Microsoft Days: Transforming the Desktop

Charles Simonyi, former chief architect at Microsoft Corp. and now CEO of Intentional Software Corp., continues his interview with Computerworld's Robert L. Mitchell. Here he discusses his work at Microsoft, which included leading the development teams for the Word and Excel programs.

What was your most important contribution at Microsoft? When you look at Word, the fact that it has been around for 20 years and the functionality when you look at it today is so wonderful ... the design 20 years ago must have captured something right, so that it can carry the load of these innovations. Basically, it's the same design, it's the same code that is carrying the fantastic innovations that people have added to Word. I think that would be my most important contribution.

What would you say was the biggest failure? I didn't have a single, spectacular failure at Microsoft. I had failures of communication with people. I had a failure to sometimes respect the market, and I'm appreciating even now the need to learn. I know that I have to be able to listen in order to learn.

When you say failure to respect the market, were you talking about your experience with the Multiplan spreadsheet program? Multiplan was a misjudgment of the market. Microsoft made two bets at the time. One was in the operating systems arena with MS-DOS, and the other was in the application arena with the Multitools.

Multiplan was done on a byte-coded interpreting system, much like Java. It was probably the most ported system ever deployed. We thought that the market would be fractured for a long time and that we would be on all of those machines -- which we were.

Interestingly enough, MS-DOS changed that and created a unified market. And, of course, Lotus 1-2-3 made their bet on creating a single, optimized, direct implementation for MS-DOS, and they cleaned up. We learned a lot from that failure. And then of course, when the next shift came to GUIs [graphical user interfaces], we cleaned their clock with Excel.

In the past, you criticized your former employer, Xerox PARC, for "biggerism" -- a "bigger is better" mentality. But products such as Microsoft Office have also been criticized for being too bloated with features. Do you see an irony in that? I thought that Xerox really made a lot of mistakes. At a time when the Alto cost $50,000, they were scaling it up in every way possible instead of scaling it down. The result was, of course, failures. I think that everyone agrees that biggerism in that instance is bad.

It's clear that people at different times want different subsets of potentially useful features. So even if 90% of the people only use 10% [of the features], over time, it's more than 10% as their needs change. It's also clear that all of those features are useful. It's been determined by market research.

The question boils down to how these large sets of potentially useful features should be distributed. Should they be distributed in one package or multiple packages? To me, that's really a cost-benefit equation, and you need to have technological solutions to lower the various costs. In some sense, it's the embarrassment of riches.

It's really a merchandising issue. The answer that you should stock fewer items is not the obvious answer.

I fully agree that there is a demand from enterprises for control, and the paradoxical point is giving an enterprise control over the features that are available as a feature. So, guess what -- that feature will also be added to products, including [Intentional Software's] products, I'm sure.

Copyright © 2002 IDG Communications, Inc.

7 inconvenient truths about the hybrid work trend
Shop Tech Products at Amazon