[Mageia-dev] Support policy

[Michael Scherer]

Le mardi 30 novembre 2010 à 13:31 +0100, Samuel Verschelde a écrit :
> Hi, 
> 
> I would like to discuss the support policy for Mageia.
> 
> It would be interesting to know (or decide) where Mageia is heading, given our limited resources :

can you first explain what are our ressources so you can explain how
limited you think they are ?

> 1) focus on stability and security : few very well equally supported packages. Apparently, this is 
> where we're going for now.  May be wise as a start, but I hope this is not our final destination, 
> because it means either very limited choice, or progressive diminution of quality of support if the 
> number of packages increases faster than the dedicated resources.

If we let people upload lots of packages without taking care of them,
yes, this will result in a decrease of quality. 

And if we do not increase ressources, this is what will happen. But I
said in lots of mail that increasing ressources must be our priority.

And so basically, the question can be summarized as "do people prefer
packages that work or lots of packages" ?

Quality or quantity ?
Yes, it is a black and white choice, but whatever the way you present,
you will ultimately have to do this choice at some moment. 

Choosing one doesn't mean that we should not have the other, obviously,
but when there is a choice to make, you can't avoid it.

So let's compare what would they bring :

Quality focus would bring :
- more confidence for admins to use the distribution
- more confidence for people that rely on it for various reasons
- less unhappy users on forums ( or at least, not unhappy because the
distribution is doing something bad )

Quantity focus would bring :
- arguments to convince people to use the distro "we have lots of rpm,
come here" ( as expressed in the previous thread )
- likely more users ( because this would fullfill more needs )

In my opinion, focusing on quality will likely attract more technical
contributors. People that value something that work, that are willing to
devote time in bug reports, sending patchs because they know we also
focus on quality. People that are skilled enough to become packagers, QA
team member, or to give useful advice on forums, etc.

And I think bringing contributors to fuel the development of the
distribution should be our first concern. First concern because that's
what allow to offer a good system for users, because that's what make it
sustainable in the long run, because the work of everybody will be
easier with a good system backed up by a solid technical community.

Of course, that doesn't mean that we should not strive to have lots of
packages, and to try to provides choice to everybody. Just that this
need to be done in scalable and sustainable fashion. This usually mean
1) later and 2) slowly.

Quality and quantity are not incompatible in the end. 

But we can't start to aim both quality and quantity at the same time.
Deciding to import all package or to review them is for example a case
where you can't focus on both at the same time.

And so I think that to have quantity in the end, we must first choose
quality for the start to make the distribution grow on solid foundations
( and whith growth bring quantity ).

Starting with quantity over quality doesn't increase quality at all in
the long run. This will bring more unhappy people ( see the various
complaint on Mandriva forums about "there is too many low quality
packages"), will instill a mindset that would be less than ideal for
contribution ("bug reporting is useless, people do not answer" ), and
therefore may not allow us to get enough ressources to grow and survive.

So to me, the goal should be first to aim for quality, and then add more
packages, as I said, in a sustainable way ( ie no mass import without
being first sure that we can manage them, no rotting rpms because people
no longer care ).

Which mean option 1.

Yes, option 1 will likely frustrate people, because their favorite
packages might not be there. 

But we must realize that when we started this fork, we started a long
journey. And that things will be different at least for the start, and
that even when the first stable will be released, we will not have
finish our work, just really started it.

-- 
Michael Scherer