I've said for over a year that with Oracle in charge, OpenSolaris would be toast. I had people tell me that I was just spreading FUD. Even after it became abundantly clear that Oracle was going to let it die off, I still had people swearing up and down that Oracle would keep OpenSolaris going.
It's over. OpenSolaris is dead.
In a leaked Oracle internal memo to Solaris developers, Oracle management wrote, "We will not release any other binary distributions, such as nightly or bi-weekly builds of Solaris binaries, or an OpenSolaris 2010.05 or later distribution." As the OpenSolaris software engineer, Steven Stallion, wrote, "This concludes over four years of effort that I (and many other external contributors) have worked on the OpenSolaris project. This is a terrible send-off for countless hours of work -- for quality software which will now ship as an Oracle product that we (the original authors) can no longer obtain on an unrestricted basis."
Stallion concluded, "I can only maintain that the software we worked on was for the betterment of all, not for any one company's bottom line. This is truly a perversion of the open source spirit."
To which I can only reply, welcome to the Larry Ellison school of open-source thought. As I'd been trying to tell OpenSolaris developers all along, the god-king CEO of Oracle doesn't give a damn about any open source that doesn't directly benefit Oracle. The moment Oracle acquired Sun, OpenSolaris' fate was sealed.
This is all of a piece with Oracle attacking Google's open-source Android. Oracle couldn't care less that it's a partner with Google in the Linux Foundation. Oracle is all about Oracle winning and devil takes the hindmost.
Red Hat, whose Red Hat Enterprise Linux (RHEL) forms the basis for Oracle's' Unbreakable Linux, could tell you all about that. Oracle is no friend to other Linux companies or the Linux community. Unbreakable Linux's real point is to break Red Hat. Why would anyone who actually looked at Oracle's business history think it would take a different approach with Sun's open-source projects that didn't fit into Oracle's plans?
That's why I think almost all of Sun's open-source programs are in deep trouble. Oracle has already cut the lifeline for projects that Sun had been friendly toward, specifically the open-source PostgreSQL DBMS. By the end of the year, I suspect we can start writing open-source MySQL DBMS's obituary. Other old Sun open-source projects will start getting their official pink slips around October.
Why October? Because that's when Oracle can claim the projects aren't profitable, and it'll be long enough after Oracle acquired Sun to avoid any troublesome merger and acquisition legal questions. I believe Oracle always intended to let most, if not all, of Sun's open-source portfolio die. I'm now surer of this than ever.
OpenSolaris itself may live on. The Illumos open-source project, founded by former Sun developers, is seeking to keep OpenSolaris going.
Evan Powell, CEO of Nexenta, the company behind Illumos, wrote, "We've been planning for this contingency for a long time. We have the team to continue to support customers and partners and to continue our development. We look forward to picking up the appropriate pieces of Solaris when they are made available with Solaris 11 as well."
This won't be easy. But enough of Sun's top Solaris engineers have left Oracle for greener pastures that creating a truly open OpenSolaris is possible. Whether they can do this without having Oracle attack them, well, as Google and Red Hat can tell you, that's another question.