Future of Ac... or even of PLD

[Marcin Król]

> btw,
> What I really miss is *single* good source of informations about security 
> holes in Linux software - such one that could be coupled together with cvs to 
> inform about brokeness in software which we have on HEAD and AC-branch.

Yes, that would be great thing.

> I hope that fork is a joke. I don't see any benefit from a fork in comparison 
> to simply working on AC-branch.

Unfortunately this is not a joke. Keeping Ac up to date means dealing
with architectures that I don't need (alpha, sparc, ppc, amd64) and
updating packages I don't use. Keeping my own fork means to care only
about i686 and a set of specs that I use. Calculation is simple. So if
at some point no one except me will be interested in working on Ac and
Th will not be usable for me, I'll have to drop Ac. Of course I hope
that will not happen.

> I regret that Th switched to gcc 4.2 so early but at this moment I don't 
> consider switching back to gcc 4.1 (note that gcc 4.2 rc1 (more beta than rc) 
> was recently released and we already using it + most of the problems is in 
> software, not in gcc itself).

Problems are in software, yes. However most of them appear only in
combination with gcc 4.2 AFAIK?

> Note that Th is no longer (well, for a few months) tracking latest available 
> stuff. The goal is to keep it fresh but with stable releases (yes, there are 
> some exceptions). What Th needs is some polishing, kernel and installer.

The same things are missing in Ac - kernel and installer :) Current Ac
kernel still oopses for me, but I think it will not get any better with
all those patches applied. I've switched for my own build and other
users doesn't seem to be affected (or they don't submit bug reports).
Installer is different story and not for this topic ;)

M.