Darcs throwing in the towel
So it looks like I have been right for the last two years, and darcs is basically giving up. After the 2.0 release, the maintainer went on record that he will no longer read bug reports or mailing lists, and will only review submitted patches. On top of this the GHC (the major FOSS Haskell compiler, darcs is written in Haskell) is switching off darcs due to performance issues and an uncertain future of the project. Basically they are throwing in the towel in favor of git/hg/bzr.
While the darcs patch calculus concept had a lot of interesting elements, and the mathematical ramifications were quite nice, the software just didn't work that well. The core code had to be implemented in such a bizarre and arcane fashion that few outside the core contributors can work on it. This is very similar to what happened with the Ruby interpreter, and last I heard they also gave up on the current codebase and refocused on YARV. I feel like this also reflects a lot on Haskell as a whole. Lots of really great algorithmic flexibility, but it all ends up breaking down around the edges when you try to write real software. Maybe one day someone will resurrect the darcs concept in a system that actually works well enough to use,
While the darcs patch calculus concept had a lot of interesting elements, and the mathematical ramifications were quite nice, the software just didn't work that well. The core code had to be implemented in such a bizarre and arcane fashion that few outside the core contributors can work on it. This is very similar to what happened with the Ruby interpreter, and last I heard they also gave up on the current codebase and refocused on YARV. I feel like this also reflects a lot on Haskell as a whole. Lots of really great algorithmic flexibility, but it all ends up breaking down around the edges when you try to write real software. Maybe one day someone will resurrect the darcs concept in a system that actually works well enough to use,