as the primary application layer. Two major camps will be Microsoft-centric, and Internet-centric. Customers will want a full range of choices in all application areas. The cost of these systems will be high enough to support a reasonable level of integration and support, so that a locked-in suite of applications is not needed. Indeed choice is required, since the customer may have parts of a telephony server, such as a conventional PBX, a fax server, an e-mail environment, in place and will be unwilling to replace what he has.

Channel: Flexibility and diversity reign here in the mid-range. Everything from one-stop shopping where a technology giant might provide almost all the product content and delivery on its own, to completely ad-hoc integrations by mid-sized resellers will be seen in the mid-range. The locus of system integration will move around, depending on customer need and market segment.

Partnering: Lively partnering will characterize the mid-range. More products are ready for this market, and more combinations are possible in the mid-range than in the low-end or the high-end of telephony servers. This is where the flexibility required by customers plays to suppliers' advantage. Those willing to stick to their core competencies and partner most effectively for the rest of the system will get to the business first.

Low-end telephony servers give the little guy big telecommunications capabilities at a friendly price. And in the mid-range, telephony servers will replace conventional PBXs deployed in businesses with more powerful, productive, and flexible systems. But what is a high-end telephony server? Who uses it, for what? Picture this: Your cable company launches a broadband Internet service. At their head-end they will, for a reasonable fee: Take phone calls originated from your PC or your household phones, carry them over their network locally, and connect directly to a long-distance carrier at their facility (perhaps even bypassing the long-distance carrier); Eliminate the need for a conventional modem for your PC to make data or fax calls (and, by the way, enable you to receive in-bound faxes without using a modem or tying up a phone line); Back up your PC's data; Take your calls when you do not answer, and deliver the messages by e-mail; Enable you to fetch your e-mail from any telephony and have it read to you, or faxed to you; Provide you with a single, unified identity in both switched-circuit and packet-switched worlds so that people can reach you by whatever means is most convenient to you. All this in addition to Internet access. All this is the function of a high-end telephony server. Of course, the phone company is up to much the same thing, only it will use xDSL technology over the phone line.

At the high end, these are the hardware, software, channel and partnering possibilities:

Hardware: Customers are big ISPs and LECs. At this level, the difference between the way ISPs view the world and the way LECs at it can make all the difference - the difference between needing to meet NEBS standards or not, for example. ISPs are more likely to buy off-the-rack, and to look for what computer people regard as standard. LECs have their own world of standards. The trajectory of hardware has moved from the cost-optimized at the low end, to the flexible in the mid-range, to the customer-specified in the high end.

Software: The Internet didn't need to bring it back - this is where UNIX never left. Though NT, as everywhere, is growing in importance. The biggest PC OEMs have carrier-class telephony servers in their sights. This is in no small measure due to the fact that systems in this range are customer-specified. One consequence of that is that computer people, who have had trouble formulating products for computer telephony can rely on the customer in this segment to lead the way. This is almost the exact opposite of the situation at the low end of the market, wherein the customer may not know or care what a telephony server is - they just want solutions and convenience.

Channel: Direct. You can write all the customer names on one side of one sheet of paper. On an index card if LECs are the target market. Any significant deals, especially for innovative products, are an executive selling situation.

Partnering: Partnering at the high end is situational. Many sales will be made in response to formal requests for proposal. So partners are picked - and sometimes dictated - by customer demands. As in hardware selection and software, some vendors will find the high-end more congenial just because the customers here are much more likely to know, in detail, what they want. Companies on the outside of this market can make use of partnering to become insiders.

Those are the market segments. Each has its own characteristics. At the low end, producing a telephony server means taking an understanding of telephony and coming up with a product that is much like a computer peripheral. It has to be cheap, easy to install, and virtually maintenance free. At the high end, it’s the telephony that matters, and the considerations that go into creating telephony servers at the high-end are nothing like those that go into successful mass-market computer products. But no matter if a telephony server is low-end, mid-range, or high-end, this is what a telephony server does:

A telephony server provides dial tone to handsets.

A telephony server connects handsets and Internet conferencing clients to switched-circuit PSTN trunks.

A telephony server enables computers to look like extensions on a PBX.

A telephony server enables handsets to call in to Internet telephony endpoints.

A telephony server provides a platform for the full range of telephony applications: unified messaging, call-center functions, remote-access, Internet router, fax server, and personal productivity applications.

A telephony server is, to sum it up, a way of blending and sharing packet switched and circuit switched resources so that the users of a telephony server get their work done without having to think about sharing, using, balancing, and optimizing those resources themselves. It still is an application platform, but the role a telephony server plays in blending circuit-switched and packet-switched network capabilities and uses is now its most important aspect. Sharing resources and blending utilization have been and will remain the foundations of value in communications products. This fact has moved IP telephony gateway technology into the heart of the telephony server. Even if the telephony server sits outside the router and outside the PBX, its job is to make the usefulness of the router and the PBX look seamless to the user.

Copyright 1997 Zigurd Mednieks. May be reproduced and redistributed with attribution and this notice intact.

Telirati Newsletter #1

Over the next few weeks I will be posting my old Telirati newsletters. This is a potentially embarrassing exercise in digging up old predictions, but since the Internet never forgets anyway, it's best I conduct the inquisition myself.

So here goes... This is the first of the Telirati newsletters. It dates from 1997, and talks about unified messaging. Fortunately for my prognostication cred, it took a skeptical view of proprietary email as a basis for unified messaging. Skepticism, however, was not carried far enough, and even the Web-based unified messaging I predicted did not come to pass.

What does that say about my current pursuits? It says, at the very least, that I should pay attention to the skeptics and to my own sense of the odds.

Here is what I said back then:


Newsletter #1: Where to now?

Microsoft revolutionized PC software with Windows95. It brought 32-bit power to all PC software. Microsoft also brought comprehensive standards to messaging, networking, and telephony. This made sophisticated unified messaging systems possible: using Microsoft standards, you can mix and match messaging capabilities in e-mail, voice, fax, and pager media. The result is that, although the market penetration of unified messaging has not yet burned up the voice mail business, that fire has been ignited, and is inextinguishable.

When you read of, for example, Lucent's buy-out of Octel, which had been the leading independent maker of voice mail, you read nothing of the billions of dollars of Octel hardware and software in the installed base of systems providing conventional voice mail. You read of unified messaging, and how both Lucent and Octel have leadership positions in technology in this domain. When you think of Active Voice, you think of their innovative TeLANophy unified messaging system, even though their Replay and Repartee conventional voice mail systems make up most of their revenue.

A consensus is in place: unified messaging is the way to do store-and-forward business communications in voice and fax media, which had up until now been separated from e-mail, the other main store-and-forward medium. The first wave of unified messaging technology is also in place. Octel's Unified Messenger, the NT-based version of TeLANophy due out soon, AVT's products, and other systems take Microsoft's standards and create widely deployable systems. Unified messaging is now possible in most corporate network and PBX environments.

That leaves the question "What next?" Having struggled up to that first level of implementation, which was fairly clearly marked by the standards Microsoft set, where do makers of unified messaging systems and other computer telephony software go next? These are some of the areas where the action will take place:

1. The shifting importance of standards. Microsoft's standards in networking, messaging, and telephony will be eclipsed in importance by Internet oriented standards. "IP on everything" means the complex multi-protocol networking facilities of Windows RPCs and DCOM will be less important than Sockets and Java client/server systems. You will be able to count on the presence of a Web browser, TCP/IP, and a Java runtime environment on everyone's desktop (palmtop, smart phone, set-top box etc.) than you will be able to count on Windows 95 or NT being there. Just the fact that Web browsers, TCP/IP, and Java capabilities are easier to put on a 16-bit Windows desktop than it is to upgrade that desktop means "IP on everything" is the future of networked software. The MAPI standards will also give way to IMAP, NNTP, HTML forms, and Internet mail servers as the basis for store and forward messaging and groupware. Finally, while TAPI will continue to displace proprietary APIs and grow in importance in the near future, eventually JTAPI will challenge TAPI as the basis of telephony applications.

2. Conferencing, which encompasses Internet Telephony, is an emerging medium. As conferencing capability becomes ubiquitous -- it is already part of the Internet Explorer bundle -- there will come a need for software agents that act on your behalf when someone wants to conference with you and you are unavailable. The rich, multimedia interaction of real-time conferencing will require a companion that is to it what voice mail is to phone calls. Multi-party conferences will also require a bridge from real-time to non-real time interaction in innovative ways. Some analogies between voice mail as a store and forward medium for voice calls and groupware as a store for conference transcripts might be drawn. The winner