For that reason, despite his understandable Schadenfreude over the current Java debacle, Miguel de Icaza's slightly tongue-in-cheek suggestion that Google should adopt Microsoft's C# instead is simply jumping out of the frying pan into the fire. Probably more fruitful would be to join efforts to find prior art that might be used to get the Oracle patents reviewed and invalidated.
The other clear message is that the open source world needs to fork all the main open source projects that Oracle owns, or to transfer energies to a replacement. That's already happening with MySQL, which we can now officially declare dead, or at least moribund thanks to Oracle's taint. OpenSolaris was already well on the way to forking, and I'm sure that the people concerned will be working feverishly to bring their plans forward.
The one that really worries me is OpenOffice.org. I've been using this for many years now, and watched how it has grown from a rather clunky alternative to Microsoft Office into a real rival that satisfies the needs of the vast majority of users. And that despite rather lukewarm support not just from Oracle, but from Sun before it. It's true that it might fit rather better with Oracle's overall enterprise strategy, but that doesn't necessarily mean it will thrive as an open source project. Forking OpenOffice.org would presumably be a major undertaking, but I think it's something that people need to start considering seriously.
As I've noted, I don't claim to have any insight into how this lawsuit will conclude, whether it will blow up into a long and bloody battle between Oracle and Google, or whether it will be concluded by some quick and relatively amicable solution. But I do believe that whatever happens, whatever it might nominally “win”, Oracle has certainly and irrevocably lost more in terms of trust and goodwill within the free software world than it will ever understand.