00:00:08 you can't reach the level i'm talking about. the absolute bottom of informational computationality. 00:00:22 so. 00:00:29 i think the fundamental imperative computational is something like: 00:00:39 in other words i think i should sleep 00:00:48 Perform an operation on two values and put the result somewhere, then go somewhere else depending on the result. 00:01:07 so let's say 00:01:12 sjz A B C 00:01:16 A and B are memory addresses 00:01:23 "go" "put" "perform" 00:01:26 subtracts the value in A and the value in B 00:01:34 and puts the result in B 00:01:38 then, if the result is zero 00:01:40 it jumps to C 00:01:44 otherwise, execution continues as normal 00:02:00 good? 00:02:19 equality comparison: 00:02:21 that's so compicated i can't understand it, let alone read or see it. 00:02:25 sjz A B notequal 00:02:26 *complicated 00:02:28 [its equal] 00:02:30 notequal: 00:02:32 [its not equal] 00:02:36 Notwithstanding, there is no Turing-complete language simpler than BCT. 00:02:42 maybe. 00:02:47 but i"m talking about instruction sets 00:02:49 imperative unz 00:02:49 oklopol: is "equals" too complicated for you? 00:02:52 kerlo: you're wrong 00:02:55 ocrap 00:02:57 I invented subleq 00:02:58 Execution of one instruction A B C subtracts the value of memory in A from the content of memory in B. If value after subtraction in B less or equal to zero, then execution jumps to the address C; otherwise to the next instruction. 00:03:05 well, less or equal to zero 00:03:07 No, simplicity is objective and I'm always right. 00:03:10 so mine's kind of less complex 00:03:25 kerlo: testing equality is pretty hard, yes 00:03:37 I never said testing equality. 00:03:43 so umm. 00:03:48 how about continuous computation 00:03:51 Proce. 00:03:55 ok then: 00:03:55 you need an infinite amount of it to get anything done 00:03:57 ej A B C 00:04:07 Proce. 00:04:08 programs are functions from reals to instruction fragments 00:04:11 proce? 00:04:13 Reads A and B, if the values are equal, it stores 1 in B, otherwise it stores 0 in B. 00:04:16 If they ARE equal, it jumps to C. 00:04:17 RSSB is less complex. One instruction and only one operand 00:04:24 Hmm. 00:04:24 impomatic: link? 00:04:38 i'm going for conceptual simplicity here 00:04:44 so if RSSB does a lot, it doesn't qualify 00:04:47 hmm you know what 00:04:53 i think ej could possibly be TC 00:04:56 if you scratched the storing 00:05:03 kerlo: proce? 00:05:09 if the value in A is the value in B, jump to C. otherwise, continue as normal. 00:05:13 dunno how you store things ofc :D 00:05:32 kerlo: so umm proce? 00:05:35 It's a programming language where every variable thingy is a function of real numbers. 00:05:48 You can implement the sine function very easily in Proce. 00:05:54 oh that lang of yours? 00:06:04 Yes. 00:06:07 that has nothing to do with what i said, but yeah, i remember it 00:06:20 See the RSSB page I made the other day http://esolangs.org/wiki/RSSB 00:06:29 hrm 00:06:30 interesting 00:06:37 essentially you have differential equations iirc 00:06:38 but it has registers 00:06:40 so not simple 00:06:45 which are automatically solved 00:06:53 Well, it had plenty to do with what you said until you said "programs are functions from reals to instruction fragments". 00:06:56 to get definitions for funcs involved 00:06:59 What do you mean by "instruction fragments"? 00:07:19 impomatic: can't IP/ACC be in memory? 00:07:21 like, at 0 and 1 00:07:31 ah 00:07:32 it is 00:07:37 "The result is stored in both memory and the accumulator" 00:07:38 both? 00:07:41 surely that is not needed 00:07:46 can't it just be in the accumulator/ 00:07:47 kerlo: proce doesn't have continuous computation, it has computation applied on continuous things. 00:07:48 (Tough language: Programs are continuous functions from reals to reals where the image contains only integers.) 00:07:51 *applied to 00:08:03 hmm 00:08:17 kerlo: i have no idea what i mean by instruction fragments. 00:08:35 Oh. 00:08:48 impomatic: ? 00:08:48 i mean continuous computation. instructions you can have any real amount of. 00:09:09 ehird: I'm not sure why the spec said both 00:09:20 * ehird writes simplified rssb 00:09:27 say you do instruction : 3.16 times, then # 87.001 times, and... umm... maybe something happens. 00:09:35 * kerlo nods 00:09:47 impomatic: define below 0 00:09:52 -!- puzzlet_ has quit (Remote closed the connection). 00:09:56 -!- puzzlet has joined. 00:09:57 signed integers in an architechture? Perverse. 00:10:03 kerlo: i'm just trying to think outside the box, computation's always so discrete and boring. 00:10:13 below 0: if the subtraction causes a borrow 00:10:41 So, memory is a continuous function from R. 00:10:47 There's a redcode implementation at http://impomatic.blogspot.com 00:10:51 impomatic: i don't follow 00:10:54 essplain? 00:10:56 you mean like 00:11:00 1-2 00:11:00 And, um... 00:11:05 kerlo: yeah maybe, and instructions perform on single values 00:11:06 despite not being below 0 00:11:08 since there's no such thing 00:11:10 counts as it? 00:11:14 Yes 00:11:18 okie 00:11:26 kerlo: so you need an uncountable amount of them to get anything done 00:11:46 Basically if prior to the subtraction, acc > contents of memory location, skip the next instruction 00:12:10 mem locs: 0 ip, 1 accumulator, 2 always contains 0, 3 input, 4 output 00:12:11 Instruction: Subtract the accumulator from the contents of the operand, and 00:12:12 store it in the accumulator. If this caused a borrow, jump to the instruction 00:12:14 past the next one. 00:12:20 simplified 00:12:22 kerlo: also flow control isn't jumping, it's more like slowly fading into another kind of computation, and somehow implement this by finding the conceptual integral of what the fading computation will achieve during its infinite fade. 00:12:23 An instruction is a continuous function f : x real number -> such that f(f(x,a),b) = f(x,a+b). 00:12:31 For all x, a and b. 00:12:36 impomatic: so is there a reason it has to be stored in both? 00:12:59 There is an implementation of RSSB by David Tanguay, but I can't find a copy online 00:12:59 s/real number/non-negative real number/ 00:13:11 impomatic: ? 00:13:23 ehird: I just implemented as per David Tanguay's spec 00:13:26 yeah 00:13:33 I'm just wondering if only storing in the acc could be simpler 00:13:41 kerlo: that's a bit too concrete at this point. 00:13:49 Is not. 00:13:54 you don't want to have your functions be discrete objects.... 00:14:10 * kerlo shrugs 00:14:13 they need to be continuous computation 00:14:13 wait 00:14:18 storing it in just one is useless 00:14:20 hrm 00:14:21 but both is ugly 00:14:33 i think the accumulator is a hack 00:14:42 kerlo: i guess that's pretty much the definition of continuous computation though :P 00:14:48 the accumulator is a lie 00:14:52 hmm or not. 00:15:16 kerlo: but err, probably a good definition. 00:15:23 ehird: if it only stores in acc, there's no way to modify memory. if it only stores in memory, then the only thing you can do with acc is set it to zero 00:15:45 yeah 00:15:49 still 00:15:51 it's ugly 00:16:12 The set Q consists of all continuous functions Q -> Q. 00:16:27 Q is the least fixed point of that. 00:16:40 Or, if you prefer, the union of all fixed points of that. 00:16:47 impomatic: 00:16:47 sjlz a b: Subtract the contents of a from the contents of b, and store the 00:16:48 result in b. If this caused a borrow, skip the next instruction. 00:16:52 much more conceptually pure 00:16:54 kerlo: sounds feasible 00:16:58 and only one more operand 00:18:34 I is in Q, as for all x in Q, x is in Q, and if a set S of Q's is open, its preimage under I (that is, S) is also open. 00:19:13 kerlo: I as in identity? 00:19:16 For all x in Q, K x is in Q, as for all y in Q, x is in Q, and if a set S of Q's is open, its preimage under I (either Q or empty) is also open. 00:19:23 I is identity combinator, K is constant combinator. 00:19:23 right. 00:19:30 yeah you answered that already 00:20:02 kerlo: did you invent Q just now? 00:20:20 Yes. 00:20:48 "if a set S of Q's is open, its preimage under I (that is, S) is also open" i don't understand this 00:20:59 outputting a 00:21:00 a out 00:21:00 but then again, you're the abstract nonsense guy :P 00:21:01 =a 97 00:21:05 where a out -> ajlz a out 00:21:13 and =a 97 defines a bit of memory with that value in. 00:21:26 ajlz a b: Add the contents of a to the contents of b, and store the result in b. 00:21:26 If this caused an overflow, skip the next instruction. 00:21:30 The preimage of S under I is S. Therefore, if a set S is open, its preimage under I is also open. 00:21:36 And yeah, I love abstract nonsense. 00:21:47 yeah who doesn't 00:22:07 preimage of x = the set of things that map to x? 00:22:13 also 00:22:22 here's a one-instruction one-character cat program 00:22:24 in out 00:22:24 :-D 00:22:27 * ehird makes it loop 00:23:00 oklopol: the set of things that map to elements of x. 00:23:24 kerlo: that's what i meant 00:23:27 impomatic: how do you jump backwards with rssb? 00:23:31 liek, map into x 00:23:38 but yeah okay 00:24:30 kerlo: what's the significance of the "and if a set S of..." part? 00:24:36 For K to be in Q, it must be a function Q -> Q (it is) and it must be continuous. For it to be continuous, the preimage of every open set must be open. 00:25:05 oh. that's a definition for being continuous? 00:25:09 Yep. 00:25:13 ah, alright. 00:25:16 impomatic: how do you jump backwards with rssb? 00:25:23 i'm quite new to this still 00:25:32 ehird: subtract something from ip 00:25:50 ah 00:26:15 and we all know that subtracting is a special case of addition ;-) 00:26:38 kerlo: oh actually i think i see how that relates to the crucial axiom of dedekind sets 00:26:42 The preimage of a set under K is {x : K x is in the set}, of course. And, um... bah, I'm going to ask #math a question. 00:27:12 -!- CakeProphet has joined. 00:27:16 Jump requires 4 lines of RSSB: 00:27:16 rssb acc ; set acc to 0 00:27:17 rssb $+2 ; set acc to loop offset 00:27:17 rssb ip ; subtract acc from ip 00:27:17 rssb $-loop ; the loop offset 00:28:32 then my infinite cat program is 00:29:14 ajof 3 4 00:29:17 ajof 2 0 00:29:29 3=input 4=output 2=maximum int (4294967295) 0=ip 00:29:34 ajof a b: Add the contents of a to the contents of b, and store the result in b. 00:29:34 If this caused an overflow, skip the next instruction. 00:29:50 hrm. 00:29:58 actually since the whole idea there is to overflow that doesn't work. i think 00:30:36 Is there a brainfuck variant which adds a stack? 00:30:46 many. but brainfuck sucks :P 00:30:52 Darn, the answer is "no". 00:31:30 hmm 00:31:31 I thought 00:31:31 max max 00:31:33 in out 00:31:35 in out 00:31:37 max ip 00:31:39 but then when it loops, it runs it twice 00:31:45 (max max doesn't touch max, ofc, since it's constant) 00:31:53 * kerlo ponders the definition of a neighborhood of an element of Q 00:32:00 impomatic: any thoughts? 00:32:04 kerlo: yes i looked 00:32:17 kerlo: what are neighborhoods? 00:33:07 Back tomorrow :-) 00:33:11 -!- impomatic has quit ("mov.i #1,1"). 00:33:29 A neighborhood of a point is a superset of an open set containing that point. 00:33:40 -!- ktne has left (?). 00:33:48 hmm. i'm not sure #math with appreciate your set 00:34:02 okay so 00:34:05 complete spec: 00:34:06 0 is ip, 1 contains 0, 2 contains max int (4294967295), 3 is input, 4 is output 00:34:07 ajof a b: Add the contents of a to the contents of b, and store the result in b. 00:34:09 If this caused an overflow, skip the next instruction. 00:34:11 go forth and write progz >:( 00:34:20 ps ajof a b can be written as ... a b 00:34:30 kerlo: okay right. 00:34:49 :3 00:35:06 f is near g if and only if for all x, f(x) is near g(x). Unfortunately, "near" doesn't mean anything. 00:35:46 In a neighborhood F of f, for all x, F(x) is a neighborhood of f(x), I think. 00:36:31 kerlo: btw that's a pretty weird definition, like, (2,4) U {0} would be a neighborhood of 3? 00:36:51 I believe so. 00:37:03 probably would be less weird if i saw uses. 00:37:27 someoneeeeee 00:37:34 kerlo: i'm not sure you understand what Q is. 00:37:42 What do you think Q is? 00:37:56 Nonexistent due to a power set cardinality theorem? 00:37:58 well i mean; how can a function from functions to functions be continuous? 00:38:33 kerlo: well my intuition says there might be a problem like that, but i don't actually know. what i do know is i don't understand what the definition even means. 00:38:38 I believe that for all topological spaces P and Q, there is a standard topological space P -> Q. 00:38:51 kerlo: problem is i don't know topology 00:39:15 you see i started doing math this fall. i'm doing integrals and groups now. 00:39:25 * kerlo nods 00:40:26 btw. gotta sleep i think. 00:41:13 --> 00:45:57 -!- oklopol has quit ("( www.nnscript.com :: NoNameScript 4.2 :: www.regroup-esports.com )"). 00:56:12 -!- GreaseMonkey has joined. 01:17:28 -!- GreaseMonkey has quit ("Client Excited"). 01:27:00 -!- Corun has joined. 01:35:46 -!- Corun_ has joined. 01:42:40 -!- FireFly has quit ("Later"). 01:50:28 -!- Corun has quit (Connection timed out). 02:20:40 wtf is this "instantaneous computation"? 02:20:56 and "strong synchronous programming"? D: 02:21:29 You enter shit, and bam! It's computed! 02:21:34 :P 02:25:33 -!- Corun_ has quit ("This computer has gone to sleep"). 02:38:31 Uh? 03:17:40 -!- GreaseMonkey has joined. 04:11:23 -!- GregorR has set topic: a) oko; b) the swatter; c) messing with the topic; d) http://tunes.org/~nef/logs/esoteric; e) the letter game; f) gay sex. occasionally.; g) http://esolangs.org/wiki/Main_Page; h) Felching – The Freshmaker!. 04:16:32 -!- Sgeo has joined. 04:31:31 -!- psygnisf_ has quit ("Leaving..."). 06:10:41 -!- Sgeo_ has joined. 06:28:31 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 06:29:01 A Dorponnified Talliachre is a lunchbox! It makes clicking noises! 06:29:37 Gregor is a lightbulb that floats in water! It pushes things down staircases and works underwater. 06:34:46 A woman is like a normal cricket bat, but it squirts clouds of black ink. 06:34:56 A man is like a normal pogo stick, but it's inflammable. 06:36:19 A pogo stick is a hair gel that fits in your pocket! It crushes ice. 06:36:29 Hair gel is a speaker system that's great for hammering in nails! It emits dangerous radiation and repairs itself. 06:36:40 I'm done :P 06:39:54 -!- Sgeo_ has quit ("Leaving"). 07:05:21 -!- GreaseMonkey has quit ("Client Excited"). 07:13:03 good god 07:13:08 what is this /topic 07:50:57 -!- spicule has joined. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:06:05 -!- spicule has quit ("Leaving"). 09:07:25 -!- SpaceManPlusPlus has joined. 09:13:02 -!- MigoMipo has joined. 09:25:46 -!- oklopol has joined. 09:27:07 -!- SpaceManPlusPlus has quit (Read error: 110 (Connection timed out)). 10:37:59 * oklopol is listening to some quality heavy metal bashing: http://www.vjn.fi/s/black.mp3 11:11:37 -!- Mony has joined. 11:12:41 plop 11:39:39 -!- FireFly has joined. 12:02:37 -!- Judofyr_ has joined. 12:05:35 -!- Judofyr has quit (Nick collision from services.). 12:05:51 -!- Judofyr_ has changed nick to Judofyr. 12:17:59 -!- MigoMipo has quit (Read error: 110 (Connection timed out)). 12:28:55 -!- jix has joined. 12:30:52 -!- Slereah has joined. 12:35:08 -!- MigoMipo has joined. 12:42:44 -!- Slereah2 has quit (Read error: 110 (Connection timed out)). 12:49:43 -!- Corun has joined. 12:56:43 -!- MigoMipo has left (?). 13:09:43 -!- Corun has quit ("Leaving"). 13:09:54 -!- Corun has joined. 13:24:12 -!- oklopol has quit (Read error: 60 (Operation timed out)). 13:42:45 -!- impomatic has joined. 13:49:05 112 Hello, World! ++++++++++++++[>+++++>+++++++>+++>+<<<<-]>++.>+++.+++++++..+++.>++.>[<-<<+>>>-]<++.<<+.>.+++.------.--------.>+. 13:59:45 -!- ehird has left (?). 14:00:03 -!- ehird has joined. 14:00:42 112 Hello, World! ++++++++++++++[>+++++>+++++++>+++>+<<<<-]>++.>+++.+++++++..+++.>++.>[<-<<+>>>-]<++.<<+.>.+++.------.--------.>+. 14:00:44 neato! 14:02:44 Hi Ehird. 1 instruction longer than Wikipedia, but it includes the comma. 14:02:56 does it include the newline 14:03:00 No 14:03:08 you really want a newline :p 14:03:28 Some of the other examples didn't have newline 14:03:35 hrm. 14:03:39 It is standard to terminate with 10, thoug 14:03:39 h 14:03:45 err, not 10 newlines :P 14:08:50 -!- Corun has quit ("This computer has gone to sleep"). 14:30:23 -!- Judofyr has quit (Remote closed the connection). 14:34:50 -!- luqui has joined. 14:35:22 gah omg 14:35:25 2008-01-18 14:35:25 00:42:31 --- nick: oklopol -> Z 14:35:26 00:42:46 --- nick: Z -> oklopol 14:35:28 00:42:59 too late 14:35:31 oklopol you let it get away 14:35:38 uhhhh 14:35:39 hi people 14:35:42 :| 14:35:47 ^bf ,[.,]!brainfuck. etc. 14:35:48 brainfuck. etc. 14:41:06 Lol at censoring brainfuck to b****fuck :-) 14:43:33 :) 14:48:15 lol 14:48:57 Anyone want to implement Redcode++? http://corewar.co.uk/opcodes.htm 14:49:31 sweet 14:50:00 I wanted to make a esolang like this 14:50:04 but now, it's too late :o 14:51:21 -!- oerjan has joined. 14:51:30 wait 14:51:32 redcode++? ew 14:51:35 why not just 14:51:36 recode 14:51:36 :P 14:52:02 * oerjan thinks that looks like the end of a haiku 14:52:19 Yael still beats redcode 14:52:33 15 instructions 14:52:35 * AnMaster wonders... 14:52:38 SQL++? 14:52:39 :D 14:52:39 admittedly, you only get 256 bytes of memory. 14:53:03 AnMaster: Newsflash: taking SQL and bashing something on the end of it is not the source of a thousand hilarious ideas. 14:53:06 as much as you want it to be. 14:53:49 ehird, actually C++ style object orientation bolted onto SQL would be quite nasty 14:53:57 nasty != interesting and funny 14:54:01 I mean even nastier than it already is 14:54:14 in as much as the same way as a language based on the holocaust would be "nasty" 14:54:32 ehird, well said 14:54:45 speaking of which, who wants to beta-test holocaust++ :p 14:55:07 specs? 14:55:48 O_O 14:56:50 ehird: what about SQLCOATL? 14:56:52 jew: Takes a random jew off the heap. gas(x): Gasses the jew x. 14:57:11 When you have gassed as many jews as the nazis, it executes the remaining program as Perl. 14:57:16 also: QUERY foo = NEW QUERY('SELECT * FROM mytable'); foo.EXECUTE(); 14:57:17 ;P 14:57:28 -!- luqui has left (?). 14:57:52 the great database serpent 14:58:02 oerjan, :D 14:59:44 oh wait 14:59:57 * oerjan realizes that holocaust++ was made up on the spot here 15:00:06 i was starting to worry about ehird 15:00:49 hee 15:01:13 The holocaust never happened, silly man 15:01:42 oerjan, SQLULHU? 15:01:58 AnMaster: doesn't roll that well off the tentacle 15:02:08 true 15:02:16 hard to pronounce 15:02:20 Slereah: from you this is expected 15:02:21 -!- Judofyr has joined. 15:02:37 SQLHU perhaps 15:02:55 oerjan : You expect that from a jew? 15:02:57 Racist 15:03:06 oerjan, hm maybe 15:03:10 Slereah: you've done so before 15:03:32 jew bashing, that is 15:03:56 actually there is probably some jew somewhere denying the holocaust 15:04:53 oerjan, btw: wtf at iwc today 15:04:58 i mean i vaguely recall hearing about neo-nazi jews 15:05:07 AnMaster: um no spoilers 15:05:21 that's next on my schedule :D 15:05:22 I always read that as neon-nazis 15:05:27 it sounds so much cooler than it is 15:06:08 much easier to hit them when lighted up 15:07:05 ah the Espionage theory was right 15:07:38 er wait, disregard, SPOILER :D 15:08:08 oerjan, well, it could be a red hearing 15:08:24 http://www.irregularwebcomic.net/draakslair/viewtopic.php?t=3190 15:08:33 no i mean the theory that since it wasn't in the list, it had escaped unscathed 15:09:10 oerjan, that could be a red hearing 15:09:25 >_< 15:09:38 um wait are you misspelling that on purpose? 15:09:46 oerjan, no? 15:09:48 oops 15:09:51 herring* 15:10:05 oerjan, why should I misspell it on purpose? 15:10:15 another awful pun attempt of course 15:10:26 I can't see how that could be a pun 15:10:27 at all 15:10:38 I can't see how any of your "puns" could be a pun. 15:10:40 good, good 15:11:00 AnMaster: now we're just missing a Supers strip 15:11:15 ehird, Chehkov's pun 15:11:17 sorry! 15:12:02 http://tvtropes.org/pmwiki/pmwiki.php/Main/ChekhovsPun 15:12:14 it exists? 15:12:23 (standard tvtropes warning applies) 15:12:31 but of course 15:13:44 oh it's a tvtropes meta-thing 15:14:00 and self-referential too :D 15:14:55 huh, very confusing 15:14:58 -!- impomatic has quit ("http://impomatic.blogspot.com"). 15:15:20 oerjan, oh btw uf was quite funny today 15:15:32 as for the current xkcd: I don't get it 15:15:37 it's about making new trope names by making puns on completely unrelated ones 15:15:57 uf? 15:16:10 uf=userfriendly. shit comic. 15:16:39 too long running 15:16:49 no. too shit 15:16:52 isn't it over 10 years now? daily 15:17:21 ehird, would you say garfield is shit? 15:17:35 Why is that even a question? Who the fuck would say "no"? 15:17:48 I agree. But some of the early ones aren't as bad 15:17:49 AnMaster: you don't want your genetic algorithm to develop sentience and destroy the world, is the point 15:17:52 no 15:18:04 all garfield comics are irredeemably terrible 15:18:07 oerjan, true, but xkcd wasn't very good today 15:18:11 and anyone who likes them has some sort of mental defect 15:18:13 or whatever it was 15:18:14 also, xkcd was hilarious. 15:18:15 it was yesterday 15:18:18 Friday? 15:18:27 if you want to read xkcd it's best to know pop culture. 15:18:39 Monday, Wednesday, Friday, 5:00 UTC 15:18:55 http://www.google.com/search?q=skynet 15:19:00 square root of minus garfield rocks though 15:19:05 ehird, yes I did google it 15:19:17