You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Jim Jagielski <ji...@jaguNET.com> on 1998/01/09 15:11:45 UTC
Where to now
OK... Everyone agrees that changes need to be made. Right now
people are talking review-then-commit and commit-then-review as
if they are the only options.
Since many people are more upset but the fact that the current
review-then-commit process requires votes, why not go with
something we already have in place, and one which we feel
comfortable with, one which addresses concerns: Lazy Voting
Sure it's a compromise. It's not all that everyone wants, but
no one will get everything they want. It's give-and-take.
It's good for those who want commit-then-review since it avoids
the pitfall that if reviewer are lazy and they don't vote, the
code can still go in. It's good for the review-then-commit
crowd because it offers them the chance to see the code, and
comment on it, before it's committed.
If we could streamline and automate the process of updating
STATUS, all the better. Or maybe we go back to someone having
responsibility for STATUS... this also addresses some concerns.
In any case, let's try to keep everyone happy and spread the
pain around.
--
====================================================================
Jim Jagielski | jaguNET Access Services
jim@jaguNET.com | http://www.jaguNET.com/
"Look at me! I'm wearing a cardboard belt!"
Re: Where to now
Posted by Brian Behlendorf <br...@organic.com>.
At 09:11 AM 1/9/98 -0500, Jim Jagielski wrote:
[lazy voting]
>Sure it's a compromise. It's not all that everyone wants, but
>no one will get everything they want. It's give-and-take.
>It's good for those who want commit-then-review since it avoids
>the pitfall that if reviewer are lazy and they don't vote, the
>code can still go in. It's good for the review-then-commit
>crowd because it offers them the chance to see the code, and
>comment on it, before it's committed.
Just to explore this further; what's the threat if code is reviewed after
it's committed? So long as those with commit access agree to a set of
principles (as I started with yesterday, I haven't read all my messages yet
to see if there's been responses), I don't see the problem. I do see
something we could add to those principles - there should never be an
argument to the extent of "well it's already committed, so there's nothing
that can be done about it". We should not be afraid to make mistakes, and
roll back patches or fix code in the tree if one was made.
>In any case, let's try to keep everyone happy and spread the
>pain around.
What? Can't we spread the love too? :)
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
specialization is for insects brian@organic.com
Re: Where to now
Posted by Dean Gaudet <dg...@arctic.org>.
On Fri, 9 Jan 1998, Jim Jagielski wrote:
> OK... Everyone agrees that changes need to be made. Right now
> people are talking review-then-commit and commit-then-review as
> if they are the only options.
>
> Since many people are more upset but the fact that the current
> review-then-commit process requires votes, why not go with
> something we already have in place, and one which we feel
> comfortable with, one which addresses concerns: Lazy Voting
Lazy Voting requires just as much extra work on the part of the patch
writer. And since Lazy Voting essentially means apathetic voting that
work is wasted. i.e. why should I be motivated to post a patch if I know
I can just commit the sucker in 48 hours anyway regardless of whether
someone reads it? Why don't I just commit it and let them read the diff
and "cvs update" their code and test it right away easily?
>From my point of view lazy voting and voting are the same.
Dean