stdcxx-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <>
Subject Re: [RFC] commit-then-review vs review-then-commit
Date Fri, 18 May 2007 19:58:03 GMT
Liviu Nicoara wrote:
> Judgment calls could be quickly trained by peer feed-back. Although we
> cannot precisely define what simple or obvious means in the context of
> stdcxx changes, an initial period of RTC for all newcomers -- and rusty
> committers as well ;-) -- would probably be beneficial.

+1, well said.  However, the 'initial period' isn't necessary - the point
is that they have submitted enough patches, followed enough dev@ dialog
by the time they are nominated that we *trust* they are ready to follow
the project's policies *before* they are voted in with commit access.

To help them, when they submit a patch through jira or here on list, it
would be good to call out 'thank you, committed your trivial fix without
further discussion' or 'thank you for your patch, please post it on the
[jira | dev@ list] for further discussion, as this is a non-trivial change
 - I'm +1 on adopting your patch'.

If we help them see that gray line with each submission they offer, they'll
be ready to follow the process when they get the commit bits themselves.

> Other than that I agree that CTR for trivial patches is good comon-sense.


View raw message