On Fri, Apr 08, 2011 at 10:14:54PM -0000, Jamie Strandboge wrote:
> To me, this is a pretty big change less than three weeks to release. I
> appreciate the hard work that went into examining potential breakage and
> on the surface things seem generally ok and workable, but considering
> how part 1 of the mutliarch changes went, I have some concerns (not
> being critical here, just that these are rather fundamental changes we
> are discussing and I think that is worth some pause).
Yes, I recognize the cause for your concern and understand completely where
you're coming from. If this is considered too high-risk, it's not the end
of the world if we don't include this change - it would make it easier for
progress to be made on multiarch between now an oneiric opening, but it's
not critical.
I think if anything else did turn up that broke because of this change, it
would be less effort to fix that than it would to maintain an out-of-archive
linux package. And there's definitely much less risk here than with the
earlier change, because software has much less reason to probe paths to
asm/*.h than to probe paths for libraries.
> Could something not be coordinated with Lucas? I thought archive rebuilds
> for him were in the neighborhood of 8 hours (that is total hearsay-- I
> don't actually know).
That's a good idea; I've pinged him on IRC, hopefully he'll get back to me
soon.
If we can get a rebuild scheduled and I can commit to addressing any new
build failures that turn up as a result, are you happy for this change to go
forward in the meantime? If not, I think we should forgo this entirely for
natty due to the timing.
> @Steve, In all honesty, I have a hard time imagining you letting this
> change through if you were release manager. :P
I would be asking all the same questions if the tables were turned, but the
thing about being the developer proposing a change is that it changes your
perspective in ways that are difficult to filter out, which is why members
of the release team also use the FFe process. :)
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
<email address hidden> <email address hidden>
Hi Jamie,
On Fri, Apr 08, 2011 at 10:14:54PM -0000, Jamie Strandboge wrote:
> To me, this is a pretty big change less than three weeks to release. I
> appreciate the hard work that went into examining potential breakage and
> on the surface things seem generally ok and workable, but considering
> how part 1 of the mutliarch changes went, I have some concerns (not
> being critical here, just that these are rather fundamental changes we
> are discussing and I think that is worth some pause).
Yes, I recognize the cause for your concern and understand completely where
you're coming from. If this is considered too high-risk, it's not the end
of the world if we don't include this change - it would make it easier for
progress to be made on multiarch between now an oneiric opening, but it's
not critical.
I think if anything else did turn up that broke because of this change, it
would be less effort to fix that than it would to maintain an out-of-archive
linux package. And there's definitely much less risk here than with the
earlier change, because software has much less reason to probe paths to
asm/*.h than to probe paths for libraries.
> Could something not be coordinated with Lucas? I thought archive rebuilds
> for him were in the neighborhood of 8 hours (that is total hearsay-- I
> don't actually know).
That's a good idea; I've pinged him on IRC, hopefully he'll get back to me
soon.
If we can get a rebuild scheduled and I can commit to addressing any new
build failures that turn up as a result, are you happy for this change to go
forward in the meantime? If not, I think we should forgo this entirely for
natty due to the timing.
> @Steve, In all honesty, I have a hard time imagining you letting this
> change through if you were release manager. :P
I would be asking all the same questions if the tables were turned, but the
thing about being the developer proposing a change is that it changes your
perspective in ways that are difficult to filter out, which is why members
of the release team also use the FFe process. :)
-- www.debian. org/
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://
<email address hidden> <email address hidden>