And unfortunately, a farewell
By Jacob Kessler on Jul 20, 2009
And now, what may well be the hardest blog for me to write.
See, normally I've been blogging about technical things, sharing my knowledge of the work that I've done with the world at large, trying to help other people use the stuff I've written. That's easy, since I can at least pretend to be able to explain things well, insert a bit of humor, and generally not worry too much about the end result: most of the stuff that I miss I can clarify on the mailing lists anyway.
But this is a bit different. This blog is to announce that I'm leaving Sun for elsewhere, primarily because of the impending Oracle takeover. And even just saying it doesn't help much. It would be easier, I suppose, if I hadn't enjoyed working at Sun so much. Then, it would be something fairly simple, a short post "I'm leaving Sun for better opportunities", and that would be it. But it's really not. I really liked working at Sun, really liked each of the projects that I worked on, really liked working with all of the people that I worked with. I'm not leaving because I didn't like working at Sun, or even because I thought that my new job would be better. I'm leaving because Oracle is a giant cloud of uncertainty on the horizon, filled with things that there isn't any way to know about or deal with.
And it's that uncertainty that makes it so difficult to make decisions. It could be that everything Sun will be transplanted into Oracle without modification, that I'll remain with excellent management and excellent people working on the same excellent things, with the same wonderful freedom to talk and blog about everything that I'm doing, even if I haven't been doing much of the blogging recently. At the same time, it could be that nothing will survive, GlassFish will be disassembled to form some kind of Frankenstein construct with Weblogic, and everything will change (for the better, or for the worse). It could even be that GlassFish, or at least scripting, will be found not to fit with Oracle's plans and dropped entirely, leaving me either to be shifted to some new and unknown place or cut loose. And nobody can say which direction any of it is going to go in, because they don't know and the SEC says that they can't find out. Of course, changing jobs introduces uncertainty too: what will my new manager and co-workers be like? What will I be working on? What will the corporate culture of the new place be like? But there, at least, I have some control over it, and the total uncertainty is much less. So I'm leaving Sun not because I don't like Sun, or I know I don't like Oracle, but because I'm more certain that I will like my new job than I am that I will like whatever Oracle has in store for me.
That basis, of course, makes it rather difficult to actually go through with it. There is the lingering doubt, the questions, the what-ifs. Is my analysis correct? Are the potential actions that I've assigned to Oracle accurate? Does my model of the new job actually represent it well? Is it worth having to get to know a whole new set of people? What's the chance that I'd need to get to know a whole new set of people at Oracle anyway? I tell myself that I'm just nervous and that I'm making the right choice, and then I think that I'm just rationalizing and should go think about things more. Uncertainty is a pain.
Ultimately, I made the choice that I think is the right one, to leave Sun and be employed elsewhere. I had intended to try to leave at the same time as the Oracle takeover, which I felt would ease the transition somewhat, but the timelines of my job search and the SEC didn't intersect quite the way that I wanted them to, so I'm leaving a bit before it happens. For those of you who use my code, I plan to stay active on the JRuby list, so I'm not going to completely disappear. For the rest of you, I've got a gmail email address that you are welcome to contact me at: Eagle.Kessler. While I may not be good at conversation, I'm more than happy to help explain the stuff that I've done, or help with whatever.
 I say this, of course, without any bias towards the goodness or badness of Frankenstein constructs. I'm a big fan of Mad Science, and great things have come out of stitching a few things together.
People keep mistaking me for a people person. I'm really not. Speaking to people is really hard. Getting to know a new set of coworkers will likely be one of the harder parts about settling into a new job.