On Oct 5, 2010, at 2:40 PM, Don Armstrong wrote:
We probably should do a better job of identifying these packages and responding to the RFS to tell people that it's of questionable importance (or clearly no importance) and then channeling them into assistance to Debian that is of greater importance. [It's also difficult to break it to people that the work that they've done probably isn't needed and keep them positive about contributing to the project... though I think the sooner this happens, the less painful it will be.]
Could this be done as a direct response to the initial ITP bug submission?
Proposal: No RFS can be submitted to the mailing list unless the ITP is at least 14 days old and has a "sniff check" response, which can be either a pointer to the appropriate packaging team or a "first pass approval" from a DD. "First pass approval" having the meaning "I think this might be interesting to some users, it doesn't obviously appear to duplicate an existing package, and it seems to fit into the general idea of Debian", i.e. it's permission to do the free-for-all of searching for a sponsor that mentees already do now. The win for the mentees is that they can get in touch with the packaging team best suited to help them (if it exists), the win for the potential sponsors is fewer RFS for questionable/useless packages to wade through, and the win all around is that the meta-level questions of "how does this fit into the bigger picture" are answered early on before mentees have committed lots of hours creating something that no one appears to be interested in sponsoring.
If the ITP process is already supposed to be doing this, perhaps some attention should be focused on making that happen better?