FC4 state of affairs and FC5

[Gilboa Davara]

On Wed, 2005-09-07 at 11:26 -0400, Alan Cox wrote:
> On Wed, Sep 07, 2005 at 05:54:41PM +0300, Gilboa Davara wrote:
> > FC4 users cannot use their Palm and a full fix is no where to be seen.
> 
> Or bluetooth phones properly. FC4 was not the greatest Fedora, as a Red
> Hat developer I've actually gone back to FC3 for most stuff.
> 
> Having said that - long slow stable releases - see RHEL, see Centos etc

I may have misrepresented my own point.
I am looking for bleeding-edge distribution; else I'd be using
RHEL/CentOS on my private/work workstations. (Like I do on my servers).
However, the question remains, how much resources are to be allocated
for bug-fixing once a critical bug (again, in my eyes) is being found
and what tools do the community have to determine the current status of
a certain bug (beyond bugzilla) and decide (do we decide?) which bugs
take precedence.

> 
> But do the fixes depend on other incompatible updates ? Certainly some updates
> are easy for rawhide and hard for FC[old] because they require a whole chain of
> other changes

To be honest, I have no idea.
The bug was resolved to "CLOSED RAWHIDE" with no additional information.
My question (as for back-ported the fix to FC4) left unanswered.

> Nothing is stopping anyone making new ISO images with the fixes, nor putting
> up FC4 "old gnome-pilot" packages. 
> Alan

I understand and accept it.
However, this opens two new problem:
First and foremost, once I turn LFS, building my own packages I risk a
major breakage in my RPM based setup (Been there, done that, broke it); 
Essentially switching back to an older backbone library (pilot-link,
gnome-pilot) may require a full gnome rebuild (which is a major task).
More-ever, even if I succeed in doing it (and unlike KDE, building gnome
is not for the weak-hearted... even with garnome) I automatically lose
the auto-security-update capability for all the packages involved. In
short, this task is well beyond the scope of any normal user. It's much
easier to switch distribution.

Second, a user that tried Fedora (and failed) or worse, a user that
switched from Fedora to <insert another distribution name here> is a
user that will never return to Fedora.
Even much worse, a switching Windows XP user, that sees a kernel panic
due to a DRI problem in anaconda is a user that will never try Linux
again. 
We will never know how many Fedora users we have lost due to the iso
boot problem (bug 159026) or the catastrophic "Linux killed my Windows
boot problem" (bug 115980)

While it's semi acceptable in case of an unknown bug, having the known,
already resolved bug, push away potential users is problem well worth
fixing.

All of this is more of the frame of mind then a technical problem.
What is the main goal of the Fedora project? Is it a distribution bent
on defeating Windows, or is it simply a staging ground for RedHat to
field-test the next RHEL version.

Gilboa Davara