Re: What to do with "bugs forever" ?
Hello,
> Hello,
>
> Please, reply to mailing-list, so it's useful for others too.
Yes i do this now.
> On Sat, 2009-06-06 at 18:12 +0200, Karsten wrote:
>> Frank Lin PIAT schrieb:
>> > On Sat, 2009-06-06 at 17:06 +0200, Debian wrote:
>> >
>> > > here are two bugs that exists for years and where no solution is in sight:
>> > >
>> > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=252896
>> > > and
>> > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=518815
>> > >
>> > > What can be done with such problems that will be not solved and are ONLY
>> existing
>> > > in Debian?
>> > >
>> >
>> > What's the purpose of this emails? Upset someone? Actually help
>> > debugging the problem?
>> >
>>
>> Of course.
>>
>> > * Did you try to diagnose why/when it happen?
>> >
>>
>> Yes - i try to make the best diagnose to find the problem.
>
> I haven't read the but report carefully, but you didn't seem to provide
> information to identify why the bug "ONLY exist is Debian" (assuming
> it's actually the case).
I can't definitely say that this 2 bugs only exist in debian,
but i have spend much time to find a solution for them.
I only found many forum entries and discussions about this bugs from other users,
and they are always regarding debian or kubuntu.
So it seems that is a problem only regarding to debian.
>> > * Did you try a new version of the various software involved?
>>
>> I would do it if it is possible.
>
> Debian has two development version:
> Debian "Unstable" (Sid), has the latest version of a software, packaged
> for Debian. It is sometime incompatible with the stable version
> (because of unsatisfied dependency).
>
> Debian "Testing" (Squeeze), which is a "beta" version for the next
> stable version.
>
> It is possible to mix them, but I recommend installing a Virtual machine
> (see VirtualBox).
>
I know - i am using Squeeze.
>> > * Did you try with upstream version?
>> >
>>
>> What is the upstream version?
>
> "Upstream" is the name the name used to refer to the people who are
> actually writing the software.
> (Debian is a distribution, i.e an aggregate of 26000 packages coming
> from various upstream. Debian's goal is to release a distribution with
> all those packages, choosing the version that best works together).
>
You mean something like the "Original" ?
I can't see how this will help with this bugs?
It should be the intention to get a stable package and the errors occur in stable
or testing.
>> > - Did you ensure the bug was opened upstream [if you can reproduce it
>> > with upstream's version]?
>> >
>
> If the bug isn't specific to Debian, it is important that the people who
> write that software are aware of the bugs.
I already opened bugs at the source projects when i could identify them there.
Solutions where found that has also been integrated in Debian. :-)
>> > > Here the process of maintainers and bug reports is not working. :-(
>> > >
>> >
>> > The process is the same everywhere:
>> > - Diagnose the bug (Find which softwares are involved, why, when...)
>> > - Find a solution ( = Write one or more patches)
>> > - Test the patchs.
>> > - Submit the patch to the maintainer (to upstream)
>> > It takes time, effort, and user contribution.
>> >
>>
>> The problem is that there is a break after point 1 or 2.
>
> ???
> No, each point is a difficult task. It takes time and effort to solve a
> bug.
I mean that here there is a break in the analysis although the problems are not
existing in other distributions.
>> >
>> > *YOU* can help fixing the bug. (or you can pay someone to fix it).
>> >
>>
>> I am at work.
>> And there are many problems i could help to point out or solve them.
>>
>> This 2 examples are problems that are remaining for years without a
>> solution
>> *AND* they are only existent in Debian.
>
> Well great, so you can investigate what's the difference between the
> distribution that work and the distribution that doesn't work.
That maybe help.
But if the distributions use different versions of the driver or configuration
it is very difficult to interprete the differences when you don't understand
the code at all.
So you need help from "experts" at this point.
Mostly there where experts by the package maintainers.
That's what we need in this case.
>> That's the reason i write this email. ;-)
>>
>> > Thank you for reporting bugs.
>> >
>> > Franklin
>> >
>> >
>> >
>> > FYI
>> > The company I am working for is paying an awful lot of money to
>> > get support from various software vendors.
>> >
>>
>> I am doing the same.
>>
>> > And guess what, we have lots of bugs that are unsolved (not to
>> > mention the bugs that are "behavior by design", "unsupported
>> > tools" [that are only available for those who pay for the support,
>> > lol], or "fixed in new release" but we can't backport the patch
>> > because it isn't open-source).
>> >
>> >
>> Yes - i see it like you.
>> But in this case we are talking about problems that are solved.
>> Somehow they only exist in Debian.
>
> Other distro also have their own [bugs|problems].
Of course and for my purpose debian is the best.
> *THE* main difference is that all Debian problems are public. (By Debian
> Social Contract #3 http://www.debian.org/social_contract )
>
> Posting to a user oriented mailing list is unlikely to help (I still
> don't believe there is a problem, except Debian could have even more
> DDs)
Everything is working fine in Debian!
I just want to write about 2 special cases where annoying problems are not solved,
although there is no problem with it using other distributions.
So my question is how can we make things like this >more< better?
Karsten
> Franklin
>
>
Reply to: