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

Re: Agressive FUD by Fedora contributor



Le Lun 16 octobre 2006 01:53, Axel Thimm a écrit :
> On Sun, Oct 15, 2006 at 04:14:46PM -0700, Christopher Stone wrote:

>> You seem to be completely ignoring the **FACT** that Fedora users
>> who use ATrpms end up with messed up systems in which they can no
>> longer properly upgrade using yum.
>
> That is what I call FUD. I count a dozen thousand unique daily
> visitors on ATrpms' master mirror and if one could get metrics from
> the other mirrors there would be quite a huge number. So there are ~
> 15000 users with broken systems daily? Or are do they all belong to
> the category "have been very lucky" as you write in bugzilla?

Actually given the size of the Fedora community a dozen thousand is not
much (especially when some high-profile apps such as mythtv are only
available through atrpms). Though I'll concede it's way harder to find how
many of the Fedora users not using atrpms do so because of a bad past
experience, how many because of third-party advice, and how many just
aren't aware of atrpms existence.

Anyway let's forget the past and work on the future;) Since the conditions
which caused atrpms to be created arguably no longer exist (Fedora not
doing EE, slow bug fixing within FC, Axel not being part of Fedora
instances...) how about aiming for atrpms resorption by FC7 time ? (with
atrpms only providing rebuilds for releases/packages Fedora does not care
about).

For example:
1. create an atrpms tracker bug in FE and L repo
2. for every package in atrpms create an issue in those bugzilla, blocking
this tracker (issue being: review request for new package, FC/FE bug for
duplicate packages if the duplication has a technical reason)

Surely if as some claimed atrpm is causing a high number of problem for
users there should be no dearth of people working on those issues.

This way next time someone complains about atrpms, Axel can just point him
to the tracker:)

Regards,

-- 
Nicolas Mailhot


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