On Tue, May 29, 2007 at 10:51:58AM -0500, Aaron Griffin wrote:
> On 5/28/07, Jason Chu <[EMAIL PROTECTED]> wrote:
> > On Mon, May 28, 2007 at 01:59:11PM -0600, Scott Horowitz wrote:
> > > On 5/28/07, Jason Chu <[EMAIL PROTECTED]> wrote:
> > > > My fear with updating a full integer is having people say, "why hasn't 
> > > > x86
> > > > updated to -5?!? arch64 has -5, why does x86 only have -4?!? are we that
> > > > far behind?!?"  I always thought of the -1.1 notation as being a 
> > > > revision
> > > > of the revision.
> > >
> > > What makes you think people won't ask why x86 hasn't updated to -1.2?
> > > If version numbers don't match in _any_ way, people will undoubtedly
> > > be confused. And who can blame them?
> > >
> > > I find decimals in the pkgrel to be overly confusing myself. Frankly,
> > > I would just bump the pkgrel and release a new package for x86 as
> > > well, even if it's an identical package. Arch already requires lots of
> > > bandwidth (come on xdelta! :D) and I doubt these situations arise
> > > frequently.
> >
> > I'm pretty sure you'd hear just as many people complaining about having to
> > download packages that are no different than the ones they already have.
> > Also developers grumbling and arch64 forcing them to rebuild and upload
> > packages that didn't change.
> 
> Why would packages need to be rebuilt?  Just to match versions? That's
> the reason we use the cvs tags - CURRENT and CURRENT-64 point to the
> current versions in CVS.  Rebuilding simply to match versions seems a
> little daft to me.

It does.  But how many mark out of dates do you think we'll get because
arch64 is -2 and x86 is -1?  We barely know why half of the mark out of
dates happen now...

Jason

Attachment: pgpBx9dZxHgiH.pgp
Description: PGP signature

_______________________________________________
arch mailing list
[email protected]
http://archlinux.org/mailman/listinfo/arch

Reply via email to