
On Thu, May 30, 2013 at 08:15:28PM +0200, Nicolas Pouillard wrote:
Quoting Magnus Therning (2013-05-30 18:27:26)
On Wed, May 29, 2013 at 01:15:50PM +0200, Nicolas Pouillard wrote:
Quoting Ramana Kumar (2013-05-29 09:45:19)
On Wed, May 29, 2013 at 8:15 AM, Magnus Therning
wrote: Please let me know if you /do/ retire [habs-web] as there might be a number of commonly used packages in there that ought to be moved into [haskell-core]. I'm guessing you don't have any kind of usage statistics on the packages, do you? ;)
I think it's best to consider habs-web retired as of now, since it currently has no maintainer. However, as far as moving packages goes, it would probably be fine to wait until someone explicitly asks for something that was in web to be made available in core.
I am/was also a user of habs-web for roughly two use-cases: * haskell programs such as git-annex or notmuch-web [0] that I want to build once and install multiple times. * my strategy to reduce compilation time was to install all binary haskell packages available and then use cabal install for the rest.
I wolud like to also suggest that as long as storage permits we should keep the latest version of a working set of packages for each version of GHC. This does not cost much to just keep a copy before trying an upgrade and this would help users migrating.
I don't think I understand what you mean by "we should keep the latest version of a working set of packages for each version of GHC". Would you mind describing in a few more words what that means?
I mean to spawn off a frozen repo for at least the previous release of GHC. Repos could be named such as [haskell-core-7.6.2].
Is that reasonable?
No, at least not if you are suggesting having several repos, with the same set of pacakges, but compiled with different versions of GHC. Maintaining that would require a bit more time than I have to offer, and for me personally it's of no interest whatsoever to maintain a repo for an old version of GHC. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Perl is another example of filling a tiny, short-term need, and then being a real problem in the longer term. -- Alan Kay