On 05/11/2009 06:56 AM, Michael Nielsen wrote:
Hi,
I've been told this is the right place to place this debate starter.
Not to demean the fine work that has been done in maintaining fedora,
however the distribution is slowly killing itself, being destroyed by
contradicting philosophies. Many of the problems have been directly
copied from the Windows world.
The main problems are.
I don't agree with you on some of your points, but I do support your
overall point. This is something that has been bugging me for years
also, but does seem to be especially an issue lately. I have come up
with a few different ways to look at it.
It seems to me it isn't always even a server vs desktop, but a
desktop vs laptop divide. The server generally sorts itself out in a
fairly clean fashion, and that is especially true if you use something
like CentOS for servers. By the time something makes it into CentOS from
Fedora it has been polished enough to be in fairly usable shape. I still
don't always like they direction things take, like say all the anaconda
changes, but it is usable. The bigger issue is desktop vs laptop users.
Which also can be viewed as older power users vs new users, and
sysadmins vs programmers. These dichotomies cause a lot of grief in a
number of ways.
Examples:
Desktop vs Laptop:
NetworkManager vs /etc/sysconfig/network-scripts:
NetworkManager is great for laptop users, because it deals with
wireless, especially wireless with encryption in a fairly graceful way,
when it works. It is not so great for desktop users who have no
encryption needs, or mobility and find NetworkManager less reliable.
Which is why desktop users complain when it becomes the new default.
PulseAudio(esd, artsd, etc) vs ALSA(direct):
PulseAudio, or something like it, is something needed for laptop users
to do software mixing on their generic onboard sound cards. Yet for
desktop users, especially power users, who still have SB Live!, Audigy,
Turtle Beach, etc cards with hardware mixing, things like PulseAudio are
just more trouble than they are worth.
There are various goals in conflict. The programmers seem to have the
laptop/new user mindset at heart. Part of this mindset, which I have
seen in this thread, and you see all the time is "Why are you whining to
me, you don't pay my paycheck". While out of the other side of their
mouth they say, be a part of the community. But the you don't pay my
paycheck aspect doesn't quite jive with the be part of the community
aspect. Another aspect of be part of the community is, well if you don't
like it, write your own patch. The problem is these are generally power
users and/or sysadmins. While they are technical enough that they could
program, they are too busy with work and keeping up with all the changes
to the various programs. That they can't spend all their time submitting
patches to every project that is going in a direction they don't like.
On top of that, when they do submit patches, more than half the time
they developer/maintainer will say they aren't interested in the patch.
So why should they bother in the first place.
Open source largely lets the programmers do whatever they want. On
the other hand, for the users(power users) it becomes a dictatorship per
program. Fedora leaders tell us to go to upstream. In the past we have
gone to upstream, and have been ignored. The issue is also a grey area
since many of the maintainers are also the upstream developers.
What is it going to take to get the programmers to take power user
input before they go off coding some new idea? What is it going to take
to get them to not make everything a work in progress when it is
critical desktop infrastructure like sound and networking?
--
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list
