httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ben Laurie <>
Subject Re: cvs commit: apachen STATUS
Date Fri, 09 Jan 1998 16:43:03 GMT
Jim Jagielski wrote:
> Ben Laurie wrote:
> >
> > And if we operated commit-then-review there'd be more chance they got
> > cleaned up...
> >
> Implicit in commit-then-review is the "review" part... For those
> who have never bothered lately to review in the current
> review-then-commit process, where for code to be included it
> must be looked at and voted on and therefore there is a real
> PUSH for "review", how likely is it that it will increase/get better
> in commit-then-review ?
> Implicit in the Voting guide is a desire that patches get a look
> though by eyes other than the patcher. It even has the Veto
> idea which also implies that it's important that people have
> the right and the capability to stop something before it's done.
> Maybe we have progressed beyond such trivial matters... But there
> was a perceived need "awhile" ago. We should question whether
> any of that is still valid.
> commit-then-review is really, we all know, commit-then-whatever.
> That's cool, if we are all aware of that and aren't kidding
> ourselves.

I don't agree with that - I suspect that much review currently is
probably done without much conviction, simply because people know that
it has to get votes or it can't happen. I tend to read stuff when it is
committed more attentively than when it is offered up as a patch for
review. Furthermore, the bottom line is that broken code leads to
visible problems with running servers. Broken code is thus picked up

I must admit, that taking this attitude has meant that some things have
slipped into the code that I find, umm, surprising, when I come across
them later. But if there were an easy route to sorting them out, I'd be
less worried that people were surprising me!

Even if we do decide we should do commit-then-review (which we really
should), anyone who is hesitant about it should feel free to do
review-then-commit. Obviously it does put more load on the committer,
but I can only see that as a good thing.

I say we should try commit-then-review for a while and see how badly it
bites us. If others' comments about other software projects are to be
believed, it won't bite us badly.



Ben Laurie            |Phone: +44 (181) 735 0686|Apache Group member
Freelance Consultant  |Fax:   +44 (181) 735 0689|
and Technical Director|Email: |Apache-SSL author
A.L. Digital Ltd,     |
London, England.      |"Apache: TDG"

View raw message