On Thu, Jan 23, 2014 at 7:05 PM, Sven Panne <svenpa...@gmail.com> wrote: > The notion of "bug-free SW" is an illusion, no matter how long we wait.
I did not say that a bug-free software is expected, I know that there is no perfect software, only better. But starting to use the first, previously not widely tested(?), release from a branch sounds a bit risky to me. I do not recall that HP ever experimented with pulling a .1 release GHC into the set, probably for a reason. > If there is a bug in the HP, so be it: Just release a 2014.2.0.1 version > soon then. Pushing out a new release has its own associated cost, at least for packagers, I believe: if GHC has to be updated, all packages derived from Haskell Cabal packages have to be potentially updated due to the ABI hash changes. > Although this is a totally subjective feeling (based on the very > subjective impression on the installed user base), I think that a good > out-of-the-box support for Macs is more important than FreeBSD. Sure, I see. I simply had the impression that the problems for Mac OS X has been resolved and HP 2013.4.0.0 is ready to be published. Obviously, I do not want you to be hindered due to FreeBSD's apparent conservatism, if that is the way forward for Mac OS X, so be it! > And what exactly does it mean that "we still use GCC 4.6" when Clang is > the default on FreeBSD 10.0? GCC can be always installed from the Ports Collection as a dependency, even on systems later than 10.0 -- that is, the same C compiler can be kept under the hood, independently of the base system settings. _______________________________________________ Haskell-platform mailing list Haskell-platform@projects.haskell.org http://projects.haskell.org/cgi-bin/mailman/listinfo/haskell-platform
