Hi Jacopo,

I'm not for a general change to Review-Then-Commit but pleading for a more sensible way to handle these things as a responsible committer.

As a guideline: if in doubt, always provide a patch and describe what the change should do or fix and let others review.

Of course not for trivial changes or fixes. But when it comes to new modules, significantly changing the behaviour of the business logic etc. I think it's better to apply Review-Then-Commit.

And of course, if others want a commit to be reverted, we should do it first and then discuss.

Regards,

Michael


Am 27.03.17 um 10:07 schrieb Jacopo Cappellato:
On Sat, Mar 25, 2017 at 1:21 PM, Michael Brohl <[email protected]>
wrote:

+1

The lack of code documentation is not a free ticket to just change the
code behaviour without proper analysis.

The right process should be

1. discuss

2. provide a patch

3. let others review/comment

4. decide

5. commit

It is really dangerous to easily change code like this.

Jacques, please be not so hasty with committing stuff. We have had a lot
of similar cases with reverts, committing half done solutions and such
lately. And please be aware that others might not have so much time to
follow every commit in detail, analyze and comment promptly.

It really worries me because we lose quality and it's not easy to detect
errors and changed functionality in such a complex project. And don't rely
too much on the tests as we don't have such a high test coverage.

Thanks for some more patience,

Michael


I am fine with continuing with the Commit-Then-Review approach (and
leverage Review-ThenCommit approach only when more appropriate) provided
that the committer is willing to accept (by reverting or performing further
work as requested by the reviewers) negative reviews. What we should really
avoid is committers dragging their feet, pushing back reviews, trying to
argue and negotiate to keep their code as committed.

Jacopo



Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to