00:02:37 SLEREAH 00:03:57 wot 00:05:01 you are maintaining that guy's page :P 00:06:35 I no rite 00:07:05 I want to see what crazyv shinanigans he'll think up next! 00:07:19 Slereah: probably he's WILLY ON WHEELS 00:07:22 ON WHEELS! 00:08:09 Methlab on METH 00:08:32 YES 00:10:09 Slereah: I bet you're melab 00:11:04 I don't even understand that esolang of his. 00:11:52 Slereah: It does look pretty cool. 00:30:24 Fuckdick 00:30:41 I'm trying to think of a language to write the Andrei Machine in it 00:30:49 But I don't know that much. 00:31:13 I know like three real languageqs enough to program in. 00:33:00 Haskell (and other functional languages too) is considered very good for writing other languages in 00:34:00 Would doing some graph-related shinanigans be practical in it? 00:34:24 oerjan: really? hm. 00:35:02 hm graph support is a bit hairy if you want it pure... but there is a Data.Graph.* hierarchy 00:35:53 ML means meta-language - it was essentially _made_ for doing other languages in 00:36:12 (well a theorem prover originally, but still) 00:37:37 "Programming language graph" : http://www.cs.berkeley.edu/~flab/languages.html 00:37:42 I am a bit disappointed. 00:39:47 It also only goes up to 97.19% 00:39:51 Where are the other languages :o 00:41:07 Also this : http://www.levenez.com/lang/history.html 00:41:07 Slereah: zoo 00:41:12 It's actually quite nifty. 00:41:38 the TIOBE index has more (ooh, haskell is up to no. 31 now, i recall 50 or so) 00:42:23 tiobe sucks 00:42:34 hm maybe i remember wrong it only goes to 50 00:42:38 It also does not list INTERCAL 00:43:03 well it says they may have missed languages 00:43:13 truth in advertising there :) 00:44:54 "A Graph Rewriting Programming Language for Graph Drawing" 00:44:58 Let's see this 00:45:21 "The paper describes Grrr, a prototype visual graph drawing tool." 00:45:24 I even like the name. 00:45:30 I hope it's available somewhere. 00:45:52 grrrrrrrrr 00:46:02 that sounds suspiciously like UML 00:46:35 Fuck 00:46:42 I only find research papers 00:47:29 -!- kar8nga has left (?). 00:47:30 so? 00:47:35 Come on, it was out in 1998! 00:47:46 You'd think in ten years, it would be easy to find! 00:47:54 well 00:48:13 you think it's easy to find something that was published in some obscure scientific journal 300 years ago? 00:48:28 that's 10 years internet time :) 00:49:04 Hell, with Wikipedia nowadays, who knows! 00:50:55 Come on guys, how is your paper supposed to be peer-reviewed if your peers do'nt have access to the program! 00:51:01 Give us a link, something! 00:51:08 Something to work with! 00:54:55 Well, I suppose I'll never know how useful it would be! 00:59:18 # Clean 00:59:18 is a lazy, pure, higher order functional programming language with explicit graph rewriting semantics; one can explicitly define the sharing of structures (cyclic structures as well) in the language; 00:59:25 Anyone knows Clean? 01:01:08 Slereah: Clean. Is like. 01:01:14 Haskell. But with a few more weird stuff. 01:01:17 And something weirder than monads. 01:01:19 And. Nobody uses it. 01:01:25 But. Their compiler is pretty cool. 01:01:38 "Computation is based on graph rewriting and reduction. Constants such as numbers are graphs and functions are graph rewriting formulas." 01:01:46 Strange indeed. 01:02:00 But if it can hosts the Andrei Machine 9000, I'll give it a look. 01:02:02 Also, knowing too much about it causes you to speak in short punctuated spurts. 01:03:02 It's a risk I'll have to take. 01:04:29 Slereah: Oh, and. If Haskell isn't your thing for Andrei? 01:04:31 Clean won't be. 01:04:36 It is, pretty much, the same paradigm. 01:04:43 There's a questionaire before the downloading of Clean. 01:04:45 But.. nobody uses it. So you don't have the lovely #haskell people. 01:04:54 " I intend to use Clean for the following purpose:" 01:05:35 "Try to see if the graph rewriting features will help me construct the unholy Kolmogorov machine." 01:05:45 I hope this answer will help them in their marketing. 01:05:49 the underlying graph semantics is not necessarily any improvement for implementing actual graph data structures - haskell has been given graph rewriting semantics too 01:06:15 Oh. 01:06:39 Slereah: Yeah, Clean doesn't help you, in any way. 01:06:45 It hinders you, because #HASKELL IS NICE DAMNIT 01:06:54 Then what will damn it! 01:07:01 i think (but don't really know) making it explicit is more in order to support the uniqueness types that Clean uses instead of monads - to know whether things are copied you have to track them 01:07:11 Slereah: HASKELL. 01:07:14 DAMNIT. 01:07:26 I hate Haskell D: 01:07:37 And how do I shot Graph? 01:07:42 In Haskell, that is. 01:07:58 why do you hate Haskell? Clean is _very_ similar in many respects so unless it's _just_ monads chances are you'll hate that too 01:08:20 I don't want to build it out of lists. I can just go back to Python for that. 01:09:49 Why do programmers hate graphs so much! 01:09:56 Data.Graph.* i hear, although there are at least _two_ implementations inside that 01:10:47 for Haskell it is somewhat more complicated because it is hard to make an efficient _immutable_ implementation of graphs, i think 01:11:29 also, in Haskell you often define your own data structures with data 01:11:57 data structures... with data 01:11:59 (I know what you meant.) 01:12:00 but ask in #haskell for better advice 01:12:06 oerjan: he is scared of #haskell 01:12:08 (the data keyword) 01:12:15 whenever he joins he just says I'M TOO SCARED TO SAY ANYTHING!11 01:12:26 Heh. 01:12:31 also you may try ML (SML or Ocaml) 01:13:00 they are not lazy and so behave in some ways more intuitively to people from other languages 01:13:17 while still having much functional goodness 01:13:53 oerjan: his languages are lazy 01:13:55 so I doubt it's that 01:14:10 oh, and #esoteric-blah if anyone wants to see a fun spam flood test 01:14:48 no? 01:15:09 Lazy languages are good for esolangs. 01:15:19 I'm not too sure about real languages. 01:25:40 Fuck this. 01:25:47 I'll make the damn graph out of lists. 01:28:52 -!- Slereah7 has joined. 01:29:39 >:| 01:34:28 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 01:34:48 Wait, Kolmogorov graphs aren't directed. 01:34:59 I can't use cons like that 01:35:07 Does Scheme have sets? 01:35:48 no 01:35:56 scheme does not have anything 01:36:36 Fuck all that 01:36:49 I'll just write some algorithm in pseudocode for a start 01:37:10 I still have no idea how to find active zones and all. 01:37:49 Slereah7: rewriting stuff is fun 01:37:59 wot? 01:42:49 wot 01:43:12 Rewriting what 01:43:43 Your code! 99 times! And then starting again from scratch! 01:43:48 Oh wait, does the 0 node have to be in the active zone? 01:43:52 I bet that's important! 01:45:02 -!- tusho has quit (Remote closed the connection). 01:45:06 "By the active part U(S) of a state S we mean the subcomplex of the complex S consisting of the vertices and edges belonging to chains of length lambda <= N containing the initial vertex." 01:45:11 Indeed it is. 01:45:26 Writing that active zone just got a whole lot easier! 01:45:43 I just have to check the hoods of 0 01:46:47 * Slereah7 is totally scanning the article 01:46:50 Take that Andrei! 01:47:20 He's a commie, he probably won't mind. 02:25:58 -!- Slereah has joined. 02:26:35 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 06:25:13 -!- timotiis_ has quit (Read error: 104 (Connection reset by peer)). 06:28:51 -!- cherez1 has joined. 06:30:22 -!- Deformative has quit (Read error: 110 (Connection timed out)). 07:03:03 -!- timotiis has joined. 07:14:56 -!- oklopol has joined. 07:39:21 -!- cherez1 has left (?). 07:39:35 -!- cherez has joined. 07:42:37 -!- GreaseMonkey has joined. 07:59:29 -!- oklopol has quit (Read error: 104 (Connection reset by peer)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:09:25 -!- GreaseMonkey has quit ("Saliendo"). 08:09:32 -!- oklopol has joined. 08:36:54 -!- puzzlet has joined. 08:42:37 -!- puzzlet has quit ("WeeChat 0.2.6"). 08:59:25 -!- puzzlet has joined. 09:27:21 -!- kar8nga has joined. 09:27:33 -!- oklopol has quit (Read error: 113 (No route to host)). 09:31:24 -!- oklopol has joined. 09:34:10 -!- oerjan has quit ("leaving"). 09:53:47 http://www.vjn.fi/oklopol/straw.txt cool lang 09:54:17 i especially like the way mutation is done 10:37:16 -!- psygnisfive has quit (Read error: 104 (Connection reset by peer)). 10:37:17 -!- augur has joined. 10:48:59 -!- oklofok has joined. 10:49:07 -!- oklopol has quit (Read error: 113 (No route to host)). 11:00:44 -!- kar8nga has left (?). 11:01:45 -!- oklofok has quit (Read error: 113 (No route to host)). 11:16:24 -!- timotiis has quit (Remote closed the connection). 11:16:48 -!- timotiis has joined. 11:41:51 -!- kar8nga has joined. 12:02:49 -!- Hiato has joined. 12:36:41 -!- kar8nga has left (?). 13:02:38 -!- oklopol has joined. 13:09:03 -!- RedDak has joined. 13:19:48 -!- ihope has quit ("ChatZilla 0.9.82.1 [Firefox 2.0.0.14/2008040413]"). 13:43:11 -!- Slereah7 has joined. 13:44:19 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 13:51:19 -!- oklopol has quit (Read error: 113 (No route to host)). 13:51:37 -!- oklopol has joined. 13:56:27 -!- oklopol has quit (Read error: 113 (No route to host)). 13:56:33 -!- oklopol has joined. 13:57:41 -!- Hiato has quit ("Leaving."). 13:58:03 -!- oklofok has joined. 13:58:22 -!- oklopol has quit (No route to host). 14:21:00 -!- oklopol has joined. 14:21:03 -!- oklofok has quit (Read error: 113 (No route to host)). 14:27:07 -!- Slereah has joined. 14:27:07 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 14:34:33 -!- kar8nga has joined. 14:36:10 -!- ais523 has joined. 14:41:36 -!- oklopol has quit (No route to host). 15:01:21 I suppose no one knows how to use the PSP emulator? 15:05:36 Slereah, what PSP emulator? 15:05:48 Program Segment Prefix? 15:05:55 http://membres.lycos.fr/bewulf/Divers3/PSPE.jpg 15:05:58 This one here. 15:05:59 or portable Playstation? 15:06:12 The latter. 15:06:16 no idea 15:06:40 I could help with getting the n64 emulator mupen64 working 15:07:03 1) don't try on x86_64, the code is mostly x86 asm 2) thus install a 32-bit chroot 15:07:13 oh the irony 15:07:25 as it emulates a 64-bit CPU 15:12:48 * ais523 realises with horror that they had 5 different web browsers open at once a few minutes ago 15:13:16 Who are "they"? 15:13:24 You and your multiple personalities? 15:13:41 haha 15:14:04 Owait 15:14:08 On the PSP emulator 15:14:12 Slereah: I was using singular they to refer to myself 15:14:19 There's a stderr.txt : 15:14:19 in the third person 15:14:21 load H:\pspe\ms0\PSP\GAME\SOFT1\EBOOT.PBP 15:14:21 PBP format 15:14:21 not elf header 15:14:25 I need elves 15:14:30 I use they for that in several cases 15:14:56 incidentally, the browsers were IE, FF2, FF3, Konqueror, and Akregator's built in web browser 15:15:07 Slereah, ELF.... 15:15:09 I have Epiphany installed here too, but happened not to be using it at the time 15:15:11 Slereah, is a file format 15:15:33 used by, for example, Linux, FreeBSD, OpenBSD, NetBSD, and most, if not all other *nix 15:15:42 Mac OS X use a custom format I know 15:16:08 AnMaster: ELF is one of the formats that SCO claimed they owned the right to a while ago 15:16:22 nobody believes them AFAICT, but that at least implies that SCO probably use it 15:16:52 -!- RedDak has quit (Read error: 104 (Connection reset by peer)). 15:17:13 * ais523 loads up SunOS 15:17:17 So how do I shot elf header? 15:17:48 you use a linker that produces ELF 15:17:52 to link the executable 15:18:25 yes, SunOS uses ELF too 15:18:34 and as a result it seems highly likely that Solaris does 15:19:04 so IOW, it seems likely that every major OS but MacOS and Windows uses it by default 15:19:14 well, MacOS X 15:19:30 and I suspect MacOS X is capable of using it but it isn't the default 15:19:37 "“Shit happens.” This is not a PSP Emulator. It won’t play comercial games, ISO’s or even ELF’s." 15:19:53 I think my computer might not be the culprit. 15:20:18 "Fuck this software! those who are rich and can afford to buy psp FUCK YOU… we cant just buy psp it is not cheap. LOOOKK AT YOUR MAMA AND FFFUCCCKK THEEEMM UPPP, BECAUSE YOU ARE ALL MUTHAFUCKING ASSHOLES! MOST AMERICANS ARE ASSHOLES…" 15:20:31 I get the feeling this PSP emulator does not work so well as an emulator 15:20:42 well, what does it emulate, then? 15:20:46 the PSP's processors? 15:21:01 maybe it could run PSPOS, whatever that is 15:21:27 Maybe I could try to find the game on PC. 15:21:35 But boy does it not look good 15:22:18 Ah, found one. 15:22:32 Slereah: you aren't trying something illegal, are you? 15:22:39 Well, actually, no 15:22:47 I actually possess the Playstation CD. 15:22:58 ah, but you want to run it on a different system 15:23:12 Well, my playstation doesn't work so good any more 15:23:15 It's 11 years old 15:23:20 Plus, I don't know where it is 15:23:34 And last time I tried it, the memory card wasn't working. 15:23:46 those are all good reasons not to use it 15:24:17 I also don't know where the playstation CD is. 15:24:35 My mom probably put all of that somewhere mysterious. 15:25:39 If you want something illegal, try this! http://membres.lycos.fr/bewulf/Russell/Kolmo/ 15:25:50 no, I don't want something illegal 15:25:58 But beware, or the American Mathematical Society might send their goons after you. 15:26:03 Slereah, I won't click that link, what is it? 15:26:09 illegal maths? 15:26:15 * AnMaster gets confused too 15:26:15 there are a couple of illegal numbers 15:26:21 ais523, well that is disputed 15:26:23 Well, it's technically copyrighted. 15:26:23 both of which became famous on the internet 15:26:29 yep 15:26:39 It's the Uspenski-Kolmogorov machine article 15:26:50 Slereah, and why is that ileegal? 15:27:10 Well, it is copyright 1963 by the American Mathematical Society. 15:27:28 the article itself is copyright infringement? 15:27:37 "No portion of this book may be reproduced without the written permission of the publisher" 15:27:51 And I totally did D: 15:28:26 Have mercy on me, office of technical services! 15:28:38 (The guys who did the translation) 15:30:06 Your request for http://membres.lycos.fr/bewulf/Russell/Kolmo/Kolmo%2001.jpg could not be fulfilled, because the domain name membres.lycos.fr could not be resolved. 15:30:09 tor phail 15:30:31 (spelling was sarcasm with intent) 15:30:58 No need for Tor, I'm sure the AMS isn't monitoring my websit! 15:31:47 or so they want you to believe 15:32:39 well I don't have time to read it, or knowledge to understand it 15:37:18 Actually, I would recommend reading from page 11 and on first. 15:37:39 It's easier to understand. 15:37:42 ish. 15:51:56 http://www.esolang.com/ 15:51:59 Oh shit. 15:52:04 what is that? 15:52:40 Click it, and be amazed. 15:52:42 nothing to do with esolangs AFAICT 15:52:49 registered in Russia, unsurprisingly 15:52:52 Yes. 15:52:54 given its content 15:53:03 I was googling for esolang 15:53:07 And found this 15:54:48 http://forums.gleemax.com/showthread.php?t=1042139 15:54:50 Heh. 15:55:05 Magic players aren't enthusiastic at the idea. 15:55:13 hey, Esolang has an article about that 15:55:17 I even wrote a program in it 15:55:53 the cards in it are a little out-of-date, though 15:56:11 I admit to being slightly shocked that someone else went to the trouble of understanding the program well enough to fix a bug in it, though 15:58:00 Slereah: what do you really expect? It's never even explicit in that post that you're talking about a programming language. The only people that could possibly have any idea what you're talking about would be people that already code in esolangs, and I doubt a wizards-of-the-coast forum is a good place to find those. 15:58:28 That post isn't by me. 15:58:39 I don't play Magic. 15:58:41 yeah, I just noticed that 15:58:58 so apparently I don't get to ridicule you for playing Magic. :/ 15:59:10 I don't play it at the moment 15:59:16 because prices went up at the place where I used to play it 15:59:23 and I tend to desert places when they increase prices 15:59:23 I used to play it... in 5th grade. 15:59:36 it's a good idea for a game, though 15:59:54 and it can certainly feel like programming at times 15:59:57 And as Knuth said, that's the most important part! 16:00:13 Oh, game, not name 16:00:27 oh dear, we seem to have a new meme 16:02:00 lol 16:02:14 so games are now the most important part. This could generate some interesting esolangs. 16:02:35 well, there's my esolang-based text adventure 16:02:40 which is not very complete 16:02:48 and which I haven't written any of for ages 16:02:53 And there was that text adventure based esolang. 16:02:58 DUNGEONS AND DATAS 16:03:10 and Lost Kingdoms, which is a text adventure written in an esolang 16:03:16 Also, cake. 16:03:17 GregorR: I demand you add this hat to your collection: http://www.costumecraze.com/HAT37.html 16:03:20 heh, three different concepts 16:04:05 http://forums.gleemax.com/showthread.php?t=1042139 16:04:08 Oops 16:04:14 I mean, http://www.costumecraze.com/MASK54.html 16:04:21 I want to rob a bank with that mask. 16:04:23 Slereah: used the wrong clipboard? 16:05:00 Yup 16:05:07 Slereah: nah, you should rob a putt-putt golf course or a driving range with that mask 16:05:47 I found a French page on esolangs. 16:05:48 this is actually rather terrifying: http://www.costumecraze.com/MASK56.html 16:05:52 Slereah: PURGE IT 16:05:54 There's something you don't see everyday. 16:06:14 The missing teeth make it look a little goofy. 16:08:11 Heh. The Klingon Hello world is "What do you want, universe?" 16:08:24 Slereah: yes, that's about the closest you can get in Klingon 16:09:09 Those klingons are so full of boyish attitude. 16:09:59 The website is actually pretty horrible 16:19:18 HOLY SHIT NEW DRESDEN CODAK http://dresdencodak.com/cartoons/dc_053.html 16:21:12 I'm trying to tie this to the previous comics 16:21:17 But it ain't easy. 16:21:30 it makes sense to me 16:21:54 Is that Kimiko? 16:22:01 And if so, what is she doing in a jar? 16:22:04 and in the future 16:22:08 And in that robot's head 16:22:29 if you haven't read the last few pages, read them 16:22:58 I have. 16:24:29 Hm. 16:24:47 I wonder if you could make a TC language with stupid mathematical functions. 16:25:05 "stupid"? 16:25:13 like, INTERCAL stupid? 16:25:16 I was thinking stuff like Ackermann and such 16:25:25 oh, massive-stupid 16:25:31 stupid to calculate 16:25:36 You can do Ackermann with µ, but can it work the other way around? 16:25:48 how about Ramsey numbers? 16:26:31 Or that 91 function. 16:27:18 http://en.wikipedia.org/wiki/McCarthy_91_function 16:27:19 Thar. 16:27:52 interesting 16:28:07 sure, seems like a sufficiently evil way to provide a constant 16:28:33 Well, not all values get out 91. 16:28:37 and even an extremely limited form of branching 16:28:44 "constant" 16:29:06 I guess you'd have to include at least 0 in the language. 16:29:20 and perhaps 1 16:29:26 Plus, Ackermann can easily provide a successor operator, since A(0,n) = n+1 16:29:35 ah, bingo 16:29:56 so you can thus trivially generate any needed integers via ackermann 16:30:00 RodgerTheGreat: I also thought about a lang where Ackermann and inverse-Ackermann were the only operators 16:30:21 also, I was going to have it so that everything in the entire lang was unprintable characters 16:30:33 it's a very pure idea, but it might be more fun to have several functions like that, to add some variety 16:30:41 e.g. ASCII 0-31 and 127, not allowing spaces, tabs, newlines or vtabs 16:30:43 and perhaps make it more possible to be TC 16:31:22 we should make all variables greek characters. 16:31:30 oh, and my lang was deliberately not TC 16:31:39 it was an Ackermann-bounded automaton 16:31:44 hm 16:31:56 whenever it took input, it was allowed to allocate a certain amount of memory to use 16:32:04 which it could calculate with the remnants of its old memory 16:32:20 and a program can start with any amount of memory to begin with which must be specified in the code 16:32:24 also, all memory was write-once 16:32:28 -!- kar8nga has left (?). 16:32:39 I haven't worked out the rest of the details yet, though 16:33:01 anyway, the lang is definitely not TC, but likely to be powerful enough for most computations which aren't searches through an infinite search space 16:33:13 like searching for Riemann hypothesis counterexamples, for instance 16:33:26 so, we have A, R and M. can anybody think of other wildly impractical functions that could potentially be useful? 16:34:02 well, there's Malbolge's tritwise-crz operator 16:34:18 that was invented for the purpose of being wildly impractical 16:34:21 would the Jacobi Symbol be a good one? 16:34:28 http://en.wikipedia.org/wiki/Jacobi_symbol 16:34:34 RodgerTheGreat: look-and-say operator 16:35:05 oooh, nice one 16:35:07 -!- Hiato has joined. 16:35:09 ais523 : I like the way you think! 16:35:27 oh, you may as well do generalised Graham's stuff 16:35:40 because it would just be wrong not to be able to easily express A(g_64,g_64) 16:36:24 -!- Hiato has quit (Client Quit). 16:36:24 haha- check out the 71st-order polynomial related to look-and-say: http://en.wikipedia.org/wiki/Look-and-say_sequence 16:36:26 Nah, Graham's notation makes some sort of sense. 16:36:48 Unlike the Ackermann functioncreated to be specifically non-primitive recursive. 16:37:51 Inverse Ackermann, too? 16:38:05 Otherwise, I'm not sure how to get smaller numbers. 16:38:39 inverse ackermann so you can do things like subtraction 16:38:46 Yes. 16:38:55 And function definition? 16:39:08 not so sure about that 16:39:14 To have some sort of functional Malbolge. 16:39:25 aargh! 16:39:51 IMO Malbolge's main interesting features are the c++,d++ stuff and the encryption of commands when they're run 16:40:18 trying to use crazy to perform useful operations can also be interesting 16:41:34 So if no function definition, how to use all that? 16:41:59 well, with some other control structure 16:42:10 or allow definition of piecewise functions 16:42:19 the standard ones are functional, iterative with looping, imperative with goto, and the fungelike method 16:42:28 oh, declarative and concatenative too 16:42:30 which would in effect give you quite powerful flow control and make recursive procedures easy 16:42:44 what do you mean by 'piecewise functions' here? 16:43:29 make it a purely functional language, and allow the programmer to specify values for output based on properties of the input 16:44:04 wikipedia's description of Ackermann, for example, makes use of a piecewise definition 16:44:28 http://upload.wikimedia.org/math/0/a/e/0ae4053de098cc9554752b190a38bc56.png 16:44:40 So... Some conditional operator? 16:44:45 the issue with allowing users to define their own functions is that they probably wouldn't end up using all the interesting ones the language offers 16:45:02 But this language would offer no interesting functions. 16:45:17 ais523: which might be solved by allowing piecewise definitions, but not having conventional boolean operators. :D 16:45:33 how many logic levels are we having? 16:45:48 it's hard to come up with interesting boolean operators if you only have 2 logic levels 16:45:51 I don't understand 16:46:04 Logic level? 16:46:28 as in, true/false are the usual logic levels for a boolean 16:46:32 Oh. 16:46:35 and that gives you operators like AND and OR 16:46:39 What about... INFINITY? 16:46:47 but say if we had true/false/FILE_NOT_FOUND instead, that would make things more interesting 16:46:58 Doesn't intuitionist logic have an infinity of values? 16:47:29 yes, I think so 16:47:48 true/false/infinite regress 16:48:00 SELF REFERENTIAL 16:48:02 Noooooo 16:48:20 then the language would require a halting oracle 16:48:33 to implement, anyway 16:51:59 So... Piecewise definition? 16:52:20 Using our crazy functions as logical operators? 16:53:45 In a three valued logic? 16:55:32 sounds sufficiently esoteric to me 16:55:44 yep 16:55:54 What would it do if the third value hits? 16:56:16 and toss in the greek letter idea for syntactic flavor 16:56:27 Slereah: it would do the opposite 16:56:41 e.g. if you say if(boolean) {a++}, and boolean is megafalse, it decrements a 16:56:41 What would be the opposite? 16:56:55 OFC, that isn't the actual syntax 16:57:03 just an example in syntax that's well-known 16:57:07 bbl 16:57:45 Opposite seems a little challenging, especially with the 91 function. 16:58:27 hmm... what would 'opposite' mean here? 16:58:35 instead of a=91(b), do b=91(a)? 16:58:42 that would be one possibility 16:58:54 OFC, this means you have to be able to assign to constants, but INTERCAL's never found that a problem 16:59:02 and it used to be legal in Fortran 16:59:10 confusing, too, because constants were passed to functions by reference 17:01:07 But if constants become functions, they still won't be fed arguments. 17:01:18 -!- Judofyr has joined. 17:02:11 Maybe we could rotate functions if we get the third value. 17:02:27 Like A becomes 91, 91 becomes look and say, what have yous. 17:02:48 aha, yes 17:02:59 also, all commands other than conditionals should use the functions 17:03:11 if you want to assign one variable from another, you have to do it through an Ackermann 17:03:29 Why "other than the conditionals"? 17:03:44 Slereah: because they take boolean input 17:03:58 Well, you could just get the numbers mod 3. 17:04:59 OK, fine by me 17:05:02 how will loops be done? 17:05:24 Recursion? 17:06:37 how will function definition be done, then? 17:11:45 -!- Slereah7 has joined. 17:12:13 Ah shit. 17:12:58 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 17:22:40 [17:24] [CTCP] Received CTCP-PING reply from ais523: 33 seconds 17:24:39 Is there a way to compute the Ramsey's numbers? 17:24:43 Easily, that is. 17:26:13 no, I don't think so 17:26:18 most of them aren't known 17:26:30 Oh. 17:26:39 At the time of writing, even the exact value of R(5,5) is unknown, although it is known to lie between 43 (Geoffrey Exoo) and 49 (Brendan McKay and StanisĹ‚aw Radziszowski) (inclusive); barring a breakthrough in theory, it is probable that the exact value of R(6,6) will remain unknown forever. 17:27:03 Then it might not be a good idea to include it.3 17:28:02 -!- Hiato has joined. 17:28:28 there's a neat Erdos quote on the same page 17:28:53 It is quite awesome. 17:28:53 Hello, World! 17:29:04 What do you want, universe? 17:29:21 heh :) 17:29:22 Hiato, what would you think of some sort of functional Malbolge? 17:29:50 "Imagine an alien force, vastly more powerful than us landing on Earth and demanding the value of R(5, 5) or they will destroy our planet. In that case, we should marshal all our computers and all our mathematicians and attempt to find the value. But suppose, instead, that they asked for R(6, 6), we should attempt to destroy the aliens." 17:30:13 err... well, lets see. Sounds interesting, though I doubt [one] would be able to take advantage of lazy evaluation in Malbolge 17:30:39 how would the encryption be done? 17:30:46 olsne: Can you help me out 17:31:00 ais523: OOOH - Lets use my strem cipher 17:31:03 PLEASE! 17:31:15 Hiato: does it encrypt functions? 17:31:17 *olsner 17:31:34 Well, it encrypts stuff :P Anything in the range [0;255] 17:31:50 so, yeah, it would be able to encrypt any binary info :) 17:31:52 most functions aren't in the range [0;255] 17:31:58 you could encode them as such 17:32:08 but I don't think it's possible to have a unique encoding for functions 17:32:14 at least, not in a TC system 17:32:15 well, I do :) 17:32:21 because it's uncomputable to determine whether two functions are the same 17:32:43 I was thinking along the lines of, the function name is encrypted, the function process is encrypted and the output s decrypted (not being encrypted in the first place) 17:33:02 Also, what would the language be called? 17:33:07 XKCD? D: 17:33:20 Slereah7: not bad 17:33:24 and by encryption, I mean literal character encryption by means of my poly-alphabetic substitution self modifying network 17:33:30 heh, yeah, not bad 17:33:32 after all, writing the xkcd number is one of the few things it would be good at 17:33:41 :) 17:33:58 Well, if we put in the Graham notation 17:33:58 assuming it can calculate g_64 easily 17:34:07 I was thinking just G(64) 17:34:15 rather than the rest of the notation 17:34:17 Graham notation? 17:34:22 so you can get arbitrary Graham's numbers 17:34:37 ... 17:34:38 I just had an idea. 17:34:49 anyway, this is unlikely ever to be implementable 17:34:56 What if we totally went overboard with that greek character idea? 17:35:01 what's look-and-say of the XKCD number, anyway? 17:35:04 What if we used function definition, ancient greek style? 17:35:24 bringing a whole new meaning to 'lambda' calculus? 17:35:30 I dunno. 17:36:14 I feel it wrong of me to distract from the topic here, but I must ask: Who here is good at mathematics, specifically the business of odd conjectures? 17:36:46 I have one that needs proving/debating and that is as interesting as the Collatz Conjecture (in my view anyway) 17:36:46 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 17:36:47 -!- Slereah has joined. 17:37:15 ais523, Slereah, anyone? :P 17:38:51 I'm good at maths but not really at odd conjectures 17:39:01 I only like bits of maths, really 17:39:20 well, see if this tickles you then: http://rafb.net/p/9IEm5j97.html 17:39:28 there's the impl. and here's the conjecture: 17:40:30 Take any 2 positive integers, say x and why, and continually add the sum of all the digits of each number to their respective number. Unless *only one* of the numbers is a multiple of three, the two paths will merge 17:40:38 *y not why :P 17:41:11 sounds arb, bu it seems to work :) 17:41:23 *but 17:41:27 ok, that's mildly interesting 17:41:41 are there attractor numbers which always seem to end up on paths 17:41:42 * Hiato wonders if it's international typo day 17:41:52 what do you mean? 17:41:56 for divisible by 3 and indivisible by 3 17:42:18 e.g. if for some reason the paths always ended up going through 12345678 unless the original number was higher than some critical value 17:42:36 aha, I see 17:42:41 in that case, yes 17:42:55 620 for pretty much everything below 20 I believe 17:43:07 and 1003 for everything below 50 17:43:14 so I presume really 17:43:23 *below 50, above 20 17:43:45 try it, using Merge('101','595') 17:43:55 or Merge('5','1') 17:44:03 err, that is the bigger number comes first 17:44:45 -!- kar8nga has joined. 17:47:55 Holy shit 17:48:09 I have absolutely horrible ideas for output. 17:48:20 Hiato: what about Merge('9','3')? 17:48:32 ah, the divisible-by-3 can become divisible by 9 17:48:34 but not vice vers 17:48:38 so it still works 17:48:41 pretty trivially 17:49:00 err, I revise my conjecture :P If one of the numbers is a multiple of three, the paths never converge, if both the numbers are multiples of three, there is not guarantee of a merge 17:49:15 as Merge('45','3') doesn't seem to work 17:49:52 "Equality. In printed books before the modern equal sign, equality was usually expressed with a word, such as aequales, aequantur, esgale, faciunt, ghelijck, or gleich, and sometimes by the abbreviated form aeq " 17:50:09 bbiab, supper calls :) 17:51:55 http://www.ibiblio.org/expo/vatican.exhibit/exhibit/d-mathematics/images/math01.jpg 17:52:04 Come on Euclid, can't you write correctly? 18:03:58 Found the Elements in greek. 18:04:27 ?????. 18:04:31 MONAAAADS 18:07:04 µ???d?? 18:07:12 Monads just can't give me a break. 18:18:56 -!- tusho has joined. 18:19:01 hello! 18:19:11 hi tusho 18:19:21 bad timing BTW, because I leave here at 7 on Sundays 18:19:22 What do you want, universe? 18:19:26 i have settled into a routine of starting with irssi&w3m it seems 18:19:42 ais523: heh, very bad timing 18:19:52 I was just going to work on ESO... 18:20:02 we can still do some work on ESO 18:20:05 for 40 minutes 18:20:16 and you can probably manage some of it without my help 18:21:19 probably :P 18:22:38 the eso forum is difficult to use with w3m 18:22:40 w3m should do css 18:22:41 :P 18:23:33 ok, I think I'll try X 18:23:54 I think I will use esti for function definition. 18:24:05 It's what's used in Elements, apparently. 18:24:37 I'm not too sure how to do the arguments of a function next. 18:25:08 xchat go! 18:25:20 -!- tusho_ has joined. 18:25:24 <3 18:25:42 ok, now for epiphany 18:26:20 ais523: do you think this is all superstition? I have a nagging feeling the crashes are entirely random. :P 18:26:44 WHAT THE FUCK, I'm #11 on reddit 18:26:49 with my 'hello esolangs reddit' post 18:26:51 :| 18:27:06 I think they're determinstic, mostly, but loading programs up in a different order won't help 18:27:08 and link? 18:27:13 ais523: reddit.com 18:27:16 11th item 18:27:50 not to my view 18:28:04 ah 18:28:10 maybe it's because I'm (obviously) subscribed to that reddit 18:28:17 and it's hot on that reddit being the only thing there 18:28:21 so it gets on my front page 18:28:33 anyway 18:28:34 -!- tusho has quit ("leaving"). 18:28:41 phew 18:28:54 -!- tusho_ has changed nick to tusho. 18:31:32 -!- tusho_ has joined. 18:31:32 -!- tusho has quit (Read error: 104 (Connection reset by peer)). 18:31:37 lal! 18:31:41 You said 40 minutes but it looks more like 5 now. 18:31:51 more like 27, actually 18:31:59 ais523: you think it won't crash again? 18:32:06 it depends on what you do 18:32:23 ais523: open a terminal :P 18:32:29 that didn't crash it 18:33:37 ais523: by the way, I enabled user dirs on eso-std.org 18:33:47 http://eso-std.org/~ais523 -> /home/ais523/public_html 18:33:54 cgi doesn't get enabled by default but I can 18:33:58 where are they in the filesystem? 18:34:14 http://eso-std.org/~ais523 -> /home/ais523/public_html 18:34:23 you said that twice 18:34:43 ais523: I was answering your question. 18:34:50 twice? 18:35:06 ais523: I said it before you asked. 18:35:09 ah, the first one wasn't intended to be an answer to the question 18:35:11 and then after 18:35:12 but you said it after I asked 18:35:16 wrong 18:35:18 at least at this end of the connection 18:35:19 http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:19 cgi doesn't get enabled by default but I can 18:35:19 where are they in the filesystem? 18:35:28 [18:35] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:28 [18:35] cgi doesn't get enabled by default but I can 18:35:28 [18:36] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:33 whoops 18:35:36 [18:35] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:36 [18:35] cgi doesn't get enabled by default but I can 18:35:36 [18:36] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:40 [18:35] ais523: by the way, I enabled user dirs on eso-std.org 18:35:40 [18:35] where are they in the filesystem? 18:35:40 [18:35] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:35:44 [18:35] ais523: by the way, I enabled user dirs on eso-std.org 18:35:45 [18:35] where are they in the filesystem? 18:35:45 [18:35] http://eso-std.org/~ais523 -> /home/ais523/public_html 18:36:02 then you answered the same question again 18:36:12 -!- timotiis has quit (Read error: 110 (Connection timed out)). 18:36:18 * timotiis has quit (Read error: 110 (Connection timed out)) 18:36:27 [18:38] <-- timotiis has left this server (Read error: 110 (Connection timed out)). 18:36:32 [18:38] <-- timotiis has left this server (Read error: 110 (Connection timed out)). 18:36:41 [18:38] [18:38] <-- timotiis has left this server (Read error: 110 (Connection timed out)). 18:36:52 . 18:37:06 " and Akregator's built in web browser" 18:37:10 this is called 'konqueror' 18:37:16 not exactly 18:37:19 they use the same backend 18:37:25 but Akregator is not Konqueror 18:37:31 ais523: I believe the component is actually called Konqueror 18:37:34 and it uses KHTML 18:37:34 it can't do file-management AFAIK 18:37:40 it can 18:37:46 but akgregator will never end up feeding it a file url 18:37:53 why not 18:38:02 what if someone puts one in an RSS feed? 18:38:07 that might work 18:38:12 ais523: oh, and OS X can't use ELF, period 18:38:13 it's Mach 18:38:16 only 18:38:24 interesting 18:38:30 Which it inherits from NeXTStep (or whatever the CapS are) 18:38:38 I would have thought it would be able to run them 18:38:51 ais523: Basically, OS X is what happens when you take NeXTStep, and some BSDs, and collide them together with blood, guts, and battle cries. 18:38:59 Then, you nurture its wounds until it's all happy. 18:39:04 Then, you make it pretty. 18:39:26 -!- ihope has joined. 18:39:39 If it walks like a bar and talks like a bar, duck. 18:39:39 ihope ihope is the real ihope 18:39:50 ihope: that's a pretty neat antisig 18:39:58 IRC so needs join messages 18:40:01 as well as quit messages 18:40:07 ais523: that's SO abusable 18:40:08 for spambots 18:40:24 tusho_: well, they wouldn't show if you couldn't make a commen 18:40:30 s/commen/privmsg/ 18:40:34 speaking of spambots, I wrote a script that takes a text file and outputs a script that my okotterance co-ordinator can use 18:40:37 as in, you don't get a join message if unvoiced 18:40:38 you can configure how many clones it uses 18:40:57 I don't see how it would be any more abusable than privmsg, then 18:41:03 it wouldn't 18:41:09 my script is very abusable though 18:41:16 so why did you say it would be? 18:41:17 since you can have an infinite send speed 18:41:22 well 18:41:24 given infinite clones 18:42:06 can't you do that with privmsgs? 18:42:33 "You have been guided to this site by a divine light of healing and creative energy of Reiki. " 18:42:34 tee hee 18:42:44 ais523: no, because of the forced delay when you spam 18:42:48 and of course the flood protection 18:42:56 if you have 100 bots, you can make messages 100x faster 18:43:07 I don't see why join messages would be any different 18:43:20 they would presumably be force-delayed for spammers too 18:44:10 I like how that esolang site contains pages called 'Rebirthing' and pages called 'Cross-cultural communication in English' 18:48:27 is it just me, or has anagolf died? 18:49:21 What's anagolf? 18:49:21 ais523: it's always like this 18:49:27 if you mean 'slow' 18:49:32 we just caused a blimp of activity 18:49:35 ihope: http://golf.shinh.org/ 18:50:09 * ihope downloads tomsrtbt, sets up a virtual machine, etc. 18:51:07 Gee, I think the Linux computer has lots more free space. I'm going to go use it, I think. 18:51:30 ihope: What, are you doing rootnomic? 18:51:37 Yep. >:-) 18:51:42 ihope: Fine, fine, I'll code it. 18:51:59 If only to stop tomsrtbt being used. 18:52:07 ihope: limit its memory usage to 256.1 MB so as to annoy ehird 18:52:09 * tusho_ sets up virtualbox 18:52:18 ais523: actually, he was just trying to get me to do it 18:52:24 he's done it before 18:52:43 I did consider the possibility that he was just pretending to do something to encourage you to do it first 18:52:48 ihope: virtualbox is installing, happy now? ;) #ircnomic 18:53:07 ais523: oop, it's 5 minutes 18:53:13 that went fast 18:54:55 4 minutes now 18:55:04 ais523: 3, actually 18:55:11 still 4 by my clock 18:55:13 and it's only approximate 18:55:20 ah, /now/ 3 18:55:47 ais523: my clock is ntp'd I believe 18:55:59 mine was manually ntp'd 18:56:08 from time to time, I synched it 18:56:16 off an Internet server 18:56:21 seems that can't be done in Hardy, though 18:56:30 it's automatic NTP or none 18:56:48 ais523: 60 seconds 18:58:41 -!- ais523 has quit ("(1) DO COME FROM ".2~.2"~#1 WHILE :1 <- "'?.1$.2'~'"':1/.1$.2'~#0"$#65535'"$"'"'&.1$.2'~'#0$#65535'"$#0'~#32767$#1""). 19:07:01 -!- tusho_ has changed nick to ehird. 19:07:05 -!- ehird has changed nick to tusho. 19:16:30 HELLO EVERYONE 19:16:58 wassabi 19:18:28 What do you want, universe 19:18:32 brb 19:18:39 -!- tusho has quit (Remote closed the connection). 19:19:15 -!- tusho has joined. 19:38:04 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 19:43:02 -!- Slereah7 has joined. 19:46:55 Holy Batman. 19:47:08 What's wrong, Robin 19:47:12 I think I might actually be able to pull off some ancient greek functional language. 19:47:40 Diophantus had a whole system of it. 19:47:45 Slereah7: YES BATMAN! 19:48:08 I think that with that and some flat out invention, I might be able to do something 19:49:03 Slereah7: If you do, I'll implement it, 19:50:05 ihope: Oi. 19:50:05 (diff) (hist) . . N Chris Barker‎; 18:01 . . (+78) . . 58.56.109.6 (Talk) (RipfxvQbmrCYuf) 19:50:07 tusho : If you missed the previous discussion 19:50:07 Ban. Now. 19:50:16 Slereah7: Probably? 19:50:22 The idea was to make a functional language with stupid functions 19:50:39 zomg 19:50:48 Like Ackermann, McCarthy 91, Look and say 19:50:48 Stuff like that 19:51:05 Then, someone proposed greek letters 19:51:05 And I had the idea of doing it totally in greek, dude 19:51:22 you can represent all the rational numbers in an infinite list in which ALL rational numbers are reachable 19:51:29 :o 19:51:58 Well yes 19:52:00 They're countable 19:52:08 i know, but i just figured out how to do it :) 19:52:21 Actually, every definable number is countable 19:52:42 defineable? 19:52:46 Uncountable sets are for non-constructive math. 19:52:47 Definable. 19:52:53 oh i see. 19:52:55 constructive. ok. 19:52:56 anyway 19:52:59 i figured out how to do it :D 19:53:02 im pleased with myself 19:54:09 Slereah7: should we revive unikitten 19:55:03 Slereah7: the set of all definable numbers, you mean? 19:55:22 aleph_1 and its buddy omega_1 aren't countable, but they're definable. 19:56:16 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 19:57:35 -!- Slereah7 has joined. 19:59:25 augur: now figure out how to represent every rational number as a number with a finite decimal expansion such that if one rational number is bigger than another, its representation is bigger than the other's. 19:59:29 >:-) 19:59:41 lol 20:01:05 infact 20:01:33 i think the way i figured out makes it possible to not even need to store the list but just compute the nth rational of the list on the fly 20:02:17 not that the list for doing that would be efficient in space usage, since it'd have repetition of /value/ but 20:03:13 1/2, 1/3, 2/3, 1/4, 2/4, 3/4, 1/5, 2/5, 3/5, 4/5, etc.? 20:03:47 no, that misses the rationals > 1 :) 20:04:10 True. 20:04:37 Hm. 20:04:47 1/1, 1/2, 2/2, 2/1, 1/3, 2/3, 3/3, 3/2, 3/1, 1/4, 2/4, 3/4, 4/4, 4/3, 4/2, 4/1, etc. 20:04:47 Maybe I should make two languages in greek 20:04:55 The silly one, and a srs one. 20:05:00 1/1 -- 1/2 ; 2/1, 2/2 -- 1/3, 2/3, 3/3 ; 3/1, 3/2, 3/1 -- ... 20:05:13 Because Greek arithmetical notation seems eso enough 20:05:17 yeah, ihope 20:05:21 there you go. 20:05:46 tho redundancy as i have it works to make it trivially findable 20:06:47 atleast i think so.. lol 20:08:36 I think xaě means "multiply" 20:11:54 I THINK 20:11:56 THEREFORE i ESO 20:13:29 i thunk, therefore noone ever disputes me. 20:13:53 "Notation and definition of Diophantos" 20:13:55 Oh yeah 20:25:23 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 20:26:34 are we sure that real numbers are uncountably infinite? 20:27:35 Yep. 20:27:43 hm. 20:29:39 yes 20:30:26 yes 20:32:24 thats because there are reals which aren't repetitive, nor finitely long, right? 20:32:25 like pi 20:32:37 right o - third and final conjecture for this evening :) 20:32:42 ok. 20:33:03 ok im out 20:33:04 see ya 20:33:05 augur: more like, what's 'twixt 1/3 and 1/4 ... say 1/5 etc etc etc 20:33:19 http://rafb.net/p/Q0NgAz40.html 20:33:20 what? 20:33:27 here is the implementation 20:33:31 what? 20:33:44 augur: a real/floating point/non-intergral number is a fraction, right? 20:33:54 sure 20:33:57 gotta go tho bye :) 20:34:05 so, you can arbitrarily sub-divide any range between any two fractions 20:34:35 into any more arbitrary ranges 20:34:35 ad iniftum 20:34:35 i know 20:34:35 ok bye 20:34:35 *infinitum 20:34:35 ok, cheers 20:35:06 tusho? 20:35:07 you there? 20:36:02 -!- Slereah has joined. 20:36:13 /summon tusho 20:36:17 hi Hiato 20:36:27 * Hiato breathes a sigh of relief 20:40:07 hahah 20:40:07 tusho, can I ask you to glance quickly over my latest and greatest conjecture? 20:40:07 http://rafb.net/p/Q0NgAz40.html? 20:40:07 yep 20:40:07 want an explanation as to what it is and why it's so cool? :P 20:40:07 I am not Tusho 20:40:07 heh, well spotted slereah 20:40:07 however, Slereah, would you like to know? :P 20:40:07 * Hiato needs opinions 20:40:07 I'd like to know 20:40:07 but your code style sucks ;P 20:40:07 heh, tusho, I just scraped out of delphi :P 20:40:44 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 20:40:58 Hiato: remove the space before function arguments 20:41:06 and use do_it instead of DoIt for function names 20:41:20 anyway, it goes like this. Take any arbitrary integer, x, and if it is positive, find the sum of its digits and one and subtract that from it. If x is negative, find the sum of its digits and one and add that to it. Recursively apply this procedure and it is conjectured that for all of x, it will reduce to the pattern: -1;1;-1;1... 20:41:28 tusho: roger that 20:41:33 apart from that, just add some more whitespace, really 20:41:38 heh :) 20:41:40 it's better than a lot of python! :-P 20:42:21 thanks :) 20:43:11 so, on the conjecture? 20:43:29 Hiato: I'll think about it 20:43:29 btw. http://www.python.org/dev/peps/pep-0008/ <-- PEP8, python style guide 20:43:32 written by guido himself 20:43:34 awesome :) 20:44:06 -!- Slereah7 has joined. 20:44:06 http://svn.browsershots.org/trunk/devtools/pep8/pep8.py <-- a script that complains if your code doesn't conform to it 20:44:06 nice :) 20:44:28 -!- RedDak has joined. 20:48:42 -!- oerjan has joined. 20:51:33 Slereah7: Give me somethign to feed that mu function will you? 20:55:31 Hmmmmmmmmmm. 20:55:34 I have ideas for le eso os! 20:55:53 who wants to hear 20:57:18 .... 20:58:22 -!- Slereah has joined. 20:58:27 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 20:58:28 Hello people sirs. 20:58:56 Slereah: Ideas for an eso os. Wanna hear? 20:59:04 Sure, why not 20:59:53 Slereah: Well, you know the lazy, strongly-typed self-rewriting language we talked about yesterday? 21:00:01 MIN? 21:00:21 Slereah: Yeah, but ... before we went totally crazy about it. 21:00:32 Right? 21:00:35 Go ahead. 21:01:06 Slereah: Well, all of that, but PURELY FUNCTIONAL and REFERENTIALLY TRANSPARENT. 21:01:21 For instance, the whole program rewrites itself into a lazy list of things like PutStr "hi" 21:01:36 which then get performed in order by the program engine (thus eventually forcing the whole list) 21:01:50 So the program is of type [IO], I guess. 21:01:51 But whatever. 21:01:54 Isn't that awesome? 21:02:32 Slereah: Rite? 21:02:36 So, you write the OS. In that. 21:02:45 I dunno. I'll have to understand it first 21:03:22 Slereah: Well, basically 21:03:24 You have a program 21:03:28 I hope so 21:03:29 and it lazily rewrites itself into something like 21:03:46 [Print "Hello world", Print "How Are You?"] 21:03:51 Slereah: So what the interpreter does 21:03:53 is take the program 21:03:54 How can something rewrite itself lazily? 21:04:00 and force it to evaluate the first element, and performs it 21:04:01 etc to the end 21:04:04 Slereah: Well, that's the hard bit. 21:04:09 But we discussed that yesterday. 21:04:15 The new part is that it's purely functional & ref transp. 21:04:30 Slereah: Oh, and installing a program amounts to the OS rewriting itself to include it. 21:04:38 Heh. 21:04:50 tusho: that's a bit awkward if you want to do input and have the rest depend on it 21:05:06 oerjan: yes, that was what I thought 21:05:07 oerjan: but no! 21:05:11 GetLine rewrites into "the input" 21:05:12 that's why the IO monad, after all 21:05:17 oerjan: because, after all, it is a self-rewriting language 21:05:24 oh 21:05:38 oerjan: clever, no? >:D 21:06:53 But right now. 21:06:55 ok but that's not referentially transparent 21:07:06 I'm a lot more excited by ARITHMETICA 21:07:13 I totally want to try it. 21:07:24 (It's the srs one, not the Ackermann one) 21:07:35 (because if you have two GetLine's they don't need to become the same thing) 21:07:39 oerjan: wrong 21:07:40 they do 21:07:47 um, that is the part I haven't worked out 21:07:55 but, basically, when the monadic action of GetLine happens, it's morphed into the string 21:07:57 somehow 21:07:57 but 21:07:59 not 21:08:00 kidn of 21:09:00 actually most of this works but not putting all the actions in a list 21:09:08 oerjan: ok then, forget that part 21:10:24 oerjan: it is an interesting concept, though 21:10:27 don't you agree? 21:10:31 self-rewriting ... but lazy 21:10:32 and _typed_ 21:11:09 hm 21:12:04 it certainly sounds like the OS TECHNOLOGY OF THE FUTURE 21:12:21 the self-morphification and reflection of lisp as the very core of the language 21:12:26 the mathematical elegance of laziness from haskell 21:12:32 and the program-assurance of types 21:14:19 oerjan: or not 21:15:21 BRB 21:20:00 Back. 21:20:13 Slereah: Opinions? 21:20:14 Hiato: you too 21:20:30 tusho : No. 21:20:36 sorry, what? 21:20:41 heh 21:20:47 READ UPWARDS :P 21:20:48 I'm working on ARITHMETICA 21:20:56 Hoping that I don't have to learn greek to do it 21:20:57 tusho: how far? :P 21:21:13 Hiato: Slereah: Ideas for an eso os. Wanna hear? 21:21:26 right o guv 21:25:19 How would it be done to use a whole set of non-standard symbols for a programming language? 21:25:24 Wow, could you, er, idiotize this (for a lack of a better word) 21:25:25 How does APL do it? 21:25:45 Hiato: no :P 21:25:57 Slereah: well 21:25:58 in ascii 21:26:00 97=a 21:26:02 and stuff 21:26:06 just make a new encoding 21:26:12 But HOW :o 21:26:12 97=FLYING SPAGHETTI MONSTER 21:26:17 Slereah: Just in your program 21:26:20 interpret the bytes 21:26:21 differently 21:26:24 e.g. 21:26:30 if c == 'a': # c = flying spaghetti monster 21:26:31 tusho: In that case, as I understand it, impossible to implement, but potentially awesome :) 21:26:40 Hiato: not impossible, no 21:26:50 in fact, this would open up a very very interesting direction in programming 21:26:55 Yes, but I mean, how would it display in the interpreter, graphically? 21:26:58 Slereah: are you SURE they aren't in unicode 21:26:59 well, tusho, then I evidently don't understand right :P 21:27:04 tusho : Some are. 21:27:06 and, well, you'd draw a picture and use a graphics library 21:27:09 Actually, most 21:27:13 That is, of course, a giant fuss. 21:27:14 Stick to uncode. 21:27:17 *unicode 21:27:18 But important ones aren't. 21:27:26 Slereah: Pick other ones. 21:27:26 Although maybe there's close ones. 21:27:34 Or ask the unicode consortium to add 'em. 21:27:38 Also, there's plenty of exposant, how to? 21:28:13 Slereah: wut 21:29:03 Hey, there's actually Diophantus minus symbol 21:29:29 tusho : Like, ˛ 21:29:33 But for greek letters. 21:29:52 Slereah: i dunno lol 21:30:02 Make code html and use 21:30:03 :P 21:30:31 Will it work for greek? 21:30:58 that was a joke 21:31:00 but yes 21:42:36 -!- Slereah has quit (Read error: 104 (Connection reset by peer)). 21:47:47 -!- Slereah7 has joined. 21:51:56 >:| 21:54:47 -!- jix has joined. 22:00:22 -!- kar8nga has left (?). 22:05:08 -!- cherez has quit (Remote closed the connection). 22:05:38 -!- cherez has joined. 22:07:00 -!- cherez has left (?). 22:07:26 -!- Hiato has quit ("Leaving."). 22:10:31 -!- Judofyr has quit. 22:37:40 Can the esowiki do superscript? 22:40:34 yes 22:40:35 22:41:01 -!- timotiis has joined. 23:02:59 z 23:05:41 I'm not too sure what to use for variables in arithmetica 23:06:13 There's the sigma for single-variables, but he doesn't seem to use anything consistent with multiple ones. 23:06:33 It's a number with two letters in superscript, but the letter change often 23:07:03 . 23:10:45 ¤ 23:10:59 wotwot 23:11:52 ESO IDEA!!! 23:12:00 'unwrapped' conditionals 23:12:01 Like: 23:12:03 if n == 0 23:12:06 then fact n = 1 23:12:10 else fact n = n * fact (n-1) 23:12:13 Cool? 23:12:54 * ihope shrugs 23:12:54 Not very 23:13:18 -!- oklopol has joined. 23:13:35 Slereah7: Why? 23:13:42 I think it's actually distantly related to prolog 23:13:43 Watch: 23:13:45 if n == 0 23:13:47 then fact n = 1 23:13:58 else if result == n * fact (n-1) 23:14:08 then fact n = result 23:14:18 see? 23:14:27 The conditionals 'predict' the future variables. 23:14:30 Maybe it interests oerjan ;P 23:15:01 Interesting. 23:15:34 ihope: I believe you can do nondeterminism with it. 23:15:36 like: 23:15:48 if x `elementOf` xs then choose xs = x 23:16:08 Yep. 23:16:41 Though I'd rather write that as if xs = x then x `elementOf` xs. 23:16:50 s/if xs/if choose xs/ 23:17:01 ihope: Well, that's confusing. 23:17:10 Mine is fairly simple: You just 'unwrap' the conditionals outside the function 23:17:20 It's more logically correct. :-) 23:18:00 if ys `sorted` xs then sort xs = ys 23:18:00 :P 23:19:01 Something like if x `elementOf` xs then choose xs = x would kind of imply that choose xs is equal to every element of x. 23:19:24 ihope: No. 23:19:26 It reads: 23:19:27 Which is... oofy unless all elements of xs are equal. 23:19:35 s/would kind of/would seem to/, then. 23:19:38 'If x is a member of xs, then choosing an element from xs results in x' 23:19:44 Makes sense to me. 23:19:57 ihope: Yours makes the membership of xs an afterthought. 23:19:59 Mine makes it the basic idea 23:20:00 s/results/can result/. 23:20:09 Or not. 23:20:25 s/Or not/Or else not/ 23:20:51 Be back in a moment. 23:20:54 s/Or/The/ 23:21:01 Uh oh. :-) 23:21:03 -!- ihope has quit ("ChatZilla 0.9.82.1 [Firefox 2.0.0.14/2008040413]"). 23:21:06 s/Of/And/ 23:28:38 -!- ihope has joined. 23:28:43 Back. 23:28:50 As I'm sure you can tell. 23:33:21 :P 23:46:37 -!- Slereah has joined. 23:46:38 -!- Slereah7 has quit (Read error: 104 (Connection reset by peer)). 23:48:56 -!- RedDak has quit (Read error: 104 (Connection reset by peer)). 23:59:57 -!- Slereah7 has joined.