by Mark Stone on March 12, 2009 09:15am


"Open source at Microsoft." My friends still find that phrase surprising. Yet for those of us who have worked so long on open source, if we really believe the principles we have espoused, shouldn't this be the expected outcome?

In 1994 I did my first Linux install. It was an early version of Slackware, running the 1.0.8 kernel. The term "open source" was still several years in the future. While I never really accepted the basic premises behind the ideology of the Free Software movement, the methodology we later called "open source" seemed obvious and sensible. Share knowledge, collaborate with others, expect and encourage others to evolve your ideas and share their innovations. In other domains, we call this the Scientific Method. Without the ability to openly share ideas, the process of scientific discovery would come to a grinding halt, and we'd be stuck in something like the medieval era of alchemy.

So I am pleased, but not surprised, at the progress open source has made in the last 15 years. And I'm happy to have had a front row view to a lot of it. That journey has taken me through O'Reilly, as the executive editor for their open source group, to Editor-in-Chief of the brief-lived Journal of Linux Technology, to a long stint at VA Linux Systems (now SourceForge) initially leading the web arm of their open source evangelism efforts and later running their developer relations program. Along the way I worked with Chris DiBona and others to get a couple of important books out on open source (Open Sources, and Open Sources 2.0).

SourceForge's developer relations program introduced me to a lot of technology companies eager to reach out to SourceForge's community of open source developers. About six years ago one of the companies we worked with was Microsoft. I worked with Stephen Walli (then at Microsoft) and others to help get Microsoft's first open source projects up on SourceForge. At the time this was a big deal. Few at Microsoft had much familiarity with open source licensing, and there was unease about opening up intellectual property in this way. And Micrsosoft had no experience with the long term benefits of "paying it forward" with this kind of investment in the open source community.

Today is a different story. Microsoft has its own open source project hosting site, Codeplex. Codeplex is growing steadily, and hosts about as many projects today as SourceForge did in 2002. Microsoft has OSI approved licenses that are used by many projects. And Microsoft has an entire group under Sam Ramji that works, among other things, to improve open source offerings on top of Windows.

Some view this turn of events at Microsoft with suspicion and hostility. I do not. Indeed, it would be hypocritical for any true open source believer to view Microsoft that way. If we genuinely believe that the collaborative practices inherent in open source are an important part of software development methodology, then we have to believe that (a) the world's largest creator of commercial software would benefit from contributing to open source, and (b) the world's largest creator of commercial software would be smart enough to recognize those benefits. So it should seem natural, not surprising, that Microsoft's evolution has turned in this direction.

Six months ago I was curious -- fascinated even -- watching Microsoft's recent open source efforts from the outside. For the last six months I've had the privilege of working first-hand with Sam's team, and getting an inside view of what open source is like at Microsoft. I've also had the distinct privilege of getting to know some of the developers and projects on Codeplex. Mine has been an unusual journey from SourceForge to Codeplex, but one I'm happy to have made. And I look forward to sharing some of my experiences with these open source projects here on Port25.