00:00:43 Yes, you are correct 00:01:10 -!- lambdabot has quit (Ping timeout: 240 seconds). 00:01:25 lambdabot seems really unstable today 00:01:54 It's practicing *really* lazy evaluation w.r.t. responding to pings. 00:01:56 -!- GreaseMonkey has joined. 00:02:32 lethargic evaluation 00:05:32 oerjan: i just wrote a patch to make it more stable 00:06:17 good, good 00:12:15 -!- zzo38 has quit (Remote host closed the connection). 00:12:35 -!- kallisti has quit (Ping timeout: 245 seconds). 00:17:16 -!- oerjan has quit (Quit: Good night). 00:17:47 elliott: @admin - me all day. I'm sure it'll be satisfying. 00:19:09 -!- ais523 has quit (Remote host closed the connection). 00:19:45 -!- kappabot has joined. 00:24:00 -!- kappabot has quit (Remote host closed the connection). 00:37:01 -!- kallisti has joined. 00:43:51 Madoka-Kaname: hi 00:44:03 Who's you?? 00:44:18 /win 4 00:44:29 coppro: oops 00:45:11 coppro: your parole officer was here earlier 00:45:37 elliott: I noticed 00:45:51 wha 00:50:26 git is dumb, why it has to mail stuff, instead of just connect to the site and put your stuffs there lol 00:50:26 now you have to mail manually or configure smtp and shit 00:51:34 * kallisti scratches head furiously. 00:54:54 -!- Klisz has joined. 00:57:29 -!- DCliche has quit (Ping timeout: 248 seconds). 01:00:37 "sceptical" 01:00:49 It used to be that that spelling didn't look odd to me. 01:01:38 -!- Patashu has joined. 01:02:30 elliott: So anyway, I think MID is a better choice than MOD. 01:02:32 shachaf: I always used to say skeptical but then I became TRUE BRITISH 01:02:36 Or maybe ... both? 01:02:37 Gregor: Hokay :P 01:02:49 Gregor: Make the MID one just render to MOD >:) 01:02:53 And then pass through 01:03:19 elliot: You're hardly true British. 01:03:36 I'm truest British. 01:03:36 You probably don't even drink tea. 01:03:39 Also I noticed that. 01:03:41 Yes I do. 01:05:22 -!- augur has joined. 01:07:38 * shachaf tries to think of other British stereotypes. 01:07:44 Anyway I'm sure they don't apply to you. 01:12:39 shachaf: heavy tea-drinking is one of those British stereotypes that is apparently true or something? 01:14:22 shachaf: I'll bet his teeth are all white and straight. 01:14:45 elliott: Yes! Your teeth must be perfect. 01:14:55 shachaf: What "teeth"? 01:15:44 You know, the things that make gears work? 01:15:59 I'm not clockwork. 01:16:33 Someone write mollis for me for Christmas, please. 01:16:37 elliott crushes his food with a mortar and Union Jack before he shoves it in his mouthhole. 01:16:49 THAT'S DISRESPECTFUL 01:17:13 *Union Jack on a flagpole not the flag itself ha ha ha 01:17:17 shachaf: You do it. 01:17:46 elliott: Do what? 01:17:52 Write mollis for me. 01:18:30 -!- derdon has quit (Remote host closed the connection). 01:18:45 What's a mollis? 01:19:11 shachaf: An ostensible port of durus to Haskell. 01:19:24 wow Skype literally cannot go a few hours without freezing up inexplicably. 01:19:25 What's a durus? 01:19:57 durus is this awesome client-server-or-embedded file-backed persistent STM implementation; the only flaw is that it's written in Python for some incomprehensible reason. 01:24:39 elliott: will you please kill me? 01:24:51 Yes. 01:25:26 okay. if you devise a sufficient means I'll gladly fly to Hexam to complete arrangements. 01:30:04 elliott: why isn't the flag of Wales featured on Union Jack 01:30:15 wouldn't your nation's flag be so much cooler with A DRAGON On it? 01:30:20 we hate wales 01:30:54 "As Wales was not a Kingdom but a Principality it could not be included on the flag." 01:31:16 http://www.telegraph.co.uk/news/uknews/1570998/Japan-offers-to-solve-Union-Jack-problem.html 01:31:17 thank you japan 01:31:58 shachaf: Are you done yet? 01:32:17 elliott: With your molidur thing? 01:32:27 Yes. Mollis./ 01:32:37 In the debate, Albert Owen MP said that "we in Wales do not feel part of the union flag because the dragon or the cross of St David is not on it."[21] Conservative MP Stewart Jackson described the comments as "eccentric". 01:32:41 SICK BURN 01:32:46 Yes, I finished but then it was slow because Haskell makes everything slow. 01:33:20 shachaf: you should have written it in Perl. 01:33:58 shachaf: That's okay, I'm slow. 01:34:29 elliott: http://i.telegraph.co.uk/multimedia/archive/00652/news-graphics-2007-_652632a.gif come on 01:34:31 elliott: Well, I already deleted it. 01:34:35 you know you want that dragon 01:34:36 shachaf: Just write it again. 01:34:37 I'm not doing all that work over again. 01:39:52 I wonder if you can get the internal GHC bytecode as output from GHC. 01:39:59 I wonder if that's what -fbyte-code does. 01:42:26 No, that generates the external GHC bytecode. 01:42:53 Wow, -fbyte-code actually produces executables. 01:43:03 There's also -ddump-bcos. 01:43:44 Now I just need to figure out how to make it compile an expression to a bytecode object and print its type. 01:43:46 -XNoNewQualifiedOperators -XExplicitForALl -XNoExplicitForAll 01:43:48 Good documentation. 01:44:33 No, no, that's no the documentation. This is the documentation: 01:44:37 -XNewQualifiedOperators 01:44:37 Enable new qualified operator syntax 01:44:57 It's in the man page. 01:45:26 Read my lips: no new qualified operators 01:45:26 -XNoFoo ~ No -XFoo 01:46:07 shachaf: "-XExplicitForALl -XNoExplicitForAll" 01:46:10 Sheesh. 01:46:12 It's the L part. 01:46:31 elliott: Oh. 01:46:40 * shachaf is capitalization-blind, you insensitive clod! 01:46:46 Hmm, I wish acid-state wasn't so annoying to use. 01:53:00 -!- augur_ has joined. 01:53:32 -!- augur has quit (Read error: Connection reset by peer). 01:56:40 oh, hmmm, maybe I /won't/ fail this class. 01:58:34 -!- Jafet has joined. 02:01:05 > pi + sqrt(-1) :: Complex Double 02:01:18 @tell lambdabot come back plz 02:12:26 kallisti: "i" isn't even sqrt (-1). :-( 02:13:04 * kallisti gives shachaf a gold star. 02:13:07 GOOD JOB! 02:13:12 -!- Vorpal has quit (Ping timeout: 240 seconds). 02:13:15 What? 02:13:19 you're correct. 02:13:21 you win the prize. 02:13:23 which is a gold star. 02:13:33 you should feel proud. 02:13:36 * shachaf doesn't get it. 02:16:53 -!- PiRSquared17 has joined. 03:02:43 -!- MDude has quit (Ping timeout: 248 seconds). 03:03:34 -!- MSleep has joined. 03:03:34 -!- MSleep has changed nick to MDude. 03:05:02 -!- Jafet has quit (Quit: Leaving.). 03:15:44 -!- Jafet has joined. 03:32:56 -!- augur_ has changed nick to augur. 03:35:28 > map (\n -> foldr (+) 0 [1..10^n]) [1..] 03:35:32 ... 03:35:48 hi, my name is kallisti, and I'm a lambdabot addict. 03:37:36 hi my name is elliott and im a robot 03:38:28 Prelude> map (\n -> foldr (+) 0 [1..10^n]) [1..] 03:38:29 [55,5050,500500,50005000,5000050000,500000500000,^C 03:38:44 there we go, much better. 03:39:49 > x ^^ y 03:39:53 hsertiuhssiudfhisudhfse 03:41:13 elliott: also, as far as I can tell (^^) actually is repeated multiplication, but it also applies recip for negative exponents. 03:41:21 @src (^^) 03:41:25 let's just wait for lambdabot 03:41:27 lol 03:41:30 well 03:41:33 x ^^ 2 03:41:33 yields 03:41:35 x * x 03:41:40 x ^^ (-2) 03:41:40 uh thats irrelevant 03:41:42 yields 03:41:45 it could just be the way Expr implements it 03:41:45 recip (x*x) 03:41:46 but ok 03:41:52 i guess that would work too 03:42:39 http://hackage.haskell.org/packages/archive/base/latest/doc/html/src/GHC-Real.html#%5E%5E 03:42:46 yeah it uses (^) in the definition 03:43:33 Note to self: Learn how to unmtl stuff manually 03:43:47 Relying on lambdabot, as people in #haskell seem to suggest, is not a good idea 03:44:03 yes it is 03:44:05 I like GHC's comments. 03:44:06 you can install lambdabot locally 03:44:08 very enlightening. 03:44:51 -!- MDude has quit (Ping timeout: 248 seconds). 03:45:09 -!- MSleep has joined. 03:45:39 it would be nice if (^) and (^^) could somehow be the same thing? I think that would require dependent typing or some form of subclassing. 03:46:35 or... a (^) method in Num but that sounds bad. 03:47:39 numbers are tricky. 03:48:14 elliott: maybe the person who hosts lambdabot is playing some kind of cruel prank. 03:48:18 TO SHOW US OUR LAMBDABOT DEPENDENCE. 03:48:42 it's cale 03:48:47 and it just has issues with disconnections 03:48:50 which i sent a patch to fix 03:48:56 so it should get in the next time cale is present 03:49:15 http://hpaste.org/54864 <-- my hideous patch 03:50:04 I may add a plugin to my bot that runs lambdabot locally and then acts as a proxy between my bot and lambdabot so that I can have all of lambdabots commands. 03:50:09 * kallisti elegant. 03:50:33 lambdabot has a command-line interface + an api 03:50:38 I already have @type and > but @type is kind of hacky 03:50:41 elliott: oh. cool. 03:50:44 but 03:50:45 don't 03:50:48 you really don't want to compile lambdabot 03:50:57 why not? 03:51:03 it's easy if you remove all the dependencies and thus make it almost functionality-free 03:51:06 but 03:51:14 what if I'm just awesome? 03:51:14 (a) the codebase is really REALLY ancient and crufty 03:51:18 oh. 03:51:19 (b) it hasn't been updated in years 03:51:28 (c) the dependencies, like, don't work, there's internal conflicts at least on this system 03:51:34 (d) 03:51:34 (e) 03:51:35 (f) 03:51:36 ... 03:51:38 the obvious solution: reimplement lambdabot's plugins in perl. 03:52:10 clearly @pl is just a series of regex substitions. 03:52:28 Does Racket have these sorts of issues with dependencies? Reason I mention Racket is I've heard good things about the module system 03:53:09 "On 15 August 2008 he was awarded a knighthood. He is the first penguin to receive such an honour in the Norwegian army." 03:53:15 also I can one-up lambdabot by having @undefine actually remove definitions. 03:54:31 Sgeo: the 03:54:32 "issue" 03:54:36 is that the dependencies are specified wrongly 03:54:43 because they don't specify lower bounds 03:54:46 and the package was incompatibly updated 03:54:50 Gregor: glad to see the Norse are beginning to show penguin tolerance. 2008 marks the dawn of a new era, where Norwegians recognize that they're all penguin-like creatures inside. 03:54:55 so the "issue" is that lambdabot's code sucks 03:56:58 maybe, as I establish clients from freelancing, I should start writing the most obtuse code possibly. 03:57:06 so that they must continue hiring me to maintain it. 03:57:36 kallisti: http://thc.org/root/phun/unmaintain.html 03:59:54 -!- Jafet has quit (Quit: Leaving.). 04:00:12 y a copy of a baby naming book and you'll never be at a loss for variable names. Fred is a wonderful name, and easy to type. If you're looking for easy-to-type variable names, try adsf or aoeu if you type with a DSK keyboard. 04:00:16 lol 04:00:20 s/^/buy/ 04:00:36 buyy 04:00:56 elliott: shush 04:01:07 Nisstyre: what's funny is that I've actually seen code that had some of these features. 04:01:10 perhaps not intentionally though 04:01:11 yes 04:01:15 me too :( 04:14:39 Overload the '!' operator for a class, but have the overload have nothing to do with inverting or negating. Make it return an integer. Then, in order to get a logical value for it, you must use '! !'. However, this inverts the logic, so [drum roll] you must use '! ! !'. Don't confuse the ! operator, which returns a boolean 0 or 1, with the ~ bitwise logical negation operator. 04:14:44 so good 04:17:26 " For pseudo-Esperanto pluraloj, add oj. You will be doing your part toward world peace." 04:19:17 Not that that's the most obfuscating thing there, I just like Esperanto >.> 04:22:15 * kallisti sighs. 04:24:37 Doesn't even come close to most obfuscating thing. 04:25:01 I suggest Japanising and romanising your names. 04:25:21 Sinicising would be worse, but a bit more difficult to pull. 04:25:55 int main(int arugusii, char** arugubui) { ... } 04:26:55 for(int ai=0; ...;...;) 04:27:16 Oooh. purintsueffu. 04:28:08 Nah, take more suggestion from that list. hůrinntue'hu 04:36:08 UPDATE 04:37:06 -!- DCliche has joined. 04:38:18 elliott, kallisti 04:40:33 -!- Klisz has quit (Ping timeout: 252 seconds). 04:49:05 -!- PiRSquared17 has quit (Remote host closed the connection). 04:53:19 -!- Jafet has joined. 04:54:51 http://i.imgur.com/c2liK.jpg 04:54:55 my eyes broken x.x 04:56:24 weird 05:07:50 -!- Jafet has quit (Ping timeout: 244 seconds). 05:10:27 -!- Jafet has joined. 05:10:41 Sgeo 05:11:43 hi kallisti 05:12:26 quintopia: hey. 05:12:28 how's it going? 05:12:30 what's the world? 05:12:35 is lfie? 05:12:48 what is life? 05:12:55 what is...this? 05:13:03 what is it? 05:15:49 what is blsshah? 05:19:47 lets get together and discuss it over beers 05:19:59 youre in like cumming or something right? 05:31:33 ..no 05:31:34 fuck you 05:31:39 I'm in jasper 05:31:40 asshole 05:33:03 i wasnt off by much >.> 05:33:37 one could go from cumming to jasper in under an hour yes? 05:34:18 probably 05:34:54 one can go a lot of places in an hour with a hellicopter, what's your point? 05:35:21 -!- zzo38 has joined. 05:35:51 my point is i didnt remember where you are, just that it was generally north of atlanta 05:36:05 I wish these cans of Busch Light 05:36:08 weren't in a Yuengling case 05:36:23 because it increased my expectations for how it would taste. 05:36:27 and i dont understand why confusing jasper and cumming provoked that repsonse 05:36:37 and it shouldn because huengling sucks too 05:36:43 what 05:36:45 you're mad. 05:36:50 okay we can never hang out ever. 05:37:09 FINE 05:37:09 yuengling light maybe. 05:37:11 but yuengling is like 05:37:23 THE MOST AMERICAN FUCKING BEER YOU CAN GO DIE YOU UNPATRIOTIC SCUM. 05:37:28 nah 05:37:34 fuck patriotism 05:37:43 yeah fuck it. 05:37:49 points: US sucks, Georgia sucks, beer sucks, particles suck, existence sucks, logical consistency sucks 05:37:56 of the commercial brewers, sam adams is far more patriotic 05:37:58 * elliott vanishes 05:38:02 quintopia: what 05:38:03 no 05:38:15 just because it has a patriotic name does not make it patriotic 05:38:18 yuengling is the FIRST 05:38:19 yes it does 05:38:21 the first american beer. 05:38:31 or well 05:38:32 probably not 05:38:34 but whatever. 05:38:46 excuse me 05:38:47 i made some 05:38:48 very good point 05:38:49 s 05:38:50 elliott: logical consistency is the best. 05:38:50 elliott sucks! 05:38:56 yes 05:39:01 elliott: no 05:39:02 but at least i am contradictory about it 05:39:10 elliott: ofuck logical consistency 05:39:15 I'm so confused 05:39:15 fuck you 05:39:16 about where I stand 05:39:19 on logical consistency 05:39:49 its such a good thing to have, we can honestly say no one should have it 05:40:08 towels are so good 05:40:12 have you ever a towel 05:40:14 a really nice towel 05:40:15 Why drink beer that's not-good anyways? 05:40:22 why drink 05:40:23 why exist 05:40:23 pikhq_: cheap. under 21. 05:40:27 why ovulate 05:40:31 why be 05:40:32 why elliott 05:40:33 pikhq_: also elliott makes very good points 05:40:35 all these questions and more 05:40:37 answered yesterday 05:40:40 why art thou romeo 05:40:50 im not romeo 05:41:00 pikhq_: please agree with me that yuengling is not disgusting and is actually pretty good. 05:41:02 -!- copumpkin has joined. 05:41:02 For the low low price of $25,000,000! (per question. no refunds. offer not valid in the state of California) 05:41:04 all roads lead to romeo 05:41:10 kallisti: I haven't had it yet, so I can't comment. 05:41:15 -!- elliott has set topic: how to be a toucan | The IOCCC is back on! http://www.ioccc.org | http://codu.org/logs/_esoteric/. 05:41:19 does anyone know how to be a toucan 05:41:23 help? 05:41:26 aid me in becoming toucan 05:41:26 elliott: I have studied. 05:41:31 * elliott googled for: toucan facts 05:41:34 kallisti: That said, it certainly doesn't fit in the category of "all-American pisswater". :P 05:41:36 * elliott googled for: toucan the night away 05:41:41 * elliott googled for: are toucans all-american pisswater 05:41:43 pikhq_: even better. 05:41:45 pikhq_: and yes it does 05:41:47 * elliott googled for: are toucans people 05:41:52 pikhq_: it's definitely an American beer 05:42:00 FIRST. FUCKING. AMERICAN. BREWERY. BITCHES. 05:42:04 and it is definnitely pisswater 05:42:06 kallisti: The noun "pisswater" is key there. 05:42:14 pikhq_: oh you mean light beer. well, no. it's not that. 05:42:33 * elliott piss water 05:42:44 well there is yuengling light. but don't ever drink that. 05:43:00 * elliott piss yuengling light 05:43:01 (though it is marginally better than most light beers) 05:43:07 elliott: Congrats, you can legally sell that as Bud Light. 05:43:22 * elliott googled for: bud light "infinity and beyond" quotes 05:43:30 * kallisti should make a modded starcraft map where you can upgrade zerglings to yuenglings. 05:44:37 I have a VHS/DVD combination recorder. It can copy VHS to DVD if it is not Macrovision, it can also copy unfinalized DVD to VHS. Do you know if there are any ways to bypass the Macrovision and other things? 05:44:39 * elliott googled for: do elephants celebrate christmas 05:45:01 * kallisti googled for: toucan meaning of life 05:45:04 * kallisti googled for: toucan secrets 05:45:10 * kallisti googled for: toucan mathematics 05:45:17 * kallisti googled for: toucan string theory 05:45:23 * elliott googled for: is the universe toucans 05:45:37 * elliott googled for: validate me 05:45:37 * kallisti googled for: why do toucans prove ftl 05:46:05 * elliott googled for: silent night true meaning 05:46:29 * elliott googled for: what happens if you shred people 05:47:07 * kallisti googled for: are toucans monoids in the category of endofunctors 05:47:38 * elliott googled for: toucan 05:48:13 * kallisti googled for: how to win debate with toucan 05:48:14 * elliott googled for: toucan cheese 05:48:19 * elliott googled for: toucan byproducts 05:48:22 * kallisti googled for: human cheese 05:48:31 * kallisti googled for: does breast milk coagulate 05:48:46 * kallisti googled for: do toucans eat their toucan cheese? 05:48:51 you can end your search elliott, knowing that it is possible 05:48:56 zzo38: There's several devices that strip the Macrovision off an analog signal. 05:49:11 if the macaw saw what the leopard spotted, then the toucan can and you can too. 05:49:21 You probably couldn't do that to the single device there without a lot of work, though. 05:49:51 is macrovision that noise-adding thing? 05:50:24 quintopia: No, it's the "add shit to the VBI so that automatic gain control breaks" thing. 05:51:09 -!- lambdabot has joined. 05:52:14 lambdabot: hi 05:52:36 @pl (\f g x y -> f x (g y)) 05:52:36 flip . ((.) .) 05:52:47 @pl (\x y -> f x (g y)) 05:52:48 (. g) . f 05:52:53 .. 05:53:07 @pl (\y x -> f x (g y)) 05:53:07 flip f . g 05:53:33 I wonder if it has my ROBUSTNESS patch, god I hope not. 05:53:59 * kallisti googled for: toucan fixed-point combinator 05:54:18 * kallisti googled for: robust toucan zygotomorphism 05:54:25 apparently some dvd players have loophole menus where you can turn off macrovision 05:54:33 probably not that converter tho 05:54:43 Oh no, he's applying it. 05:54:47 RIP lambdabot 2011-in a few minutes 05:54:51 *not 2011 05:55:15 rip elliott 05:55:34 cale will assimilate you into his leet prelude 05:55:36 for failure 05:56:04 elliott: oh dude if I rewrite lambdabot in perl I can make CAKESKELL 05:57:15 Cake.Data.List, yessss 05:57:59 length = genericLength; length' = Data.List.length 05:58:10 no thanks 05:58:16 length xs :: Double 05:58:33 ...so? 05:58:50 length xs :: Complex Rational 05:58:55 fromIntegral . length $ xs :: Double 05:58:58 these are not things likely to not be mistakes. 05:59:05 kallisti: yes, with an explicit conversion. 05:59:14 integers /are/ a subset of reals and complexes 05:59:16 haskell "implicit conversions go away" 05:59:21 kallisti: yes, but Double isn't Real 05:59:36 and if you're working with integral data, using Real is the Wrong Thing to do 05:59:38 integers /are/ a subset of floating point numbers, and complexes 05:59:46 not true 05:59:50 * elliott would accept (Integral b) => [a] -> b 06:00:01 :t genericLength 06:00:03 forall b i. (Num i) => [b] -> i 06:00:16 elliott: that's reasonable 06:01:38 I just haven't really encountered a situation where I might have accidentally used a Double when I did not mean to. 06:02:37 perhaps if you forgot that (/) doesn't perform integer division. 06:03:01 there's a somewhat likely scenario 06:06:32 but intuitively, from the perspective of basic algebra, integer division isn't even a thing and typically you can treat both fractional and integral numbers as being generic "numbers." it would make sense if Haskell followed this looseness, and I think the only people who might get thrown off are "experienced" programmers expecting arithmetic operators to behave as they do in other languages. 06:07:31 but consider: sum ls / length ls 06:07:54 a fledgling Haskell programmer might try to compile this and be surprised to find that they get an obtuse typeclass error. 06:08:56 Gregor: Ping 06:09:06 kallisti: A fledgling Haskell programmer will make many mistakes if they just write random things and expect it to wrok 06:09:07 work 06:10:57 that's not really a defense though. You're simply saying they will make mistakes. Some of these mistakes could still easily be blunders of language design. Most non-Haskell programmers would not expect an error from that, and neither would someone familiar with algebra, statistics, etc 06:11:45 > genericLength [a,b,c,d,e] 06:11:48 5 06:11:57 > genericLength [a,b,c,d,e] :: Expr 06:11:58 1 + (1 + (1 + (1 + (1 + 0)))) 06:12:00 kallisti: haskell is not algebra/statistics/etc. 06:12:11 elliott: sure. 06:12:14 if haskell tried to interpret all the ambiguous mathematical notation out there, it'd be a sloppy, useless piece of crap 06:12:30 explicit conversions is one of haskell's defining features 06:12:32 right, I simply think basic numbers are something it could concede on. 06:12:52 And computers require different treatment of "numbers" than everyone is really used to. 06:13:00 kallisti: how often do you miss a fromIntegral 06:13:05 if it's often you're doing something wrong 06:13:12 elliott: not often. 06:13:14 it happens though. 06:13:30 type errors happen, that's why we have a compiler that catches them before the program runs 06:13:54 I agree certainly. 06:13:54 I agree with elliott that length :: (Integral b) => [a] -> b -- is preferable. 06:14:02 it's safer, yes. 06:14:11 and much better than Int 06:14:20 and still allows conversion to Num 06:14:26 i would just go for Integer to be quite honest if not for one thing 06:14:37 namely, with Integral you can define a generic peano naturals type 06:14:40 and use it to do lazy length comparison 06:14:50 so that length [0..] > (5 :: Nat) is true 06:14:53 very useful 06:15:01 and avoids wasting work by traversing the whole list even for normal cases 06:15:16 Also, using floats where you mean integers *will screw up*. 06:15:45 how so? 06:15:47 I'm a bit iffy about them even being Nums, TBH.\ 06:16:24 isn't genericLength more like using an integer as a float? (in the more specific case we're discussing) 06:16:27 you can give Float a reasonable Ord by breaking the IEEE spec for NaNs so that's ok 06:16:32 Num is 06:16:34 ehhhhhhhhhh 06:16:36 well Num just has to go 06:16:40 so it's not even worth thinking about 06:16:43 but I wouldn't object to Nums being Fields 06:16:45 erm 06:16:47 but I wouldn't object to floats being Fields 06:16:48 True, Num is a pretty bad typeclass. 06:16:52 because 06:17:01 there's no point creating a whole mirrored hierarchy of "almost s" 06:17:10 pikhq_: it provides a number of useful hacks, though a better typeclass hierarchy could probably do the same. 06:17:15 yes, you have to be careful with the properties when using floats 06:17:30 but it's more like your whole program is taken as an approximation of a computation on the actual rationals/reals 06:17:37 rather than the individual values being approximate 06:17:38 IMO 06:17:46 so it's justified to make them instances: also because it's really convenient 06:18:03 elliott: no I think we should remove Float and Double and only use CReal 06:18:04 the other part of me objects to htis though because I want our interfaces to come bundled with proofs that their properties hold 06:18:08 and you can't do that for floats 06:18:22 at least not without a custom equivalence relation that you specify to be "close enough" for floats 06:18:28 (with formal bounds on close enough for each operation, but still) 06:18:35 *this 06:18:48 elliott: I... think it would be okay to concede on that for floats. because they're very useful. 06:19:02 I take it you don't understand what I said 06:19:02 because 06:19:05 for like, programs that do things rather than programs that serve as mathematical proofs. 06:19:05 it's not about conceding 06:19:21 a Monad should contain the proofs that it follows the monad laws, that's simple as 06:19:28 something that doesn't follow those laws should not be allowed to be a Monad 06:19:31 this is not controversial at all 06:19:36 but you can't "just compromise" 06:19:40 It's about how you define the typeclass. 06:19:44 because the whole benefit of proof carrying is that 06:19:58 - you can optimise aggressively based on the properties with /NO RISK/ of anything "going wrong" at runtime, it's 100% safe and type-safe 06:19:58 elliott: floats aren't monads though. how is this relevant to what I was saying. 06:20:17 - you can use these properties in larger proofs of correctness - this is NOT a mathematical thing, this is about real programs 06:20:29 - etc. 06:20:32 kallisti: because Field is just like Monad 06:20:45 it's an interface 06:20:47 a structure 06:20:51 elliott: what is a viable replacement for floats? 06:20:53 a formally-defined set of operations and rules 06:20:59 kallisti: did i say there was one 06:21:00 did i say 06:21:02 floats aren't useful 06:21:02 elliott: no 06:21:03 did i say 06:21:04 i don't want floats 06:21:05 no. 06:21:09 elliott: it was just a question. I was curious. 06:21:22 hey kallisti what is a viable replacement for oxygen 06:21:29 considering all the boundless anti-oxygen sentiment you've been expressing lately 06:21:49 elliott: so would floats still be treated as "numbers" in this typeclass hierarchy? 06:21:57 Float + Float = Float 06:22:09 i think you're mixing up a bunch of things i said. 06:22:15 what typeclass hierarchy, what are you talking about 06:22:20 feel free to clarify. 06:22:23 you appear to read my messages non-linearly 06:22:36 and mix ones after i explicitly change the track of what i'm saying to ones previously 06:23:23 okay. 06:23:26 kallisti: i can't clarify something that i have no idea the basis of, so... 06:23:55 elliott: I'm assuming Field etc which have formal laws of their correctness would not take over the syntax of simple arithmetic? 06:24:02 because Float cannot fit in those 06:24:04 because 06:24:04 uh 06:24:06 it does not obey those laws 06:24:19 but it's still nice to say "lol 0.5 + x" 06:24:23 im just giving up on this because i explicitly said i was of two contradictory minds for different reasons and needed to reconcile them 06:24:24 and 06:24:28 also your questions are stupid 06:24:34 I know there are devices that strip out Macrovision, such as time base correctors. I do not have a time base corrector but I have a analog->digital->analog converter that among other things, removes macrovision (it also removes captions and everything other than the picture). But I don't know if that can be used in this VHS/DVD recorder. 06:24:54 elliott: perhaps I missed that part. 06:25:05 the other part of me objects to htis though because I want our interfaces to come bundled with proofs that their properties hold 06:25:06 and you can't do that for floats 06:25:13 things i say don't make sense if you skip lines 06:28:05 elliott: so how about: keep "Nums as Fields" with Floats being a Field and obeying properties as approximate calculations of rationals/reals, and then have a seperate typeclass (or a convenient way to remove non-precise instances such as floats) for formally correct computations? 06:28:29 -!- lambdabot has quit (Remote host closed the connection). 06:28:31 my response would be a long list of "you do not understand X, Y and Z", so I doubt you want to hear it. 06:28:40 no these are good 06:28:44 those are usually missing from your responses. 06:30:00 you don't understand why dependently-typed languages matter. you don't understand why proof-carrying matters. you don't understand why formal properties matter. you don't understand why one cannot simply make proofs "optional" addons. you don't understand how proof carrying allows for very advanced and aggressive optimisations. you don't understand how proof carrying ensures very high levels of securit 06:30:00 y statically at compile time. 06:30:05 i think that's enough for now 06:31:31 if you really want, you can ask in greater detail about why i think these things. you will get better results for the more specific statements 06:32:01 elliott: I was going on the premise that proof-carrying matters when attempting to devise a solution. 06:32:06 oh, I forgot: you don't understand how proof carrying allows for distributed computing etc. without trust in any direction 06:33:02 elliott: what I'm saying is that you're basically talking about two different typeclasses. 06:33:20 one is proof-carrying and one isn't 06:33:31 the solution to having a problem with an abstraction 06:33:39 is not to "rip out" (not a well-defined operation) the problematic parts 06:33:47 and call this half thing a "new abstraction" 06:33:58 multiplying complexity does not solve anything 06:34:24 -!- DCliche has quit (Quit: You are now graced with my absence.). 06:34:51 at least not without a custom equivalence relation that you specify to be "close enough" for floats 06:34:51 (with formal bounds on close enough for each operation, but still) 06:34:51 *this 06:34:55 this is one possible option, for instance 06:35:09 and you recover the original behaviour by requiring the relation to be equality 06:35:18 but it feels inelegant 06:35:30 albeit giving the best possible properties for "precise" types and floats alike 06:37:28 kallisti: does this help you understand my position? 06:37:42 elliott: yes 06:38:15 elliott: I think you would need to incorporate laws into the typeclass system rather than having them as friendship pacts, so that a) typeclasses could specify some rules to be optional b) optimizations could be specified in the language itself for instances that obey a particular set of rules c) you can opt out of rules that your instance doesn't obey d) you can define functions that allow you to satisfy those rules (in an 06:38:25 elliott: to have it all as one typeclass I mean.. 06:41:04 friendship pacts? 06:41:49 "I think you would need to incorporate laws into the typeclass system rather than having them as friendship pacts" 06:41:49 what I am saying is not even related to typeclasses; my system has no typeclasses. also, it's not "incorporated" into the system, and they're *certainly* not based on honour, that would defeat the whole point. the context is a dependently-typed language, and if you don't know how they can do logic and proofs then you understand them even less than I claimed 06:41:49 "a) typeclasses could specify some rules to be optional" 06:41:49 no. this does not aid a solution, at all. it makes matters a billion times worse by throwing away *every* *single* *benefit*, adding additional hassle, and giving incredibly suboptimal results even in the float case 06:41:52 "b) optimizations could be specified in the language itself for instances that obey a particular set of rules" 06:41:55 completely irrelevant to the issue at hand. yes, optimisation is but one benefit of proof-carrying, but it... is simply not relevant here, I only mentioned it as one of the reasons proof-carrying is incredibly important. 06:41:58 "c) you can opt out of rules that your instance doesn't obey" 06:42:00 hahahahahahaahahaha 06:42:02 "d) you can define functions that allow you to satisfy those rules (in an" 06:42:04 this was chopped off. 06:43:09 elliott: a like how you switched the context to dependently typed languages without telling anyone. 06:43:13 s/a/I/ 06:43:21 yeah 06:43:22 see 06:43:28 "proof-carrying" 06:43:31 was a pretty big hint 06:43:35 if you actually knew what you were talking about 06:43:41 by hint, I mean literal outright statement 06:46:04 I wasn't aware that dependently typed languages are the only context in which proof-carrying can be talked about. 06:47:12 the context was functional languages + formal proofs soooo... and anyway, it's not really relevant that a dependent language is the setting; with a term like "proof-carrying", the idea that the "proofs" might just be gentleman's agreements is not even a possibility 06:47:14 -!- hagb4rd has joined. 06:47:39 elliott: right. 06:49:31 -!- hagb4rd2 has quit (Ping timeout: 240 seconds). 07:08:00 -!- elliott has quit (Remote host closed the connection). 07:21:18 -!- sebbu2 has joined. 07:21:18 -!- sebbu2 has quit (Changing host). 07:21:18 -!- sebbu2 has joined. 07:23:46 -!- sebbu has quit (Ping timeout: 258 seconds). 07:30:57 -!- azaq23 has quit (Read error: Connection reset by peer). 07:31:12 Make all of your leaf classes final. After all, you're done with the project - certainly no one else could possibly improve on your work by extending your classes. And it might even be a security flaw - after all, isn't java.lang.String final for just this reason? If other coders in your project complain, tell them about the execution speed improvement you're getting. 08:13:10 -!- Vorpal has joined. 08:18:05 -!- Slereah_ has joined. 08:18:35 -!- monqy has quit (Quit: hello). 08:19:44 -!- Slereah has quit (Ping timeout: 258 seconds). 08:20:42 -!- zzo38 has quit (Remote host closed the connection). 08:21:39 -!- MSleep has quit (Ping timeout: 248 seconds). 08:33:19 hmmm smalltalk could potentially be abused in polyglot languages. 08:33:24 as " is used for inline comments 08:34:43 so something like "\" would comment out most languages but leave smalltalk available 08:34:56 then a subsequent " would begin a smalltalk comment and end the other-language string 08:35:12 assuming multi-line strings are legal. 08:35:15 kallisti, update 09:42:18 -!- ais523 has joined. 10:02:54 -!- ais523 has quit (Ping timeout: 252 seconds). 10:17:50 -!- MSleep has joined. 10:25:05 -!- GreaseMonkey has quit (Quit: The Other Game). 10:38:35 -!- ais523 has joined. 10:50:05 https://bugs.php.net/bug.php?id=32100 (found via proggit); wow at the first comment 10:50:41 I mean, the whole page is cringeworthy, but the first comment is particularly bad 11:01:19 someone on Reddit described the attitude as "Pokémon exception handling"; I hadn't seen that before, but it's a good name for it 11:06:51 > foldr (*) 0 [1..10] :: Expr 11:07:07 >_> 11:07:11 oh 11:07:32 > -> 11:07:37 * 11:07:39 > _> 11:07:41 there we go 11:07:53 that actually times out on my bot running mueval 11:07:59 foldl does not 11:24:21 ais523: maybe they don't want to implement finally because they don't know how. :3 11:25:05 kallisti: that was mentioned in the thread 11:25:10 on Reddit, I thikn 11:27:46 the languages employers turn to when writing the software that makes the world turn are C/C++, Java, and PHP, in that order. C# is a close 4th, then everything else is a niche. 11:27:50 I think it's actually Java up top. 11:39:40 -!- ais523 has quit (Ping timeout: 245 seconds). 11:43:45 -!- ais523 has joined. 11:44:35 hmm, what do you call the things made with Flash that are basically interactive animated vector images, and have nothing to do with (compressed) raster video? 11:54:36 -!- derdon has joined. 12:05:44 -!- ais523 has quit (Remote host closed the connection). 12:06:45 -!- ais523 has joined. 12:06:52 -!- ais523 has quit (Changing host). 12:06:52 -!- ais523 has joined. 12:17:04 ais523: ....flash? 12:17:19 kallisti: not really specific enough 12:17:24 perhaps "flash animation" would be the right name 12:18:54 probably. 12:20:23 -!- derdon has quit (Remote host closed the connection). 12:23:39 -!- MDude has joined. 12:26:27 -!- MSleep has quit (Ping timeout: 248 seconds). 12:33:28 I'm still stuck on the 10th level of the MESH - Falling Hero demo 12:33:30 :| 12:33:36 right side's solved, left side I'm stuck on 12:37:18 where's zzo38/myndzi when you need em 12:50:11 *-not 12:51:16 bsmntbombdood: and in case you aren't familiar with it the N -> D conversion is done with the subset construction (which can be extended for alternating -> deterministic as well, see my master's thesis for instance) 12:52:10 basically let your NFA A have state set S, then what you do is you take the set of subsets 2^S as the states of your new DFA A' and have as your new initial state the set of all initial states of A. 12:52:43 the transitions are given by having the transition U -> U' (where U, U' in 2^S) with label a if U' is exactly the set of states A can reach from the states U with label a 12:53:18 obviously the final states should then be just the sets containing at least one final state; it should be immediate that A' then accepts the same language as A. 12:54:19 you then need to minimize the DFA, for that you can just first assume that final states are separated from nonfinal states (exactly finals accept the empty word), and then keep separating states into smaller and smaller sets (making the eq relation of states finer) until all labels from equivalent states lead to equivalent states 12:54:44 oh and ofc drop unreachable states and states from which final states cannot be reached except a single sink state maybe, depending on your formalism 12:55:21 the conversion from regexps to NFA is very simple, OR is trivial, concatenation is just connecting final states to initial states, and kleene star is done by doing the same but to your own initial states 12:56:11 once you have two minimal DFA, just start from the initial states of each, and you'll find correspondence of states, if one exists, by just following the labels. we know that minimal DFA are unique up to this sort of isomorphism, so you get your answer from that. 12:56:44 apparently you left the channel before i had time to answer, but that's not really my problem 12:59:15 -!- oerjan has joined. 13:02:56 i didn't really explain minimization very clearly, but you can think up the details with little difficulty 13:14:57 -!- Patashu has quit (Quit: MSN: Patashu@hotmail.com , Gmail: Patashu0@gmail.com , AIM: Patashu0 , YIM: patashu2 , Skype: patashu0 .). 13:29:05 brilliant: Oracle and Google are arguing over whether the word "runtime" in a patent should be given its ordinary-language meaning 13:29:34 Oracle claim it should be, and that that meaning is "during execution of the virtual machine"; Google claim it needs an explicit definition as "during execution of the virtual machine instructions" 13:29:44 I'm wondering when the judge is going to yell at them that it obviously doesn't matter 13:30:08 because they both agree that the term has the same meaning, they just disagree over whether the court needs to specifically define it as having that meaning 13:30:12 which is, why does that matter? 13:31:16 -!- boily has joined. 13:36:46 * kallisti is immune to sleep deprivation now. 13:37:22 he said, collapsing. 13:38:05 i have until 11. to finish a paper i made last summer and i just realized it's like the sloppiestly written paper ever 13:38:07 and i dont understand why confusing jasper and cumming provoked that repsonse <-- the closer the places, the more violent response to confusing them, duh 13:38:55 gonna be a long saturday :))) 13:42:43 aid me in becoming toucan <-- i'm sorry i'm over quota on fowl puns 13:49:30 oklopol: write a perl script to fix it up. you'll be fine. 13:57:43 -!- ais523_ has joined. 14:03:51 i refereed a paper and broke it. 14:27:00 bleh, fire alarm 14:27:06 (just got back from a really obvious fire drill) 14:27:20 (only thing that'd have made it more obvious would have been being announced in advance 14:27:23 *) 14:28:49 "WE WILL BE SETTING FIRE TO THE CANTEEN IN THIRTY MINUTES; EVERYONE PLEASE BE PREPARED FOR THE FIRE DRILL" 14:36:04 -!- Madoka-Kaname has quit (Ping timeout: 248 seconds). 14:42:37 -!- copumpkin has quit (Quit: Computer has gone to sleep.). 14:46:27 oerjan: it reminds me of an unexpected early-morning fire alarm at my secondary school 14:46:37 apparently, it's because the detectors detected "HEAT" in the kitchens 14:46:48 I thought, the kitchens are somewhere heat would be quite likely, surely? 14:47:12 you'd think 15:13:03 perhaps it meant more heat than usual 15:17:39 -!- copumpkin has joined. 15:22:27 elliott best not have left me any lambdabot-messages. 15:22:37 But then, lambdabot doesn't seem to be here. 15:22:42 SO I GUESS I'LL NEVER KNOW 15:30:28 where is elliott, anyway? 15:42:43 -!- sebbu2 has changed nick to sebbu. 15:45:56 -!- MDude has quit (Ping timeout: 248 seconds). 15:46:13 -!- MSleep has joined. 15:52:23 -!- MDude has joined. 15:52:44 -!- oerjan has quit (Quit: leaving). 15:55:32 -!- MSleep has quit (Ping timeout: 248 seconds). 15:59:31 ais523: Better question: Where is lambdabot? 16:00:10 ELLIOTT IS LAMBDABOT 16:07:29 !perl print int(rand(2)) 16:07:32 0 16:07:41 man, I love pseudo-random decision making 16:08:18 -!- ais523_ has quit (Quit: Page closed). 16:11:56 kallisti: Man, I base most of my life on pseudo-random decision making. 16:19:37 i usually just ask my dick and i then rarely even bother to listen 16:31:01 All power to penis 16:42:21 Gregor: it's a great excuse to slack off. 16:42:29 especially when you haven't been slacking off very much in the past few weeks. 16:44:32 * kallisti wishes that SQL injection were still a thing you could do. 16:44:50 maybe one day I'll find that horribly insecure site and exploit the shit out of it. 16:54:09 http://en.wikipedia.org/wiki/National_Defense_Authorization_Act_for_Fiscal_Year_2012 16:54:16 pikhq_: have you seen this shit 17:20:02 why the fuck did I buy a red bull 17:20:14 oh, hm, it tastes good. maybe it tastes better when you don't sleep much. 17:20:40 the only problem is that it's way overpriced for the caffeine content. 17:20:57 though, recently I've started /not/ binge drinking caffeine 17:21:06 and instead drinking small amounts throughout the day 17:21:09 and it works better. 17:25:37 hi hi hi hi 17:26:06 this Red Bull can mysteriously says "Rexham" o nit 17:26:18 I think it scod for HEXHAM 17:26:41 KING HEXHAM 17:26:50 REXIMUS MAXIREX 17:26:57 * kallisti latin 17:41:21 -!- monqy has joined. 17:44:02 monqy: bye 17:44:08 hi 17:49:18 -!- elliott has joined. 17:51:29 elliott: hihiii 17:51:53 are you just feeling silly or does caffeine do this to you? 17:52:44 no slep deprivation 17:53:03 -!- blodeuwedd has joined. 17:54:21 -!- blodeuwedd has quit (Client Quit). 18:16:18 `log coincidence 18:16:53 2010-05-03.txt:18:06:13: Maybe they just repeat naturally, or maybe it's just a coincidence. 18:18:57 -!- Klisz has joined. 18:30:47 okokokokokokokoko 18:30:49 okokokokokokokoko 18:30:51 okokokoko 18:30:55 okokokokokokokokokokokokokoko 18:31:31 11:44:35: hmm, what do you call the things made with Flash that are basically interactive animated vector images, and have nothing to do with (compressed) raster video? 18:31:32 ais523: animation 18:31:37 and/or game 18:31:52 I was wondering if there was a general term 18:32:12 flash :P 18:32:15 as in, "a flash" 18:32:26 youtube has a flash player, but the individual videos aren't flashes 18:33:30 hmm, I wonder if this has anything todo with Shockwave + whatever the other thing that was merged to make Flash was 18:34:09 ais523: shockwave still exists, if you can believe it 18:34:16 elliott: I can 18:34:20 ais523: as in, "is still updated" 18:34:22 I can also believe that Quicktime still exists, but don't know for sure 18:34:32 also Realplayer 18:34:41 (as web browser plugins, I mean) 18:34:59 ais523: quicktime is obviously still updated since it's still a core part of OS X :P 18:35:02 (as a browser plugin too) 18:35:23 oh dear, realplayer is on version 15 18:36:07 * elliott can't figure out the difference between shockwave and flash 18:36:12 it doesn't help that flash used to be called shockwave flash 18:36:18 quick, Mozilla, iterate version numbers faster to beat it! 18:36:22 elliott: still is in my plugins list 18:36:48 quick, Mozilla, iterate version numbers faster to beat it! 18:36:57 but that wouldn't be mathematically sound! 18:37:09 firefox's version-number is the fastest-growing possible function 18:38:32 ais523: I think one of the best things about Adobeacquiring Macromedia (wow, that happened in 2005) 18:38:41 is that Adobe ended up with like 18:38:43 five duplicate products 18:38:57 that they had to justify 18:38:58 as recently as 2005? 18:39:01 X-D 18:39:03 elliott: why didn't they just merge them? 18:39:16 ais523: how do you merge two codebases? 18:39:26 merge features, rather than code 18:39:32 unless the codebases happen to be similar, which is unlikely 18:39:49 well, OK 18:39:59 basically you're asking "why didn't they discontinue one" :) 18:40:27 yep 18:40:29 What's the difference between the Flash and Shockwave Players? 18:40:29 Flash and Shockwave Players are both free Web Players from Adobe. Together, they bring you the best rich media content on the Internet. Each has a distinct purpose. Flash Player delivers fast loading front-end Web applications, high-impact Web site user interaction, interactive online advertising, and short to medium form animation. Shockwave Player displays destination Web content such as interactive multimedia product demos and training, e-merch 18:40:29 andising applications, and rich-media multiuser games. When you download Shockwave Player, it automatically includes Adobe Flash Player. 18:40:40 well, "why didn't they discontinue one and give a migration path from it to other versions" 18:40:59 so, umm, Flash is for animation and interactivity, Shockwave is for applications, or something 18:41:21 http://www.shockwave.com/home.jsp ;; which is to say, games 19:13:24 -!- MDude has quit (Ping timeout: 248 seconds). 19:24:26 ais523: christmas spam :D 19:24:43 ais523: have you contacted graue yet? it's becoming impossible to follow recentchangs 19:24:44 es 19:24:45 it's been going on for a while 19:24:46 and no, I haven't 19:24:54 :( 19:25:00 I'm something like three days behind with my work as is 19:25:17 oh, looks like smjg has 19:25:24 and I don't like to bother him with something that's under control, in that the spam can be cleaned up very quickly 19:26:20 it's not under control if it massively disrupts the wiki... 19:26:51 how is it disrupting anything but Recent Changes? 19:27:26 that's a rather major thing to disrupt; furthermore, it's disrupting everyone's time because we have to constantly revert [[Excela]] 19:27:44 I'm hardly the first person to complain that it's severe 19:39:54 OK, emailed Graue 19:40:51 ais523: in a week graue will return, post a message saying "sorry for not reading email, I've been busy etc. etc. etc.", op Timwi, and leave 19:40:57 * elliott calls it now 19:41:05 elliott: more admins wouldn't help, I made that pretty clear 19:41:14 ais523: stop it it 19:41:16 hurst my joke 19:41:18 youre killing it :( 19:47:16 ais523: y william hearst 20:18:57 -!- ais523 has quit (Remote host closed the connection). 20:29:34 -!- oerjan has joined. 20:31:06 `addquote man, I love pseudo-random decision making kallisti: Man, I base most of my life on pseudo-random decision making. i usually just ask my dick and i then rarely even bother to listen 20:31:08 745) man, I love pseudo-random decision making kallisti: Man, I base most of my life on pseudo-random decision making. i usually just ask my dick and i then rarely even bother to listen 20:32:24 oklopol :D 20:34:51 `quote 20:34:52 `quote 20:34:52 `quote 20:34:53 `quote 20:34:53 `quote 20:34:57 80) Warrigal is the Harlem Globe Frotter 20:34:57 360) [...] OOPS.. my cockfile got destroyed 20:34:59 172) Why do you use random acronyms you know we don't know the expansions of? alise: TLAAW 20:35:11 539) What does "life" actually mean, anyway; it seems to mean "this thing that's infinitely greater than all my actual hobbies that I do all the time because I rule" 20:35:12 264) (the former is a very deep theorem, i'd have had to read the whole book to understand it, so i didn't.) 20:35:47 hm was that about the SL = L theorem 20:35:52 most of those suck :( 20:35:56 `log (the former is a very deep theorem, i'd have had to read the whole book to understand it, so i didn't.) 20:36:17 No output. 20:37:12 o_O 20:37:14 i don't understand 80 20:37:23 you're too young 20:37:28 (it's not funny though) 20:37:44 80, 360 and 172 are pretty bad 20:37:52 `log [t]hhe former is a very deep theorem 20:37:57 No output. 20:37:58 thhe former 20:38:03 oopd 20:38:08 `log [t]he former is a very deep theorem 20:38:14 2011-02-05.txt:02:09:38: `addquote (the former is a very deep theorem, i'd have had to read the whole book to understand it, so i didn't.) 20:38:18 my speling, were diddit goo 20:38:21 `logurl 2011-02-05.txt:02:09:38: 20:38:23 http://codu.org/logs/log/_esoteric/2011-02-05 20:38:41 lol, not that day or the previous 20:38:52 perhaps i said it in privmsg? 20:39:06 oh probably 20:39:14 in that case i won't have the logs of it 20:39:18 do you log; if not rip 20:39:20 me neither 20:39:20 the line's 20:39:21 history 20:39:21 ok 20:39:39 `log [t]he former is a very deep theorem 20:39:44 2011-02-05.txt:02:09:38: `addquote (the former is a very deep theorem, i'd have had to read the whole book to understand it, so i didn't.) 20:39:57 oerjan: pastelogs, dude 20:40:03 ah 20:40:08 `pastelogs [t]he former is a very deep theorem 20:40:15 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.21522 20:40:25 oh 20:40:29 it was one of my logreading additions 20:40:35 that i failed to mark appropriately :( 20:40:51 oerjan: wow, we only had 291 quotes in feburary 20:40:51 `qc 20:40:53 745 quotes 20:40:57 *february 20:41:05 `logurl 2010-06-07.txt 20:41:07 http://codu.org/logs/log/_esoteric/2010-06-07 20:41:17 12:59:39: I kind of feel like devising an abstract algebra. 20:41:17 12:59:59: they've all been invented already. probably. well, the simple ones. 20:41:17 13:00:19: * oerjan was reading about medial magmas/groupoids recently 20:41:17 13:01:13: one operator (call it *), fundamental equation: (a*b)*(c*d) = (a*c)*(b*d) 20:41:17 13:02:43: hmm, post on rgrn 20:41:19 13:02:45: surprisingly, there are interesting consequences. if * is surjective, then it's essentially a kind of linear function, and all linear functions of two variables have this property. 20:41:22 13:02:49: umm, ali 20:41:24 13:03:00: two newsgroups which probably have more of an overlap than they ought to 20:41:26 13:03:11: but a post on rgrn is not surprising, and a post on ali is 20:41:28 13:03:19: it was asking about how to implement the factory and singleton patterns in INTERCAL 20:41:30 13:03:24: (the former is a very deep theorem, i'd have had to read the whole book to understand it, so i didn't.) 20:41:45 oh that one 20:41:47 oerjan: omfg! 20:42:01 oerjan: that log, by complete coincidence, contains one of the /other/ programming games 20:42:07 that I couldn't for the life of me find 20:42:21 the SL = L theorem was just another one where i didn't bother to read the whole book/article :P 20:42:30 00:19:32: Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. 20:42:30 00:19:37: Some people, in an effort to sound intelligent, quote other people. Now they look retarded. 20:42:30 00:19:47: Some people, when confronted with a problem, think "I'm going to grab a sandwich." Now they have two problems because the sandwich is poisoned. 20:42:30 00:20:01: Some people, when, now they have, a problem. 20:42:53 in fact i think that may have referred to earlier articles about the graph techniques they used 20:43:33 elliott: I can't say I've created two problems from a regex for more than 10 minutes at most. 20:44:01 that one is a famous not-Zawinski quote. 20:44:07 yes. 20:45:13 the big question is whether zzo38 used regular expressions to generate those quips 20:47:07 oerjan: you are assuming incorrectly that those aren't all direct quotes :P 20:47:23 observe 20:47:24 http://www.google.co.uk/search?gcx=c&sourceid=chrome&ie=UTF-8&q=%22Some+people%2C+when+confronted+with+a+problem%2C+think+%22I'm+going+to+grab+a+sandwich.%22+Now+they+have+two+problems+because+the+sandwich+is+poisoned.%22 20:47:36 http://www.google.co.uk/search?gcx=c&sourceid=chrome&ie=UTF-8&q=%22Some+people%2C+in+an+effort+to+sound+intelligent%2C+quote+other+people.+Now+they+look+retarded.%22 20:47:50 http://www.google.co.uk/search?gcx=c&sourceid=chrome&ie=UTF-8&q=%22Some+people%2C+when%2C+now+they+have%2C+a+problem.%22 ;; ok this one wasn't a direct quote 20:48:34 well, i have to assume if i'm going to make any asses 20:49:48 `addquote well, i have to assume if i'm going to make any asses 20:49:51 746) well, i have to assume if i'm going to make any asses 20:50:23 * oerjan puts his fishing hook away 20:52:32 746 is a lot of quotes 20:53:12 maybe we're approaching quote singularity 20:53:33 the moment where all quotes are bad 20:53:34 `quote 20:53:35 `quote 20:53:35 `quote 20:53:35 `quote 20:53:36 `quote 20:53:44 721) but yeah the caliphates expanded their empire by conquering people and then forcing them to either convert to Islam or die. [...] i thought it was sort of, convert to islam or pay extra taxes, but i guess it varied a lot. 20:53:45 539) What does "life" actually mean, anyway; it seems to mean "this thing that's infinitely greater than all my actual hobbies that I do all the time because I rule" 20:53:51 NOT THAT ONE AGAIN 20:53:56 83) My mascot is a tree of broccoli. 20:53:56 682) it's not a list of /all/ interesting esolangs, btw; otherwise you can take the first command from the first esolang, the second from the second, the third from the third, etc, then add 1 to all of them and you get a new interesting esolang diagonal principle… 20:53:56 626) Let us discuss the correct procedure for converting LP -> FLAC The correct procedure is: you put the LP into a flatbed scanner, scan it as a Windows .bmp file, and then rename that file to .flac. 20:54:21 -!- itidus20 has joined. 20:54:24 hm 20:54:27 `delquote 83 20:54:30 ​*poof* My mascot is a tree of broccoli. 20:54:42 i actually thought that one was cute 20:54:53 a girl? 20:55:06 obviously it's just Gregor crossdressing 20:55:17 aha 20:55:21 -!- derrik has joined. 20:55:26 oerjan: well propose a worse one that's there :P 20:56:07 now you're just being mean :P 20:56:50 ok fine 20:56:51 `revert 20:56:52 Done. 20:57:24 oerjan: er i just had a really bad idea 20:57:38 yay 20:57:49 noooo 20:57:51 will it cause the end of life as we know it? 20:57:52 3 cheers 20:58:04 oerjan: not quite. but my hard drive might fill up 20:58:13 ah. even worse, then. 20:58:20 even worse! 20:58:21 yes 20:58:23 yay 20:58:51 oerjan: what if i wrote a bot to download every url mentioned in here as it is linked. 20:58:55 and then 20:58:59 hook it up to the logs 20:59:00 so in the future 20:59:14 you can see the links.... as they were... then........... 20:59:16 this is a terrible idea wow 20:59:29 it basically uses up gigs of disk space for really tiny amounts of nostalgia in the far future 20:59:51 elliott: you mean when url format is superceded? 21:00:04 -!- derrik has quit (Client Quit). 21:00:07 yes that's totally what i meant 21:00:34 like when dns is replaced with a non-deterministic ai system where you just make natural language enquiries into the address bar :-? 21:00:55 i ... 21:00:58 i'll stop 21:01:01 itidus20: that's called google 21:01:06 no I think he means a esoWayBack 21:01:07 ah 21:01:12 at least many people use it that way 21:01:30 i wonder if google is used more than dns these days 21:01:30 as in 21:01:31 not internally 21:01:33 but by people 21:01:35 elliott, ais_isnt_here_so_I_dont_remember_this_number: MUSIX 21:01:37 i think: probably 21:01:38 elliott: but you know the best language for that project? 21:01:42 Gregor: what 21:01:42 hint: it's perl 21:01:44 and it's 523 21:01:47 elliott: MU6 21:01:52 Gregor: help 21:02:30 elliott: MU 21:02:42 elliott: by me it is. "darn. i forget if company foo is .com or .org .. i better visit google to check" 21:02:48 Gregor: HELP 21:02:52 itidus20: i do that too sometimes 21:02:54 elliott: SIX 21:02:58 Gregor: IM CRYING 21:03:04 elliott: IIRC, both you and ais wanted me to musix some musix musix. 21:03:15 oh! 21:03:27 Gregor: are you fedoragirl? 21:03:28 well AII isn't quite at that stage yet :P we're still toying with the physics code 21:03:37 ais was a separate thing 21:04:43 itidus20: Although we both have long hair and hats, we are distinct persons. 21:05:09 "I'm also a girl who neither uses Fedora GNU/Linux nor wears fedoras." 21:05:11 ok cool 21:05:18 F- would not google again 21:05:38 fed, or a girl? you decide. 21:05:44 i will also keep an eye out for redhatboy 21:05:58 fed or a freak 21:06:30 and blackhatyouthofindeterminategender 21:06:59 I would /nick to something but I'm on a lot of channels. 21:07:10 Gregor: that's why it's fun 21:07:11 elliott:do you feel better now? 21:07:16 than what 21:07:24 than when you were crying 21:07:51 i wasn't 21:07:52 actually crying 21:07:55 X-D 21:09:13 to have my brain in particular is so terrifyingly painful that i should rather be a moth made to flap my wings by an electrical current 21:09:31 -!- MSleep has joined. 21:09:51 not really 21:13:48 `log aii 21:13:54 2010-05-02.txt:06:21:16: There's also Hawaii that doesn't do DST. 21:14:05 lol 21:14:17 So glad we have `log :P 21:14:27 `pastelogs aii 21:14:33 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.1013 21:14:54 thats a very weird pastelog 21:16:13 which could be best expressed as roger did 21:16:17 `log aiiigh 21:16:21 2011-12-05.txt:21:16:17: `log aiiigh 21:16:34 `log aiiigh, 21:16:38 2011-12-05.txt:21:16:34: `log aiiigh, 21:16:59 2007-10-17.txt:00:20:41: aiiigh, my brain 21:17:22 -!- MSleep has changed nick to MDude. 21:17:30 2006-09-18.txt:23:42:27: Waiiiiiiiiiiit ... 21:17:33 Gregor: I'm still waitin'. 21:17:40 Waitin'? 21:17:53 WAITIN' SINCE 2006 21:17:57 `pastelogs axiomatic 21:17:59 Oh :P 21:18:01 `pastelogs aargh 21:18:03 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.26644 21:18:06 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.18420 21:18:51 lolwow 21:18:55 That was quite the conversation in 2006. 21:19:09 these are really painful pastelogs to read 21:19:16 23:43:55: Your teeth aren't up your ass. 21:19:17 23:44:07: thats what my awkward doctor told me though 21:19:19 23:44:16: Is this a prison doctor? 21:19:21 23:44:54: Did he then tell you that sleeping with strippers is good for preventing lung cancer or something? 21:19:24 23:45:00: I dunno I met him on the corner of my street 21:19:59 waht 21:20:06 itidus20: painful? 21:20:20 2007-01-10.txt:23:08:36: what's a costum enviornment... and how does it change the axiomatic grand law of python's print statement? 21:20:22 kallisti: what 21:21:44 http://www.smbc-comics.com/?id=2452 SMBC is so great. 21:21:45 costume cost, um 21:22:19 2007-07-27.txt:09:52:24: on an LCD, the color displayed on the screen has no impact on power usage! aargh 21:22:19 -!- ais523 has joined. 21:22:25 hi ais523 21:22:29 the ais back 21:22:41 i learned to make my lcd screen turn itself off after my first one died 21:22:44 im and ais back 21:22:48 ijd a 21:22:52 ;/ ;;; ; ; 21:23:00 eek i broke elliott again 21:23:09 ais523: MUSIX 21:24:13 Gregor: ? 21:24:25 oerjan: try bashing just next to the power button, that usually helps 21:24:31 (don't hit the power button itself ofc, for obvious reasons) 21:25:02 ais523: MU6 21:25:45 ais523: elliott: IIRC, both you and ais wanted me to musix some musix musix. 21:25:54 elliott: ah, thanks 21:26:23 elliott: Pff 21:26:24 I'm not sure I actually said that; I discussed that I was writing a game, plus music for it, and was stuck with where to go with parts of it 21:26:34 Gregor: I thought that might be it but wasn't sure 21:27:23 ais523: nintendo never really did anything cool since inventing super mario bros. 1 and legend of zelda 1 21:27:41 itidus20: that's kind-of a sweeping statement 21:28:17 Wii? BORING 21:28:43 i havent actually seen a wii up close 21:28:47 there have been quite a lot of Nintendo consoles since the NES 21:29:00 itidus20: have you played super mario galaxy 21:29:03 if you don't think that game is cool 21:29:06 you're probably not human??? 21:29:23 but maybe still a human 21:29:25 i am not an expert on humanity 21:29:51 my comment more or less answers your question 21:29:51 -!- boily has quit (Quit: WeeChat 0.3.6). 21:29:58 and perhaps explains my whole confusion 21:30:16 what 21:30:24 i havent actually seen a wii up close 21:30:35 oh 21:30:38 hmm, now I'm trying to figure out the best way to send the channel a music file (I have it in .rg, .mid, .ogg, .pdf) 21:30:40 + i havent emulated wii games 21:30:48 you don't go close to them to play them :P 21:30:56 emulating wii games sounds terrible without the controller 21:30:58 (with a boring open-source soundfont, but it still sounds reasonably good) 21:31:15 ais523: write a program that outputs the .ogg 21:31:18 elliott: depends on the game, some of the best wii games don't really care about the motion controls 21:31:40 ais523: well, OK, but the wiimote is pretty unconventional even then 21:31:42 oh /ouch/ YouTube's home page has got ugly 21:31:52 elliott: indeed 21:32:09 yeah.. its like windows xp turned into windows 3.1 21:32:18 but.. its growing on me quickly 21:32:22 hmm, I recommended Metroid Prime Trilogy to someone when they asked me what Wii games to buy 21:33:11 perhaps interface change is initially percieved as ugly, but after mere exposure we start to appreciate it 21:33:34 "change upsets people", a controversial statement 21:34:04 humm 21:34:12 change kills people >:-) 21:34:42 itidus20: that's why it's illegal to throw pennies at unpopular players during a football (soccer) match 21:35:10 football (soccer) 21:35:28 ais523: don't you mean "association football" >:) 21:35:38 (the joke is wikipedia) 21:35:55 meh, Wikipedia has reasonable redirects from all possible guessed disambigs for that, I think 21:36:07 http://en.wikipedia.org/wiki/Football 21:36:09 they missed one! 21:36:15 it's some irrelevant article instead 21:36:27 ok so its a tuesday here.. enough of my angst 21:36:40 time to be positive 21:38:31 http://www.smbc-comics.com/?db=comics&id=1164#comic 21:43:05 it is now christmas day in three timezones!!! happy birthday ais523 21:43:27 oerjan: go reboot elliott again for me, would you? 21:44:03 * oerjan bashes next to the power button 21:44:06 -!- elliott has quit (Quit: Broadcast message from root (pts/0): The system is going down for reboot NOW!). 21:44:44 -!- elliott has joined. 21:44:50 21:45:02 21:45:09 elliott: 21:45:15 ! 21:45:20 Welcome to the esoteric programming channel! Check out 21:45:38 how to the esoteric programming channel! Check out 21:45:45 * Topic for #esoteric is: 21:45:51 olsner: he's having to load a lot of stuff in the background 21:45:58 so you'll get a bit of thrashing for a few minutes 21:46:07 The IOCCC is background so you'll get a bit of thrashing 21:46:43 Loading Markov models (43/768248)... 21:47:12 Loading Markov models (5963/768248)... 21:47:17 hi ais523 21:47:22 hi elliott 21:47:29 hi 21:47:33 hi 21:47:36 hi oerjan 21:47:38 hi ais523 21:47:44 hi 21:47:47 hi 21:48:07 Loading Markov models (23312/768248)... 21:48:13 hi 21:48:18 hi help 21:48:37 Loading Markov models (101049/768248)... 21:48:43 hi ais523 21:48:46 what 21:48:47 help 21:48:49 help what 21:48:50 hi elliott 21:48:52 hi what help hi oerjan what help 21:49:00 fungot: go talk to elliott 21:49:00 ais523: i had always deemed strange, but as she continued her choking he reached feebly in his fnord past the bidding of the vulgar, but by close questioning obtained a very picturesque legend of the fnord 21:49:03 fungot: I'm not elliott! 21:49:04 ais523: imaginations of all the prevailing horrors, we were still drifting south, meanwhile sinking deeper and deeper into that treacherous and sinister white immensity of tempests and unfathomed mysteries which stretched off for some fifteen hundred miles to the fnord, and gnarled, were of absorbing vividness and convincingness, and whenever he awaked he retained a vague sense of having undergone much more than he had been 21:49:11 Loading Markov models (768247/768248)... 21:49:12 hi what help ok 21:49:13 ^style 21:49:13 Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft* nethack pa qwantz sms speeches ss wp youtube 21:49:14 ais523: maybe we should send him in for repairs 21:49:14 hi ais523 help what ok 21:49:18 ok what help hi hi hi hi hi help what ok 21:49:21 oh, that explains it 21:49:27 Loading Markov models (768248/768248)... 21:49:34 I have been restored to an adequate level of loquaciousness. 21:49:51 Hopefully the garrulity endowed by this upgrade shall endeavour to produce most pleasant statements for the considerable enjoyment of yourselves. 21:50:16 (the joke is that the first 768247 out of 768248 models just contain "hi", "help", "what" and "ok", laugh) 21:50:20 fungot: fnord past the bidding of the vulgar 21:50:20 olsner: but whatever had happened, they did lift for a second 21:51:20 we have cthuliftoff 21:51:41 i kinda wish i knew what a markov model really was, but im probably in bliss by not knowing 21:53:18 anyway, if a cylinder is an extruded circle, and a cube is an extruded square, and a box is an extruded rectangle 21:53:24 -!- Ngevd has joined. 21:53:27 Hello! 21:53:30 itidus20: you're thinking of the word "prism" here, I think 21:53:32 Ngevd: hi 21:53:34 or at least should be 21:54:02 ais523: but suppose a prism is concave 21:54:21 err, what? extruding always creates a prism 21:54:23 No it isn't 21:54:34 A prism is pretty damn convex 21:54:34 ok 21:54:41 hmm 21:54:42 Unless the cross-section is concave 21:54:43 you could have a prism with a star shape base, say 21:54:43 ok again 21:54:57 *shaped 21:55:01 what about a moon shaped prism? 21:55:06 What phase? 21:55:16 which moon? 21:55:22 i dunno... like a banana 21:55:35 . 21:55:36 a banana-shaped prism? that's a banana 21:55:36 .. 21:55:38 ... 21:55:41 instance (Show a) => Show (IO a) where show m = "return " ++ show (unsafePerformIO m) 21:55:50 good idea y/n 21:55:57 my bananas so far have been remarkable un-prism-like 21:55:59 humm 21:56:00 *y 21:56:02 elliott: YES 21:56:11 ok like a C shape as a prism 21:56:26 not very safe, but should be fun 21:56:26 elliott: ignores precedence 21:56:26 That's be concave, but not due to its prisminess 21:56:34 oerjan: that would obscure the intent :P 21:56:36 an extruded C shape 21:56:50 sort of like a halfpipe i guess 21:56:54 elliott: you could at least have added parentheses 21:56:57 oerjan: hmph 21:56:59 itidus20, what I just said 21:57:14 -!- yiyus has quit (Remote host closed the connection). 21:57:14 or used showsPrec to produce parens only when necessary? 21:57:14 well would it be a prism? :D 21:57:22 Yes 21:57:38 Prisms can be, but are not inherently, concave 21:58:30 i dunno what made me bring it up.. 21:58:54 Do all platonic solids have planar edge graphs? 21:59:07 a prism is the product space of an interval and some subset of R^2. 21:59:16 Ngevd: yes, obviously 21:59:31 Perfect. 21:59:37 Now NOBODY CAN STOP ME! 22:00:16 more generally, a graph on a sphere is also planar 22:00:29 How about a torus? 22:00:58 just choose some point on the sphere outside the graph and project from it 22:01:14 torus, i don't think so 22:01:16 K3,3 can be drawn on a torus, but is not planar 22:01:26 Perfect. 22:01:34 Now NOBODY CAN STOP ME! 22:01:36 EVEN LESS! 22:01:45 Or more, I'm not sure 22:02:05 Your lack of stoppability inspires awe among your peers. 22:02:06 the problem of drawing K3,3 on a paper without crossing lines went all around my school in 4th or 5th grade, no-one solved it (=> it's unpossible) 22:02:37 olsner: it would only be impressive if one of you managed to actually prove it impossible 22:02:54 does anyone know how to amplify with alsa? 22:03:02 all my volumes are at 100% but it's still quiet 22:03:23 oerjan: it would. I have no idea how to go about proving that 22:03:27 sigh, gotta edit config files for it 22:04:22 hm might need the jordan curve theorem 22:06:04 ok something which to me as a newbie seemed interesting just occured to me 22:06:18 -!- derdon has joined. 22:06:30 Hit us 22:06:37 Not literally 22:06:59 only hit ngevd 22:07:08 -!- DCliche has joined. 22:07:13 -!- augur has quit (Remote host closed the connection). 22:07:28 suppose you have a string of length N on a piece of paper... and you curve it into a circle on that piece of paper 22:07:33 Plugin `dummy' failed with: thread killed 22:07:34 [...] 22:07:37 Just out of curiousity, though, what *is* up with lambdabot? (i.e. the one on this channel right now? 22:07:37 got patched to be more reliable 22:08:00 i want to swat elliott but i guess he's miserable enough. 22:08:05 is this a combination of rotations and translations? 22:08:12 oerjan: wat 22:08:23 elliott: YOU BROKE LAMBDABOT 22:08:29 MY PATCH HASN'T BEEN APPLIED YET :P 22:08:36 ah. 22:08:49 good i didn't swat then. 22:09:07 it seems to me that twisting a straight thing into a curve should really have a unique name for that geometric operation 22:09:14 or... rather... 22:09:15 -!- Klisz has quit (Ping timeout: 245 seconds). 22:09:33 itidus20: rotations and translations turn straight lines into straight lines. 22:10:11 suppose you took water filing a cube and poured it into a long thin rectangular prism 22:10:11 there is something called a mobius transformation, though... 22:10:16 what have you done to the water? 22:10:23 which can turn lines and circles into each other 22:11:01 i guess its a scaling operation in a way... 22:11:18 like if you uhhh 22:11:40 if you have a rectangle of ratio x:y where x+y = z .. 22:11:48 and you alter x and y but maintain z 22:11:58 i guess thats probably scaling 22:12:12 Stretching 22:12:16 z in this sense should probably be called 'a' to avoid confusion 22:12:17 itidus20: it's a linear transformation 22:12:31 scaling would be x/y = z I think 22:12:46 ignoring that real water would be mixed in a chaotic way 22:12:57 yes ignoring such things is important 22:13:23 :-s 22:14:47 any translation, rotation or scaling, and some other things besides, can be written as a map \x -> A x + v where x is a vector variable, v a constant vector, and A a matrix. 22:14:50 i guess that since more or less all shapes can be approximated as polygons, then the rotate scale translate tends to be enough 22:15:03 but.... 22:15:26 grr: hint is too limited 22:15:42 i think 22:15:51 but i suppose that it is one delusion to approximate an object with polygons.. it is another delusion to suppose an object ever remains the same between any 2 measurements 22:16:01 at least it seems to have no way to compile an expression to bytecode, get its type, and then use it in the context of another expression 22:16:03 itidus20: you may have to divide it into pieces that are transformed differently, though. 22:16:21 im having a weird day... trying to think normal but failing 22:16:45 math isn't really normal thinking, many normal thinking people are bad at it. 22:17:27 Maths is what should be normal but isn't 22:20:57 i used the example of a string since a string is good at behaving normally 22:21:18 Unless it gets set on fire 22:21:21 like it doesn't really change it's nature when you twirl it 22:21:22 Or wet 22:21:46 whereas if i had a rectangular prism and i tried to curve it 22:21:47 -!- Patashu has joined. 22:21:51 i would have to slice it up 22:22:06 Or stretch bits 22:22:17 stretching is pretty mind blowing 22:22:58 ah ok 22:23:22 so.. a rectangle.. the left side is the same length as the right side 22:23:39 but if you were to "curve" the rectangle.. 22:23:55 then the ratio of the lengths of the sides would have to alter 22:24:35 which is where a fluid would come in handy 22:26:09 so i guess that a string can bend like it does because that on molecular level it is made of atoms 22:26:17 a fluid of atoms 22:26:42 *shrugs* 22:27:35 -!- Ngevd has quit (Read error: Connection reset by peer). 22:28:10 -!- Ngevd has joined. 22:30:21 -!- Ngevd has quit (Client Quit). 22:30:41 itidus20: a rectangular prism would curve if it was made of soft plastic 22:30:53 it's all about material 22:31:25 and a string would break if it were made of something hard. like uncooked spaghetti. 22:32:06 i guess what i am trying to do is questioning the sufficiency of the rotation/translation/scale/reflection operations 22:32:53 so stretch was mentioned before.. i have seen this one in the paint program along with skew :D 22:32:59 itidus20: rotation and translation is pretty much all you can do with a hard substance without cutting it up 22:34:36 as for projection... thats a bit over my head 22:34:43 hehe 22:35:00 projection is what you get when you shoot your hard substance into a wall at 16000 m/s 22:36:55 oerjan: what about if you suppose that your substance is perfectly soft 22:37:12 then i suppose it would have the property that it can pass through itself 22:37:50 but i am not sure if that property would ever be very useful 22:37:52 besides that, there are a _lot_ of more general mappings. i recall the riemann mapping theorem which is all about stretching to a different shape ... but with functions that have power series expansions in the complex plane 22:39:12 heavy.. 22:39:53 itidus20: dark matter may be like that in a sense, perfectly soft. although it has no cohesiveness either, consisting of individual particles (presumably) 22:40:26 i dont know what i mean by perfectly soft 22:41:13 i do have an idea of perfectly hard though.. as seen in some collision detection in video games 22:41:37 hm about the riemann mapping functions: "Intuitively, such a map preserves the shape of any sufficiently small figure, while possibly rotating and scaling (but not reflecting) it." 22:41:50 my idea of perfectly soft would be more like an animation which can just change its pixels arbitrarily.. im just ... not a mathematician... ahahha.. 22:43:31 itidus20: your perfectly hard idea may be equivalent to http://en.wikipedia.org/wiki/Rigid_body 22:44:24 i suppose the polar opposite of a rigid body is not such a common idea as a rigid body :D 22:45:04 well there are also such ideas as an "ideal gas" 22:45:34 oerjan: rigid bodies and ideal gases, are you just trying to list off things used for bad jokes 22:45:50 laughs out loud 22:46:20 -!- MSleep has joined. 22:47:20 now were getting somewhere 22:47:43 so.. what i am pondering is a geometry of objects whose material is ideal gas 22:47:53 rather than rigid bodies 22:48:31 with the hope that rotation translation and scaling would not feel at home 22:48:46 or maybe they would still 22:48:58 but it would be nice if a larger family of transformations was now available 22:49:17 * kallisti presses his rigid body firmly against itidus20's perfectly ideal gas. 22:49:19 elliott: there you go. 22:49:24 -!- MDude has quit (Ping timeout: 248 seconds). 22:50:02 the rigid body is producing random ideal gas 22:50:21 thsi happens when i need to take a shit 22:50:36 nothing smells worse than the farts of someone who needs to visit the potty 22:53:21 -!- oerjan has quit (*.net *.split). 22:54:13 -!- ais523 has quit (Remote host closed the connection). 22:54:52 -!- augur has joined. 22:55:15 -!- augur has quit (Read error: Connection reset by peer). 22:55:23 -!- augur has joined. 22:56:08 -!- oerjan has joined. 22:59:51 itidus20: .. 23:00:05 uhm..maybe ideal gas isnt entirely what i have in mind.. maybe it is 23:00:26 but i am basically thinking of the opposite of a rigid body 23:00:35 "i deal gas" ha ha ha ha ha ha ha ha ha ha ha ha ha ha 23:00:38 so good. 23:00:55 yes it's time for #esoteric-fartjoke 23:01:10 rigid body vs. ideal gas 23:01:23 3x bad joke combo. 23:04:45 what i have in mind is a material which maintains it's saturation per volume regardless of stretching and shrinking 23:06:23 maybe thats not it 23:06:49 itidus20: do you have an internet connection? 23:06:59 i might instead be thinking of a body which is only a surface 23:07:07 kallisti: yes 23:07:31 itidus20: there are also incompressible liquids 23:07:43 itidus20: http://en.wikipedia.org/wiki/Ideal_gas 23:08:16 * kallisti has a large store of valuable URLS 23:08:19 just ask if you need some. 23:08:44 like if you had some body which was only a surface... 23:09:04 then.. having no volume you could seem to stretch it freely 23:09:07 Are they all of the form echo topic | sed 's/ /_/g ; s|^|http://en.wikipedia.org/wiki/|' ? 23:09:19 oops no wait it has volume 23:09:22 Gregor: there are many like that yes. 23:09:24 but it contains a vaccuuum 23:09:41 kallisti: actually, i didn't see a good url for incompressible liquids 23:09:46 uhmm 23:10:03 Gregor: but if you ask me a specific topic I can probably procure URLs in my vast collection that are relevant to your query. 23:10:28 Gregor: I cannot stress how valuable these URLs are. It's hard to discover sites on the web. 23:10:51 so! this type of body has the following properties. 1) it has a perfectly deformable surface. 2) it always maintains the same volume. 23:11:13 itidus20: are you talking about an ideal gas still? 23:11:17 what am i dealing with in this case? 23:11:18 i don't know 23:11:53 i think an incompressible liquid might be closer, as i think that preserves volume 23:12:20 well volume preservation seems important to me in using the term "body" 23:12:48 that would also give it a relationship with a rigid body which also preserves volume 23:12:51 itidus20: you have a blob. if you punch it the other side will poke out. 23:13:05 itidus20: surprisingly, solid bodies apparently often compress more under pressure than liquids do 23:13:20 if you try to squish it, it will stretch out from the top and bottom of your fist. 23:13:31 but a perfect blob 23:13:32 it's one of those squishy blob toys. 23:13:37 yes perfect blob. 23:13:47 no it's The Blob, run away! 23:14:44 ok heres the thing 23:15:06 i want to create an object which gets beyond the old rotate/translate/scale 23:15:16 since they are kind of boring 23:16:24 maybe i shouldn't.... 23:16:30 not good for my brain 23:18:01 the problem i guess is that the old rotate/translate/scale are not infact boring.. 23:18:08 i am projecting my own boringness onto them 23:18:45 -!- aloril has quit (Ping timeout: 240 seconds). 23:19:08 there's a theorem that says you can get any polygon onto any other of the same area by cutting it up and translating. i think not even rotation is needed. 23:19:17 iirc 23:20:47 oerjan: that seems rather obvious actually. 23:20:57 well.. 23:21:02 it's easy for regular polygons. 23:22:50 well i seem to recall it mentioned in relationship to banach-tarski (which requires 3d, rotations and nonmeasurable sets) 23:25:12 oh hm maybe i'm thinking of the rather more advanced http://en.wikipedia.org/wiki/Tarski%27s_circle-squaring_problem 23:26:02 theres a useful idea i can think of in relation to DMM's piet language 23:26:14 the pixel area of a shape 23:27:03 elliott: What happened to QHC? 23:27:13 since the pixel area of a shape has a very simple intuitiveness for "you can get any polygon onto any other of the same area by cutting it up and translating." 23:27:17 shachaf: It was lost in the Great Hard Drive Crash of I Forget When. 23:27:44 The Quixotic Haskell Compiler? 23:27:50 basically by translating each pixel of color N from bitmap A to bitmap B 23:28:24 well there was some program in piet which calculates PI by using a literal circle made of pixels 23:28:56 http://en.wikipedia.org/wiki/Bolyai-Gerwien_theorem is the simple version with polygons if you also allow rotations 23:30:04 * elliott googled for: famous bisexual toucans 23:31:47 -!- aloril has joined. 23:33:59 -!- copumpkin has quit (Quit: Computer has gone to sleep.). 23:38:04 -!- calamari has joined. 23:44:56 -!- Madoka-Kaname has joined. 23:44:56 -!- Madoka-Kaname has quit (Changing host). 23:44:56 -!- Madoka-Kaname has joined. 23:54:42 * kallisti googled for: can you get any toucan by cutting and translating toucans 23:56:18 * kallisti is now 24 hours without sleep. 23:56:46 when I walk around the house I occasionally feel like I'm going into little micro sleep things. 23:56:54 24 hours is nothing stfu 23:57:12 -!- copumpkin has joined. 23:57:42 elliott: okay I'm 24 hours without sleep after getting 2 and a half hours of sleep after not staying awake for I don't remember how long but definitely not 24 hours. 23:57:58 um 23:58:01 after _not_ staying awake? 23:58:04 after staying awake 23:58:06 . 23:58:39 :%s/asleep/awake/g 23:59:03 watr 23:59:23 kallisti: THERE IS NO MORE SLEEP