On Mon, 13 Jan 2003, David Dawes wrote: >>Nope. There are clearly times XFree86 should say: this shouldn't be fixed >>in XFree86: the distro should fix their breakage. For example, I think >>it would be correct to do this when a distro has a distro specific compiler >>bug: making XFree86's code worse forever for a transient problem >>is a *very* bad idea. >> >>In my personal experience, I build XFree86 on both RH and Debian, typically >>with no problems on either. > >The problem I see here is that XFree86 is not only a Linux-based >system and this triage method would give Linux bugs a leg up >over OSs.
Of course XFree86 is not just a Linux based system. I seriously doubt that anyone would consider it such. And a bugzilla system for it would not be Linux specific either. If someone tried to make it Linux specific, I'd be one of the first ones on top of them. >This would skew all work done on XFree86 to Linux at the expense >of other, not has heavily supported, platforms. How so? >Another problem I forsee is that the bugs would be geared >towards new or newer hardware, because most of the people >reporting bugs have just gotten their Tablet, Laptop or >whatever. People don't "gear bugs towards" things. The software either has bugs, or it doesn't. If it does have bugs, then hopefully people will report them, regardless of the OS that they use. The question really is if developers want to know if bugs exist or not, or even care, and that is OS nonspecific as well. Very very few bugs should be OS specific, and while someone might file a bug and click on the OS field and choose say "FooIX", it would be foolish for a Linux developer to skip over this bug without investigating it and seeing wether or not it is OS generic - which I believe most bugs are highly likely to be. So, I see this claim as poppycock. -- Mike A. Harris _______________________________________________ Devel mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/devel
