Karsten 'quaid' Wade escreveu:
On Thu, 2008-10-23 at 13:43 -0300, Luis Felipe Marzagao wrote:
  
I think this is an important issue. I must say it´s impressive how 
Fedora has been changing and giving much more attention, on each 
realease, to the needs of end-users. We don´t need to go very far. If we 
compare FC5 and F10, we can see there are huge differences in terms of 
end-user support. NetworkManager is one great example.
    

In many cases, Fedora is just the face for upstream software
development.  Take GNOME as an example -- there is a project with a
strong opinion about user interface design and human interaction.  
We're not adding much value on top of that, except to channel bugs back
to GNOME.

I think we need to be clear about which problems we can solve within
_this_ project.
  
Exactly! It's clear it's not in the project objective to alter upstream software. And in fact I agree it shouldn't be!

The trouble is the end-user doesn't even know what upstream means. In fact, I think the end-user won't even want to know what it means, as long as the system is running fine. For him, Fedora is an operating system. And the GNOME example is very good for this matter. There are somethings that don't imply altering the core of upstream projects in order to make the "out of the box" user experience more happy :)

A single line, for example, could improve the user experience when entering GNOME on Fedora:

gconftool-2 -s -t bool /apps/nautilus/preferences/always_use_browser true

Bingo! A single line (maybe with some other adjustments) in any rc.local file or any other place specifically designded by Fedora Project should make the end-user experience a lot better. And it does not require any upstream intervention or any opinion change by GNOME upstream team. And is very simple to maintain. And that's it, the end-user would enter Fedora for the first time and would not complain about a zilion windows opening every time he clicks on a folder.

It's just an example, but my point is there are small things that makes the user experience better and does not require huge changes on coding. I think this is the kind of problem Fedora Project should pay attention to, as you adequately put.

  
Although this has been really nice, I think there are things to improve. 
I´m always hiting on the same key when it comes to end-users. I know 
Fedora cannot (and should not) go away from its objectives (which is the 
very reason of its success), but paying attention to the end-user voice 
should be added as one *serious* project objective also. I mean, side by 
side with "smooth updates", "rapid evolution" and the other good Fedora 
principles.

I don´t have any doubts about the quality of the software behind Fedora. 
And I´m aware a lot of people don´t mind if Fedora has a lot of 
end-users or not. And that´s what I think should be changed. Fedora 
should seek for increasing the number of end-users. And that´s only 
possible when it receives, with an open spirit, and put to 
consideration, the end-users demands. I´m not saying it´s not done at 
all. I´m only suggesting it should be added as a main goal.
    

I think we are safer leaving that goal to the upstream projects we adopt
from.  Let us channel user experience back to GNOME, KDE, et al, and all
the smaller applications.

I do think that having something such as, "Grow quantity of consuming
end-users by focusing on their demands," is going to be at direct odds
with the rest of the project objectives.  We cannot effectively test the
  
Yes, of course. Not so much to heaven, not so much to hell. Other goals should not be replaced by this one. But I think this matter should be taken into consideration more often. Otherwise the "voice of community" balloon on Fedora logo would be just an idea without any concrete results. Not all demands should be applied, but only those that wouldn't hurt/break other Fedora objectives.
quality of the latest GNOME and KDE interface, for example, if we are
not willing to put it out there as-is and see what happens.  If we feel
we need to "adjust" such software to match end-user demands directly, we
begin a very slippery process.  It might appear successful at first, as
it is for other distros, but in the long run, it is a failing
proposition.

This is an evolutionary process.  We need to *first* improve the quality
of tools that Fedora contributors use.  We need to make it simple for
people who help in #fedora to file and track bugs, etc.  As we have
success there, we can push those tools closer and closer to the edge
where the end-user lives.

- Karsten