00:04:12 -!- sebbu2 has joined. 00:04:12 -!- sebbu2 has quit (Changing host). 00:04:12 -!- sebbu2 has joined. 00:04:17 -!- sebbu has quit (Ping timeout: 248 seconds). 00:21:14 -!- carado_ has joined. 00:21:38 -!- carado has quit (Ping timeout: 256 seconds). 00:21:48 -!- carado_ has changed nick to carado. 00:22:40 -!- kallisti_ has quit (Ping timeout: 250 seconds). 00:31:51 How does Free (Cont r) behave? 00:58:25 -!- oonbotti has quit (Ping timeout: 252 seconds). 01:03:00 :t runState 01:03:02 State s a -> s -> (a, s) 01:03:15 @src State 01:03:16 Source not found. :( 01:03:38 The order of applicatives matter, yet there's no nice syntax for rearranging the order when using a nonmonadic applicative 01:03:59 :t (<**>) 01:04:00 Ambiguous occurrence `<**>' 01:04:01 It could refer to either `Control.Applicative.<**>', 01:04:01 imported from `Control.Applicative' at State/L.hs:4:1-26 01:04:08 cool 01:04:18 @hoogle (<**>) 01:04:18 @hoogle (<**>) 01:04:18 Control.Applicative (<**>) :: Applicative f => f a -> f (a -> b) -> f b 01:04:19 Control.Applicative (<**>) :: Applicative f => f a -> f (a -> b) -> f b 01:04:29 (not flip (<*>)) 01:04:32 ambiguous occuence 01:05:04 I still think a do-like syntax for applicatives could be reasonable 01:05:32 Sgeo: i've been thinking comprehension syntax is more fitting 01:05:57 Good point 01:06:40 all you need afaict is to check that none of the comma separated <- declarations use each other 01:06:48 But, if there's a monad that let's you put applicatives together (maybe returning Nothing if you attempt an illegal operation), monad comprehensions could be used 01:07:14 Although direct syntax support would be better 01:10:44 oerjan: well comprehensions and do notations are basically the same. though the free "pure" in comprehensions suits applicatives well 01:11:13 yes, the latter is what i was thinking 01:15:04 so [f x y z | x <- ..., y <- ..., z <- ...] ==> f <$> ... <*> ... <*> ... 01:15:20 hmm it would work well with record puns, too bad record puns are evil 01:15:29 or can you pun that way... 01:20:52 seems backwards 01:21:26 or wait i'm thinking of record wildcards 01:23:29 I was thinking if you could do [Rec{..} | field <- ..., field2 <- ...] 01:23:49 that's wildcards 01:24:32 and also backwards, i think 01:24:41 what do you mean by backwards 01:25:03 that .. is a pattern you get the names _out_ of, not put them into 01:25:22 oh wait 01:25:42 that seems to be allowed 01:26:04 that should work then 01:27:30 I wonder if you get a fun type error if you e.g. forget to bind field3 01:27:37 and it tries to do field3 = field3 01:27:41 where field3 is the _accessor_ function 01:27:44 (because punning is awful) 01:27:53 probably it just detects that I guess :( 01:28:31 it says that particular binding is excluded 01:36:56 Dependently typed languages don't have idiotic numbers of zip functions, right? 01:37:05 zip3 zip4 zipOkThisIsDumb 01:37:10 :t zip4 01:37:11 [a] -> [b] -> [c] -> [d] -> [(a, b, c, d)] 01:37:20 Just use HLists instead of tuples 01:38:27 HLists ~ tuples 01:38:39 that was how they were originally written 01:38:41 nothing to do with dependent types, also 01:38:52 ZipList abstracts away the zip pattern 01:39:19 HLists are tuples structured like lists though, rather than flat, iirc 01:39:39 (1, (2,3)) rather than (1,2,3) 01:39:46 or, erm 01:39:51 more like (1,(2,(3,()))) 01:39:53 (1, (2, (3, ()))) 01:39:54 that's some old school shit right there 01:39:58 anyway it's a bad representation because you get O(n) access 01:42:31 -!- Phantom__Hoover has quit (Remote host closed the connection). 01:45:54 > zip (zip (zip (repeat ()) (repeat ())) (repeat ())) (repeat ())) 01:45:55 :1:64: parse error on input `)' 01:45:59 > zip (zip (zip (repeat ()) (repeat ())) (repeat ())) (repeat ()) 01:46:01 [((((),()),()),()),((((),()),()),()),((((),()),()),()),((((),()),()),()),((... 01:48:55 > join $ fix $ (["","()"]++) . ap (zipWith (\x y->"("++x++y++")")) tail 01:48:57 "()(())(()(()))((())(()(())))((()(()))((())(()(()))))(((())(()(())))((()(()... 01:49:14 it's like i'm really using an esolang 01:49:57 you're not, Bike. because you don't know haskell 01:50:01 because you're barely even a person!!!! 01:50:07 i'm barely even a bike 01:50:35 what kind of bike are you 01:50:49 a fixie 01:51:07 > fix it 01:51:07 but you don't fix anything 01:51:08 Not in scope: `it' 01:51:08 Perhaps you meant one of these: 01:51:08 `id' (imported from ... 01:51:17 @google fixie 01:51:19 http://en.wikipedia.org/wiki/Fixed-gear_bicycle 01:51:19 Title: Fixed-gear bicycle - Wikipedia, the free encyclopedia 01:51:22 > let bike = const "Bike" in fix bike 01:51:24 "Bike" 01:51:28 thanks 01:51:36 :t const "shitshitshit" 01:51:38 b -> [Char] 01:51:40 i know what fixies are!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! FUCKER 01:51:43 oh nice 01:51:56 I did not know what a fixie s 01:51:57 is 01:52:26 hope that helped 01:52:35 elliott, what's that thing that was in lambdabot that made a joke about fixing? 01:52:47 @fix 01:52:47 Maybe you meant: bid faq ft id thx 01:52:52 @id 01:52:56 @thx 01:52:56 you are welcome 01:53:00 @id foo 01:53:00 foo 01:53:00 :-) 01:53:04 @id "honk 01:53:05 "honk 01:53:23 i think i must be tired because im laughing at Bike's smiley 01:53:29 and i usually only laugh when i use that smiley 01:53:42 are you okay bro 01:53:44 !underload (:aS(:^S^:)Sa:):^S^:(:aS(:^S^:)Sa:) 01:53:45 no :( 01:53:46 (:aS(:^S^:)Sa:):^S^:(:aS(:^S^:)Sa:) 01:53:55 :( 01:54:02 :t ( :) 01:54:04 a -> [a] -> [a] 01:54:05 elliott, I laughed when you said that what I said was the saddest thing you've heard 01:54:16 the saddest thing he'd heard you say 01:54:53 Sgeo: were you tired 01:54:59 I don't remember 01:55:06 insufficient data for meaningful answer 01:57:32 @pl (\x -> x x) (\x -> x x) 01:57:35 ap id id (ap id id) 01:57:35 optimization suspended, use @pl-resume to continue. 01:57:42 @pl-resume 01:57:48 ap id id (ap id id) 01:57:48 optimization suspended, use @pl-resume to continue. 01:57:53 uh...? 01:57:56 i think i see where this is going 01:57:57 optimization? 01:58:20 its optimising the points away 01:58:54 But the points are gone 01:59:00 Why is it still trying to get rid of more? 01:59:50 :t app 01:59:51 ArrowApply a => a (a b c, b) c 02:00:01 wait i thought arrows sucked 02:00:49 (#(% %) #(% %)) 02:00:50 Clojure 02:00:56 elegant 02:01:07 its ap not app 02:01:12 :t ap 02:01:13 its trying to make the code shorter 02:01:14 oh 02:01:14 Monad m => m (a -> b) -> m a -> m b 02:01:17 whoa 02:01:58 :t id<*>id$id<*>id 02:02:00 Occurs check: cannot construct the infinite type: a0 = a0 -> b0 02:02:00 Expected type: (a0 -> b0) -> a0 02:02:00 Actual type: (a0 -> b0) -> a0 -> b0 02:02:04 uh? 02:02:11 :t (\x -> x x) (\x -> x x) 02:02:13 Occurs check: cannot construct the infinite type: t1 = t1 -> t0 02:02:13 In the first argument of `x', namely `x' 02:02:13 In the expression: x x 02:02:21 did you expect that to work i'm confused 02:03:54 :t \x -> x x 02:03:56 Occurs check: cannot construct the infinite type: t1 = t1 -> t0 02:03:56 In the first argument of `x', namely `x' 02:03:56 In the expression: x x 02:04:06 this is why fix exists surely 02:04:45 There are functions you can pass to themselves 02:04:58 So why can't I write a function that can only take such a function? 02:05:07 what functions can you pass to themselves 02:05:11 id 02:05:21 Sgeo: what type do you expect (\x -> x x) to have 02:05:24 well that's because it's polymorphic isn't it 02:05:29 note that it can be assigned a type but not in haskell 2010 02:05:45 oh? what type would it be assigned? 02:05:58 Bike: i won't spoil 02:06:05 you dick 02:06:10 ok i will but only if everyone gives up 02:06:27 how can we give up if we're too incompetent to even guess 02:06:33 does it count as giving up if i already know the answer? 02:06:39 yes 02:06:43 you don't count as part of everyone, oerjan 02:06:44 ok then 02:06:56 a priori i assume that anything i could possibly say is something oerjan already knows 02:06:59 and i'm ok with that 02:07:14 I don't know anything about type theory outside of Haskell. Well, a little subtyping I guess 02:07:21 its not really anything to do with type theory 02:07:30 Typeclasses? 02:07:35 no 02:07:46 surely \x -> x x not being typeable in HM is type theory something 02:07:47 Some sort of or type? 02:07:56 (\x -> x x) is a function so must have type $A -> $B for some $A, $B 02:08:11 are those perl style variables 02:08:18 we apply x inside so we know $A must be a function type, ($C -> $D) -> $B for some $C, $D, $B 02:08:30 the function's result is the result of applying x 02:08:32 so $B = $D 02:08:39 Recursive types 02:08:40 ($C -> $D) -> $D for some $C, $D 02:08:50 $C is what we pass in i.e. the type of x 02:09:01 but oops, that's infinite 02:09:03 $C = $C -> $D 02:09:09 02:03:56 Occurs check: cannot construct the infinite type: t1 = t1 -> t0 02:09:34 from a rank2 pov we can assign the type (\x -> x x) :: (forall a. a -> a) -> b -> b 02:09:38 but it's not inferred 02:10:15 Bike: btw fucking learn haskell you god damn fixie 02:10:17 @where lyah 02:10:18 http://www.learnyouahaskell.com/ 02:10:20 go. go do it 02:10:24 that fuckin book 02:10:31 i dont care 02:10:36 about your opinions of the book 02:10:46 well i'm certainly not doing it now but that's because my head is underwater 02:10:58 how about this tomorrow i will learn agda and then shoot you 02:10:59 buy hutton's programming in haskell if you want I hear that's good!!!! 02:11:10 fuck it give up and read real world haskell if you want to be the worst person slash bike 02:11:14 i don't care it just needs to happen 02:11:17 why 02:11:27 well rwh is kind of crappy as a `first read' 02:11:39 no i mean why does it need to happen 02:12:02 look 02:12:06 Why does Bike hate LYAH? 02:12:13 either you'll learn and understand 02:12:15 or you'll die trying 02:12:18 and it won't matter any more 02:12:22 i don't hate it, i just think it's kind of irritating 02:12:27 headlines FIRST BICYCLE KILLED BY HASKELL 02:12:31 SIMON PEYTON JONES ARRESTED 02:12:32 okay fuck fine i'll try it right now you fucking shitter 02:12:37 thx <3 02:12:40 fucking sun 02:12:43 fuck you, sun. 02:13:02 Bike: have you read why's (poignant) guide to ruby 02:13:02 Sun's dead. Try Oracle. 02:13:12 lyah is the tamest shit in comparison (and i love wpgtr with all my heart) 02:13:39 i read some of wpgtr but i thought it was funnier than lyah 02:13:39 iirc wpgtr says something and that something is obsolete 02:13:48 Some flaw that Ruby had that was fixed 02:13:50 i forget why i don't like lyah, maybe it's because i hate joy 02:13:55 well wpgtr was last updated like a billion years ago 02:13:56 do i hate joy? i forget 02:14:01 i think its for ruby 1.8.something 02:14:04 Joy is a different language. 02:14:14 but it doesn't matter because nobody has ever learned ruby from wpgtr 02:14:16 and that's the beauty of it 02:14:26 i like joy the language i mean joy the happiness 02:14:35 did you know wpgtr has a soundtrack 02:14:46 yes 02:14:54 good bicycle 02:15:25 -!- kallisti has joined. 02:15:26 -!- kallisti has quit (Changing host). 02:15:26 -!- kallisti has joined. 02:15:30 is there a haskell mode for emacs that isn't bullshit 02:15:39 yes its called the haskell mode 02:15:44 cool 02:15:46 just get the latest one from git or whatever since your package might just be shit 02:15:54 oh i set it into stupid indentation mode i think 02:15:59 since the smart indentation mode irritates me 02:16:05 like it knows better than me 02:16:12 and wants to cycle thoruhg every possible indentation when i press tab 02:16:16 wow it's actually in a repository gosh 02:16:21 "did you know it's theoretically possible for you to write code at this indentation here???" 02:16:26 haha 02:16:37 also in reality 02:16:38 i just use vim 02:16:38 elliott: those are the best indentations you know 02:16:45 and type my indentation manually 02:16:49 by pressing the spacebar 02:16:55 i don't know when i regressed like this but i am happy 02:17:10 oh right 02:17:17 it's because i tried the fancy alternative haskell mode thing chrisdone had 02:17:30 and it annoyed the hell out of me and wanted me to make a .cabal for everything when i just wanted to load a file into ghci 02:17:39 so i gave up and started doing vi foo.hs instead 02:17:39 so what's the usual mode for editing haskell 02:17:49 like uh 02:17:52 "workflow" that's the ticket 02:17:53 well chrisdone just merged the good stuff into real haskell-mode i think 02:17:56 oh 02:17:59 well you write stuff in a file 02:18:00 C-c C-l 02:18:05 it loads in the ghci in the other emacs window 02:18:09 oh good, i'm all over that. 02:18:09 and you switch over and evaluate stuff 02:18:16 its basically like lisp but we don't have anything fancy like slime 02:18:37 suddenly i have the horrible idea of writing haskell swank 02:18:51 like it's basically literally the same as running ghci in another terminal and pressing :r 02:18:54 which is what i do now 02:19:08 well i'll miss C-c C-c but i can deal 02:19:11 :r still wipes bindings 02:19:13 :( 02:19:27 i've forgotten what c-c c-c does 02:19:28 it's been so long 02:19:29 why would it not 02:19:38 holy fuck haskell-platform installs a lot 02:20:00 Because I want to test a function with some data that took some time to produce and don't want to re-create the data each time I change the function 02:20:03 everything haskell platform installs for you is something you cant fuck up installing manually with cabal-install 02:20:08 accept the blessing 02:20:43 Sgeo: yeah sounds like C-c C-c would be good for that, but a lot of times being able to destroy old bindings would be kind of nice too 02:21:11 yo 02:21:14 what does c-c c-c do!!! 02:21:16 don't leave me in the dark 02:21:20 elliott: until you need to upgrade a package 02:21:25 slime-compile-defun 02:21:31 Jafet: that's when you throw out your computer and buy a new one 02:21:33 just compiles and loads the current form 02:21:36 instead of like, the whole file. 02:21:46 % format / 02:22:22 whole-file vs single-function isn't relevant, it's what happens to old bindings made at REPL that's relevant 02:22:33 Bike: the real cool emacs stuff starts when you give up on ever writing programs and just go to coq and agda 02:22:43 good, i hate programs 02:22:44 where you get tightly-integrated ~HYPER-INTERACTIVE~ emacs environments 02:22:49 it's pretty much the best 02:23:48 Maybe I should be a teacher instead of a programmer 02:25:34 hey Bike should i sleep 02:25:38 god yes 02:25:45 ok but why 02:25:50 is it so you can have peace 02:26:04 because you're whiting out every few seconds and at this rate you'll become a p-zombie 02:26:20 ok becoming a p-zombie is now on my list of life goals 02:26:30 just before becoming a bicycle (how do you do it???) 02:26:34 read more smullyan then 02:26:38 it's an ancient secret i'm afraid 02:26:45 ancient wisdom of the bicycles 02:27:06 "so for reasons that will remain unclear i have an account on a novell network which uses windows 7 and i wish to get the sam and system files or the password hashes and saved onto a flashdrive" ##asm is a great channel 02:27:15 nice 02:27:53 it's like 98% actual asm questions 2% random scriptkiddies asking for hacking help 02:28:24 plot twist: 40% of the actual asm questions are people trying to wargame. 02:28:31 97% homework questions 02:28:36 well as I understand it the point of assembler is to bypass the security checks in the Micro-Soft Windows operating system to achieve a lower level of control with which to extract the necessary passwords from the system code 02:28:37 Yeah XD but wargame stuff sounds really cool 02:28:46 lol 02:29:15 that is why microsoft do not supply an assembler. it is too risky 02:30:11 do you ever get the feeling someone in the channel is taking your bullshit joke seriously 02:30:17 i'm getting that feeling real strong right now 02:30:37 you think someone in this channel would say "Micro-Soft"? 02:30:48 bike have you SEEN the people in thnis channel 02:30:53 is it me 02:30:56 am i the narc? 02:30:57 it's everyone 02:31:02 oh shit 02:31:12 and noöne (← DIÆRESIS) 02:31:13 -!- wareya has quit (Read error: Connection reset by peer). 02:32:08 -!- wareya has joined. 02:32:10 damn Bike you are so right. i should be sleeping 02:32:14 see 02:32:22 `welcome Bike 02:32:28 Bike: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.) 02:32:33 this is the advantage of the ancient wisdom of the bicycles. 02:33:12 right britland 02:33:50 bike have you ever even seen a britland 02:34:03 Only once, when I was young. 02:34:22 wow you're looking at my time without permission 02:34:27 that's confidential Bike 02:34:38 @localtime elliott 02:34:39 Local time for elliott is Thu Mar 7 02:34:38 2013 02:34:47 GUYS?? SOME PRIVACY??? 02:35:03 what if ghosts find me 02:35:05 /say @localtime had better just ctcp people 02:35:34 @localtime words more words 02:35:51 @localtime words Sgeo lambdabot words 02:35:54 /say hello Bike 02:35:58 good command. 02:36:09 /help 02:36:20 /sleep 02:36:42 02:36:39 -!- Irssi: Unknown command: sleep 02:36:46 my irc client sucks 02:36:53 /msg nickserv ghost elliott 02:37:08 kmc: whats your password 02:38:05 bonghits4jesus 02:38:42 i didn't believe you but tried to ghost you nonetheless 02:38:47 basic principles of professionalism 02:38:54 you sure did 02:39:06 did it tell you 02:39:11 you're one of the worst sleepers i've ever seen. 02:39:16 did it say hey kmc im nickserv the snitchand that elliott guy is the WORST 02:39:16 nobody will guess my password because i used the numeral '4' instead of the word 'four' 02:39:24 yes 02:39:24 you lied again 02:39:25 im so upset 02:39:33 Bike: wait am i asleep 02:39:37 is this all a dream 02:39:39 no not you that other guy 02:39:40 don't you trust me kmc 02:39:42 with your password 02:39:47 no 02:39:48 fuck do i look like zhuangzi to you 02:39:48 fuck you already have mine since i use mosh 02:39:49 hth 02:39:56 and im sure you have some great backdoors in that crypto 02:40:00 and don't forget that sharing is caring 02:40:04 no i just have a backdoor for AES 02:40:06 don't you acre 02:40:08 about anything 02:40:09 'much better' 02:41:04 Bike: what does it feel like to be able to decide to go to sleep and then enact that decision successfully 02:41:25 narcoleptic. 02:41:50 i wish i was narcoleptic 02:41:55 It feels like being dead tired at midnight instead of 2pm 02:42:11 It's... weird. 02:43:06 I think... my sleep schedule has actually been quasinormal lately 02:44:40 im starting to become irritated at everything for existing 02:44:49 especialyl Bike because bicycles feel more like tangible objects than humans to me 02:45:32 FireFlys aren't humans either. 02:45:54 And humans being the dual of pumpkins seems weird 02:46:50 elliott may already be a dualist! 02:47:13 im offended Bike 02:47:29 What if I threw in being a duelist as well? 02:47:58 like galois 02:47:59 Bike: do you know haskell yet 02:48:20 Almost. 02:48:26 Just need to figure out these monad things. 02:48:33 i know some good tutorials for that :-) 02:48:44 (-: 02:48:50 (-:-) fusion 02:48:58 that's sick 02:48:59 superglued them eyeballzzz together 02:50:50 dual & duel smiling power 02:50:55