Tom Wijsman wrote: > For really severe bugs I think that pinging just anyone who is around > will do, alternatively you could ping the proxy maintainers herd. In > both cases there is most of the time someone available; so, it would be > a matter of minutes to have the patch applied.
You're changing the subject. Nobody was talking about really severe bugs, and those aren't really the common case. Normal bugs which have to wait for the maintainer is quite likely to take much longer than simply committing the (correct) bugfix. > And for bugs that can wait, it doesn't really matter that there is a > slight delay; with a ping where you provide a patch it is still fixed > faster than the average bug on bugzilla (assuming maintainer has time). Maybe delays don't matter to you, but you can't posit that the same is true for every other developer. I think the potentially significant delay matters a lot. The whole point is that we cannot assume that maintainer has time. The scenario is that I have taken a little time right now to fix a bug and get the fix committed so that I no longer experience said bug. I'm fixing this bug because it is blocking me in some way. It matters a whole lot if I have to wait for someone else to unblock me, in practice that completely demotivates me to contribute back, and I would simply work around the block. > Please note that the average commit takes longer than 24 seconds as it > involves testing the change, repoman checks and similar QA matters. QA has nothing to do with committing, don't confuse Gentoo policy with actual meaning of the terms. Testing, repoman and QA matters are of course part of creating the patch in the first place. Maybe you can understand the point that I wanted to make even if I was off by a factor of say 4. The delay is still three orders of magnitude longer. //Peter
pgpLI2fv88jiG.pgp
Description: PGP signature
