[Mageia-dev] Bye Bye Mageia

Donald Stewart watersnowrock at gmail.com
Mon Nov 26 19:48:03 CET 2012


On 26 November 2012 18:25, Joseph Wang <joequant at gmail.com> wrote:
> One reason that I kept rather quiet is that e-mail conversations on
> mailing lists
> have a tendency to become flame wars, and I really want to avoid that.  It's
> very easy in public conversations to come across as a jerk, and as a newbie
> I really wanted to avoid that.  I've tried whenever finding myself in unfamiliar
> social situations to just keep my mouth shut, and listen to what's going on
> for a while before saying anything.
>
> The thing about communications is that it really doesn't scale.  One
> thing that's
> nice about the open source model is that you can go into your garage, work on
> something, and you don't need anyone's permission or to communicate or
> coordinate
> with anyone.  Coordination is a real pain, and you really want to
> structure things so
> that you can minimize coordination. Communication is also a problem.  You say
> one thing, someone replies, pretty soon you have a flame war, and you
> aren't doing
> and "real work."
>
> The thing that I'm working on in my garage is a linux workstation that
> is set up for
> hard-core astrophysics theory.  Packaging is a missing piece of
> scientific software
> since there are hundreds of scientific software packages that are not packaged.
>
> Rather than engage in vaporware, I just need some stable distribution
> that is very
> open to adding new items into some "bleeding edge" repository.
>
> I ended up with Mageia partly for historical reasons, but partly out
> of a sense that
> because it was a community distribution, it would have some easy mechanism
> for accepting "bleeding edge" packages.  Once I got through initial
> packaging learning,
> I was planning to add things like a stellar evolution code and hard
> core CFD code.  I'd
> like to add some professional astronomical telescope tracking software
> (like IRAF or
> DS9), and to hard core astrophysics research.
>
> My assumption was that as a community driven project, there would be
> some mechanism
> for adding new packages to the system, and that I wouldn't have to
> worry about getting
> permission, I would just do it.  Fedora already has a mechanism for
> doing that, but sense
> I was already using Mageia, I had thought that Mageia would be at
> least as open as
> Fedora, and that it would be easy to add large numbers of new packages.
>
> Now if I'm mistaken about this, and this is not the goal of the Mageia
> maintainers, then
> I just need to find some other platform to work on.
>
> Again, it makes perfect sense to me not to put cinnamon into Mageia 3
> core.  It's
> unstable and buggy and it's going to be a pain in the rear end to get
> it to work smoothly.  The
> problem is that if it's not possible to put Cinnamon *somewhere* in
> the Mageia tree
> so that "bleeding edgers" can work on it, then it's going to be
> impossible to use Mageia
> as a distribution mechanism for even more bleeding edge experimental
> software, and if
> that's the intent, then I've just got to find another distribution to work on.
>
> I'm not trying to be a jerk or to blackmail anyone.  It's just that if
> there is no mechanism
> for Mageia users to share bleeding edge software with each other, then
> it's not going to
> work for what I want to do with it.  urpmi and cauldron is a great
> mechanism for two nuclear
> physicists to share say the latest nuclear equation of states
> libraries, and as something
> that advertises itself as a community distribution, I was hoping that
> Mageia could be the
> center of that.  One problem that we have here is that everyone wants
> to copy Apple OSX
> and Android.  The Ipad and MacOS is a slick piece of software.   The
> trouble is that it
> only lets you do what Apple wants you to do, because if you do
> something really different
> you might break the box.  This is really, really bad for real hard
> core, scientific research
> since what you are trying to do is to push the machine to the point
> where you are breaking
> the box, and it's hard to communicate in advance what you are trying
> to do, since you don't
> know.
>
> If that's not what people want to do with Mageia, then I just have to
> accept that and move
> elsewhere.......

The thing about communication and coordination is that unless it is
there, and people are aware of what others are doing, then huge
amounts of work will get repeated. Now I am not saying that you need
to keep in communication over every single thing that is going on, but
it is good to have a background knowledge and just to check in. The
better informed we are about what you are doing, then the more that
others can help you.
As you say, sometimes it is hard to know what you are going to do, but
it is still important to be aware of the projects goals, now it is
unfortunate that you have done work on cinnamon, however, as in the
other thread, it seems that there is a possibility that it can be put
into updates/testing. And as you said, it is not ready from the prime
time, so that offers a very good solution.
Now I feel that there would be a far greater reward to start on the
other packages, as there is far less in there way, and infact we
welcome scientific packages, there was a drive a few months back to
get as many included in the distro as possible, there is a lot of GIS
stuff, and I am sure that the astrophysics software you have in mind
will have a place.
Bear in mind that we have switched to a new rpm group policy, so
please make sure that the groups that you put the software in fall
into them, similarly for the libs policy. This is not trying to
conform or limit your work, but trying to guide things so that they
fit in with the rest of the distribution, it is really important that
we offer as uniform method on these things to aid the user in
installing the software.

Schultz


More information about the Mageia-dev mailing list