03:06:41 <dleedev> hi, I read online that happstack was going to switch over to wai/warp
03:06:46 <dleedev> did that already happen?
03:07:14 <stepcut> nope -- we are going to switch to hyperdrive which is in development and based on pipes
03:11:29 <dleedev> is there somewhere I can read up on that?
03:11:35 <dleedev> is hyperdrive based on WAI?
03:13:11 <stepkut> there is not a singular place to read up on hyperdrive yet
03:13:26 <stepkut> it is not based on WAI -- mostly because wai requires conduit
03:16:55 <dleedev> is pipes much different from conduit?
03:18:35 <stepkut> not much, but it is different in important ways
03:21:52 <stepkut> for example, pipes for a proper category and conduits do not
03:48:45 <dleedev> why did WAI choose to go with conduits/
03:50:48 <stepkut> because WAI and conduits were written by the same purpose (ignoring the fact that idea wai was borrowed from hyena)
03:52:25 <dleedev> what is the ETA on hyperdrive?
03:55:32 <mm_freak> pipes 4 is very different
03:56:06 <mm_freak> conduit is really unstructured stream processing, while pipes gives you many ways to add structure to your streams
03:56:28 <mm_freak> this changes your coding style a lot
03:56:55 <dleedev> is there somewhere I can read up on that?
03:57:39 <mm_freak> dleedev: there is a tutorial in the haddock documentation
03:57:54 <mm_freak> for the more advanced stuff you may want to follow this blog: http://www.haskellforall.com/
03:59:43 <mm_freak> pipes 4 code reads almost like list code
03:59:50 <mm_freak> count (lines stdin) >>= print