News Of Solaris's Death Is Greatly Exaggerated
Sun's got a long, hard road ahead of it as a new sibling in the Oracle family, but I'm not inclined to believe the recent doomsaying that Solaris, or OpenSolaris, is about to be kicked out of the house. If that happens, it won't be for years yet, if at all.</p>
Sun's got a long, hard road ahead of it as a new sibling in the Oracle family, but I'm not inclined to believe the recent doomsaying that Solaris, or OpenSolaris, is about to be kicked out of the house. If that happens, it won't be for years yet, if at all.
Steven J. Vaughan-Nichols blogs about open source over at ComputerWorld, and recently attracted a lot of attention for his piece "Is Oracle getting ready to kill OpenSolaris?", wherein he speculated that Oracle with its Unbreakable Linux had no long-term use for Solaris and would probably strangle the OpSol baby in its crib before it grows up to threaten Daddy. He cites words from one of Oracle's chief software architects: "Oracle definitely runs on Linux" (yes, but it also runs on Solaris as well) and "Our test platform is Oracle Enterprise Linux".
Sure, but what about the customers? The biggest reason why OpenSolaris isn't going to be an early kill is simple: to do that means Oracle is sending all of their existing Solaris/Sun software/hardware customers -- the folks who run Oracle on the Sun stack -- a very negative signal. Nobody really wants to be told that the hardware and software they invested quite a bit of money in for the next five to ten years is being sidelined by its own maker.
Another thing to consider: OpenSolaris and Solaris itself are theoretically different animals, and that if OpSol is killed off the full-blown Solaris will continue as before. This doesn't make sense given my own talks with folks in Sun. Their POV is that OpSol is the next version of Solaris in all but name, and that offering it as open source benefits everyone on both sides. It's marginally plausible that Oracle could order that future iterations of Solaris not be open source -- but if they were going to do something that imperious, I don't think it would be worth the risk. Not at least until they get an idea of how at least iteration of the whole OpSol-to-Solaris cycle fares for everyone involved.
One other thing Vaughan-Nichols points out is that open source projects like MySQL, VirtualBox and OpenOffice (not mentioned by name in the article, but it's fairly obvious) will continue with or without Sun's help. My feelings about each of those projects are entirely separate.
VirtualBox benefits from being under Sun's wing, thanks to its parent company's know-how. The project's progressed quite dramatically in the year-plus that I've been working with it, and for most basic virtualization use is as good as VMware, if not better. I'm hoping VirtualBox stays with Sun where it belongs, and prospers.
I had good hopes for MySQL as a Sun Project, but now it seems the only real future for MySQL is thanks for a fork (MariaDB) under the auspices of MySQL founder Michael Widenius himself. Maybe Red Hat can add MariaDB to its roster of officially-supported projects?
The development of OpenOffice was glacial to begin with, and if Oracle decides to discontinue work on it, it might be the best kick in the pants for it possible. Maybe the team of folks at Novell who make the Go-OO build will do the right thing with it, or maybe IBM's Lotus Symphony project can really take off and run.
But Solaris may be much tougher to kill. Not just because open source is forever (well, it's only forever as long as someone's actually working on it, but that's another essay), but because there are a whole mess of paying customers with support contracts who say that Solaris is forever, too. Or as close to forever as their IT budget allows.
InformationWeek Analytics has published an independent analysis of the next-generation Web applications. Download the report here (registration required).
Follow me and the rest of InformationWeek on Twitter.
About the Author
You May Also Like