Suggestions for improvement of the emulator
rob at codeweavers.com
Wed Sep 7 13:28:25 CDT 2005
Robert Lunnon wrote:
>On Wednesday 07 September 2005 19:35, Alexandre Julliard wrote:
>>Francois Gouget <fgouget at free.fr> writes:
>>>Yeah, maybe a very generic 'Needs review' email to wine-devel would be
>>>enough. It would also be the clue to the other Wine developpers:
>>> * that you're not going to be duplicating Alexandre's work if you
>>> review this patch
>>> * to look at the patch, dissect it to see what is wrong
>>> * if it is in your domain of competence and it looks good, post an
>>> approval message
>>> * to test the patch
>>> * and help its author get it accepted
>>That should really be the default behavior, all patches need review;
>>there's no reason to wait until I have looked at a patch to look at
>>it. If you see a patch in an area that you know anything about, please
>>review it, don't wait to see my reaction first.
>There is a problem here, you are presupposing the submitter is interested in
>reviewing the patch to the projects specification. This subverts the value of
>collective development if the submitter is unwilling then you lose the value
>of the improvement AND potentially the developer.
If the developer isn't willing to get the patch up to a high enough
level of quality, then they aren't going to get their patch in unless
someone else takes over their patch. I don't see how anything is going
to change that.
>It would be better to commit it to a branch to open it up to all to consider.
There isn't really much of a difference between having a branch and
having a patch in the wine-patches archives. Having it in a branch may
in fact be detrimental as it makes it easier to not do the work to get
it committed to the main tree. You could end up with 5 different
branches, all with different goals and none of the developers being
bothered to merge anything into the main tree so all of the features can
be used together.
More information about the wine-devel