R Statistical Software
Slashdot has an interesting (and old) item about R which features a New York Times article where various analysis sing the praises of the R statistical software package.
The gist of it is that there is a diverse group of users in both academic, research and corporate spaces. SAS (a provider of commercial statistical software) tries to badmouth it by playing the trust card (would you trust your life to it?), but the article pretty much leaves that as the only point of contention to the remarkable rise of an open source product in a technical and complex space.
I love seeing such tools rise for two reasons. First, it lowers the barrier to entry for students and others with limited budgets and second it forces the commercial vendors to "up their game" with new features and better performance. Open source provides a great baseline for what the market must exceed to succeed. As I can't think of any space where open source has completely eliminated the commercial vendors, it appears they are up to the challenge.