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

Re: Review queue/FESCo after the merge



On 19.11.2007 00:55, Jesse Keating wrote:
> On Sun, 18 Nov 2007 21:43:05 +0100
> Nicolas Mailhot <nicolas mailhot laposte net> wrote:
> 
>> I stated when I started the SIG I didn't want the glorious leader
>> position, so I won't pretend I speak for it, and I'll only object for
>> the three affected packages I maintain or co-maintain.
>> If this change is to be done I'll insist his author explain and defend
>> it on the SIG list, then write a formal FPC proposal, and get this
>> proposal approved all the way up.
>> This is a "fix"¹ in search of a problem², and while I've been known to
>> make gratuitous changes to make a point too³, I've limited myself to
>> my own packages, and I'll appreciate the same restraint in my fellow
>> Fedorans.
> So it would seem we're /not/ in agreement on this point.

Nothing new -- I leave that debate to Nicolas and the Font-SIG together
with you and the other members of the FPC, as I don't care much about it.

>  Thorsten, are you trying to fix a real problem,

Well, I run into problems once due to the wrong exit code. That was
likely a corner case, but there are likely other corner cases with lead
to the rule spot mentioned ("script should not exit with non-zero exit
code"), which IMHO is a good one.

> or just trying to ensure everything
> follows a guideline regarding exit failures in %post scriptlets?

I just want to raise the point that FPC creates or modifies rules that
are often not really (or very slowly) being realized in existing the
packages.

Someone in a lot cases (like this one) could just realize many of the
changes in the Packaging Guidelines directly in CVS for all effected
packages. I think we should do that way more often and get away from the
mantra "this is my package and nobody else is allowed to touch it" --
that's why I took this particular issue as example to just show how it
could be done. Now I expect from FESCo advice for this particular
example as well as the general concept where one persons realizes
changes in all effected packages directly in CVS.

CU
knurd


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