[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: FC2 initial schedule posted

On Mon, Dec 15, 2003 at 02:59:20PM -0500, Gene C. wrote:
> 1.  I see only three test cycles.  Previous release work (granted for Red Hat 
> Linux and not Fedora Core) involved around five cycles including both private 
> and public beta cycles.  Will three be enough?  [Perhaps it will but new 
> gnome, new kde, plus new kernel, plus all the other things that will get 
> updated sounds like a lot.]

That's why we mention trying to keep rawhide installable.

And, actually, they weren't necessarily five.  Three to five, most
commonly four or five.  But we're trying to make changes; one change
that has been explicitly mentioned is that in general we are trying
to move from a pretty rigid 6-month cycle to a 4-6 month cycle, a
little faster.  That pretty much implies being able at least sometimes
to move a bit faster.  I think that we can do that since we stand a
chance of being more efficient this way (no one-month wait for the
shelves).  I really think that focussing on ways to encourage rawhide
use is an important part of moving faster.

New KDE is due out well before, so it really doesn't add pucker factor,
and beta reviews are good.  GNOME/Gtk+ updates add more pucker factor
since the schedules are closer, but also good reports there.

> 2.  While the three test cycles will enable test-fix-test the anaconda 
> installer, the use of yum/up2date to get new packages can provide for 
> additional test-fix-test cycles for other packages (including the kernel).  
> What is the thinking about how yum/up2date (or whatever it will be called) 
> during the FC2 cycle?

Sync to rawhide!

> 3.  With new/updated  packages being placed in fc-development and that being 
> the basis for fc2, would there be some benefit if some of us (those who 
> really want to be on the bleeding edge) to downloading and installing 
> fc-development?  I noticed that at least one person has done this but would 
> it be beneficial for a "bunch" of us to do this?  By beneficial I mean, can 
> we help debugging packages before the 2 February test1 date and thus improve 
> the chances that the overall schedule can be met?

Hmm.  Good question.  I think that from the installer point of view,
Jeremy will post at times that will be particularly useful for the
installer to be tested via rawhide.


 "He that composes himself is wiser than he that composes a book."
 Linux Application Development                     -- Ben Franklin

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]