There's a list here:

http://www.welton.it/freesoftware/

With the most important one right now being Hecl. Rivet was moderately successful in the past, and got me into Apache. I've done lots of minor patches for things, including the linux kernel (lameasses never accepted nor rejected it...), Erlang, Tcl, OFBiz, etc...

I would probably spend all my time hacking on open source software if money wasn't a concern... it's a lot of fun.

I did GSoc for Freenet and hopefully will find more time to contribute to it.If money werent an issue,i would be working full time on the kernel and Freenet, but unfortunately it is :P

I wrote TagLib, JuK, a couple other small KDE applications, and a handful of the library classes (inline spell checking, search widgets, maintained the listview classes for a few years) and more recently implemented a number of algorithms and classes for ChucK (music orientated programming language).

I'd recommend working at least tangentially in one of the larger projects since I feel like hooking into those cultures is a really instructive and challenging environment. Don't underestimate the social component of OSS projects.

In a big project the easiest way to get started is to pick something you already use, start running the unstable version (from CVS, SVN, whatever) and sign up for some mailing lists. You'll hit bugs, and if you've been on the lists for a bit you'll get a feel for how they're handled in that project. IRC can also be useful. Often you can find someone there that will kind of informally mentor you in the project if they like what you're working on.

One big caveat that I always point out: Don't worry about your skills. You'll learn them if you're motivated, latch on to a good community and have moderately thick skin.

I work on a ton of stuff: god, chronic, fuzed, grit, erlectricity, yaws (those first four are mine or co-mine). You can see them all (and clone the git repos or look at a feed of my commits) at http://github.com/mojombo/

Mostly my PHP projects lately: my take on ORM (ORMer); a database abstraction layer (dbFacile); and some others. They're all geared at making things easier on the developer.

Also started a console multiplexer app called Knox a few years ago that's been sitting dormant.

They're all at: http://greaterscope.net/projects

In addition to those I submitted a patch to Xynth a few months ago for Debian support (http://xynth.org)

Avant Window Navigator, http://awn-project.org/ (also, wiki here http://wiki.awn-project.org)

It's a dock for Linux with Python-based applets.

I used to contribute to yaita.sourceforg.net. It's a GUI testing tool for swing app, using AOP. It started as a class project me and a couple of friends had while we were taking our degrees.

Apache Xap and Dojo Toolkit.

In addition I have a few things that I've built that I would love to open up. I just haven't found the time to spend to get the last 5% done needed around packaging, clean up etc.

Blender and Second Life. I also started some work on GIMP but I haven't submitted anything for GIMP yet. To get started I just sent an email to the relative mailing lists.

Orange data maining software. http://www.ailab.si/orange/

started the returnable project, over at http://returnable.org

I tend to contribute to projects that benefit whatever else I'm working on at the time. If an open source project gets me half way to fulfilling a feature, I have no problem taking it the rest of the way and contributing back. It's better than starting from scratch!

None!

I have never seen the point why highly qualified developers work for free instead of turning their skills into real money.

Maybe someone can enlighten me (and others who read this and don't understand too).

I've contributed some icons to The Fedora Project in the past, but not nearly as much code related contributions.

the Cairo graphics library

I really want to see some sort of iPhone open sauce apps (:

panda3d - python c++ graphics library http://panda3d.org

i manage http://newskicks.com

Banshee.

I started TrakEM2: register/edit/analyze/3D model terabytes of images. For scientific image processing -mostly neuroscience, mostly electron microscopy images, but also other fields and imaging techs.

http://repo.or.cz/w/trakem2.git

http://www.ini.uzh.ch/~acardona/trakem2.html

It's based off ImageJ. And yes it is in Java, with plenty of jython scripting on top.