[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: No answer to easy bug policy (and hal examples)



On Mon, 2008-07-14 at 10:10 -0400, Adam Jackson wrote:
> Every package owned by xgl-maint that I know of has cvsextras+ set.
> If there's a trivial bug in one of these packages, please, by all
> means, fix it.

Generally, is there a place to pool "easy" bugs that members of the
cvsextras group could fix or escalate? (I couldn't find one).

Specifically, there are a couple of F8 hal crasher bugs (#431377 and
#452701). I've recently isolated existing upstream patches for them, but
they need an errata/backport to F8. I reckon accepted upstream patches
probably means they're "easy".

Is this the sort of thing the cvsextras group can deal with?

I'm doubtless being unfair to the hal package maintainer - it may well
all be in hand - but I'm tainted by my experience of a pilot-link fix
being blocked for 5 months on a hal update; even advice/a response from
the hal maintainers would have been endlessly helpful (it was left
NEEDINFO). I find it particularly odd given that hal is key to systems
operation.

I'm sure it happens because of the very understandable swamping reasons
Adam mentions... but nonetheless, as someone who's been trying to solve
the odd problem and provide the odd patch here and there, it's quite
frustrating and disheartening to have your efforts stalled - or even
undone - by these non-technical issues. Yeah, I know that's just how
life is ;)

I say, I only dug up the upstream patches recently. Perhaps I'd have
done this at an earlier point, and would certainly be encouraged to do
so again in the future, if I were more confident they'd make it into the
package in a timely manner.

So perhaps what's needed is a mechanism for a trusted group (cvsextras?)
to enact what are essentially solved bugs? If they don't feel confident
about applying and pushing a fix, then at least the package maintainer
should be left with a smaller "non-easy" bug list.

Regards,

kev.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]