Jason Perlow in eWeek is suggesting that Microsoft open source Windows NT 4.0 -- Microsoft, Let Go of My NT! That could be interesting. I mean, just watching Microsoft going through the process of opening a code base of that size and building a community could be quite entertaining. Whatever.

Then Jason says:

And unlike Sun's release of Solaris under CDDL, which will be of no interest to most of the Linux folks and will likely generate only a small amount of new development activity, even a quasi-open-source or a restricted open-source license for NT would generate lots of developer excitement.

An "Open NT" could become a very successful platform on its own for open-source software development. Sounds hypocritical? Think of the numbers -- few Solaris developers, mucho Windows developers. Do the math.

I don't know about this. To me, communities come in all different shapes and sizes and personalities. Diversity is what I'm after, not massive numbers. I'm after quality, not quantity. And I think the OpenSolaris community will be plenty interested in this project. Don't you?

Comments:

I would agree, as I blogged recently (http://bobsnook.blogspot.com).
Given it's current market, OpenSolaris is almost certain to get many what you might call tier 1 developers given the number of Fortune 500 IT professionals that rely on it every day.
Bob

Posted by Robert Lunnon on January 22, 2005 at 04:07 PM JST #

First, Microsoft will not release anything under a true open source license. It's anathema to them. Second, if they were to release NT4 under some deceptive licensing terms like their shared-source program or perhaps BSD without patent grants, they will find that no open source developer will touch their code. Third, there's very little interest in the technology anyway except possibly for interoperability purposes. NT4 is woefully obsolete; even Microsoft has recognized this, and truth be told it has never been competitive with even contemporary versions of Unix and Linux (to say nothing of the lead enjoyed by current versions). The only value would be in understanding all the undocumented protocols in it, most of which have already been reverse-engineered. Worse still, most of these are probably patented, and so anyone using the code would just be opening themselves up to lawsuits, much like those using the court-ordered "open" specifications (that require payment of royalties and/or have been licensed specifically to prevent open source implementations). Microsoft is terrified of open source software because it has no cost basis and can effectively undercut Microsoft no matter how large its cash reserves. Therefore their entire strategy is based on lock-in, using patents, lawsuits, and undocumented interfaces as weapons against those whose products offer a meaningful migration path away from Microsoft's. Open source makes this strategy impossible and forces vendors to compete on features, quality, and price. Opening even obsolete junk like NT4 would only serve to lend credibility to a distribution philosophy that neutralizes Microsoft's entire strategy.

Although I believe uptake of OpenSolaris will be slow and steady, I believe uptake of this hypothetical "OpenNT" would be virtually nonexistent. A handful of companies with existing relationships with Microsoft and perhaps a few third-world governments might have some interest, but most companies that would be in a position to make use of the technology under any of the likely licenses already get what they need from the court-ordered licensing program. Mainstream developer attention would continue to focus on Linux and a handful of other popular open source projects. Even if Sun stumbles badly, OpenSolaris will be far more successful as an open source effort than anything Microsoft could hope for. And, as I've discussed above, Microsoft doesn't hope for anything of the sort anyway.

Posted by Keith Wesolowski on January 24, 2005 at 10:32 AM JST #

Post a Comment:
Comments are closed for this entry.

This blog copyright 2009 by jimgris