It's been said from time to time that Dr. Dobb's has a love/hate relationship with a little software company up in Washington state, a company you may have heard of. Some people who read the magazine often think we're biased in the company's favor because we mention it so often, or because we haven't published a scathing indictment of the bugs in one of their compilers (lately). Other people, who presumably don't talk to the folks in the first camp, charge that we spend too much time covering Microsoftian developments and urge that we spend more time covering (fill in the name of your favorite product category).
While there seems to be almost a tradition of Microsoft "bashing" in the industry, that's not what's going on here. What's really going on is that the company has grown so much in the last few years that we're all becoming victims of shifting paradigms. Although it may be hard for us (especially the old-timers) to appreciate, there's no escaping it: If you're a serious developer in the personal computer marketplace, you have to analyze Microsoft's products and strategies and how that impacts your products and plans. Often that means reading both for what was said, and what wasn't; analyzing both for tactics and strategy.
Which brings us to the point of this little essay: For most developers, Microsoft's role can be seen today as primarily of strategic rather than tactical importance. Increasingly, the details of specific announcements are becoming less important than strategic concerns.
Case in point: OS/2. It took years to develop-big companies move slowly--and it'll take years for the new generation of applications to appear. But despite the tactical opportunity for other operating systems to penetrate the market (in relatively small numbers), the strategic reality is that OS/2 will undoubtably prevail in the coming years unless IBM pulls the plug. OS/2's victory will have little to do with technical merit, and everything to do with the incredible resources of Microsoft and IBM. OS/2 is the Ada of personal computer operating systems: less attractive than other options, but massive and impossible to assail.
In reflecting on this state of affairs you might be disappointed, like many of us, that Microsoft's strategy ignores the 386 market. It is ironic that most of the scheduled acceptance of OS/2 will occur in a year or more, just when (rumors say) we'll all be hip-deep in 386 machines, but this just further emphasizes the strategic choices developers have. You can take the safe route and write applications for OS/2, or consider all those 386 machines crying for new software. A great deal of successful 386 software is being written by small companies, by people who are able to respond quickly to changing market conditions.
And finally, there are times when it's to everyone's benefit to have a big company leading the way. Microsoft's involvement in CD-ROM is a perfect example. The company's extensions for MS-DOS are an important tool for everyone, but the most important concern right now is making CD-ROM viable. In a market deadlocked with the hardware companies are waiting for the software companies and vice versa, Microsoft has taken the laudable step of continually pushing ahead with CD-ROM products and support. Which means we all might get to reap the benefits of CD-ROM before the turn of the century-presumably even those of us running MS-DOS as a task under a 386 version of Unix.
Tyler Sperry editor