I guess this is where David's words about darcs having to target the
small-to-medium-sized project niche ring true.  Hopefully the GHC team
will have a better experience with git/hg (*)!

Following the bugtracker, I think we can sum up the GHC experience with:

[a1] conflicts problems, which could be fixed by switching to darcs 2,
except for [b]

[a2] various warts which could be fixed by switching to darcs2, except for [b]

[b] on a practical front, darcs2 is slower than darcs1 on some stuff.
It deals *much* better with conflicts (much reduced risk of
exponential blowup), but day-to-day things are slow enough to be
annoying

[c] various warts not yet fixed, even in darcs2

Hopefully darcs can attract enough developers to work on the
day-to-day performance issue [b] so that we at least catch up with
darcs 1.  We certainly do not want largeish projects to be forced to
use darcs 1 :-(

(*) And hopefully, darcs has done some things well enough that it will
be missed, and there will be a future "let's switch back to darcs now
that it's worked out its issues"

-- 
Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow>
PGP Key ID: 08AC04F9
_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users

Reply via email to