I've had 24 hours to ruminate on the Web Services and Open Source panels at AlwaysOn, and integrate the comments with my own thoughts on the subject. Herewith a few extracted issues (apologies for unsourced quotes - I wasn't always able to keep up with who said what).

Why do developers work on open source? There are the highly visible individuals, out for some combination of fame, joy of craftsmanship and community, and resume polishing. There are also the commercial interests, who are often less remarked when the open source phenomenon is analyzed.

Some of them are using open source to enable another revenue model. IBM is the most notable practitioner. Their Linux enabled integration services business strategy has been exemplary, and the effort put into the platform repaid many time. Some commercial interests are just out to deflate their business opponents' bottom line by commoditizing the competitors' dominant positions. Microsoft is the biggest target, but Oracle is now obviously on the list as well, and Sun is wandering somewhere around ground zero.

Lastly are those trying in some fashion to directly build commercial products as counterparts or extensions of open source (rather than using it as a platform.) mySQL certainly seems to fit this mold. I was intrigued by the CEO's comments that buying from him could in some fashion circumvent the GPL restrictions on derivative products. If true, one hopes that the firm is adding significant value on top of the open source version. Regardless of the defects of the GPL, an arbitrage play on the commons only works so long as the commons puts up with it.

How does a commercial interest align itself successfully with the opinion leaders and developers who make open source more than another doomed commercial 'alliance'? Does it work best to try and spin off something internal to the community, e.g., Mozilla? Or find a train that's already moving in a useful direction and jump on? There's an interesting study to be made of the success, failure, distribution of contributors, and promotional strategies for open source projects with commercial synergy.

And how about the users of this software? Let's do call them customers, because even if they don't pay directly, they are going to sink a lot of collateral effort into their projects It's that old TCO argument. I heard several interesting points:

• Most programmers work in-house, on one-off projects, not on glamorous platforms and Internet services. They have a job to get done, not a platform war to fight.
• Open source gives them more flexibility on chosing what to borrow, what to build, rather than having to accept some vendors' view.
• Nonetheless, the combination of convenience and lock-in presented by commercial vendors is not going away now or ever, and they will continue to win their share.
• There is a consolidation in operating systems that will put some part of the customers up for grabs when IT investment revives, e.g., HP consolidating its seven OS' down to three.

Chris Stone of Novell a couple of times raised the question of how many architectural layers would eventually be successfully supported by open source, and how long would it take to build a viable 'stack' supporting various commercial markets. His take was that it would take quite awhile - I never heard the point engaged by others. Pointing back to my question about how to promote open source projects, I'd also ask how predictable the process can be made from the point of the view of the customers, given its inherent component of chaos (or emergent order if you prefer.)
5:51:51 PM