This post was discussed for fifteen minutes or so on LugRadio series 5 episode 5. There is a forum page to discuss the issue over there.Part of this is based on a response I wrote to Evan Martin's post about upgrading to Gutsy. Evan said that there was a lot of what jwz has called the "Cascade of Attention-Deficit Teenagers" model, in that people had fixed things that were fun to fix and added new features while ignoring routine fixes that needed work. Part of what he touched on involved window managers, but not particularly at a level I have to deal with; nevertheless, I thought I should respond there.
See, I work on a window manager. Sometimes people ask me what I do in GNOME, and I say "I work on the window manager," and then they say, "What's a window manager?" And then I know I've succeeded. I don't want to work on one of those window managers where it's right in your face taking up mindspace with million-dollar special effects. I want to work on a window manager which gets out of your way and gives you more time to live your life. If the window manager is noticable, the window manager has failed.
There is a bunch of sysadmins who administer various Anglican/Episcopalian things worldwide, including the anglican.org domain, and they are known as the Society of Archbishop Justus. St Justus of Canterbury (died 627) was an extremely minor archbishop in the big scheme of things, but he kept the church running in his day; similarly, sysadmins are doing their job best when nobody knows who they are. And it's a little like that with the Board elections currently; as Federico said, "If you are a rock star hacker... you will not be a good Board member." (and no, this isn't hustings, I don't have any spare time, and I am bad at persuading people to vote for anyone, let alone myself; but if you're planning to stand, think on that.)
Now in the end, it turned out not to be a Metacity problem at all. But for a while I was a bit put out by the idea that Metacity had succumbed to jwz's CADT model, because I don't think I've written much new code at all in the last year; all the spare time I've had (which has been squeezed a little in the last few months) has gone on bug fixing and patch review. But this is the thing that we really need: we don't need rock stars. You won't get your name in lights and your own Wikipedia page saying "Jane Q. Hacker is a rock star programmer", by fixing a ton of minor bugs and making the users happy, but really, in the final analysis, that's not the important thing. The important thing is to make the users happy and (as much as possible) not to let them know you're there. After all, one day you won't be.
So, anyway, Metacity still has a largish bug queue. Some things like patch review need a knowledge of the overall shape of the code, but some can be approached by people with various levels of experience. If anyone reading this has a little spare time, some simple programming experience, and is not particularly hungry to win riches, glory, and the Turing Award just this year, come and find me and I can probably find you some way to help, or take you to someone who can; you'll be making the users happy.