It's time to state the bleeding obvious: The OpenSolaris project is on life support. Or dead already.
I couldn't have put it better myself, but that's actually the conclusion of Dennis Clarke, one of the members of the OpenSolaris Governing Board (OGB).
There's been a huge question mark hanging over the future of the Sun-sponsored open source UNIX project ever since the unfortunate company was taken over by Larry Ellison's Oracle. Oracle clearly has different plans for Solaris -- evidenced by changes made to the licensing terms that make it free for only 90 days. It's not clear, however, how OpenSolaris fits in with Oracle's UNIX plans. If at all.
Back in February, Oracle executive Dan Roberts assured users at the OpenSolaris Annual Meeting that "Oracle will continue to make OpenSolaris available as open source, and Oracle will continue to actively support and participate in the community." The next release of OpenSolaris was meant to be 2010.03, due -- as the name suggests -- in March. Except it hasn't been released yet; the community is still waiting. Clarke, for one, is not the least bit surprised. In a message on the OpenSolaris mailing list last week, he said:
The OpenSolaris project as a whole was a very nice idea under the old Sun management but it serves no useful purpose under Oracle... Under Oracle I see zero contributions thus far. Zero actions of real merit that befits a multi-billion dollar sponsor of an open source project ... We are all gathered around the still warm body (of OpenSolaris) poking the toes and asking questions.
Of course, it could be that Oracle is so busy digesting Sun that it simply hasn't yet gotten around to thinking about OpenSolaris and how it can benefit the official Solaris in the way that Fedora provides cutting-edge functionality that is mature and later included in Red Hat Enterprise Linux, or the way OpenSUSE does the same for SUSE Linux Enterprise Linux.
To quote Vladimir Ilyich Lenin,the question that the OpenSolaris community is asking is "What Is to Be Done?" Oracle owns the OpenSolaris name and hosts the community sites, so despite its inactivity it still has the project by the goolies.
Desperate times call for desperate measures, and Clarke for one is thinking the unthinkable: He's raising the possibility of a fork. Forking frequently leads to conflict, extreme unpleasantness, as well as wasted or duplicated effort, so it's a measure of the frustration in the OpenSolaris community that it's being mooted at all.
"If we need to hold a wake then fine," says Clarke. "If we are waiting for a Lazarus effect, then that is not so fine. People need to take actions that have merit, and I am not one for sitting on my hands while people stand around looking for leadership ... Personally I am all for a fork ..."
There's certainly a sense that forking the project would be a landmark decision. Without the sponsorship of a multi-billion dollar company, OpenSolaris, which was once touted as a possible open source challenger to the dominance of Linux, could fade into obscurity. On the same mailing list, fellow OGB member and influential programmer Joerg Schilling had this to say of the situation:
"We ... should however not ignore the current situation and thus be prepared to fork in case this is needed. Once we do this, we most likely will harm us and Oracle, so our goal should not be a fork but a much better collaboration, and collaboration means being open and giving information."
It's a nice thought, but Oracle is not known for its openness or its giving, and something tells me things are not going to be the same as they were in the good old Sun days. It's more likely we will end up seeing a fork, and a new project with a name like "the project formerly known as OpenSolaris."
But without Sun or Oracle as a sponsor, it's difficult to see what relevance it will have.
Paul Rubens is a journalist based in Marlow on Thames, England. He has been programming, tinkering and generally sitting in front of computer screens since his first encounter with a DEC PDP-11 in 1979.