[openstack-dev] auto-abandon changesets considered harmful (was Re: [stable][all] Revisiting the 6 month release cycle [metrics])

James E. Blair corvus at inaugust.com
Tue Mar 3 15:49:03 UTC 2015

John Griffith <john.griffith8 at gmail.com> writes:

> ​Should we just rename this thread to "Sensitivity training for
> contributors"?

I do not think that only new contributors might feel it is negative.  I
think that both some new and long-time contributors do.

My oldest patch is from July -- it's still relevant, just not important.
I just picked up and finished a series of four patches that Jay Pipes
left sitting for a month without updates -- but they are really good and
should be merged.  Yesterday, Monty went through and either cleaned up
or abandoned his patches that had been sitting for several months.

None of these things offend me or anyone else involved, and they are all
enabled by the fact that those changes were not abandoned.  And none of
these people are new contributors (the opposite, in fact).

If I had to deprioritize something I was working on and it was
auto-abandoned, I would not find out.  It would simply drop from my
personal list of patches I am working on, and I would never think about
it again.  Until, one day perhaps, I see something and think "hey, I
thought I fixed that".  And after a long time digging, I find a patch
that someone else abandoned for me.  I would be furious that they had
made the decision on my behalf that I would do no more work on that.

Not everyone's workflow is like this.  Not everyone feels the same way
about the word "abandon".  But we have a diversity of contributors in
the community and we have a diversity of workflows.

We need to help core reviewers focus on patches that are useful to them
-- no doubt about it.  We need to let new contributors know that we
expect them to engage.  We have better tools to do both than abandoning,
which has negative impacts beyond the immediate purpose of its use.

This thread has uncovered some information about how people use Gerrit
and where we need to put effort to make sure patches are prioritized
correctly.  That is really useful information, and I hope other folks
will chime in so we can make sure we cover all the bases.


More information about the OpenStack-dev mailing list