Re: [Gnu-arch-users] a brief note on "delta compression"

[Tom Lord]

    > From: Matthieu Moy <address@hidden>

    > Tom Lord <address@hidden> writes:

    > > Interestingly enough, though -- nobody has bothered yet.  Why not?  I
    > > As far as I can tell, it's because 90+% of the time arch's facilities
    > > for archive-cached revisions and revision libraries solve the same
    > > problem, better, and more simply, albeit at the cost of more disk
    > > space.  While arch's alternatives use a little more disk space, in
    > > dollar terms, the excess is trivial.  Simpler is better, in this case.

    > But they could be complementary. 

Absolutely.  No doubt.   There's nothing _wrong_ with summary deltas.
It's just that the need for them in arch seems, empirically, to be a
second or third order need, presumably because archive caching and
revision libraries offer many comparable benefits.   The absense of
summary deltas is an itch some people (especially Miles) have -- but
not so bad an itch that anyone has gotten off their ass to do anything
about it.    In and of itself that wouldn't mean much but you can
contrast it with other performance hack itches that have come up over
the past six months or so and gotten scratched real quick.

I still think it's likely to happen.   Probably some time after
abentley's backbuilder gets merged and we watch it slowly morph from a
nice little tweak to an algorithm into a little A.I. exercise that can
cleverly build revisions 10 different ways from tuesday.   That's
where it'll fit in cleanly.

-t

p.s.: Does "10 different ways from tuesday" make any sense to a
non-native speaker (it just means, "lots of different ways")?  For
that matter, does anyone know the origin of that phrase?