Ten simple rules for the open development of scientific software

I got hit by this in the early days of working on Krita. There were a number of papers and dissertations that were quite interesting. Mostly, there was no source code, though it was already amazing I could download the papers.

There was Bill Baxter's work on paint simulation which now has ended up in Microsoft's Freshpaint. No source code, and while the videos looked interesting, no way to reproduce from the dissertation and papers alone, at least not for me. (http://www.billbaxter.com)

I managed to get Tunde Cockshott's Wet & Sticky code, under the GPL, and managed to build it, which was amazing. Unfortunately, the actual results of that application were a bit disappointing since it kept crashing, but then, that was work from a different age. (http://www.valdyas.org/fading/index.cgi/books/hacking/wet...)

Clara Chan's Chinese brush simulation was easy enough to port to Qt, and was a good basis for work, so that was an excellent resource (http://www.valdyas.org/fading/index.cgi/hacking/krita/wac..., original paper seems to have disappeared).

The MoXi paper in contrast was also on Chinese brushes, but no source code, so useless.