[openstack-dev] [nova] minimum review period for functional changes that break backwards compatibility
sean at dague.net
Thu Jan 2 12:53:34 UTC 2014
On 01/02/2014 06:43 AM, Day, Phil wrote:
> I don't really see why this thread seems to keep coming back to a position of improvements to the review process vs changes to automated testing - to my mind both are equally important and complementary parts of the solution:
> - Automated tests are strong for objective examination of particular points of functionality. Each additional tests adds one more piece of functionality that's covered
> - The review process is strong for a subjective examination of changes, and can often spot more holistic issues. Changes / improvements to the review process have the potential to address whole classes of issues.
100% agreed. I just think that we're tapped out on "humans do better" on
the review side, especially as there is constant pressure to add more
people to review teams, which on average means reviews get worse as
those reviewers are less experienced in certain issues that can crop up.
It's a constant onboarding problem.
Which is why I don't find that a particularly fruitful conversation, and
keep trying to steer us away from it. :)
And steer us on to "with 40 extra hours someone could add validation
approach X". Which would make the machines better, and by removing
something the humans had to think about, would actually make the humans
better as well. Basically the answer, in my opinion, to "Humans do
better" is actually "humans do less", especially when those humans are
already working pretty hard.
But I think we disagree on the mental load problem, and we just keep
going back and forth on that. So there's probably no more value in
discussing that. Let's just agree to disagree on that one.
Samsung Research America
sean at dague.net / sean.dague at samsung.com
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 547 bytes
Desc: OpenPGP digital signature
More information about the OpenStack-dev