On Tue, Aug 20, 2002 at 12:41:01AM +0300, Giorgos Keramidas wrote:
# On 2002-08-12 14:56 +0000, Jens Schweikhardt wrote:
# > Ladies (yo!) and Gentlemen,
# > The current number of open PRs seems to be inversely proportional to the
# > Dow Jones, NASDAQ and <your national stock index>. See
# > http://people.freebsd.org/~bsd/prstats/ for a chart.
# You've done some excellent pieces of work lately. Now I have to work
# hard to keep up with this amount of PR closing. It sure sounds like
# this is gonna be a niiice Autumn.
Thanks, but I think some of this praise belongs to iedowse as I have
only closed many he has put to feedback in the past. I tried to make up
for that by requesting feedback from a few open ones.
Some in my area of expertise I tried to address. Of others I thought
"hey I can get this to work" which then turns out to be not that easy. :-)
Johan has started to assign PRs to either individuals or mailing lists,
which I have done a few times in the past as well. We (=bugbusters, in Cc:)
should consider being more aggressive here. Assigning responsibility of
PRs to individuals instead of freebsd-bugs might give them some audience
at all. This is assuming many developers don't even browse the GNATS db;
I have no stats on this, so could be wrong, but this is my working
hypothesis right now.
If the problem is related to some file or the PR contains a patch, I
have found it useful to use the "whodid src/what/ever/file.c" utility on
freefall to get an idea who is working on some file or at least who has
done the most commits.
One question I have is this: Some PRs describe problems that are very
hard to fix in the Right Way(TM), and even then will be only a marginal
improvement for some corner case or weird hardware. For example some of
the sysinstall related PRs are in this category. Beating a Dead Horse,
IMHO. Shall we be honest and close with "Yes it's icky, but it won't get
fixed." or leave them open and rot in the hope someone from the outside
picks it up?