Hi Laca,
Thank you Laca for putting together this nice and simple code review
process
http://www.opensolaris.org/os/project/jds/documents/code_review/
I need some clarifications on these statements when taken together:
* bug category owners must review all patches in their category and
respond. If the patch looks good and you have no comments, just
reply with "approve"
* reviews are non-blocking: you can go ahead and commit if you are
confident enough that the fix it good, but be aware that you may
need to revert or amend your changes if your peers or the upstream
community have issues with it
* you must address all questions or concerns
Q1. Who is the final 'authoritive' approver? (e.g. bug category owner
approve a patch, but non category owner opposes it or raised questions
and concerns so is the patch approve or not?)
Q2. If the submitter is going ahead to submit the patch anyway, what is
the point of addressing all questions and concerns?
(Okay, I am over-stressing here a bit, but this is one possible
interpretation of the statements).
With regard to this statement:
* your changes are considered approved if no issues are raised
within 3 working days or if all issues raised are addressed
This implies the lead time for a patch approval is 3 days. This is too
long.. I think.
I think if the category owner has approved it then it is considered
approved, however, others can still raise concerns and questions
(regardless of time) that still need to be addressed but do not let the
process stopped. Ultimately any concerns raised on this list will most
likely to be raised by the community.
-Ghee
Laszlo (Laca) Peter wrote:
> The JDS Core review process posted earlier is now in action.
> See http://www.opensolaris.org/os/project/jds/documents/code_review/
> for the description of the process.
>
> A new mailing list called jds-review was created and all Sun
> contributors were subscribed. Anyone is welcome to join at
> http://mail.opensolaris.org/mailman/listinfo/jds-review
>
> Everyone, please start submitting your svn diffs to jds-review
> prior to committing.
>
> Thanks,
> Laca
>
>
>
