[openstack-dev] [nova] A modest proposal to reduce reviewer load

Sean Dague sean at dague.net
Wed Jun 18 12:31:22 UTC 2014


On 06/18/2014 08:26 AM, Duncan Thomas wrote:
> On 18 June 2014 10:04, Thierry Carrez <thierry at openstack.org> wrote:
> 
>> As an aside, we don't really need two core reviewers to bless a trivial
>> change: one could be considered sufficient. So a patch marked as trivial
>> which has a number of +1s could be +2/APRVed directly by a core reviewer.
>>
>> That would slightly reduce load on core reviewers, although I suspect
>> most of the time is spent on complex patches, and trivial patches do not
>> take that much time to process (or could even be seen as a nice break
>> from more complex patch reviewing).
> 
> 
> I think removing the need for two +2s is higher risk that you think -
> the definition of 'trivial' gets stretched and stretched over time
> because it allows people to get patches in quicker/easier and we end
> up in a mess. I'm all for adding the tag, but reducing the review
> requirements is, in my view, dangerous. If a change is truly trivial
> then it is only going to take moments for the second core to review
> it, so the saving really is negligible compared to the risk.

Agreed.

Even with 2 +2s you do the wrong thing. Yesterday we landed baremetal
tests that broke ironic. It has a ton of +1s from people that have been
working on those tests.

People throw +1s around with 'please do this thing', and miss the part
about 'and this current way of doing this thing is actually the correct
way to do it'.

	-Sean

-- 
Sean Dague
http://dague.net

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 482 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20140618/11e09826/attachment.pgp>


More information about the OpenStack-dev mailing list