Enterprise Unix Roundup: Sun Hits Community-Building Road Bumps
Sometimes I know how to call it, and sometimes I completely miss the mark. My November 2007 prediction that the then-small OpenSolaris naming kerfuffle would fizzle out quickly is definitely an example of the latter.
By a mile.
For those of you just surfing in, here's the quick summary: Last year, when Debian GNU/Linux founder Ian Murdock came to work for Sun Microsystems, he was put in charge of a team that would essentially make a binary-based distribution out of the somewhat scattered bits of code needed to put together an OpenSolaris install. This "distro-ization" was referred to as Project Indiana.
From the start, Project Indiana was meant to be a distribution of OpenSolaris, containing "a core operating system, kernel, system libraries, a desktop environment and a package management system," according to Sun. In other words, one-stop downloading for those who want to try out OpenSolaris. But in October, when the first public beta was released, Sun revealed that this new distro was going to be rewarded the OpenSolaris name. The distro would be getting the name of the operating system.
This is akin to one of the Linux distributions say, Red Hat taking the name "Linux" for itself. Except that could not happen because no one company owns the Linux operating system. But in Solaris-land, that is not the case: Sun owns Solaris and OpenSolaris, and therefore saw fit to name its official distro accordingly.
I believe the debate fundamentally comes down to a question of identity: Is OpenSolaris a code base that others (including Sun) use as the basis for their operating environments, or is OpenSolaris an operating environment in its own right? Given that much of the world already assumes OpenSolaris is an operating environment namely the community version of Solaris one answer to that question is clear to me: OpenSolaris must be something new users can download and install.
Indiana is the first, and so far only, distribution created on OpenSolaris.org containing only bits from other OpenSolaris projects. It is, in a sense, a delivery vehicle for their work. For all intents and purposes Indiana is OpenSolaris in binary form. For all of the discussion that has gone on around the name, very few people seem to disagree with this.
Well, apparently not. Since that late October message, and my November recap of the incident, the rank and file in the OpenSolaris community have grown increasingly frustrated with Sun's level of control over the project. Various public statements have been made, not just about the naming issue but also about an alleged lack of real openness on Sun's part.
Most recently, Apache co-founder Roy Fielding (rather loudly) announced his Valentine's Day resignation from the project, saying:
Sun agreed that 'OpenSolaris' would be governed by the community and yet has refused, in every step along the way, to cede any real control over the software produced or the way it is produced, and continues to make private decisions every day that are later promoted as decisions for this thing we call OpenSolaris.
This kind of thing has been said by other developers before. One thing to keep in mind is that as near as I can tell, this is still the feeling of a minority of the OpenSolaris development community. Many coders seem to be happily writing away on their projects.
I don't want to sound like an apologist, for either side of the argument, but I feel compelled to sincerely ask the question: What did people expect?
No slam on Sun, mind you, but I think any software project with a central ownership or authority position is going to be like this. Surely the OpenSolaris developers knew what they were getting into when they signed on. When you give someone or some company that much ownership, even the most liberally licensed project is going to shift more control to that central body. Let's face it, OpenSolaris isn't as openly licensed as BSD or Linux.
That said, enough comments have been made about unkept promises Sun has made to its development community that one has to wonder if the problem goes above and beyond the normal authority-rebellion concerns. Despite the inevitability of central control in a corporate-run project like this, there are still better ways to run such a railroad. In its haste to build a community, has Sun forgotten to apply the open lessons that make such communities stronger?
Time will tell, and we'll have plenty of time to find out. This situation isn't going away soon.
Brian Proffitt is managing editor of JupiterWeb's Linux/Open Source channel, which includes Linux Today, LinuxPlanet, and AllLinuxDevices.