[gst-devel] Re: thomasvs gstreamer: gstreamer/ gstreamer/gst/

[Thomas Vander Stichele]

On Mon, 2004-04-05 at 16:48, Benjamin Otte wrote:
> Disabling something because it doesn't work and you don't know how to fix
> it is not ok, especially not if you just do it without asking the people
> who did this first.

Commiting something that breaks the distcheck is also not ok, and was
done prior to my revert.  I realize this is a difficult topic and that
people's opinions differ, but as long as we don't discuss this
thoroughly and come to some conclusion on the topic it's basically a
difference of opinion.  The difference in our opinion obviously being
that I don't think it's ok to commit stuff that breaks regression
testing, and I think it's ok to revert stuff that broke the regression
testing.

> And for the record: I consider a working test much more important than a
> distcheck that leaves files somewhere.

On an unstable branch, *maybe* I'd agree.  On a stable branch however,
there is NO WAY we can do regression testing if we feel it is OK to
commit changes that break the regression testing.  For the simple fact
that such a commit can easily obscure all ensuing commits that could be
breaking it even worse.

Thomas


Dave/Dina : future TV today ! - http://www.davedina.org/
<-*- thomas (dot) apestaart (dot) org -*->
I feel like I'm at something you always say you need more time
Well I'll stay right here and I'll wait for good
until I find a love worth mine
<-*- thomas (at) apestaart (dot) org -*->
URGent, best radio on the net - 24/7 ! - http://urgent.fm/