
On Tue, Nov 8, 2011 at 13:20, Nicolas Pouillard
On Tue, Nov 8, 2011 at 12:53 PM, Magnus Therning
wrote: On Tue, Nov 8, 2011 at 11:11, Nicolas Pouillard
wrote: On Tue, Nov 8, 2011 at 10:32 AM, Magnus Therning
wrote: On Mon, Nov 7, 2011 at 21:01, Bernardo Barros
wrote: On Mon, Nov 7, 2011 at 11:18 AM, Magnus Therning
Any thoughts or comments on this?
But there is also cases like qtHaskell that are not straightforward to build by hand, and it is not in hackage. That should have a [aur] if not present in [haskell]
qtHaskell is special, because it's not on Hackage. That's why it isn't in [haskell].
In most cases since cabal does a better job, [aur] packages should be strongly discouraged.
IMNSHO we should never look at cabal as a replacement for an Arch repo or [aur].
How do you make the distinction between using cabal-install and [haskell] in you day-to-day use.
I don't use cabal-install, so it's simple to make the distinction :)
OK, so I misunderstood «we should never look at cabal as a replacement for an Arch repo or [aur].»? From this I read that you do want make cabal-install useless in Arch?
No, I meant no such thing. pacman (with repos) and cabal-install are at best complementary, and often they aren't even that. This means we can't say that dropping [aur] OK because users can always turn to cabal-install. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus