httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <>
Subject Re: cvs commit: httpd-2.0 STATUS
Date Wed, 16 Apr 2003 16:42:13 GMT
At 03:55 AM 4/16/2003, wrote:
>          R-T-C:   trawick, wrowe, jerenkrantz, stoddard, rederpj, striker, nd,
>                   gregames(lazy consensus OK for mainline), jwoolley
>          Abstain: 
>  -       C-T-R:   BrianP, aaron, jim, gstein, bnicholes
>  +       C-T-R:   BrianP, aaron, jim, gstein, bnicholes, minfrin

>  +       minfrin: I see no point in adding any bureaucracy between a bug
>  +                and its fix.

Actually, your commits yesterday absolutely convinced me.  If they
were not in head - I would have finished the work I had on my plate
instead of being siderailed into reviewing code that I shouldn't have dealt
with until I had free cycles.

You committed two patches - one correct, one not, and that sounds
to me like you were batting .500 (good for baseball, bad for CVS stable.)

I'll apologize that not every commit seems to be vetted, there are
simply some areas of the code with too few eyes, and some that happen
in the sequestered chambers of security@httpd.  But the code that goes
into 2.0 needs to be spotless.

I agree that the segfault fix you committed shouldn't have layers of
cruft to pass through - but the Date: fix clearly needed it.  The problem
is that one developers idea of a 'simple' fix may be damming breakage
in the eyes of another developer - and the simple answer is post it.

We once had the concept of lazy consensus.  If that's factored into
the R-T-C policy, everybody benefits.  It gives the contributor of patches
in far-away reaches of the code some time to sleep on their *own* patch,
even if nobody else's eyeballs are on it.

My most lucid realizations of how my commit will f*k something up
always comes when I'm away from the keyboard - especially on three
hour drives.  Something about really boring stretches of Illinois farmland 
that does wonders for free association.

While you are busy committing to status, please take a minute to
revert the Date patch to APACHE_2_0_BRANCH.


View raw message