[webkit-dev] On frequently hosing other port's WebKit2 builds

Antonio Gomes tonikitoo at gmail.com
Wed Oct 6 21:33:00 PDT 2010

Maybe making EWS bots build already r+'ed patches could help here as well.

On Wed, Oct 6, 2010 at 7:13 PM, Andras Becsi <abecsi at inf.u-szeged.hu> wrote:

> Dear WebKit Community,
> In the past few weeks there were several occasions where Apple-WebKit2
> developers ignored the red Qt EWS on Bugzilla and nonetheless committed
> their patches and let the Qt bot stay red for several hours without any sign
> of trying to fix the issues (mostly introduced by thoughtlessness), or
> asking for help on IRC, or even the webkit-qt mailing list, or CC'ing
> somebody to the bug who might be able to help fixing the problems before
> hosing the tree, like with these bugs:
> https://bugs.webkit.org/show_bug.cgi?id=47281
> https://bugs.webkit.org/show_bug.cgi?id=47239
> https://bugs.webkit.org/show_bug.cgi?id=47097
> https://bugs.webkit.org/show_bug.cgi?id=46585
> https://bugs.webkit.org/show_bug.cgi?id=46432
> https://bugs.webkit.org/show_bug.cgi?id=46054
> https://bugs.webkit.org/show_bug.cgi?id=46043
> This is making the lives of the maintainers miserable because if a bot is
> already red it is completely ignored by other developers, which makes it
> really hard to catch up.
> I respectfully think that this kind of arrogant and ignorant way of
> development does real harm to the community.
> We have a really fast Qt EWS and the build failures weren't complex ones,
> the EWS would have only needed a minute more than the style-bot to have a
> nice output where the errors caused by left-in old API calls or old includes
> or left-out extra compiler definitions would have shown themselves, and I do
> not think waiting this few minutes would hurt anyone. Especially because
> this kind of behaviour is not only disrespectful to us, but also to Eric and
> Adam who are working hard to make these awsome QA technologies possible.
> Comments like this https://bugs.webkit.org/show_bug.cgi?id=47097#c10 are
> really frustrating, because all contributors are supposed to bother with the
> Mac xcodeproj files and their cryptic hashes, too (not to talk about the
> other nearly dozen build systems) if they implement something, and
> stringently have their patches checked on all the EWS.
> Surpassing all my private disappointment, I think IRC, Bugzilla, the EWS's
> and other buildbots aren't there to force some kind of an unwanted ritual on
> developers, in which they only upload patches to Bugzilla to conform some
> policy, but to make collaboration possible.
> I do not want my letter to be looked at as an offense or attack, I'd rather
> like to kindly ask the responsile ones to stop the ignorant and
> discriminative way of handling other ports, and at least try to show some
> respect on other developers work.
> Sorry, if this souded like an offense.
> WBR,
> Andras Becsi
> On behalf of the QtWebKit Team at the University of Szeged
> _______________________________________________
> webkit-dev mailing list
> webkit-dev at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

--Antonio Gomes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20101007/869b0a69/attachment.html>

More information about the webkit-dev mailing list