[rant] An essential part of the domestication of the personal computer happened in the early 80's at Moscone Center in San Francisco. As a teenager with limited Apple Basic skills, the room was filled with unlimited possibilities... 10 megabyte hard drives, 640k memory cards, high resolution color, touch screens, and voice activated computing.

Looking back at the list, it is amazing to see the huge advances made in some many of those technologies, but what is more fascinating is too understand why some of them never grew beyond their original niches.

Voice Computing: It seems that every year I go to Comdex, there is some young guy sitting in the IBM booth with a BETA version of OS/2 equipped with voice recognition. I let him give me the standard demo of him dictating an email to a co-worker and the accuracy is always impressive. I follow with a request: "How 'bout if I try?" and I always get the same negative answer.

The best recognition I've see gets approximately 60% of my words when I speak much slower than I type. If I get ahead of the voice parser, I'm invariably misinterpreted and end up using the keyboard to go back and manually edit the mistake -- the whole process doubling the time to write a simple three paragraph e-mail. Voice recognition stalwarts retort: It's getting better every year.

It's been 14 years by my count, people. That is not progress, it's denial.

Touch Screens: Touch screens were a hit at user conference during the early 80s and popularized in movies like Tron when it was realized that you could completely rid yourself of your keyboard and make your desk one big touchable monitor.

The rise of touch screens never peaked above the occasional info kiosk at the airport because the feedback was never there. A common illustration of this problem is: What would happen if the elevator button didn't light up when you pressed it? The same issue arises for touch screens -- how do we visually convey that the correct part of the screen has been selected?

Flashing or audible clicks often help, but we've grown to love our clicky keyboards, the physical movement reassurances us that our action has been noticed -- if you don't believe me move to the upper right-hand corner of your screen and click away.

[rant] The plight of the computer science major at The College on the Coast was that they were had to learn Unix. For undergrads, you had to make sure your Pascal would compile on Unix. For upper-division work, your C++ programs had better work under Motif. Finally, when you left to enter the "real world", everything was Windows... thanks for all the prep work, U.C. Santa Cruz.

Unix continues to exist because it's 20 year old technology that is extremely stable. Even Microsoft continues to run much of it's site using Netscape Unix web servers for the very simple reason, it just doesn't crash. So, why isn't there a reliable Unix box on every desk?

The answer: user interface. Since the early Mac era, we've come to expect that a computer shouldn't be hard to use. Anyone who has sat down at a command prompt of a Unix box and tried to be productive usually ends up calling the one Unix guru in the building who demonstrates via a series of archaic commands how to "get things done" and promptly vanishes -- leaving the employee with the unenviable task of having to reverse engineer what the guru just did.

When the UltraSparc ended up on my desk, it was the first time I had MIS install a computer for me. When I wanted to get Netscape running on that machine, it took three phone calls to figure out exactly how to GUNZIP and UNTAR the application. When the application launched, I figured out why any platform supported by Netscape still has a chance.

At least in the Solaris version of Netscape, the application is functional clone of the Windows version. This means within a few seconds of sitting down at the Sparc, I was productive because I didn�t need to learn a new interface.

A question: How is surfing the web productive? The answer: Web applications. Currently, Java remains too immature and too new to provide the complex applications which will replace our spreadsheets and word processors, but the first wave of web applications such as contact tracking systems or phone number look-ups are already available and they�ll work on any version of Netscape you want.

Bill Gates is big this idea of minimizing the time to productivity except he likes to put in such a way that corporations will understand � zero administration . The idea being that in the world of an IntrAnet, all the software exists in one place � presumably MIS�s application server and is viewed via one interface, in this case, a version of Windows which seamlessly integrates Internet Explorer into the OS.

The flaw in the Microsoft strategy happens to be the strength of Netscapes. From day one, Netscape didn�t care what operating system you had, they�d provide a browser for you. Microsoft has gone long on dominating the Intel architecture primarily because they�ve owned it forever.

But, suddenly, the hardware architecture becomes irrelevant because someone has finally create a software architecture that is. You can bet Microsoft is going to have a healthy dip in it�s 7 billion dollar war chest in order to convince multi-OS shops that it�s browser will run on all their platforms.