The one that interests you. We can't tell you that.<p>I'll tell you what I did, and you can take it or leave it. About a dozen years ago, after a few months of using Linux on my desktop, I knew I wanted to be involved in Open Source (it was still just called Free Software at that point...the Open Source moniker was yet to be coined for the more pragmatic arm of the movement). So, I started tinkering with all of the projects I found interesting. Squid was one of them, and I spent several years as a developer on the project. Webmin was another, which turned out to be the one I really had an affinity for. I also submitted small patches to yum, various Perl modules, Extropia (before it was called Extropia), LVS, quite a few audio and sound packages, and a few others I don't remember at this point. I also maintained my own kernel patches for several years, until the functionality I needed made it into mainline.<p>I found the various projects more or less amenable to my input, and more or less suiting my personality and strengths (some projects have a strong leader, others expect lots of discussion over even minor changes, etc.), and I found myself working more and more on the ones that suited me and less on the ones that didn't. And sometimes, I'd submit a single patch to a project for purely selfish reasons (I needed the problem fixed, and could see a way to fix it), and never looked at the code again. That's OK, too.<p>There are obvious "rock star" projects...the ones that impact almost everyone everywhere. Apache, Linux, Firefox, any of the P languages (PHP, Perl, Python, and the honorary member Ruby). With Perl you could work in a pseudo high level language called NQP (Not Quite Perl) to work on the language itself...which is pretty cool. I've never felt an urge to work on anything that low-level for fun before, but that's something that does appeal to me. A lot of interesting low-level stuff seems to be happening in the Haskell world, as well, but it's still esoteric enough to be unlikely to reach a lot of people for a few years...but being involved early in something that explodes later is a good career move.<p>So, my advice: Don't pick one project. Try many. You'll eventually find the one that suits you, and the one that you can be a valuable contributor to. Don't try to settle on one too soon, because you might miss out on some great learning experiences.