Re: merge should copy-with-history

Branko Čibej <brane@xbc.nu> writes:
> I don't agree. Merge was broken by design, and Philip's patch unbreaks
> it somewhat. See my other post.
>
> >Most bugs are in Subversion now were there in our previous tarball,
> >there are just fewer of them now. But we don't have to fix
> >*everything* before releasing a tarball -- it just has to be an
> >improvement over the last one :-).
>
> Not everything. Just this. :-)

Yes, but *why*? :-)

Why this tarball, as opposed to the next one?

However long the fix takes to apply, that's how long it takes. So
we're not going to get the fix faster by delaying the tarball, but we
could end up delaying the tarball by waiting for the fix. To put that
more bluntly: waiting can do harm, but not much good. So why bother?

If we can get this fixed by EOD tomorrow, then it goes in. If it
needs more time, then it should go into the next interim tarball,
which won't be that long from now anyway.

The point of the interim tarballs is to be *frequent* -- not to
maximize the amount of improvement from tarball to tarball.

-Karl

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Aug 8 09:18:15 2002