Re: Proposal: Debian release numbers
On Wed, Jan 08, 2003 at 09:19:29AM -0200, Gustavo Franco wrote:
> On Tue, 2003-01-07 at 15:32, Matt Zimmerman wrote:
> > On Tue, Jan 07, 2003 at 09:20:54AM +0000, Scott James Remnant wrote:
> >
> > > Does the gcc 3.2 transition constitute a major enough change? How about
> > > the inclusion of d-i? KDE 3.1? GNOME 2? Linux 2.4?
> >
> > Since Linux 2.4 was included in woody (despite claims by the popular press
> > to the contrary), I'd say that is definitely not a major change.
>
> With gcc 3.2, KDE 3.1, GNOME 2, Linux 2.4 kernel(by default), and
> XFree86 4.3(maybe?) we can't disagree that sarge is a major upgrade,
> calling 4 or 4.0.Or not?
Note that of that list, only GNOME 2 itself warranted a major number change.
All the others were considered by their upstream developers to be minor
versions.
The main reason GNOME bumped its major version was *incompatibility*.
Signaling compatibility breaks and large restructurings is the historical
(pre-marketing) function of major version numbers, and that's how Debian
has been using them until at least woody. (I'm undecided about woody.
Did the introduction of the testing distribution count as a large
restructuring? It didn't affect the release itself much, but it's a
significant change in the way we do releases.)
Richard Braakman
Reply to: