I obviously disagree with Michael’s characterizations of the project but I do think it’s particularly unfair to say that most or all of the OpenSolaris development processes are done inside of Sun. It’s true the vast majority of commits are done by Sun employees and it’s true that the “master” SCM for the core kernel/library/utility “consolidation” still lives inside Sun’s firewall.

However, there has been a great deal of work the past three years in pushing large amounts of development work to be *outside* of Sun’s infrastructure. Much of the project work that takes place occurs on the public mailing lists housed on opensolaris.org, architecture reviews are almost completely done out in the open nowadays, public code reviews are routinely done, etc. Clearly, we need to do better on this but it’s certainly neither a closed development model nor is it an aquarium model where folks are simply looking in. We’re in an awkward transition phase but that transition continues to take place and I expect things to only improve on that score.

The fact that one of the repositories in currently inside Sun’s firewall and is instead mirrored to the outside is a reality today but that’s considered a high-priority bug that’s being fixed. The issue is complex since it’s not simply a matter of putting the Mercurial repository out there but including all of the relevant change acceptance required for commits to be done (someone needs to sign off on a request to commit, the commit needs to be done and then there’s various “paperwork” like bug tracking that needs to be updated). Once those things come together later this year, the community will gladly throw out the “request sponsor” method and folks outside of Sun will be able to commit directly into the master source base (I should note that other parts of OpenSolaris are already outside the firewall and have outside folks doing commits today).

So while OpenSolaris may not look like Linux in terms of how the project is run, I suspect it won’t look like MySQL either. Perhaps it will be something in between or

I do bemoan about all the heat and the very little light that comes from these discussions. And I think it’s unfortunate that it at times overshadows what I think are the interesting technical things being done by the Indiana project and OpenSolaris as a whole.

I don’t know how much the CDDL does to hinder that, but ultimately if it’s not forkable it’s not open source. MySQL has always used the model they use, so people are used to it, and at the same time their code has always been GPL, which reassures people that if it becomes a problem they *can* fork; this knowledge in turn probably keeps the MySQL folks from pulling any proprietary-style moves with their database. And they’ve never really courted anyone’s approval.
But with Sun, we’re looking at a process of change, so there is no status quo situation of long standing that everyone’s learned to accept. And they *are* clearly trying for community approval. People, and I’d say Free Software people in particular, react worse to someone claiming to be doing something wonderful and then putting in caveats than to someone who just did what they did and let the situation speak for itself, even if the substance is similar. So a serious fork is not out of the question; Sun is unlikely to go the MySQL route because while MySQL can get away with it, Sun quite likely cannot.

From where I’m sat, any hesitation on the road to freedom (heh) can be blamed on that.

The Indiana shambles seems to be because Sun is a company rather than a dictatorship.
Statements coming out of certain mid-management were pounced on as though they were the word of God rather than one opinion among many.

Some of Sun’s previous mistakes and statements have led to confusion of where Sun stands in relation to Open Source, granted. However, since Jonathan Schwartz has come to the leadership position, not only can you listen to what Mr. Schwartz says about Open source, but look at what Sun has done since he is steering the ship.

Sun has release Java under GPL. They are releasing other key technologies under GPL as well as other Open Source license. They have been more than generous with the Open Source community, with Open Office and other vital key components to Open Source success. Schwartz has stated many times that Open Source is Sun’s future and given interviews on why he believes this.

Technologically, both Sun’s Solaris and Linux are superb OSes. Linux is better and far more advanced on the Desktop, while I would give Sun’s Solaris a slight edge on the Server.

Solaris has been running the Enterprise for years and they have done a great job of situating Solaris for Enterprise large scale deployments. Solaris has a little better monitoring tools.

Solaris has more recognition in the corporate world, but that is changing. I love both and I would have no hesitation deploying either with confidence.

However, Open Solaris has some work to do. No doubt it is getting there, but is not to the level of Solaris 10 yet. In the current case I would feel better deploying Linux than Open Solaris. Open Solaris !=Solaris, currently.

As you stated, competition is a good thing. It will only serve to make each better and make Microsoft servers almost irrelevant.