So, it seems my Summer of Code proposal to work on java.lang.management has been approved. I have to say I was quite surprised. Given how successful it was last year, and the amount of additional organisations available this year, I didn’t think I stood that much of a chance. But, here we are…

The initial confusion over who got in didn’t help either. I didn’t wait up on the Tuesday for the results (the Google times being in PST means that it’s some eight hours later here in the UK), and so I was greeted with all the controversy on Wednesday morning. I checked my section on http://code.google.com/soc and found that the application had been accepted, but I wasn’t sure if this meant the GCC guys or Google had. Checking my mail didn’t bring any joy, until I later realised that the mail would, of course, have gone to my gmail account rather than my more usual one. There I did find an accepted e-mail, and this was further backed up by a quick chat with Mark Wielaard (who will also be my mentor, it turns out).

I guess I now need to think about making a start on things. Already being a GNU Classpath contributor makes things a little odd as I’ve already posted quite a few patches over the last few years. The SoC project will mean concentrating on one particular thing on my own, which is something new, and also involves a fair bit of playing around with gcj, which I haven’t yet had chance to do… so things should still be pretty interesting. But now doubt I still find myself committing the odd bit of other stuff as well, if just the merges between HEAD and generics.

I’ll probably start by tackling some of the easier bits, especially as I can’t do any low-level stuff until all the stuff goes through for GCC access. The package is basically made up of several disparate beans, each of which I feel can be worked on in isolation. The obvious initial target is to do the OS one, as this doesn’t require any native code (it just wraps up a few calls that already exist inside a bean). By the time I get on to the second bean and need to do some native stuff, hopefully I should be sorted for GCC…. which reminds me I also need to find some hard disc space for this at some point.

Should be fun… :)