15:35:17 <mm_freak> hi there
15:36:00 <mm_freak> what's the correct way to set the content-type of my response?  for example i'd like to send a CSS stylesheet as text/css instead of text/plain
15:36:16 <mm_freak> but i'm not using file serving, it's generated on the fly
15:38:42 <donri> mm_freak: use one of toResponseBS or a ToMessage instance
15:40:26 <mm_freak> but how do i actually set the header properly?  setHeader/setHeaderBS?
15:40:44 <donri> toResponseBS takes the content type as a parameter
15:40:54 <donri> in ToMessage it is set in a method
15:42:28 <alpounet> ToMessage defines "toContentType"
15:43:04 <mm_freak> oh!
15:43:07 <mm_freak> thanks
15:59:09 <bergmark> *phew* almost gave an answer on how to do that in snap
16:00:19 <HugoDaniel> :)
16:00:23 <stepkut> :)
17:08:29 <donri> stepkut: you still considering operational for happstack 8? http://stackoverflow.com/questions/14263363/is-operational-really-isomorphic-to-a-free-monad
17:10:02 <stepcut> donri: I am experimenting with both free and operational. Hoping to get a post on that soon -- though not one that draws any strong conclusions
17:10:56 <donri> edwardk: ^^^ got time? :) i think it was you who noted the isomorphism
17:11:20 <edwardk> i'm pair programming ATM, no real bandwidth
17:11:23 <kxra> stepkut: ah, i see. but you're still going to stick with the bug tracker you built? that's not necessarily a bad thing. maybe you could make something better
17:11:25 <donri> roger
17:11:45 <donri> i wish tekmo was on irc more so i could bug him all the time!
17:17:14 <stepcut> :)
17:17:55 <stepcut> kxra: 'stick with' is perhaps too strong of a term. 'too busy to take alternative action' might be a better description :)
17:19:26 <kxra> stepcut: aha, i see
17:19:43 <kxra> perhaps it makes sense to use darcs.net for new reports then?
17:20:48 <donri> my take on that: dogfooding is great, but in this case it might be better to use an existing bug tracker (and darcs hosting) at least until the plugins in clckwrks are more complete. i think for example people are more reluctant to report bugs on foreign but trackers
17:21:38 <donri> the one on darcshub also integrates a bit with darcs so you can close issues with patch names for example
17:22:22 <donri> and maybe the rest of clckwrks can grow more if people don't need to focus on the darcs/bugs plugins? :)
17:29:07 <stepkut> kxra: sure
19:02:11 <stepkut> i need to write shorter blog posts.. this upcoming one is 5800 words :-/
19:04:28 <bergmark> like one of my teachers said "KISS - Keep it simple and stupid"
19:05:11 <stepkut> I did! The long version would be 20,000+ words :)
19:06:04 <stepkut> all I did was take a very simple web routing system that only has: match, capture, choice, and fail, and implement it 4 different ways..
19:06:26 <bergmark> oh only 4 :P
19:06:29 <stepkut> I almost completely avoided in discussion about category theory and abstract algebra
19:06:59 <stepkut> yeah, traditional monad transformer stack, a direct interpretive approach, using the Free monad, and using the operational monad
19:53:42 <donri> longer is better!
19:53:49 <donri> make it a paper
19:53:54 <donri> the 20k version
19:53:57 <donri> bbl dinner
21:30:20 <hpaste> donri pasted “LOL couldn't figure out why my free state monad was evaluating to 6 not 7” at http://hpaste.org/80553
21:30:45 <donri> i was liek waat, is it stopping after the first "put" or WHAT
21:31:18 <donri> and i thought hey, it's a bit worrying that it type checks despite this misbehavior!
21:31:39 <donri> hint: on line 5, i re-used "x" instead of the new "y"
21:31:57 <donri> le facepalm ^_^