Re: Gtk/Gnome release schedules

[Jeff Waugh <jdub perkypants org>]

<quote who="Jody Goldberg">

> > > If gtk needs longer than 6 months for 2.4 fine, its nice to know that
> > > up front.  Why not make gnome 2.4 a longer cycle to correspond to the
> > > planned 2.4 period.
> > 
> > Why do we need to? In the case of GNOME 2.4, we don't need to wait for
> > any particular features / API, as we did with 2.0 and 2.2. No more
> > handcuffs! :)
> 
> new file selector.  We've waited so long.

Sure, this is a dark cloud hovering above us, but it's not *that* important.
Look at it this way:

  - GNOME 2.4 comes out built on GTK+ 2.2
  
  - GTK+ 2.4 comes out with new file selector, which is still accessible
    through the old, arsey API/ABI
  
  - Users upgrade to GTK+ 2.4 and most likely get a pretty new - but not the
    super dooper GNOMEified one-size-fits-all - file selector (there'll be
    breakage in places, sure, I don't know how Owen is going to deal with
    compatibility in this case -> perhaps "undefined behaviour"?  yuck,
    yuck, yuck)
  
  - GNOME 2.6 comes out, requires GTK+ 2.4, does all sorts of funky modular
    stuff with the file selector, everyone is happy

It seems reasonable to assume that distributors will a) wait for GTK+ 2.4
and b) do the work to get GNOME using it. We'll already be ahead for 2.6.

Is it worth waiting for, or doing a whole load of extra painful work (which
adds to the QA load, and keeps people away from significant desktop hacking)
for the purposes of a single GNOME release? I don't think so.

- Jeff

-- 
              Self-assertive pants are filled with confidence.