hi devels,
without much thinking about tools and structure, I have started building a database of bugs and buglets in the core of Pd (aka Miller's sources). Before shaping it into a form suitable for a more general use, I would like first to collect as much data as I can, from pd-list, and from my experience. Which is the easy part.
The hard part, I think, is not just gathering reports, but managing them afterwards. Usually, there would be a person `responsible' for validating a particular category of a report, and guiding the report through its lifetime in the database. That person would either actually maintain the source, or closely cooperate with a maintainer. Such setup might prove too formal, though, in case of the ``Miller's & co(mmunity)'' development model. I do not know... perhaps, that model requires a `sync-person' tracking the releases, with Miller occasionally getting forwarded mailings or digests?
The tool is gnats, currently, chosen for the sake of being able to start easily and locally, without bloating my system with apache, mysql, and what not. Gnats is nicely designed, but for some odd reason hardly maintained anymore -- hence buggy, and largely abandoned outside of the freebsd world. The transition of my local database into a public service, will probably mean importing data into some other bug tracker (requiring, hopefully, only a simple script to be written, or adapted).
When more of the actual contents builds in, I will let you know. For the time being -- comments, please (and bug reports, of course).
Krzysztof