On Mon, 2010-08-30 at 22:33 -0700, Jesse Keating wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 8/30/10 10:23 PM, Adam Williamson wrote: > > Wow, and everyone gets categorical again. You can disagree with people, > > but this is a bit different. Yes, amazingly, things do matter on a six > > month timeframe in the 'real world' (whatever that is). Example - I got > > a laptop. The graphics on it don't work, except for vesa, which is very > > slow (can't play video, and yes, that matters to me) and doesn't run at > > native resolution (yes, that matters to me). If Ben doesn't land a fix > > for this in Fedora 13's kernel I have to either wait six months to be > > able to use my laptop, upgrade it to F14 (which kind of obviates the > > point of having a stable release), or run the proprietary driver. None > > of these feel like great options. > > > > Just saying, yes, sometimes, it really is reasonable to not want to wait > > six months for something to get fixed, and please try to at least > > vaguely consider the possibility that your theoretical principles may > > not always apply perfectly. Maybe we don't want Fedora to be Rawhide, > > but maybe we also don't want it to be a distro where nothing ever gets > > updated either... > > Where do you see somebody proposing that no updates be issued? Where do > you see somebody proposing a setup where fixing a graphics card can't be > done in the stable release kernel? You've built up a nice strawman that > you've lovingly kicked down. It's implicit in what Jon said; I was pointing out that he was, possibly inadvertently, suggesting a principle that was far too strict. I'm not saying anyone seriously proposed not issuing updates, but the arguments being used in support of slowing down updates are starting to sound extremely categorical and in the 'you're just wrong and no-one wants that' line. That's what I don't like. Jon's quote, exactly as I quoted it in my email (you removed it): "Instead, we have people so opposed to a little sanity and so scared of 6 months wait - like anything really matters in that small a timeframe in the *real* world" My point is, yes, a 6 months wait *does* matter in the real world, please don't use that argument, or qualify it properly. > I don't think anybody wants to prevent fixing a graphics chip issue in a > kernel update. What we wouldn't like to see is dragging in a brand new > xorg major release and kernel major release and dri major release in > order to make that happen. Do you see a difference there? Of course, now you're the one building straw men. You know I see a difference there, get off the rhetorical horse. I also know that there's a continuum, not nice clean distinctions like that every time. What if the patch is quite an invasive one and we don't want to backport it? Distros with conservative update policies can and do reject updates that fix bugs, and only fix bugs, because the changes are too big; Ubuntu and RHEL have both done this in multiple cases. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
