00:00:24 Remember when you were GregorR? 00:01:35 Yes. Yes I do. 00:01:45 I don't 00:01:48 I don't even own that nick any more. 00:01:50 -!- RocketJSquirrel has changed nick to ungroup. 00:01:53 ... 00:01:54 X_X 00:01:57 -!- ungroup has changed nick to RocketJSquirrel. 00:02:05 Apparently /nick and /nickserv aren't the same. 00:02:40 Anyway, apparently I do still own that nick. 00:03:04 I still own "Taneb" 00:03:06 :) 00:04:04 i don't own elliott any more 00:04:40 what do you mean by owning a nick? 00:04:52 Having it registered with nickerv 00:05:00 nickserv, rather 00:05:12 can that be done?! 00:05:21 Yes 00:05:35 I own 7 00:05:42 -!- Taneb has changed nick to Taneb|Kindle. 00:05:45 -!- Taneb|Kindle has changed nick to Ngevd. 00:05:50 -!- Ngevd has changed nick to Taneb|Hovercraft. 00:05:53 -!- Taneb|Hovercraft has changed nick to ettioll. 00:05:56 -!- ettioll has changed nick to noqmy. 00:06:10 nortti: /msg nickserv help register 00:06:22 Oh no I'm stuck 00:06:35 -!- noqmy has changed nick to marapreykus. 00:06:39 -!- marapreykus has changed nick to Taneb. 00:06:40 -!- ais523 has quit (Remote host closed the connection). 00:06:43 There we go 00:08:08 (by the way, the Piet spec does specify Unicode, but goes no more detailed than that) 00:08:25 Honestly, I'm kind of disappointed to find that RocketJSquirrel was free, given how big this network is. 00:08:31 Are all the people on this network terrible? 00:08:43 Yes. 00:08:43 -!- Taneb has changed nick to RocketKSquirrel. 00:09:12 Even terribler 00:09:15 -!- RocketKSquirrel has changed nick to Taneb. 00:10:29 Also, what's the difference between a conjecture and a thesis? 00:10:46 Taneb: The ratio between glucose and fructose. 00:11:55 -!- Taneb has quit (Read error: Connection reset by peer). 00:12:42 -!- Taneb has joined. 00:13:41 Bah, internet's dying 00:13:42 Bye! 00:13:44 -!- Taneb has quit (Client Quit). 00:15:13 -!- tzxn3 has quit (Quit: Leaving). 00:15:22 -!- nortti has quit (Quit: ChatZilla 0.9.88.1 [Firefox 10.0.2/20120216213642]). 00:18:03 -!- graue has quit (Quit: Leaving). 00:21:48 my mother accidently deleted the internet past week 00:28:10 Sgeo: "is not a quine" is not a quine. 00:29:23 -!- derdon has quit (Remote host closed the connection). 00:29:58 I have read about a QR code quine on this channel once, I think it was using PNG format. 00:30:16 I have also seen a ZIP quine, which contains a picture as well as a copy of itself. 00:31:23 you somethin like a recursive compression? 00:31:28 +mean 00:31:55 Yes, like that. 00:32:05 hell 00:32:35 awesome mindfuck 00:32:38 I also saw a gzip file which when uncompressed, resulted in two copes of the original concatenated together. If you then uncompress that, it will be four copies of the original, and so on. 00:35:30 Infinite compression ratio, eh? 00:41:12 -!- nortti has joined. 00:44:06 http://research.swtch.com/zip a detailed writeup about the zipfile quine 00:45:40 Do you think, that esoos should be based on an existing os with esoteric userland or a completely new kernel written in combination of asm and some esotetic language? 00:47:15 the former is quite boring 00:47:19 since you skip all the OS parts 00:48:05 Yeh. 00:48:26 exiy 00:48:37 -!- nortti has quit (Quit: nortti). 00:55:25 nortti: Second one. 00:58:15 -!- dnm_ has quit (Read error: Connection reset by peer). 01:03:24 -!- nortti has joined. 01:04:01 what language would you suggest? 01:05:47 -!- pikhq_ has joined. 01:07:42 There's no place for PSOX, right? 01:07:43 >.> 01:07:48 -!- pikhq has quit (Ping timeout: 252 seconds). 01:09:41 Sgeo: actually I might b the api for esoos 01:10:10 s/b/use it for/ 01:11:36 good luck with tha 01:11:37 t 01:11:57 There aren't many esolanguages suited to that low-level fun. 01:12:15 Checkout is one, but you can't compile it efficiently to today's instruction sets. 01:12:20 *that kind of 01:12:58 Note that the file domain is incomplete. And that you'll need to write your own implementation, unless you want to get CPython running on the OS 01:13:08 If I use PSOX I will propably leave networking out 01:13:23 Sgeio: 01:13:31 Sgeo: 01:14:10 Ah 01:14:23 Why the fuck does android keyboard havs enter and backdpace so close?! -.- 01:14:33 I'm trying to think, is there anything truly useful in there besides networking and files? 01:14:39 Besides the general framework 01:15:10 Which as elliott said may be less suited for languages other than Brainfuck and Checkout (haven't heard of Checkout until now) 01:15:19 What? 01:15:25 No, Checkout isn't even remotely suited to PSOX. 01:15:29 I was replying to what language would you suggest? 01:15:35 Oh 01:15:48 But it's true that PSOX limits you sorely with its binary requirement. 01:16:34 Sgeo: If I use PSIX I will write my own implementation with asm 01:16:53 *PSOX 01:16:59 elliott, at least it doesn't limit you to cell-based languages 01:17:07 >.> 01:18:32 Sgeo: Does any esoteruc api limit you to cell based languages? 01:18:50 I feel compelled to point out http://catseye.tc/projects/befos/ 01:19:15 nortti, iirc, PESOIX 01:19:20 Which was PSOX's inspiration 01:19:26 Some things about PESOIX ticked me off 01:20:50 Or was it EsoAPI 01:20:57 PSOX was named after PESOIX 01:21:17 "EsoAPI 1.0 calls replace the normal output functions. To perform an EsoAPI operation, write the necessary NUL escape character and function call bytes to the output stream. Some function calls may return a status value in the current memory cell. Any other data will start in the cell following the current memory cell. This should be implemented in a way consistent with the language." 01:21:21 http://esolangs.org/wiki/EsoAPI 01:22:47 Oh. That is just stupid 01:23:49 I think my problem with PESOIX was more subtle 01:27:13 or maybe I just create another Esoteric API which needs input and output byte functions 01:28:06 "Hello! ebdcked interesting ebdcked site! I'm really like it! Very, very ebdcked good!" 01:28:16 Come to think of it, Glass could be a decent language to do it with. 01:28:53 It has simple semantics without great runtime requirements (just a GC), and has enough abstraction support to allow you to bolt on talk-to-hardware stuff without much effort. 01:30:16 Does it need dynamic memory allocation? 01:31:01 Sure, but the allocation is explicitly controlled (i.e. you'll know when a piece of code is allocating). And writing a simple allocator+GC can be done in a few hundred lines of assembly code. 01:31:42 Yeah. I'll look at Glass when I get to my computer 01:33:34 -!- augur has quit (Remote host closed the connection). 01:34:47 I actually have pretty simple best fit linked lisr based allocator ready, but it is in 16bit cide 01:34:57 *code 01:43:38 -!- kwertii has joined. 01:43:47 did #haskell change significantly after i left 01:43:58 i have some paranoid fantasy where i'm secretly responsible for all of the dysfunction in that channel 01:45:39 not really 01:45:56 you can pgdown a few times on http://tunes.org/~nef/logs/haskell/12.03.10 if you want empirical evidence 01:46:54 Write an operating system using a variant of INTERCAL. 01:47:51 intercal is kinda boring bacause the please ststem 01:50:30 danharaj: In GHC you can make an identifier that doesn't even have a type. 01:50:33 Er. 01:51:25 fizzie: Hey, that Korpela guy answers questions on Stack Overflow. 01:51:28 shachaf: You can? 01:52:09 Who is danharaj and how can you make an identifier that doesn't have a type? 01:52:23 kmc: Nope. #haskell has only deteriorated in your absence. :-( 01:52:52 Or maybe it's just me. 01:53:03 shachaf truly has deterioriated in kmc's absence. 01:53:12 how has it deteriorated 01:53:13 elliott: Data T = froall a. T { unT :: a } 01:53:30 I just mean that I've been getting annoyed with it more often. 01:53:56 s/D/d/ 01:54:12 ok 01:54:13 O, yes, you are correct, I think unT doesn't have a type (or a use). 01:54:18 let's have all the reasonable people leave #haskell 01:54:56 zzo38: It has a use. 01:55:08 Pattern-matching and record construction. 01:55:28 shachaf: Oh, right. 01:55:44 * Sgeo factory resets his phone 01:55:54 shachaf: IIRC GHC will display unT's type as T -> a if you convince it to. 01:55:58 (I think Haddock will document it as such.) 01:56:07 (If you don't expose the constructor but expose unT separately.) 01:56:20 Cannot use record selector `unT' as a function due to escaped type variables 01:56:26 "if you convince it to" 01:56:30 (You can test it in ghci these days!) 01:56:37 "if you convince it to" 01:56:40 I don't know of any way to convince it to. 01:56:47 (I think Haddock will document it as such.) 01:56:56 But that statement is tautologically true. 01:57:33 kmc: If I was reasonable I'd've left #haskell a long time ago, surely. 01:58:31 i'm still wondering if i should post my screed about #haskell 01:58:48 the channel has many redeeming qualities and i don't want to just walk away 01:58:58 but it pisses me off for reasons which are a combination of its and my fault 01:59:28 Which parts are your fault? 02:00:11 I can practically guarantee posting such a thing will have no benefits other than catharsis... but maybe if you're still annoyed at an IRC channel after not being in it for months that's what you need. 02:00:31 -!- augur has joined. 02:00:38 Catharsis is the word I want there, right? I hate words. 02:00:44 seems correct 02:01:01 i'm not sure it has no other purpose 02:01:08 it will get people talking 02:01:15 It might have a purpose, but I doubt that purpose will be fulfilled. 02:01:31 But I'm pretty cynical about the ability of online communities to improve. 02:01:40 yeah 02:02:16 Oh well, toodles. 02:02:18 -!- elliott has quit (Quit: Leaving). 02:03:59 i think #haskell is friendly, and they think friendly = helpful 02:04:03 and so they think they are very helpful 02:04:13 perhaps if i tell them that friendly ≠ helpful they could be more helpful 02:04:36 well, do you want them to be helpful or friendly? 02:04:49 preferably both 02:04:54 i am not saying they are mutually exclusive 02:05:02 i am saying that being friendly is not itself sufficient for being helpful 02:05:16 for example, elaborating every beginner question into an endless argument about monad tutorials is not unfriendly 02:05:36 it's done with good intentions 02:06:51 Some pepole have said the programming language I wanted to invent is similar to a combination between Lisp and Haskell; that is probably part of it but not all of it. 02:07:33 -!- nortti has quit (Quit: nortti). 02:07:45 standards on IRC are so low that people think #haskell is an amazing channel just because they do not overtly flame noobs 02:08:43 Although they should. 02:09:10 really? 02:09:19 i don' agree with that 02:10:08 i don't think #haskell needs to get significantly less friendly in order to be more helpful 02:10:18 perhaps less friendly in some particular ways 02:12:22 I don't think I could ever flame a newbie 02:12:32 There was one person I flamed in here a while ago. 02:12:44 Someone who pretended to know more than he did, iirc 02:12:56 can you flame on IRC? I thought that only happened over mail/newsgroups... 02:14:01 i told someone directly that they had no idea what they were talking about 02:14:07 i don't think i was particularly mean about it 02:14:52 monqy, tswett has been UPDATEd 02:15:00 Sgeo: hi 02:15:51 monqy: hi 02:15:55 shachaf: hi 02:50:49 Lies. 02:52:59 hgeo. honqy. hachaf. hswett. 02:53:41 hon? 03:02:53 -!- elliott has joined. 03:03:22 02:03:59: i think #haskell is friendly, and they think friendly = helpful 03:03:22 02:04:03: and so they think they are very helpful 03:03:22 02:04:13: perhaps if i tell them that friendly ≠ helpful they could be more helpful 03:03:29 kmc: I think they think #haskell is friendly *and* helpful. 03:03:40 If you tell them friendly =/= helpful, they'll just ask you what about all the help they do. 03:04:12 elliott: Didn't you just repeat what kmc said? 03:04:29 No. 03:04:32 Well, they're probably more helpful than most IRC channels, at least. 03:04:53 The sheer misanthropy of a lot of programming language channels I've seen prevents any attempts at help. 03:04:58 kmc is saying that #haskell people think {friendly(#haskell), friendly = helpful}. 03:05:11 If you convince them that instead friendly =/= helpful, they won't have the belief helpful(#haskell). 03:05:17 I'm saying they think {friendly(#haskell), helpful(#haskell), ...}. 03:05:27 If you convince them that friendly =/= helpful, that won't change their belief that helpful(#haskell). 03:06:41 Ah. Fair enough. 03:07:10 I'm writing a Haskell book. Almost finished the 3rd chapter. Would like feedback. (bit.ly) 03:07:20 Let me guess! 03:07:31 They learned Haskell by reading a book and writing three toy programs a few months ago. 03:08:00 They think they "get" monads, but aren't quite sure about "functors". 03:08:07 OK, I hate being such a cynic. Please let me be wrong. 03:08:26 "To #haskell, where all questions are answered in majestic stereo." 03:08:30 shachaf!! You'll never believe it! 03:08:33 I wasn't wrong. 03:09:17 elliott: :-( 03:09:59 * elliott takes a look at the actual book, as if he needed "evidence". 03:10:07 "This is not a serious project." 03:10:28 Every language (human or computer) is unique. But there exists a special breed of languages ? those that 03:10:28 challenge and shape the way one thinks. Haskell is one of them ? lost innovation in a sea of clichés. Un- 03:10:28 fortunately, the only people apparently interested in Haskell are academics who blindly push the boundaries 03:10:28 and gurus who want to learn ?just one more language?. 03:10:28 On a more concrete note, if Haskell were to have a list of prerequisites, it would be very unusual indeed ? 03:10:31 at least two of the following: 03:10:33 03:10:35 Extensive programming experience 03:10:37 03:10:39 A background in mathematics 03:10:41 03:10:43 An IQ over 130 03:10:45 03:10:47 Perseverence 03:10:49 03:10:51 Hard work 03:11:37 Perse is Finnish for arse. 03:12:25 elliott: Thanks to you, I found the book, read some of it, and became annoyed. 03:12:34 Thanks, elliott. Thellileioieliottttt. 03:12:39 http://www.reddit.com/r/haskell/comments/qquc5/im_writing_a_haskell_book_almost_finished_the_3rd/c3zqomv 03:13:07 "c = [( f x) ^2 | x <- a , x < 10] -- this really works !" 03:13:09 OH MY GOD IT'S JUST LIKE PYTHON 03:13:14 THANK YOU HASKELL, I'M LEAVING PYTHON FOREVER 03:13:28 People really need to think of bettter examples of declarative style than list comprehensions. 03:13:36 "The mathematical applications of Haskell are endless. It's even possible to de?ne and work with monoids [XREF]!" 03:13:38 List comprehensions... 03:13:41 THE INCREDIBLY COMPLEX MATHEMATICAL TOOL "MONOID" 03:13:52 FINALLY HASKELL SOLVES ALL THE PROBLEMS FOR TODAY'S MATHEMATICIANS, LETTING THEM USE "MONOID"S 03:14:53 "If we try to mix wrong types, Haskell throws a type error." 03:15:07 Are the things highlighted in red supposed to be wrong? 03:15:19 "I should actually think before coding, but the type system is so good :)" - Cale 03:15:49 Warning! Backquotes work only with two-parameter functions. 03:17:04 This is all because Haskell is riddled with complex, counterintuitive or simply mind-boggling elements. Let's 03:17:04 take a look at something interesting. 03:17:04 max 2 3 03:17:04 -- works 03:17:04 max (2 3) -- doesn 't work 03:17:06 ( max 2) 3 -- works !! 03:17:08 1 03:17:10 2 03:17:12 3 03:17:14 Similarly, in C, 03:17:17 max 2 3 /* doesn't work */ 03:17:25 (max,2)(3) /* doesn't work */ 03:17:29 max(2,3) /* works!! */ 03:17:37 PROBLEM Z 03:17:52 More like a feature, right? 03:17:53 For starters, ++ concatenates two lists. It's one of the most basic operators. It's associative, so (a ++ b) ++ c is equivalent to a ++ (b ++ c)^7. / ^7 Without this basic property, lists would be stupid. 03:17:55 FEATURE Z 03:18:08 "GHCi treats 03:18:08 min -3 4 03:18:08 as 03:18:08 min (-) 3 4," 03:18:15 > min -3 4 03:18:16 Overlapping instances for GHC.Show.Show (a -> a -> a) 03:18:16 arising from a use... 03:18:17 > min (-) 3 4 03:18:18 Overlapping instances for GHC.Show.Show (a -> a) 03:18:18 arising from a use of `... 03:19:00 Technically all functions accept only one parameter, but it's not healthy to think like this, at least for now ? remember 03:19:04 Problem Z 03:19:06 (introduced in 1.2.3)? 03:19:24 I like the part where he gets it right paragraphs later: 03:19:24 [[ 03:19:25 Also, function application has the highest precedence, so if you write 03:19:25 (for more details see A.1). 03:19:25 foo 10 + 8, 03:19:25 (foo 10) + 8 03:19:26 it means 03:19:28 ]] 03:19:35 this book isn't real is it 03:19:40 it's just an elaborate prank right 03:19:41 right 03:19:42 No. 03:19:43 google docs is super bad at copying 03:19:49 It's a prank meant to make elliott go insane. 03:20:08 Unfortunately elliott was already insane. 03:20:11 "Unlike most languages, in Haskell a zero-parameter function and a constant are really the same. 03:20:11 strangely enough, has something to do with Problem Z ? we'll understand what that means soon enough. 03:20:11 This," 03:20:14 But it's idempotent. 03:20:26 elliott: :-( 03:21:21 FYI: the author is brisingr in Freenode. 03:21:32 Oh. 03:21:40 oh no 03:22:01 ion: Oh god. 03:22:04 That guy has been in here before. 03:22:08 I don't want to make fun of people who are in IRC. :-( 03:22:18 RocketJSquirrel: CENSOR THE LOGS CENSOR THE LOGS 03:22:45 No, don't do that! 03:22:56 I like how it gets to unsafeCoerce on page 14. 03:22:58 Superb pacing. 03:23:32 /msg brisingr elliott is making fun of you in #esoteric 03:24:13 ion: You ruined all our fun! 03:24:14 Ruined. 03:24:20 what fun 03:24:20 Don't you have better channels to make fun of people in? 03:24:21 By "fun", I mean "pain". 03:24:24 oh 03:24:24 And by "ruined", I mean "saved". 03:24:33 And by all of that I mean a turtle. :( 03:24:37 LOGO? 03:24:42 logo. 03:24:47 Yes. 03:25:28 elliott: You should write a better Haskell book. 03:25:31 That'll show 'em. 03:25:37 Haskell sucks. 03:26:08 HASKELL IS THE BEST LANGUAGE EVER 03:26:24 "It's not unlike if-else in other languages ? if the statement is true, the 03:26:24 else 03:26:24 branch won't evaluate and viceversa." 03:26:32 Tell that to my speculative Haskell evaluator! 03:26:52 * ion tells that to elliott’s speculative Haskell evaluator. 03:27:01 Thank you. 03:27:03 It feels wanted. 03:27:46 To be fair, a CPU might mispredict the branch in evaluating the C code too! 03:27:55 why are you still reading that book 03:28:09 by the way have we invited brisignre here yet 03:28:13 monqy: Because it involves no effort and is an easy way to feel superior. 03:28:30 Exactly. shachaf gets it exactly right. 03:28:43 Except that I'm actually doing it as an intellectual strenghtening exercise. 03:28:55 It's like beating your head against a wall, except I'm beating my *mind* against a wall. 03:29:00 Which amounts to the same thing, but one sounds better. 03:29:27 It's not actually strengthening. 03:29:48 Yes, that too. 03:30:32 The point is that elliott is mean-spirited and blackhearted. 03:31:37 Yes. 03:32:20 Fortunately, Haskell has a strong type system. That means that however similar their internal representations are, the compiler won't allow us to perform illogical calculations on them, such as multiplying an integer with a boolean. This may seem restrictive, but it helps avoid certain types of errors[1] (type errors). 03:32:26 [1] Imagine working on a long, difficult physics problem asking for some velocity -- but after hours of calculations, the result is in kilograms. That can't be good. 03:32:39 I like how Haskell prevents that error. 03:32:39 -!- azaq23 has quit (Quit: Leaving.). 03:33:17 The joke is that doing dimensional things in Haskell is painful. 03:33:20 Ha ha ha ha ha ha 03:33:38 > 1 {- miles per hour -} + 2 {- kilograms -} 03:33:39 3 03:33:41 no haskell no 03:33:47 I think it's ignoring your comments. 03:34:29 > 1 :: miles per hour 03:34:30 Could not deduce (GHC.Num.Num (miles per hour)) from the context () 03:34:30 aris... 03:34:45 help 03:35:30 > 1 mile per hour + 2 kilograms 03:35:32 3 03:35:34 no haskell no 03:36:03 "mile" 03:36:06 i found your error 03:36:18 > 1 miles per hour + 2 kilograms 03:36:19 3 03:36:22 no haskell no 03:36:35 kilometre is too many letters for poor haskell 03:37:04 Is there anything that makes doing dimensional stuff *easy*? 03:37:19 frink 03:37:32 (ok, it's not statically-checked) 03:37:35 (but it does units excellently) 03:37:41 λ> 1 *~ (mile / hour) + 2 *~ kilo gram 03:37:48 Couldn't match type `Numeric.NumType.Zero' with `Numeric.NumType.Pos Numeric.NumType.Zero' 03:38:09 How astonishingly nice. 03:38:20 FSVO nice. 03:38:38 Well, nice for unit handling. 03:38:44 I don't see any other merits. 03:39:21 Wait, what are you replying to? 03:39:26 ion or me? I assumed ion. 03:39:34 frink. 03:39:39 You, elliott. 03:39:41 Oh. Frink is astonishingly nice in general. 03:39:53 It's the bestest calculator. 03:40:05 λ> let u = 240 *~ volt * sqrt _2; i = u / r; r = 2 *~ kilo ohm; p = u * i in p /~ watt :: CReal 03:40:07 57.6 03:40:08 Merely being good at unit handling makes it quite nice. 03:41:29 @hackage frink 03:41:29 http://hackage.haskell.org/package/frink 03:42:07 `frink 2 lightseconds -> "meters" 03:42:27 599584916 meters 03:42:36 light am fast 03:42:46 whoa, light, slow down there 03:43:55 no light no 03:47:22 * ion makes the light bounce around in fiber, reducing its speed by a third. 03:48:04 thanks ion 03:49:30 elliott: I don't like do-notation and list comprehensions anyways; my own one it doesn't have such things (but might be possible to make up something like that using macros) 03:50:12 ion: thanks ions 03:50:37 ion: OH NO YOU BROKE SCIENCE 03:50:42 elliott: ion broke science :-( 03:51:57 Rest in peace , science . we loved , you . 03:53:34 -!- Nisstyre has quit (Read error: Connection reset by peer). 03:55:51 -!- Nisstyre has joined. 04:01:44 -!- MoALTz_ has quit (Ping timeout: 240 seconds). 04:02:13 -!- MoALTz_ has joined. 04:03:52 -!- MoALTz_ has quit (Client Quit). 04:04:20 -!- MoALTz has joined. 04:09:36 -!- MoALTz_ has joined. 04:09:44 -!- MoALTz has quit (Ping timeout: 240 seconds). 04:16:30 stop monqy im science 04:16:47 hi 04:16:52 im stope 04:17:25 thank goode. 04:25:40 monqy: welcome to floor! 04:28:45 hi, floor 04:30:02 monqy: hi 04:30:22 monqy++ 04:30:42 floor++ 04:31:11 @karma floor 04:31:11 floor has a karma of 1 04:31:13 @karma monqy 04:31:13 monqy has a karma of 2 04:31:22 monqy > floor 04:32:26 @karma science 04:32:27 science has a karma of 0 04:32:32 rip 04:32:56