Open Stack

On Tue, Oct 02, 2012, Sean Dague <sdague at linux.vnet.ibm.com> wrote:
> So I realize the review queue is short right now, but it seems like
> perhaps we should have more than 7 minutes between first code
> submission and final approval for Jenkins to kick off gating jobs
> (https://review.openstack.org/#/c/13962/). Not saying that the patch
> is bad in any way, but it just seems like time to let more eyes on
> things is a good thing.
> 
> 7 minutes is often shorter than the time it takes to get the
> notification emails.
> 
> Was there previously a convention on minimum review time? I vaguely
> remember a similar conversation when Folsom opened up.

I'm generally against a minimum review time. Patches generally sit in
the queue for too long as it is.

However, not all patches are alike. For large patches, new features or
subjects that I'm not an expert in, I tend to let them sit for a little
bit longer to allow more eyes to look at them. Especially if someone had
previously raised a concern with it (allowing them to review before
approving it).

Even then, it's not the end of the world if a mistake happens from time
to time. I've seen patches go in that probably shouldn't have, but I
can't think of a situation where it was because it was reviewed too
quickly.

They've always been found out pretty quickly and either reverted or
fixed up. As long as this is fairly rare, I think we have an appropriate
amount of pace.

JE