Fedora Core 3 Transferred to Fedora Legacy

[Ralf Corsepius]

On Tue, 2006-01-17 at 22:58 -0800, jdow wrote:
> From: "Ralf Corsepius" <rc040203 at freenet.de>
> 
> > On Wed, 2006-01-18 at 09:34 +0530, Rahul Sundaram wrote:
> >> Hi
> >> 
> >> >>  It will just get lost in the list noise.
> >> >>    
> >> >>
> >> >Well, RH could have somebody monitoring this list on a professional
> >> >basis and communicate the issues reported here to their management.
> >> >  
> >> >
> >> This list is mostly noise at the management level.
> > Didn't you read what I wrote?
> > 
> > I said: Pay somebody for reading this list and filter the noise into
> > reports to the management.
> 
> Would you rather waste money on that critter, who'd go insane within
> two months, or on another bug fighting developer? Seems to me you may
> have just volunteered yourself to right up weekly summary reports to
> feed to the developers? {^_-}

I am not working for RH, and am not on "RH's hidden channels to the
decision makers being relevant here" ... but others on this list,
probably comprising son having been involved into this threat, probably
are. ;)

> > Such are report would then contain: "There is a strong demand to extend
> > the life time of Fedora releases".
> 
> So appear on the devel list and report that point.
You are barking up the wrong tree. I am around there for a long time,
but actually I don't think it will make a difference.

All I can do is "to shout into the woods, loudly" and hope somebody will
listen. It's essentially the same as you do here and will have the same
impact - Except that the bias will be different:
User tend to be "demanding", developers tend to be wanting to
"concentrate on the interesting things", management tends to "find
things too expensive" - And all of them prefer not to change anything ;)

> > BTW: I do agree on this topic - FC(current+1)t2 is too early to
> > discontinue FC(current-1). FC(current+1) + 1 month would be better.
> 
> I can see Rahul's standpoint on the problem. The weeks before release
> are hectic. 
In practice, it doesn't make much of a difference to most
maintainers[1], if a distro is formally discontinued or not.

Most maintainers will have to go after bugs/PRs in any case, no matter
how old the release/distro is, somebody complained about, and
maintainers will have to decide case-by-case if and how to apply changes
to a package in any case.

One of such decision can be not to apply/skip a change to packages in
older distros or not to update a package there-in. Similar effects can
also be caused by a maintainer's time constraints/resources - Most
fixes/updates happen when "he has gotten around to it"[2].

Whether a distro is formally discontinued or not, is secondary.

This even more applies to FE than it does to FC: There, maintainers
decide themselves on whether to update a package or not. As you might
have noticed, many packages in FC3 already don't receive regular
updates, "just because the maintainer didn't", for whatever reasons ;)

I.e. unless RH is going to turn of the FC3 buildsystem and stops hosting
FC3 extras on their hosts, the impact of discontinuing FC3 on
maintaining FE packages is close to zero. 

What will have a significant impact, is RH moving devel to FC5 and
adding devel for FC4, inside of the build system.

Ralf

[1] Exception: Fundamental packages in Core, such as kernel, glibc, etc.
[2] Exception: Security fixes.