Oh, man, did I ever had a bad feeling something like this would happen. Sun went from being a possible acquisition for IBM -- a company that could have done right by them across the board -- to being snapped up by Oracle, who are most likely going to take the hardware portion of Sun and throw the software over its shoulder.

I say this not because Oracle don't make good software, but because all of the most important and vibrant things that were happening at Sun in terms of software development were being done as open source projects. It's a shame none of it started happening on the scale that it did a good decade or more earlier, but at least it was happening, period.

And now along comes Oracle, whose biggest achievement to date re: open source is a Linux distro that has most people who take the subject seriously snickering up their sleeves. (At the last Red Hat Summit I saw a fair number of T-shirts sporting the moniker UNFAKEABLE LINUX, and Oracle was most conspicuous in their absence.) What are they likely to do with OpenSolaris, or OpenOffice, or any of the other projects that were flowering (however belatedly) under the "old" Sun? Or, for that matter, MySQL? Or Java, which is almost entirely open source at this point but desperately needs a good core engineering team more than ever to help it move foward?

(Side note: I don't think MySQL is as direct a competition for Oracle as people like to think. If I had to pick an open source database solution that was more directly competitive I'd vote for Ingres or PostgreSQL. But I suspect Oracle isn't going to care.)

No matter how I turn this over in my mind, it looks bleak. I don't see much coming from this other than Sun's hardware division essentially becoming a marketing platform for Oracle boxes -- and the software division becoming an irrelevancy, if not cannibalized entirely. Yeah, it's all well and good to say that the FOSS projects they've stewarded can always be spun off under different owners -- but that doesn't guarantee that they'll thrive or get even a fraction of the same development impetus. You can recycle code, but you can't recreate good development teams by snapping your fingers.

One commentary I read, courtesy of industry analyst Theresa Lanowitz, put it this way: Sun is not a software company and never has been (debatable, but I see where they're coming from); and Oracle is not a hardware company, either. Merging the two of them is a little like adding three mediocre guitarists to a band and expecting them to collectively add up to one good one.

It's not Sun I'm worried about here. It's all the good things Sun has made and pushed forward that now stand a chance of stagnating: Java, mainly, but also OpenSolaris and to a lesser degree OpenOffice. I have very little faith in Oracle to trust in Sun enough to leave well enough alone -- and even that would still make me uneasy.

Learn about all the latest Enterprise 2.0 technologies at TechWeb's Enterprise 2.0 Conference, Boston, June 22-25. Join us (registration required).

Follow me and the rest of InformationWeek on Twitter.