On 4/29/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> Quoting Jan de Groot <[EMAIL PROTECTED]>:
>
> > IMHO we shouldn't stuff binary files like this in CVS, same goes for the
> > gzipped kernel patches we have to update alsa, those are big and
> > gzipped, CVS can't diff it and will store a completely new file for
> > every revision. Also, you can't tell what's different between two
> > versions of the patch.
> >
> > IMHO we should put binary things like big gzipped patches or tarballs on
> > an HTTP or FTP url.
>
> the update of changes to the runtime will become troublesome since it changes
> pretty much everytime. The main reason for storing is, is actually to reduce
> downloads.

But does it reduce downloads for all of those who will never build it?
Having abs fetch it for everyone isn't practical if only 1% of people
are going to build it on their own anyway.

-Dan

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

Reply via email to