[Mageia-dev] what is this ????

[AL13N]

Op zaterdag 24 november 2012 20:09:05 schreef Romain d'Alverny:
> 2012/11/24 PhilippeDidier <philippedidier at laposte.net>:
> > Sander Lepik a écrit :
> >> 24.11.2012 18:55, PhilippeDidier kirjutas:
> >>> NO !
> >>> You mean someone is paid and create a third repo for something Mageia
> >>> doesn't want to import...
> >>> 
> >>> If that thing brings a mess into Mageia,  we will see bug reports in
> >>> bugzilla and lot of time lost by Mageia bug-triagers, devs, packagers,
> >>> before discovering that is not a Mageia problem but that this stuff
> >>> brought some shit !
> >>> We suffer of lack of time... and this will consume more time from
> >>> voluntary contributors to repair something badly done by someone that
> >>> was paid for it !
> >> 
> >> Well, we have such repos already today and you can't stop something like
> >> that. But you saw my example the wrong way. Mageia doesn't have to
> >> support those repos and problems caused by such packages.
> > 
> > That has been a problem for Mandrake with Thac repo, a problem for
> > Mandriva with MIB repo... and sometimes it took a long (wasted) time to
> > understand that a reported bug was induced by something imported from
> > third repo.
> > NB plf repo was something else : mandriva devs worked on it !
> 
> Still, we can't prevent that *. So perhaps should we:
>  - analyse it and find a clear model that explains it, so we can
> change what could be changed;
>  - even embrace it, that is, have a strategy to have more such
> external repositories to merge back into ours, or to better
> tag/mark/recognize what can come from something we do support as
> Mageia (because we produced it, or because it matches our
> requirements) and from something we don't.
> 
> * it's a symptom that means that most of Mageia (as a whole product)
> works for some people, but some parts of it don't (so they
> duplicate/manage things on their own to make it quick, or more
> controlled for them).
> 
> Still. The bottom line of having calls for bounties, people ready to
> pay to have some requests in Mageia answered in a more controlled way,
> is perfectly fine. Having them including in their requirements that
> the work be merged into (if possible), or made available to the parent
> project is even better.

i guess it needs to be policy'ed:

like: requiring that the fixes are done in Mageia itself (no 3rd party 
repository), and as usual that the fixes are done in accordance to our 
policies.

then, structure needs to be clear with a status:

[requested] To fix bug #X (preferably option Y); <user> donates Z
[taken by <user>] To fix bug #X (preferably option Y) <user> donate Z
[solved by <user>] To fix bug #X (preferably option Y) <user> donate Z
[accepted] To fix bug #X (preferably option Y) <user> donate Z

any other communcations or status, reject or whatever, doesn't need to be here 
and is between the parties involved.

I would also ask that if possible the user who solves it donates a part to 
mageia, not required.


lastly, i note that the page is a draft, so this could very well mean that the 
person who wrote this planned to bring this in a meeting.


in short, i don't really see any issue with it.