00:00:28 -!- azaq231 has quit (Ping timeout: 240 seconds).
00:05:22 -!- elliott has joined.
00:05:36 <pikhq> elliott: Have you watched the Princess Bride yet?
00:08:36 -!- azaq23 has joined.
00:10:20 <elliott> i should reaaaally switch isos
00:10:41 <pikhq> elliott: Have you watched the Princess Bride yet?
00:11:59 <Sgeo> If I say I've watched the Princess Bride, that will mean elliott will not watch it. And thus I deprive elliott of entertainment.
00:12:15 <Sgeo> [Note: elliott does not seem to avoid enjoying the fiction I enjoy]
00:13:04 <pikhq> elliott: If not, why aren't you watching it right now?
00:13:12 <pikhq> And you *better* have a good reason.
00:14:03 -!- TLUL has joined.
00:14:30 <pikhq> I'm not even kidding.
00:15:06 <Sgeo> Because Randall Munroe likes it!
00:15:10 -!- TLUL has changed nick to Jjjj_J_J_J.
00:15:48 <Sgeo> elliott, at any rate, go watch Princess Bride.
00:15:53 <pikhq> Sgeo: I don't think there's anyone who's seen it who *doesn't* like it.
00:25:07 -!- cheater- has quit (Ping timeout: 240 seconds).
00:25:11 <quintopia> you're not watching the princess bride
00:26:16 <elliott> http://i.imgur.com/KEfZv.pnghttp://i.imgur.com/KEfZv.pnghttp://i.imgur.com/KEfZv.pnghttp://i.imgur.com/KEfZv.pnghttp://i.imgur.com/KEfZv.png
00:38:24 <pikhq> You're not watching the Princess Bride.
00:38:53 -!- cheater- has joined.
00:43:41 -!- Mannerisky has left (?).
00:43:46 <Sgeo> I know being good at arithmatic isn't really something that's... entirely necessary for understanding math, but surely simple shortcuts are kind of necessary
00:44:47 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.16/20101130074636]).
00:47:53 <pikhq> elliott: No, but seriously.
00:48:59 <Gregor> DAMN IT FLUIDSYNTH X_X
00:49:13 <pikhq> Gregor: What say ye?
00:49:22 <Sgeo> elliott, if you don't watch it now, I'll spoil it for you
00:49:29 <Gregor> fluidsynth is the worst piece of excellent software I've ever used X_X
00:50:18 <pikhq> Gregor: No, regarding the movie nobody dislikes.
00:50:35 <Gregor> I'm not reading the conversation :P
00:50:53 <pikhq> Gregor: elliott hasn't watched The Princess Bride. We are attempting to get him to.
00:50:53 <elliott> Our next aim is to make amends to our god for all our bad behaviour
00:50:54 <elliott> over the previous turns. Angry gods are generally quite hard to
00:50:54 <elliott> please, demanding the sacrifice of a single very poweful monster to
00:50:54 <elliott> make amends (rather than several sacrifices of anything like they
00:50:54 <elliott> usually do, which is typically easier to come across). Luckily, we
00:50:56 <elliott> happen to have a shopkeeper corpse handy; unluckily, they're very
00:51:16 <elliott> Gregor: I haven't watched The Princess Bride yet, because I have a ten-year-long backlog of fiction to consume, and now nobody will shut up.
00:51:18 <Gregor> elliott: Nobody gives an eff. It's a cult classic, which is to say a bad movie.
00:51:32 <pikhq> Gregor: Something is wrong with you.
00:51:37 <elliott> Gregor: I have a feeling you're wrong, but thank god there's somebody in the world who won't bug me about it.
00:51:38 <Gregor> WHERE'S MY DAMNED TROLLFACE UNICODE CODEPOINT ALREADY
00:52:01 <pikhq> elliott: Top of the queue. It's the most important.
00:52:02 <Sgeo> Why is everyone bugging elliott _now_?
00:52:11 <Sgeo> pikhq, no, it's not?
00:52:21 <pikhq> Sgeo: What movies should he see first?
00:52:34 <elliott> pikhq: My backlog is very, VERY long.
00:52:35 <Sgeo> pikhq, I don't know what's on his list
00:53:00 -!- m00barax has joined.
00:53:41 <pikhq> elliott: Well, you've probably seen most of the other movies on the "Movies you've seen if you haven't been living under a rock" list, so. :P
00:53:56 * Sgeo still hasn't seen Inception
00:54:03 <elliott> pikhq: I really *haven't*.
00:54:11 <elliott> My backlog is _extremely_ long.
00:54:31 <elliott> pikhq: I don't consume fiction, I put fiction on my "to consume" list.
00:54:40 <elliott> pikhq: For instance: The entire Culture series is high up on that list.
00:54:49 <pikhq> elliott: Have you seen Star Wars?
00:55:02 <elliott> Well, yes. (I dislike the Star Wars films.)
00:55:02 <fizzie> elliott: You put fiction on your list but don't inhale?
00:55:14 <elliott> fizzie: It depends on the meaning of "on".
00:55:29 <pikhq> elliott: Lemme guess. Prequels ruined it for you?
00:55:54 <elliott> pikhq: No... I actually don't like Star Wars Episode IV: A New Hope. Independently from any other film.
00:56:16 <elliott> It's entertaining in parts, but I don't really think it's a good film. And the other films in the series don't appeal to me at all.
00:56:54 <elliott> pikhq: BUT THE HOLIDAY SPECIAL, I LOVE THAT
00:57:06 <pikhq> Anyways, *that's* an example of a movie I'd accuse you of having lived under a rock for not having seen.
00:57:41 <pikhq> Even if you didn't *like* it, well, WTF how could have not seen it.
00:58:44 * Sgeo has not seen it
00:58:52 <elliott> pikhq: Star Wars was a lucky guess though, guess again and you'll probably hit something I haven't seen.
00:59:33 <pikhq> elliott: Uh. Heck, let's just go with "Pixar" here.
00:59:45 <elliott> pikhq: Why are you guessing properly :P
00:59:46 <Sgeo> I've.. seen a flash animation that has vaguely the plot
01:00:01 <elliott> I haven't seen Cars or Ratatouille or WALL-E, though.
01:00:08 <pikhq> elliott: Because I'm actually guessing ones that seriously everyone has seen and should see.
01:00:12 <elliott> I HAVE NOT SEEN THE ENTIRE PIXAR BACK CATALOGUE
01:00:19 <pikhq> elliott: START WATCHING.
01:00:26 <pikhq> elliott: GET THROUGH YOUR BACK CATALOGUE.
01:00:31 <pikhq> elliott: I EXPECT TO SEE YOU NEXT YEAR.
01:00:36 <elliott> WALL-E is on the backlog, the others aren't.
01:00:45 <elliott> I've seen every Pixar film other than those three :P
01:00:49 <pikhq> Okay, Cars is probably skippable.
01:00:58 <Sgeo> Dagobah. That's a Star Trek thing, apparently. Where else have I seen it?
01:01:15 <elliott> Hmm, stone or wood-bordered glass...
01:01:24 <elliott> m00barax: this is a channel about programming
01:01:28 <Sgeo> >.> I KNOW THE DIFFERENCE BETWEEN THE TWO DAMMIT
01:01:55 <Sgeo> Prime numbers are more than welcome here, though
01:02:14 <pikhq> elliott: Hmm. Having trouble thinking of more movies that both *must* be seen by everyone and *have been* seen by most everyone.
01:03:19 <pikhq> Cause you start getting into movies that everyone should see but are too old to have been seen by just about everyone alive. :P
01:03:34 <pikhq> Perhaps everyone above $age, but hey.
01:03:36 <Sgeo> m00barax, have you heard of Brainfuck?
01:03:41 <elliott> m00barax: esoteric programming languages
01:03:48 <Sgeo> m00barax, computer languages with limited practical use.
01:03:51 <elliott> pikhq: I haven't seen Casablanca. Or Citizen Kane. Or ... yeah :P
01:04:07 <Sgeo> m00barax, yes. Well, BF is a famous example of an esoteric programming language.
01:04:11 <pikhq> elliott: See, movies that are too old to have been seen by just about everyone alive.
01:04:34 <m00barax> but to me basic is esoterrik too
01:04:41 <pikhq> elliott: But, anyways, I expect to see you next year with a bunch of movies watched.
01:04:55 <pikhq> And, heck, games played, too.
01:04:57 <elliott> pikhq: YOU HAVE NO IDEA HOW LONG I HAVE BEEN KEEPING THIS BACKLOG
01:05:01 <pikhq> Hope you don't have anything else to do!
01:05:23 <Sgeo> People actually once used BASIC for stuff. Maybe it deserves to be counted as esoteric, but it isn't, I think.
01:05:31 <m00barax> {0,1}-langage very standard so?
01:05:43 <Sgeo> {0,1}-language?
01:05:45 <elliott> pikhq: Yeah, there are a shitload of games on the list...
01:05:49 <elliott> pikhq: Like any game by Valve ever.
01:06:07 <Sgeo> elliott, you're running in Valve time? That explains it.
01:06:12 <pikhq> elliott: Oh fuck, there's probably RPGs on there too.
01:06:25 <elliott> pikhq: I'm not much of an RPG person... but Chrono Trigger is on the list :P
01:06:34 <Sgeo> m00barax, you may be referring to machine code?
01:06:41 <pikhq> elliott: I didn't say "many". Just some.
01:06:52 <elliott> Feel free to suggest additions to the list :P
01:07:02 <pikhq> elliott: Shin Megami Tensei: Persona 3 FES.
01:07:04 <pikhq> elliott: Shin Megami Tensei: Persona 4.
01:07:12 <elliott> GEE, I COULD NEVER HAVE GUESSED
01:07:15 <Sgeo> No one sane programs directly in machine code. Instead, they program in a language that translates very easily to machine code.
01:07:19 <elliott> I meant non-RPG suggestions, but kay :P
01:07:36 <pikhq> Heck, let's just go for "The entire series". Hope you know Japanese and have a decade free. :P
01:07:37 -!- Jjjj_J_J_J has changed nick to toenail|toe.
01:07:44 <Sgeo> Maybe a long time ago people programmed directly in machine code, I don't know. But assembly to machine code is very simple
01:07:54 <pikhq> (there's like 50 fucking games. God.)
01:08:08 <m00barax> yep.... but 0-1 its the BF of all BF langages
01:08:16 <elliott> ANYONE ELSE WANT TO EXTEND MY LIST
01:08:41 <pikhq> The Dresden Files. It'll probably be up to 20 books by the time you get around to it.
01:08:46 <Sgeo> m00barax, machine code is actually used. Not by humans, but it's what's _actually running_ on your machine
01:09:00 <Sgeo> BF, by contrast, is only used by people messing around with BF
01:09:06 <pikhq> Though it'll probably not take you too long to finish — you're not likely to eat or sleep until you finish after starting.
01:10:16 <m00barax> RPG is'nt a langage, but a stone-level of minding
01:10:21 <quintopia> does anyone in here want a hand-written letter?
01:10:56 <Sgeo> Although.. the people here aren't entirely sane [no offense, people who might take offense, I didn't mean it literally]. People here sometimes work somewhat directly with machine code if they have to becase they're working on a project that, strictly speaking, does't need to exist
01:12:20 <Sgeo> m00barax, sometimes, we go off topic. Much more rarely, we go on topic.
01:12:36 -!- Tritonio has joined.
01:12:50 <Sgeo> I'm beginning to think m00barax is a bot.
01:14:42 <m00barax> i do usr g++ and concurrent-Clean
01:15:26 <Sgeo> I've looked at Clean
01:15:39 <Sgeo> I don't know how to avoid being bored when looking at it though
01:15:48 <Sgeo> I just want to learn the differences between Clean and Haskell
01:15:59 <elliott> are you actually talking in english?
01:16:07 <elliott> are we talking Clean the language that's almost exactly like Haskell?
01:16:22 <m00barax> clean and haskell.... uniquness of typing maybe :)
01:17:05 <m00barax> arrows... category approch of minding
01:17:11 <elliott> why is a function "(a -> b) -> c" written "(a -> b) c" in clean
01:18:44 <quintopia> m00barax: the word you want it thinking, not minding
01:19:09 <m00barax> yep.... i do tink in french... that wath for :)
01:20:47 <m00barax> (a -> b) -> c not same as (a -> b) c ... in cclean
01:21:50 <m00barax> because 'c' is not linked the same way...
01:22:28 <elliott> that should be a singleton list of (cons x xs) or something
01:23:41 <m00barax> (cons x xs) is lambda thinking... not modern langage ...
01:25:06 <m00barax> why clean is in this channel of "esoteric langage" ?
01:25:21 <pumpkin> or precedence probably makes that [(5:[]):[]]
01:25:58 <m00barax> its one of the better langage with.... haskell ghci ?
01:27:17 <m00barax> (5:[]) is an item of the list [...:[]] .... i think
01:27:41 <elliott> <m00barax> (cons x xs) is lambda thinking... not modern langage ...
01:27:49 <elliott> I was just using it as unambiguous notation
01:27:54 <elliott> so that you could see what i'm talking about
01:28:50 <m00barax> yes... (cons a b) is directly term of lisp/scheme or any untyped lambda-langage as is
01:29:10 <elliott> suited towards laptops and simialr systems). A little-known feature
01:29:19 <elliott> m00barax: sure, but you could easily define that in e.g. haskell
01:29:47 <m00barax> yes in haskell... not so easy in lisp
01:30:12 <pumpkin> it could easily be a typed language
01:30:20 <m00barax> and what about precedence of operators ?
01:30:54 <m00barax> yes... lisp easy typed langage than ... basic... lol
01:31:08 <pumpkin> I'm saying the syntax could be applied to any language
01:31:45 <pumpkin> operator vs. function is purely syntactic
01:32:40 -!- pumpkin has quit (Quit: Computer has gone to sleep.).
01:32:57 <m00barax> but fondments of operators are easyer to define in function and curry...
01:34:55 -!- toenail|toe has changed nick to Jjjj_J_J_J.
01:35:03 -!- Tritonio_Droid has joined.
01:38:17 -!- Tritonio_Droid has quit (Client Quit).
01:40:02 -!- Tritonio_Droid has joined.
01:45:01 -!- m00barax has quit (Quit: Leaving).
01:47:34 -!- oerjan has joined.
01:49:50 -!- Tritonio has quit (Quit: Leaving).
01:51:02 <oerjan> <elliott> again... "Asking a linguist how many languages they know is like asking a doctor how many diseases they have."
01:51:16 <oerjan> yes, but once again, it reminds me of Lockhart's Lament
01:51:48 <oerjan> hm but then again the _opposite_ could also remind me of lockhart's lament
01:52:10 <oerjan> people rote memorizing vocabulary with no overarching understanding
01:53:46 -!- Jjjj_J_J_J has changed nick to TLUL.
01:53:54 <pikhq> AKA "the usual means of learning languages"
01:56:31 <elliott> synchronise target offiufiofoijfoijfoifoifofojfofojfoifiojfoijfoijfoifoif
01:56:34 -!- elliott has left (?).
01:56:54 <oerjan> parting the channel just to poop seems a _tad_ excessive
01:57:52 <oerjan> <Sgeo> I saw something on Fark saying that Wikipedia had a redirect...
01:58:18 <oerjan> wikipedia has a policy to use the spelling (am/br) of the region relevant to the topic
01:58:49 <oerjan> so it would obviously be World Trade Center
01:58:55 -!- copumpkin has joined.
02:00:43 -!- Tritonio_Droid has quit (Quit: Bye).
02:06:15 <oerjan> <elliott> he didn't know of "share and share alike"
02:06:28 <oerjan> despite not remembering the actual conversation
02:09:24 <Gregor> Man, sox' "earwax" filter is creepsiloo.
02:09:47 <oerjan> ...that sentence is almost poetry
02:10:22 <oerjan> of the lewis carroll type
02:10:36 <Gregor> Supposedly it makes it easier to listen to the audio on headphones by adding "cues" that make your brain hear it outside your head instead of inside your head.
02:10:41 <Gregor> Which it most certainly does.
02:10:45 <Gregor> But for some reason that's just creepy.
02:11:51 <oerjan> <m00barax> i'm an esotermik programmers
02:12:08 <oerjan> beware of esothermic programmers exploding
02:12:39 <oerjan> wait that's exothermic
02:13:00 * oerjan watches his pun stretch to the breaking point, then snap
02:14:55 <oerjan> actually exothermic puns _should_ crash and burn, shouldn't they
02:30:51 <Sgeo> Oh, I remember where I've seen "Dagobah" from! It's a website!
02:31:02 -!- augur has joined.
02:32:09 <Sgeo> Why do I always do things the complicated way?
02:32:47 <Sgeo> Professor was suggesting that putting code for randomness in the Card class was a bad idea, so I "compromise" and initialize a Random lazily
02:43:13 -!- poiuy_qwert has joined.
03:10:20 <Gregor> Horribly Inappropriate Movies LLC presents: Ein Führer und sein Hund, Hitler's story as told from the imagined perspective of his toy poodle Schnitzel.
03:12:16 <oerjan> i'm pretty sure hitler had a real dog
03:13:07 <oerjan> which mind-blowingly enough, probably loved him
03:13:38 <Sgeo> I think that's Gregor's point
03:13:46 <oerjan> http://en.wikipedia.org/wiki/Blondi
03:14:06 <oerjan> well it was not a toy, nor a poodle, nor named Schnitzel
03:14:25 <Gregor> I did not know that, and it somewhat diminishes from the joke :P
03:14:48 <Gregor> Also, I don't think any German speaker under any circumstances would name their dog "Schnitzel" :P
03:15:31 <Sgeo> "Hitler's nurse, Erna Flegel, said in 2005 that Blondi's death had affected the people in the bunker more than Eva Braun's suicide had"
03:16:12 <Gregor> Sgeo: That ... makes no sense ... wtf. Eva Braun's suicide coincided with Hitler's ...
03:16:34 <Sgeo> http://www.guardian.co.uk/world/2005/may/02/artsandhumanities.secondworldwar is the linked reference
03:27:48 <kfr> Phantom_Hoover> [22:17:37] I just saw the word "zygomorphism".
03:27:48 <kfr> <Phantom_Hoover> [22:17:42] Is kfr being an idiot?
03:27:53 <kfr> What? I never used that word
03:31:49 <oerjan> i'm not sure those lines were connected...
03:33:53 <kfr> 5 seconds, they better be!
03:46:45 <Gregor> Tomorrow is clear ... with a high of 13°F (-11°C) ...
03:48:14 <pikhq> Gregor: You lucky bastard.
03:48:40 <pikhq> Oh, wait, it'll be getting up to 30°F tomorrow.
03:49:04 <pikhq> (note: right now, not only no but fuck no)
03:49:22 -!- Sgeo has changed nick to No.
03:49:36 -!- No has changed nick to Sgeo.
03:51:24 -!- azaq23 has quit (Ping timeout: 240 seconds).
04:08:30 -!- azaq23 has joined.
04:29:54 -!- leonid has left (?).
04:43:24 -!- augur has quit (Remote host closed the connection).
04:45:00 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!a(~^)*~**)~a((, )S:^)**^):^
04:46:04 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!a(~^)*~**)~a((~aS:^):^(, )S:^)**^):^
04:46:04 <fungot> 3((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!a(~^)*~**)~a((~aS:^):^(, )S:^)**^)((~(:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S:((a(~^)*~**)~a(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)**~:((:)~(*)**)~a*~(^))~*^^ ...too much output!
04:46:34 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!a(~^)*~**)~a(!(~aS:^):^(, )S:^)**^):^
04:46:34 <fungot> 3((~(:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S:((a(~^)*~**)~a(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)**~:((:)~(*)**)~a*~(^))~*^^^!^)~a*^:)~^(((~(~(:a~*):^)))))((::**)~^)() ...out of stack!
04:48:45 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:48:45 <fungot> 3, 13, 1(::**)~^ ...bad insn!
04:49:15 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:49:15 <fungot> 13, 1(::**)~^ ...bad insn!
04:49:39 <oerjan> ^ul (()~^()~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:49:50 <Sgeo> When do I stop making fun of coppro?
04:50:49 <oerjan> ^ul (()~^()~^)((((:((0)(!(1)(!(2)(!(3)((~a( )*S:^):^))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:51:50 <oerjan> where the heck can the bad insn come from
04:53:35 <oerjan> ^ul (()~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:53:36 <fungot> 1, 11, 1:^* ...bad insn!
04:53:45 <oerjan> ^ul ((:*)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:53:45 <fungot> 2, 12, 1(:*)~^ ...bad insn!
04:54:03 <oerjan> ^ul ((:*)~^(:*)~)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:54:25 <oerjan> it means it has somehow managed to execute a non-command character
04:54:57 <oerjan> hm that ^ should not have been printed
04:55:51 <oerjan> ^ul ((:*)~^(:*)~)((((:((a)S(0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:56:13 <oerjan> ^ul ((:*)~^(:*)~)((((:((a)S(0)(!(1)(!(2)(!(3)((b)S_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
04:56:32 <Sgeo> http://www.reddit.com/r/canada/comments/fdy76/disappointed_with_snowmageddon_its_because_theres/
04:59:55 <oerjan> ^ul ((:*)~^(:*)~)((((:(((a)S0)(!((a)S1)(!((a)S2)(!((a)S3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a(((b)S, )S:^)**^):^
04:59:55 <fungot> (a)S2(b)S, (a)S2^a ...bad insn!
05:00:39 <oerjan> ^ul ((:*)~^(:*)~)((((:(((a)S0)(!((b)S1)(!((c)S2)(!((d)S3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
05:00:39 <fungot> (c)S2, (c)S2^d ...bad insn!
05:01:09 <oerjan> it is somehow trying to execute that 3
05:02:15 <oerjan> of course this _should_ be utterly impossible :D
05:03:59 <oerjan> ^ul ((:*)~^(:*)~)((((:aS:(((a)S0)(!((b)S1)(!((c)S2)(!((d)S3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
05:03:59 <fungot> (:*)(c)S2, (:*)(c)S2(((~(~(:a~*):^))))^d ...bad insn!
05:05:07 -!- augur has joined.
05:06:08 <oerjan> ok that thing on top of the stack should be a church numeral but is an equality check instead
05:06:44 <oerjan> to be precise, a check for the number 2
05:08:08 -!- Zuu has quit (Read error: Connection reset by peer).
05:08:13 <oerjan> oh wait i have another bug
05:08:49 <oerjan> ^ul ((:*)~^(:*)~^)((((:aS:(((a)S0)(!((b)S1)(!((c)S2)(!((d)S3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
05:08:49 <fungot> (:*)(c)S2(:*)(c)S2, (:*)(c)S2((:)~(*)**(((~(~(:a~*):^))))):^*d ...bad insn!
05:09:08 <Ilari> (a<1>c|(b<2>c|))|(a<3>c|(a<4>b|))
05:11:29 <oerjan> hm it _does_ manage to get through the first number for those i've tried.
05:12:10 -!- Zuu has joined.
05:12:14 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
05:12:14 <fungot> 13, 1(::**)~^ ...bad insn!
05:12:32 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((~a(; )S:^):^(, )S:^)**^):^
05:12:32 <fungot> 13; , ...out of stack!
05:13:27 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:13:27 <fungot> 13((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^); (()~^(:a(~^)*~**(::**):a(~^)*~(()(~(~(:a~*):^))(a))~*^^*)~^); (); ...out of stack!
05:13:54 <Ilari> Better example: (0<1>0|1<1>1|)
05:15:45 <Ilari> Regexps + operator that matches anything subexpression can match isn't equivalent to deterministic pushdown automata in power (that describes language no such automaton can match) ...
05:18:38 <Ilari> In fact, it seems that Regexps + operator that matches anything subexpression can match are exactly equivalent to context free grammars in expressive power...
05:19:07 * oerjan doesn't understand that <1> syntax
05:19:59 <Ilari> Matches anything 1st subexpression can match.
05:20:10 <oerjan> and what _is_ the 1st subexpression?
05:20:44 <oerjan> um that's the whole thing there
05:21:11 <oerjan> ok so it's just S := 0 S 0 | 1 S 1
05:23:07 <oerjan> ok you can encode an _impossible_ subexpression, and use that to ensure that some branches cannot be taken, which means those branches can be used as independent nonterminals.
05:23:21 <oerjan> i believe that means you can encode any CF grammar, yes
05:25:06 -!- Mannerisky has joined.
05:25:10 <Ilari> One can eliminate impossible nonterminals from CF grammar anyway.
05:25:31 <oerjan> Ilari: um it's not in the _CF_ grammar it is supposed to be impossible
05:26:17 <oerjan> say you have an expression of the form ((0<2>)|...)
05:26:38 <oerjan> then the second subexpression is infinitely recursive, and so cannot be satisfied
05:27:09 <Ilari> Thus, one only needs to reference amount of subexpressions at most equal number of nonterminals in the CFG...
05:27:50 <oerjan> that means that you can use a subexpression of the form (<2>(whatever...)) to make (whatever...) an unreachable subexpression which can only be referred by number
05:29:12 <Ilari> AFAIK, you can match any language CFG can describe without impossible subexpressions.
05:29:13 <oerjan> hm you don't need the outer parentheses
05:30:27 <oerjan> since if you need them, they must be reachable _somewhere_ from S, and you can then put the rules for that nonterminal in any of those spots and replace the others by references to that one
05:30:34 <Ilari> Because algorithmical manner of generating those would only generate impossible subexpression if there was impossible nonterminal in the CFG. And one can eliminate those.
05:30:59 <Ilari> Or impossible production.
05:31:40 <Ilari> The elimination works because emptyness of CFG is decidable problem.
05:35:03 <Ilari> Going the other way can require more nonterminals than there are subexpressions in the original (in order to deal with Kleene closures).
05:36:25 <oerjan> (<1>+<2>|<1>-<2>|(<2>*<3>|<2>/<3>|(<4>^<3>|(\(<5>\)|(<number>)))))
05:36:42 <oerjan> or something like that
05:40:23 -!- pikhq has quit (Remote host closed the connection).
05:40:39 -!- pikhq has joined.
05:41:06 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^(~!(x)S:^):^)!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:41:50 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^(~!(x)S:^):^!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:41:50 <fungot> 13xxxxxxx ...out of stack!
05:42:43 <oerjan> ^ul (()~^(::**)~^(:*)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^(~!(x)S:^):^!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:42:44 <fungot> 132xxxxxxxx ...out of stack!
05:43:20 <oerjan> ...ok the stack is _not_ supposed to be that deep at that point, and adding a digit seems to make it deeper
05:43:36 <oerjan> ^ul (()~^()~^(::**)~^(:*)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^(~!(x)S:^):^!!!a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:43:36 <fungot> 1132xxxxxxxxx ...out of stack!
05:43:58 <oerjan> and it doesn't matter if there are consecutive equal numbers. hm.
05:46:44 <oerjan> ^ul (()~^()~^(::**)~^(:*)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!(~a(; )S:^):^a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:46:44 <fungot> 1132; ((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!(~a(; )S:^):^a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^); (~a(; )S:^(:a(~^)*~**(:*):a(~^)*~(()(~(~(:a~*):^))(a))~*^^*)~^(:a(~^)*~**(:*):a(~^)*~(()(~ ...too much output!
05:52:12 -!- Mathnerd314 has quit (Remote host closed the connection).
05:53:03 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!(~a(; )S:^):^a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^):^
05:53:03 <fungot> 3; ((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!(~a(; )S:^):^a(~^)*~**)~a((~a(; )*S:^):^(, )S:^)**^); (~a(; )S:^()~^()); ((::**)~^); (); ...out of stack!
05:54:34 -!- Vorpal has quit (Read error: Connection reset by peer).
05:54:56 -!- Vorpal has joined.
05:56:17 <kfr> There should be an initiative for using more Unicode in programming languages
05:56:19 <kfr> And colours!
05:56:38 -!- poiuy_qwert has quit (Ping timeout: 255 seconds).
06:01:23 -!- poiuy_qwert has joined.
06:06:42 -!- hagb4rd has quit (Ping timeout: 240 seconds).
06:11:22 -!- poiuy_qwert has quit (Ping timeout: 272 seconds).
06:11:55 -!- poiuy_qwert has joined.
06:19:26 -!- hagb4rd has joined.
06:29:54 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
06:30:19 -!- poiuy_qwert has joined.
06:45:13 -!- Zuu has quit (Read error: Connection reset by peer).
06:45:16 -!- Zuu_ has joined.
06:50:28 -!- azaq23 has quit (Quit: Leaving.).
07:07:57 -!- poiuy_qwert has quit (Ping timeout: 276 seconds).
07:08:27 -!- poiuy_qwert has joined.
07:15:00 -!- asiekierka has joined.
07:15:05 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
07:15:33 -!- poiuy_qwert has joined.
07:30:15 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~*:a(.B)*S*)~a)~a(**~:((:)~(*)**:a(.A)*S)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~*:a(.C)*S*)~a((, )S:^)**^):^
07:30:15 <fungot> 13(:a(~^)*~*:a(.B)*S*(::**):a(~^)*~(()(~(~(:a~*):^))(a))~*^^*).A((:a(~^)*~*:a(.B)*S*(::**):a(~^)*~(()(~(~(:a~*):^))(a))~*^^*)~^).C, 1((^)~^^).B(::**)~^ ...bad insn!
07:33:05 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~*(!=)*S*)~a)~a(**~:((:)~(*)**(=)*S)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~*(!=.)*S*)~a((, )S:^)**^):^
07:33:05 <fungot> 13:a(~^)*~*(!=)*S*(::**):a(~^)*~(()(~(~(:a~*):^))(a))~*^^*=()~^()~^!=. ...out of stack!
07:33:09 <Ilari> Full IANA IPv4 depletion in about 7 hours?
07:33:40 <oerjan> there's a press conference i hear
07:34:26 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~*(!=)S*)~a)~a(**~:((:)~(*)**(=)S)~a*~(^))**a(~*^^^!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~*(!=.)S*)~a((, )S:^)**^):^
07:34:26 <fungot> 13=!=., 1!=(::**)~^ ...bad insn!
07:36:16 <oerjan> that = test is exactly backwards of what should be
07:37:59 * oerjan suspects a missing ! that was supposed to remove the branch not taken
07:38:56 <oerjan> ^ul (()~^(::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
07:38:58 <fungot> 13, 1113, 3113, 132113, 1113122113, 311311222113, 13211321322113, 1113122113121113222113, 31131122211311123113322113, 132113213221133112132123222113, 11131221131211132221232112111312111213322113, 31131122211311123113321112131221123113111231121123222113, 13211321322113311213212312311211131122211213 ...out of time!
07:39:16 <Ilari> What sequence is that?
07:39:27 <oerjan> conway's look and say sequence
07:40:57 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
07:40:59 <fungot> 3, 13, 1113, 3113, 132113, 1113122113, 311311222113, 13211321322113, 1113122113121113222113, 31131122211311123113322113, 132113213221133112132123222113, 11131221131211132221232112111312111213322113, 31131122211311123113321112131221123113111231121123222113, 132113213221133112132123123112111311222112 ...out of time!
07:41:44 <oerjan> ^ul ((:*)~^(:*)~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
07:41:45 <fungot> 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, ...too much output!
07:41:55 <oerjan> the only value which doesn't grow
07:42:51 <Ilari> Conway's constant... Algebraic number of degree 71...
07:43:18 <oerjan> ^ul ((:*)~^(!())~^()~^)((((:((0)(!(1)(!(2)(!(3)(_))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
07:43:19 <fungot> 201, 121011, 1112111021, 311231101211, 132112132110111221, 11131221121113122110312211, 31131122211231131122211013112221, 13211321322112132113213221101113213211, 1113122113121113222112111312211312111322211031131211131221, 311311222113111231133221123113112221131112311332211013211311123113112211, 132113 ...out of time!
07:47:28 <Ilari> Degree 71 is quite high for any constant arising from actual problem. Those tend to be either of low algebraic degree or transcendal...
07:48:11 <Ilari> (all rational numbers are of degree 1).
07:48:33 <oerjan> actually from my own research i would expect degree 92, the number of "atoms" in conway's cosmological theorem. i suppose there is a degree 92 polynomial with a large degree 71 factor.
07:49:40 <oerjan> because the constant should be the largest eigenvalue of the 92x92 matrix describing how those atoms turn into each other at each step
07:50:33 <oerjan> i haven't actually read conway's original article, so i don't know how he gets a simpler one
07:50:46 <Ilari> One fun problem: Is there a power of two that doesn't contain 1, 2, 4 or 8 that isn't 65536?
07:51:08 <oerjan> that rings a bell, in which case the answer is "no"
07:52:05 <oerjan> it should be possible to prove by just calculating (mod 10^k) for a not too large k, and checking that you end up with a cycle where everything contains 1, 2, 4 or 8
07:53:11 <oerjan> hm wait actuall the atom 22 doesn't lead to any of the others, so it should be 91 at most
07:53:58 <oerjan> i've seen a list of the atom transitions, all the others lead to each other
07:54:57 <lambdabot> [2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,65536,131072,26...
07:58:11 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(!(4)(!(5)(!(6)(!(7)(!(8)(!9(_))))))))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
07:58:12 <Ilari> Any power of two that doesn't contain 1, 2, 4 nor 8 must also be power of 16.
07:58:13 <fungot> 3, 13, 1113, 3113, 132113, 1113122113, 311311222113, 13211321322113, 1113122113121113222113, 31131122211311123113322113, 132113213221133112132123222113, 11131221131211132221232112111312111213322113, 31131122211311123113321112131221123113111231121123222113, 132113213221133112132123123112111311222112 ...out of time!
07:58:47 <oerjan> ah right that's a start
07:59:03 <lambdabot> [16,256,4096,65536,1048576,16777216,268435456,4294967296,68719476736,109951...
07:59:40 <oerjan> and there we have 36 at the end regularly...
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:11 <oerjan> hm actually only 1 occurs in the last 2 digits there
08:00:57 <oerjan> !haskell take 100 $ iterate (*16) 16
08:01:22 <EgoBot> [16,256,4096,65536,1048576,16777216,268435456,4294967296,68719476736,1099511627776,17592186044416,281474976710656,4503599627370496,72057594037927936,1152921504606846976,18446744073709551616,295147905179352825856,4722366482869645213696,75557863725914323419136,1208925819614629174706176,19342813113834066795298816,309485009821345068724781056,4951760157141521099596496896,79228162514264337593543950336,1267650600228229401496703205376,20282409603651670423947251286016,32
08:03:01 <oerjan> it is of course exceedingly unlikely that you can avoid 1,2,4,8 once the numbers get long
08:04:33 <oerjan> > map (fromJust . findIndex ("1248" `notElem`) . reverse . show) $ iterate (*16) 16
08:04:34 <lambdabot> Couldn't match expected type `[[GHC.Types.Char]]'
08:05:08 <lambdabot> forall a. (a -> Bool) -> [a] -> Maybe Int
08:05:31 <oerjan> > map (fromJust . findIndex (`notElem` "1248") . reverse . show) $ iterate (*16) 16
08:05:32 <lambdabot> [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,...
08:05:45 <oerjan> > map (fromJust . findIndex (`Elem` "1248") . reverse . show) $ iterate (*16) 16
08:05:46 <lambdabot> Not in scope: data constructor `Elem'
08:05:53 <oerjan> > map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 16
08:05:55 <lambdabot> [1,2,3,*Exception: Maybe.fromJust: Nothing
08:06:18 <oerjan> > map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 1048576
08:06:20 <lambdabot> [3,1,2,2,5,4,1,4,2,4,4,1,2,4,2,2,1,3,2,7,5,1,2,4,3,3,1,2,2,3,4,1,4,2,4,3,1,...
08:06:38 <oerjan> ouch there's one with 7
08:07:13 <oerjan> > filter (> 4) . map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 1048576
08:07:15 <lambdabot> [5,7,5,5,5,5,7,7,12,7,5,7,5,6,5,5,8,5,6,5,6,5,5,5,7,7,8,7,5,7,5,7,5,7,6,8,5...
08:07:25 <oerjan> > filter (> 6) . map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 1048576
08:07:48 <oerjan> > filter (> 5) . map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 1048576
08:07:51 <lambdabot> [7,7,7,12,7,7,6,8,6,6,7,7,8,7,7,7,7,6,8,6,6,9,8,6,7,7,6,8,6,6,6,6,6,6,6,6,6...
08:07:57 <fizzie> I see you're familiar with the engineer's method of proving mathematical statements.
08:08:20 <oerjan> there's even 12 and 9 in there. this might be harder than it looks.
08:08:23 -!- hagb4rd has quit (Ping timeout: 240 seconds).
08:08:30 <oerjan> or rather it's starting to look hard
08:10:22 <oerjan> hm on the other hand those numbers are probably huge
08:11:43 -!- myndzi has quit (Ping timeout: 265 seconds).
08:13:28 -!- hagb4rd has joined.
08:14:06 -!- copumpkin has quit (Ping timeout: 240 seconds).
08:14:39 -!- copumpkin has joined.
08:16:08 <asiekierka> i modified this -> http://www.remcobloemen.nl/2010/02/brainfuck-using-llvm/
08:16:16 <asiekierka> now i'm feeding Lost Kingdom through it
08:21:05 <asiekierka> also i can port it to other platforms, then
08:21:21 <kfr> I want to do something with LLVM, too
08:21:25 <Ilari> If one considers 3 last digits, the exponent of 16 must be congruent to 3, 4, 5, 9, 10, 12, 14, 15, 16, 18, 22 or 24 modulo 25...
08:21:25 <asiekierka> actually wait i think i made the tape size too small
08:22:39 <kfr> That's a good link though, I've been looking for an introduction to LLVM
08:23:04 <asiekierka> more esolangs should have LLVM versions
08:23:06 <Ilari> Which means last digits of the power two is raised to must be congruent to 12, 16, 20, 36, 40, 48, 56, 60, 64, 72, 88 or 96 modulo 100.
08:23:21 <kfr> Originally I wanted to make a self hosting compiler, just for fun
08:23:24 <kfr> In some dumb language
08:23:38 <kfr> http://siyobik.info/misc/esolang/prime.html something like that
08:24:02 <asiekierka> i want to make FerNANDo compile to LLVM for some reason
08:25:33 <asiekierka> also Lost Kingdom in C would be quite cool
08:25:56 <asiekierka> especially as you can see by the code there's space for mass optimization
08:26:24 <kfr> But if I were going to make something in LLVM I'd go for a serious language
08:26:27 -!- poiuy_qwert has quit (Ping timeout: 255 seconds).
08:26:34 <kfr> So perhaps no Unicode then
08:26:43 <kfr> Although Unicode can really make programming more fun, hahaha
08:27:07 -!- poiuy_qwert has joined.
08:28:15 <kfr> I said serious language
08:28:20 <kfr> I also said that I'd make my own
08:28:49 <kfr> What do you mean by too far?
08:29:05 <kfr> Because I'll never finish it?
08:29:20 <asiekierka> unless you get a team of a few people the chances are quite low
08:29:22 <kfr> Yeah I've started stuff like that at one or two points in the past
08:29:33 <kfr> And I never got anywhere
08:29:37 <oerjan> ^ul ((::**)~^)(<-- Initial number as sequence of encoded digits each followed by ~^)!((((:((0)(!(1)(!(2)(!(3)(!(4)(!(5)(!(6)(!(7)(!(8)(!9(_))))))))))(^))~*^^!S)()!(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a()!(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)()!~a((, )S:^)**^):^
08:29:39 <fungot> 3, 13, 1113, 3113, 132113, 1113122113, 311311222113, 13211321322113, 1113122113121113222113, 31131122211311123113322113, 132113213221133112132123222113, 11131221131211132221232112111312111213322113, 31131122211311123113321112131221123113111231121123222113, 13211321322113311213212312311211131122211 ...out of time!
08:30:11 <oerjan> i'm just checking that my reformatting for the wiki didn't break anything
08:30:50 <kfr> primeCheck iterator number ⇾ ⌥ = iterator number 1 ⌥ = 0 ∣ number iterator 0 ↺ + iterator 1 number ◼
08:30:53 <kfr> Unicode rawr
08:31:08 <kfr> I just think it's funny to make a self hosting compiler for an esoteric language
08:31:14 <kfr> It's probably an annoying task though
08:31:21 <kfr> Especially when the language is primitive
08:31:28 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
08:32:02 -!- poiuy_qwert has joined.
08:34:45 <oerjan> Ilari: do you know if the problem has actually been solved?
08:35:34 <oerjan> when it can get up to the 12th digit from the end it looks like brute force might have some trouble
08:37:33 <Ilari> I don't know a solution...
08:39:32 <oerjan> perhaps one could show that the number of digits at the end that are not 1,2,4,8 can be arbitrarily large
08:40:35 <oerjan> say if for every k, all numbers (mod 10^k) that are divisible by 2^k must repeat
08:40:35 <oerjan> or some suitably large fraction
08:40:36 <Ilari> But if such number (power of two not containing 1, 2, 4 nor 8) exists, it must be at least 100 000 digits long...
08:41:21 <oerjan> yeah i think it's heuristically unlikely it exists
08:42:40 <oerjan> and if there is no (mod 10^k) that can be brute forced to show a cycle with no exceptions in it, then it might be among those problems that current mathematics cannot handle
08:43:40 <Ilari> The cycle length seems to go up exponentially when k is increased...
08:44:14 <oerjan> yes, but how long? is it on the order of 5^k?
08:45:00 <oerjan> (that's afaict the approximate maximum)
08:45:28 <Ilari> Something like that...
08:45:52 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
08:46:50 <oerjan> oh hm (mod 2^k) it must reach 0, so (mod 5^k) is all the variation possible
08:48:01 <oerjan> and it's relatively prime to 5, so it must be within 4*5^(k-1) multiplicative subgroup
08:49:12 <Ilari> At least 200 000 digits...
08:49:40 <oerjan> what is that 4 element subgroup, i wonder
08:49:59 <Ilari> As said, the number must be power of 16...
08:50:16 -!- impomatic has joined.
08:50:21 <oerjan> yes. could it be that 16 generates the 5^(k-1) subgroup...
08:51:06 <oerjan> hm that's among the elements that are 1 (mod 5)
08:54:31 <oerjan> hm actually the subset (mod 10^k) with the properties == 1 (mod 5) and == 0 (mod 2^k) might not necessarily contain elements without 1,2,4,8...
08:57:38 <Ilari> If the power 2 is raised to is not divisible by 4, then the last digit is going to inevitably be 2, 4 or 8.
08:59:07 <Ilari> IIRC, considering 2 digits won't help any. 3 digits yields some new restrictions in complicated-looking pattern. 4 is presumably even worse.
09:03:28 <Ilari> >352 200 digits...
09:05:39 <oerjan> > let k = 5 in filter (("1248" `notElem`) . show) [0, 2^k .. 10^k-2^k]
09:05:40 <lambdabot> Couldn't match expected type `[GHC.Types.Char]'
09:06:08 <oerjan> > let k = 5 in filter (all (`notElem` "1248") . show) [0, 2^k .. 10^k-2^k]
09:06:10 <lambdabot> [0,96,576,736,960,3360,3776,3936,5056,5376,5536,5600,5696,5760,6336,6560,66...
09:06:41 <oerjan> > let k = 10 in filter (all (`notElem` "1248") . show) [0, 2^k .. 10^k-2^k]
09:06:43 <lambdabot> [0,39936,65536,66560,70656,75776,357376,373760,399360,500736,536576,537600,...
09:07:55 <oerjan> > let k = 20 in filter (all (`notElem` "1248") . show) [0, 2^k .. 10^k-2^k]
09:07:56 <lambdabot> [0,509607936,577765376,603979776,655360000,766509056,770703360,797966336,30...
09:08:16 <Ilari> Basically, there doesn't seem to be much hope for showing an explicit example... Since any such thing must be above 2^(10^6)...
09:08:40 <oerjan> > let k = 30 in filter (all (`notElem` "1248") . show) [0, 2^k .. 10^k-2^k]
09:08:42 <lambdabot> [0,3505767055360,3703335550976,6033355309056,6039797760000,6065567563776,65...
09:09:13 <Ilari> If there isn't such number, showing some lower bound where all numbers start to have 1, 2, 4 or 8 might work...
09:09:38 <oerjan> it looks like it's not very rare to avoid 1,2,4,8 in that modulus subset
09:11:12 <oerjan> so my guess is this problem is impossibly hard
09:16:27 <Ilari> But it also seems heuristically that it is exeedingly unlikely for such number to exist...
09:17:02 <oerjan> yes, that's one side of the hard part
09:17:29 <oerjan> it's unlikely for it to exist, and it's unlikely for its non-existence to be easily proved
09:19:12 -!- fxkr has quit (Quit: Reconnecting).
09:19:17 -!- fxkr has joined.
09:20:35 <Ilari> This problem arises in context of minimal substring subset of powers of two (thought to be {1, 2, 4, 8, 65536}). For primes, the corresponding set is known to consist of 26 primes (those primes are known, but I can't right now find the page that had those listed).
09:21:13 <oerjan> i recall you mentioned that before
09:22:10 <Ilari> Some of those primes: 2, 3, 7, 11, 19, 31...
09:22:12 <oerjan> and that for those it _was_ a matter of checking moduli, i think
09:22:45 <oerjan> you're going to need 5 in there
09:23:38 <Ilari> In fact, I think those primes are the 26 first primes that don't contain lesser primes of the set as substrings...
09:24:31 -!- TLUL has quit (Quit: *disappears in a puff of orange smoke*).
09:24:43 <oerjan> hm i think such a set is unique if it exists
09:25:17 -!- TLUL has joined.
09:25:21 <Ilari> And also always finite.
09:25:44 <oerjan> you have to include all the shortest ones, and then each next shortest is either necessary or unnecessary by induction
09:26:02 -!- hagb4rd has quit (Ping timeout: 240 seconds).
09:26:20 <oerjan> hm i think you mentioned that too, but that was sort of a ramsey theory thing wasn't it, not obvious
09:26:44 <oerjan> at least i think i thought about it a bit and couldn't see an obvious argument
09:28:12 <oerjan> (or a non-obvious one, since i don't know the theory)
09:28:13 <Ilari> Heh... There is context-sensitive grammar for prime numbers... Wonder what sort of mess it actually is...
09:28:38 <oerjan> well since context-sensitive can calculate anything in linear space...
09:29:18 <Ilari> There's also regular expression for Luhn numbers. A total mess, but sure it exists...
09:32:39 <Ilari> And no, not one of those extended kins of "regular expressions", but the basic kind with only empty string matcing, symbol matching, alternation, concatenation and kleene closure.
09:33:03 -!- cheater- has quit (Ping timeout: 240 seconds).
09:33:54 -!- cheater- has joined.
09:34:53 <Ilari> Oh yeah, I actually consturcted such regular expression. Only 16 587 604 bytes...
09:36:53 <Ilari> Dunno if shorter regular expression exists...
09:37:21 <oerjan> um it's obviously given by a finite automaton, no? so clearly exists.
09:37:40 <Ilari> I think it requires 20 states.
09:38:18 <oerjan> 2 for where you are, times 10 for the modulus
09:53:44 -!- Slereah has quit (Ping timeout: 255 seconds).
09:59:45 -!- Slereah has joined.
10:16:14 -!- asiekierka has quit (Read error: Operation timed out).
10:20:15 <cheater-> anyone know what's up with oklopol?
10:21:45 <Ilari> Oh, I think I figured out what would be the technical downfall of IPv6. :->
10:27:02 <cheater-> using ip numbers as UUIDs for something that there is a lot of?
10:27:15 <cheater-> like, say, line numbers in source code?
10:33:53 <Ilari> Nope. 128 bits is certainly enough for addresses. However, 16 bits might not be enough for payload length...
10:40:29 -!- asiekierka has joined.
10:40:54 <Ilari> Okay, Lagerholm seems to be trying to change IANA prediction banner to first RIR depletion banner. Except that it is still quite buggy (now it shows June 2012).
10:41:25 -!- hagb4rd has joined.
10:46:41 <Ilari> The guy behind http://www.ipv4depletion.com/
10:48:53 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.16/20101130074636]).
10:56:38 -!- Ilari has quit (Ping timeout: 245 seconds).
10:57:28 -!- Ilari_antrcomp has quit (Ping timeout: 260 seconds).
10:57:41 -!- TLUL has quit (Quit: *disappears in a puff of orange smoke*).
10:58:08 -!- asiekierka has quit (Read error: Operation timed out).
11:02:54 -!- Ilari has joined.
11:03:30 -!- Ilari_antrcomp has joined.
11:09:38 -!- Ilari has quit (Quit: Reconnecting).
11:12:03 -!- Ilari_antrcomp has quit (Ping timeout: 245 seconds).
11:14:43 -!- Phantom_Hoover has joined.
11:16:38 -!- hagb4rd has quit (Ping timeout: 240 seconds).
11:24:23 -!- myndzi has joined.
11:26:07 <fizzie> Ilari: Re "16 bits might not be enough for payload length", it does have those jumbograms. (Using a separate option header with a 32-bit payload length.)
11:26:14 <fizzie> Oh, he's gone somewhere.
11:26:38 -!- cheater- has quit (Ping timeout: 240 seconds).
11:27:55 -!- cheater- has joined.
11:38:28 -!- Ilari has joined.
11:39:03 -!- aloril has quit (Ping timeout: 248 seconds).
11:39:46 -!- asiekierka has joined.
11:40:50 <fizzie> <fizzie> Ilari: Re "16 bits might not be enough for payload length", it does have those jumbograms. (Using a separate option header with a 32-bit payload length.)
11:41:08 <asiekierka> what would be a sane memory size to give for Lost Kingdom?
11:41:17 <asiekierka> as i want to try something and i need to know that
11:42:55 <fizzie> Perhaps you could measure it by running it under some conventional interpreter made to keep statistics on tape use.
11:44:06 <asiekierka> it doesn't go over 1024 but better be safe than sorry
11:44:47 <asiekierka> essentially i'm trying to recompile Lost Kingdom with LLVM
11:46:11 -!- Ilari has quit (Read error: Connection reset by peer).
11:46:18 <asiekierka> it took over an hour the first time i tried it
11:48:14 <asiekierka> add a for loop in place of a certain array
11:48:33 -!- impomatic has joined.
11:48:44 -!- hagb4rd has joined.
11:51:12 <asiekierka> not bad seeing as there was 70KB of BF code originally
11:51:25 -!- cheater- has quit (Ping timeout: 240 seconds).
11:51:54 <impomatic> Created my first page on Wikipedia... I wonder if it'll survive or be deleted due to lack of notability :-(
11:52:29 -!- aloril has joined.
11:52:35 -!- cheater- has joined.
11:54:01 <asiekierka> so i'm uploading the recompiled Adv (the first BF interactive fiction game)
11:54:35 <asiekierka> bf.cpp - the brainfuck->LLVM recompiler I found on http://www.remcobloemen.nl/2010/02/brainfuck-using-llvm/, modded so it outputs assembly code without other information
11:55:04 <asiekierka> adv.asm - the LLVM IR code you can compile to bitcode with llvm-as and then to either x86 machine code or C code with llc
11:55:37 <asiekierka> http://64pixels.org/adv-recompile.zip for the files
11:59:36 <kfr> What is with this channel's obsession with BF?
11:59:59 <fizzie> There's quite a little BF here, really.
12:00:06 <oerjan> kfr: um i don't think we've discussed it much in a while
12:00:06 <asiekierka> i want to learn LLVM and code "recompilers" for other esolangs
12:00:26 <oerjan> fizzie: i'm not sure that sentence means what you think it means
12:00:34 <kfr> Tons of the eso languages I've seen in here seem to be strongly inspired by BF though
12:00:36 <kfr> If only syntactically
12:00:42 <oerjan> (try removing the "a")
12:01:05 <fizzie> Well, BF *is* pretty famous, after all.
12:01:17 <oerjan> kfr: it _is_ the most famous esolang. and yes there are too many derivatives of it.
12:02:13 <asiekierka> anyone knows any other cell-based esolang which is not brainfuck nor a deriative
12:04:10 <asiekierka> well pbrain is the only kind-of-used deriative i believe
12:04:22 <fizzie> Malbolge even has cells in a tape shape, though the derivativeness is arguable.
12:05:05 <kfr> Self hosting turing machine compiler hm
12:05:53 -!- Ilari has joined.
12:06:09 <asiekierka> well i am still considering underlad even though it has a stack
12:06:45 <fizzie> Also anything that has the usual sort of random-accessable memory could be considered to have "cells".
12:07:28 * oerjan ponders if smetana has cells
12:08:04 <kfr> How are functions as first class objects usually passed around? Is it a pointer to an object which basically contains a function pointer and a container with partially applied arguments?
12:08:43 <oerjan> well that's an obvious way
12:09:05 <kfr> I wonder how ghc deals with polymorphism
12:09:20 <kfr> Does it create one new function for every tuple (t_1, ..., t_n) of type variables?
12:10:33 <oerjan> it uses a boxed representation so a function does not need to look at arguments it doesn't know the type of
12:10:41 <asiekierka> the BF factorial app after recompilation is quite fast
12:10:44 <asiekierka> (this one: http://www.muppetlabs.com/~breadbox/bf/factor.b.txt )
12:10:58 <kfr> Unboxing slows it down though, no?
12:11:11 <kfr> In C++ a template will generate a separate function for each I think
12:11:36 <asiekierka> while the recompiled is "instant" for a number like 1239083483412
12:11:44 <asiekierka> an interpreted variant with a simple interpreter takes 7-8 seconds
12:11:50 <oerjan> kfr: it will sometimes specialize functions that use type classes, i believe (there's a pragma for it)
12:12:42 <oerjan> kfr: note that it is quite possible in ghc to write a program that uses a function at an _infinite_ number of types (polymorphic recursion)
12:13:03 <kfr> I thought ghc didn't permit that :O
12:13:25 <kfr> You can stack up transformers?
12:13:29 <oerjan> so they cannot always be compiled in advance
12:13:31 <kfr> Or something like that?
12:13:49 <Phantom_Hoover> Today's \sqrt{-Garfield} was the funniest in a long while.
12:14:14 <oerjan> i vaguely recall the jhc compiler tries to compile functions separately for separate types
12:15:57 <oerjan> ghc has many primitive functions with unboxed types though, and will optimize others to them
12:16:10 -!- Tritonio has joined.
12:17:13 <oerjan> kfr: doesn't permit what?
12:17:28 <kfr> I was referring to a different issue, I believe
12:17:34 <kfr> Stuff like "a = [a]
12:17:38 <oerjan> Phantom_Hoover: BUT WILL IT END THE MEME?
12:17:46 <kfr> You know, infinite type error message
12:17:51 <kfr> During compilation
12:18:16 <kfr> http://www.mezzacotta.net/garfield/?comic=625 <- I read that btw and I totally didn't get it lol
12:18:21 <oerjan> kfr: oh that. you cannot do that directly but you can work around it with newtype declarations
12:18:54 <kfr> You need a workaround like that to make your own Y combinator in Haskell, like that?
12:19:04 <kfr> I've seen a version which did it using some hackery/pragmas
12:19:21 <oerjan> polymorphic recursion is different though, that's when a function calls itself indirectly with arguments of different types than the original ones. it requires a type annotation.
12:19:39 <oerjan> it's sometimes useful.
12:20:13 <oerjan> yes you can use a newtype to make the Y combinator
12:20:22 <Phantom_Hoover> kfr, there is an *extremely* pervasive meme on SRoMG of taking a particular comic and replacing Garfield in the middle panel with something that sounds a bit like 'minus'.
12:21:00 <kfr> Terminator sounds like minus?
12:21:24 <oerjan> kfr: well this is a meta-instance of the meme
12:21:40 -!- impomatic has left (?).
12:23:22 <oerjan> at least one guy posted to the iwc forum recently that he was stopping reading SRoMG because of the meme
12:23:38 <Sgeo> Linus is the only one I get
12:23:44 -!- Sgeo_ has joined.
12:23:45 -!- Sgeo_ has quit (Client Quit).
12:24:39 <oerjan> i don't think dame jane goodall sounds like minus, but she's of course obligatory for other inside reasons
12:27:21 <oerjan> lessee, y f = (\x -> f (x x)) (\x -> f (x x)) which doesn't type because of the infinite types rule of course
12:27:26 -!- pikhq has quit (Ping timeout: 240 seconds).
12:27:33 -!- pikhq has joined.
12:29:19 <kfr> f = fixH (Roll fixH) where {-# NOINLINE fixH #-} fixH x = f ((unroll x) x)
12:29:29 <kfr> fix f = fixH (Roll fixH) where {-# NOINLINE fixH #-} fixH x = f ((unroll x) x)
12:29:36 <kfr> http://r6.ca/blog/20060919T084800Z.html lol
12:32:47 <oerjan> !haskell newtype Self a = Self { runSelf :: Self a -> a }; y f = (\x -> f (runSelf x x)) $ Self (\x -> f (runSelf x x)); y :: (a -> a) -> a; main = print . take 10 $ y show
12:33:17 -!- Ilari_ has joined.
12:33:20 <EgoBot> "\"\\\"\\\\\\\"\\\\\\"
12:34:55 <oerjan> maybe SelfApply would be a clearer name
12:35:17 <asiekierka> 30 minutes for bf.cpp to handle Lost Kingdom
12:38:39 -!- Ilari_ has changed nick to Ilari_antrcomp.
12:39:52 <asiekierka> Lost Kingdom compiled to a recompiled, native version
12:52:13 -!- copumpkin has quit (Ping timeout: 240 seconds).
12:52:51 -!- copumpkin has joined.
12:54:53 -!- BeholdMyGlory has joined.
12:56:13 -!- cheater- has quit (Ping timeout: 240 seconds).
12:56:59 -!- cheater- has joined.
13:00:25 -!- hagb4rd has quit (Ping timeout: 240 seconds).
13:00:50 -!- FireFly has joined.
13:07:13 -!- MigoMipo has joined.
13:12:45 -!- asiekierka has quit (Ping timeout: 246 seconds).
13:13:19 -!- hagb4rd has joined.
13:27:24 -!- hagb4rd has quit (Ping timeout: 240 seconds).
13:31:08 -!- hagb4rd has joined.
13:40:37 -!- hagb4rd has quit (Ping timeout: 240 seconds).
13:49:42 -!- augur has quit (Remote host closed the connection).
13:51:10 -!- elliott has joined.
13:51:16 <elliott> 02:27:55 <oerjan> <m00barax> i'm an esotermik programmers
13:51:18 <elliott> 02:28:12 <oerjan> beware of esothermic programmers exploding
13:51:20 <elliott> 02:28:43 <oerjan> wait that's exothermic
13:51:22 <elliott> 02:29:04 • oerjan watches his pun stretch to the breaking point, then snap
13:51:26 <elliott> 01:10:14 <m00barax> hi, exothermix
13:52:42 <elliott> oerjan: it's funny, he came in speaking incoherently and mentioning UFOs and prime numbers and i feared the worst
13:52:49 <elliott> but then he turned out to be a Clean programmer with really bad English
13:53:41 <elliott> oerjan: hm what's look and say in haskell
13:53:47 <oerjan> Unmaintable Functional Objects
13:54:02 <elliott> since you did it in underload :P
13:54:07 -!- freakyfractal_ has joined.
13:54:39 <lambdabot> forall a. (a -> Bool) -> [a] -> [a]
13:55:11 <elliott> > let look [] = []; look (x:xs) = length xs' : xs' ++ look xs where xs' = takeWhile (==x) xs in look [1,2,3]
13:55:38 <oerjan> > iterate ((sequence [show . length, take 1] =<<). group) "3"
13:55:38 <lambdabot> Couldn't match expected type `[GHC.Types.Char]'
13:56:22 <elliott> > let look [] = []; look (x:xs) = let xs' = takeWhile (==x) xs in length xs' : xs' ++ look xs in look [1,2,3]
13:56:34 <elliott> @hoogle (a->Bool)->[a]->([a],[a])
13:56:34 <lambdabot> Prelude break :: (a -> Bool) -> [a] -> ([a], [a])
13:56:34 <lambdabot> Prelude span :: (a -> Bool) -> [a] -> ([a], [a])
13:56:34 <lambdabot> Data.List break :: (a -> Bool) -> [a] -> ([a], [a])
13:57:04 <elliott> > let look [] = []; look (x:xs) = let (xs',xs'') = span (==x) xs in length xs' : x ++ look xs'' in look [1,2,3]
13:57:05 <lambdabot> No instance for (GHC.Num.Num [GHC.Types.Int])
13:57:24 <elliott> > let look [] = []; look (x:xs) = let (xs',xs'') = span (==x) xs in ((length xs') : x) ++ look xs'' in look [1,2,3]
13:57:25 <lambdabot> No instance for (GHC.Num.Num [GHC.Types.Int])
13:58:25 <Ilari> Doesn't that webcast begin in something like 15 minutes?
13:58:28 <oerjan> > iterate ((concat . sequence [show . length, take 1] =<<). group) "3"
13:58:30 <lambdabot> ["3","13","1113","3113","132113","1113122113","311311222113","1321132132211...
13:59:05 <elliott> 08:23:57 <oerjan> > filter (> 5) . map (fromJust . findIndex (`elem` "1248") . reverse . show) $ iterate (*16) 1048576
13:59:05 <elliott> 08:24:00 <lambdabot> [7,7,7,12,7,7,6,8,6,6,7,7,8,7,7,7,7,6,8,6,6,9,8,6,7,7,6,8,6,6,6,6,6,6,6,6,6...
13:59:07 <elliott> 08:24:06 <fizzie> I see you're familiar with the engineer's method of proving mathematical statements.
13:59:22 <elliott> oerjan: ok now tell me why mine doesn't work
13:59:51 <Ilari> AFRINIC got 102, APNIC got 103, ARIN got 104, LACNIC got 179, RIPE got 185.
14:00:58 <elliott> 08:58:50 <oerjan> and if there is no (mod 10^k) that can be brute forced to show a cycle with no exceptions in it, then it might be among those problems that current mathematics cannot handle
14:01:06 <Ilari> Oh, the inetcore exhaustion counter got updated...
14:01:13 <elliott> oerjan: erd\Hos reference?
14:02:28 <oerjan> > iterate (sequence [show . length, take 1] <=< group) "3"
14:02:29 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:02:55 <elliott> [on Collatz] Because of the difficulty in solving this problem, Erdős commented that "mathematics is not yet ready for such problems" (Lagarias 1985).
14:03:03 <oerjan> > iterate (concat . sequence [show . length, take 1] <=< group) "3"
14:03:05 <lambdabot> ["3","13","1113","3113","132113","1113122113","311311222113","1321132132211...
14:03:10 <elliott> oerjan: reference to that? :p
14:04:01 <elliott> oerjan: what was the number where arithmetic breaks down again according to you? 10^40? :p
14:04:26 <oerjan> um i think i picked it up from someone else
14:04:39 <elliott> oerjan: I blame Zeilberger
14:04:47 <elliott> 07.08.03:14:06:22 <oerjan> numbers above 10^40 might be inconsistent...
14:05:01 <elliott> Zeilberger is a strange guy :)
14:06:47 <elliott> "While I certainly hope that there does not exist a purely human, machine-free, proof of the Four Color Theorem (if there is, it would mean that this theorem is, a posteriori, trivial!), [...]"
14:07:36 <lambdabot> 100000000000000000000000000000000000000000000000000000000000000000000000000...
14:07:42 <elliott> oerjan: maybe there isn't a 9 at the end
14:07:45 <elliott> maybe Cthulhu is there instead
14:08:31 <elliott> 14:12:35 <cherez> I'm pretty sure the halting problem is provably unsolvable.
14:08:32 <elliott> 14:12:45 <cherez> At least in the most general sense.
14:08:32 <elliott> 14:12:45 <bsmntbombdood> I'm pretty sure it's been proven
14:08:54 <elliott> 14:14:58 <bsmntbombdood> the max shifts function is computable on machines with finite memorys
14:09:34 <Ilari> elliott: Yes, IANA is fully depleted.
14:09:41 <elliott> Ilari: can we have a party now???
14:10:02 <Ilari> http://www.iana.org/assignments/ipv4-address-space/ipv4-address-space.txt
14:10:11 <elliott> we need party hats for a party
14:10:21 <elliott> 010/8 IANA - Private Use 1995-06 RESERVED [3]
14:10:38 -!- asiekierka has joined.
14:10:53 <elliott> Ilari: well, IIRC Harvard said they'd return their /8. I suspect MIT might too.
14:11:05 <elliott> possibly a few of the corporations as well.
14:11:20 <elliott> Ilari: What are those "future use" /8 blocks at the end?
14:11:29 <elliott> Surely those will end up allocated? Well, not 255/8, but the previous ones?
14:11:39 <elliott> Or are those reserved for IPv6 transition?
14:12:22 <oerjan> > let look [] = []; look (x:xs) = let (xs',xs'') = span (==x) xs in length xs' : x : look xs'' in look [1,2,3]
14:12:35 <oerjan> I SENSE AN OFF BY ONE ERROR
14:12:39 <elliott> hey how did you get that working oerjan
14:12:44 <elliott> why are you a bad person and how did you get that working
14:12:52 <oerjan> you had a ++ where you needed a :
14:13:04 <Ilari> elliott: Those "future use" blocks will never be used.
14:13:19 <Ilari> elliott: Too much hardware/software does not like using them.
14:13:29 <elliott> > let look [] = []; look (x:xs) = let (xs',xs'') = span (==x) xs in length xs' + 1 : x : look xs'' in (iterate look [1]) !! 3
14:13:30 <oerjan> > let look [] = []; look (xs@(x:_)) = let (xs',xs'') = span (==x) xs in length xs' : x : look xs'' in look [1,2,3]
14:14:00 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==) xs in look [1,2,1,1]
14:14:01 <lambdabot> Couldn't match expected type `GHC.Bool.Bool'
14:14:11 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in look [1,2,1,1]
14:14:32 <elliott> oerjan: just don't give it any reason to use double-digit counts :D
14:14:41 <elliott> oerjan: hm that's like "base infinity" look-and-say
14:15:13 <oerjan> my underload program doesn't deal very well with double-digit counts either
14:15:15 <elliott> oerjan: is it interesting in any way? :p
14:15:17 <Ilari> The new inetcore depletion counter seems to be counting down at about 150 IPs per minute...
14:15:34 <Ilari> And it is currently at ~50M IPs.
14:15:50 <oerjan> elliott: all bases >= 4 are nearly the same once they whittle down to length <= 3 repetitions
14:15:59 <kfr> Ilari why do you obsess about that stuff? :p
14:16:01 <elliott> OH SURE OERJAN, ALWAYS WITH THE EXCUSES
14:16:05 <elliott> oerjan: but this can do regular look and say right?
14:16:28 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in iterate (concatMap show) [1]
14:16:29 <lambdabot> No instance for (GHC.Num.Num GHC.Types.Char)
14:16:35 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in iterate (concatMap show . look) [1]
14:16:36 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:16:38 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in iterate (concatMap show . look) [[1]]
14:16:39 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:16:41 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in iterate (concatMap show . look) [1]
14:16:42 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:17:34 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in iterate (show . look) [1]
14:17:35 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:17:45 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in concatMap show . iterate look $ [1]
14:17:47 <lambdabot> "[1][1,1][2,1][1,2,1,1][1,1,1,2,2,1][3,1,2,2,1,1][1,3,1,1,2,2,2,1][1,1,1,3,...
14:17:54 <elliott> > let look [] = []; look xs@(x:_) = length xs' : x : look xs'' where (xs',xs'') = span (==x) xs in map (concatMap show) . iterate look $ [1]
14:17:56 <lambdabot> ["1","11","21","1211","111221","312211","13112221","1113213211","3113121113...
14:18:15 <elliott> oerjan: that's the regular sequence, right?
14:19:52 <elliott> now what's that as a foldr
14:21:04 <elliott> oerjan: yeah but it's not point-free!
14:21:20 <oerjan> ...i already did a point-free one
14:21:36 <elliott> oerjan: ...yeah but did it actually work?
14:21:58 <oerjan> > iterate (concat . sequence [show . length, take 1] <=< group) "1"
14:21:59 <lambdabot> ["1","11","21","1211","111221","312211","13112221","1113213211","3113121113...
14:22:44 <oerjan> > iterate (concat . sequence [length, take 1] <=< group) [1]
14:22:45 <lambdabot> Couldn't match expected type `[c]'
14:22:45 <lambdabot> against inferred type `GHC.Types...
14:23:25 <oerjan> > iterate (sequence [length, take 1] <=< group) [1]
14:23:26 <lambdabot> Couldn't match expected type `GHC.Types.Int'
14:23:52 <oerjan> > iterate (sequence [length, head] <=< group) [1]
14:23:53 <lambdabot> [[1],[1,1],[2,1],[1,2,1,1],[1,1,1,2,2,1],[3,1,2,2,1,1],[1,3,1,1,2,2,2,1],[1...
14:24:21 <oerjan> finally got read of that concat too
14:25:55 <lambdabot> sequence (x:xs) = do v <- x; vs <- sequence xs; return (v:vs)
14:25:55 <lambdabot> sequence xs = foldr (liftM2 (:)) (return []) xs
14:26:07 <elliott> oerjan: surely you can eliminate that sequence there
14:26:11 <lambdabot> No instance for (GHC.Show.Show (m [a]))
14:26:11 <lambdabot> arising from a use of `M20581693...
14:26:18 <elliott> oh god i fucking hate you lambdabot
14:27:44 <oerjan> well i could use some liftM2 in the list monad instead but i don't think it would be prettier
14:32:10 <oerjan> @hoogle m x -> m (x -> y) -> m y
14:32:10 <lambdabot> Control.Applicative (<**>) :: Applicative f => f a -> f (a -> b) -> f b
14:32:10 <lambdabot> Control.Applicative (<*>) :: Applicative f => f (a -> b) -> f a -> f b
14:32:10 <lambdabot> Control.Monad ap :: Monad m => m (a -> b) -> m a -> m b
14:32:26 <elliott> x and y in types? are you a monster?
14:32:38 <elliott> > iterate ((length <**> head) <=< group) [1]
14:32:39 <lambdabot> Occurs check: cannot construct the infinite type:
14:32:56 <lambdabot> forall (m :: * -> *) a. (Monad m) => [m a] -> m [a]
14:32:57 <oerjan> > iterate ((<**>) [length, head] . group) [1]
14:32:57 <lambdabot> Couldn't match expected type `([GHC.Types.Int] -> GHC.Types.Int)
14:33:02 <elliott> @hoogle m a -> m a -> m [a]
14:33:02 <lambdabot> Text.ParserCombinators.ReadP endBy :: ReadP a -> ReadP sep -> ReadP [a]
14:33:02 <lambdabot> Text.ParserCombinators.ReadP endBy1 :: ReadP a -> ReadP sep -> ReadP [a]
14:33:02 <lambdabot> Text.ParserCombinators.ReadP sepBy :: ReadP a -> ReadP sep -> ReadP [a]
14:33:18 <oerjan> > iterate ((<**> [length, head]) . group) [1]
14:33:19 <lambdabot> [[1],[1,1],[2,1],[1,2,1,1],[1,1,1,2,2,1],[3,1,2,2,1,1],[1,3,1,1,2,2,2,1],[1...
14:33:40 <elliott> it's the two-element list that's irksome
14:33:56 <oerjan> huh i thought that's what makes it nice
14:33:57 -!- augur has joined.
14:34:47 <elliott> oerjan: length <@> head <=< group would be nicer though :P
14:35:23 -!- cheater- has quit (Ping timeout: 240 seconds).
14:35:42 -!- poiuy_qwert has joined.
14:35:44 <elliott> :t let (f <@> x) g = (f <=< g) <*> x in (<@>)
14:35:44 <lambdabot> forall b a b1 a1. (b -> a -> b1) -> (a1 -> a) -> (a1 -> a -> b) -> a1 -> b1
14:35:50 <elliott> :t let (f <@> x) g = (f <=< g) <**> x in (<@>)
14:35:50 -!- freakyfractal_ has left (?).
14:35:50 <lambdabot> forall b (m :: * -> *) c a b1. (Monad m) => (b -> m c) -> (a -> m c -> b1) -> (a -> m b) -> a -> b1
14:35:59 -!- cheater- has joined.
14:36:09 <elliott> especially that monad-taking, non-monad-returning one
14:37:43 <elliott> :t let (f <@> x) g = (f <=< g) <**> (g x) in (<@>)
14:37:43 <lambdabot> forall a c b. (((a -> c) -> b) -> a -> c) -> a -> (a -> a -> (a -> c) -> b) -> a -> b
14:38:15 <elliott> lambdabot needs a quote command like @v but for ridiculous types
14:39:25 -!- poiuy_qwert has quit (Client Quit).
14:41:20 <oerjan> quack are you talking about
14:41:44 <elliott> haha i heard that as an actual quack, that's proof that oerjan is a duck
14:42:36 <oerjan> well if it fits the bill
14:43:12 <elliott> 12:05:03 <asiekierka> but on the other hand it's quite fast
14:43:12 <elliott> 12:07:16 <asiekierka> 43KB for the compiled executable
14:43:13 <elliott> 12:07:25 <asiekierka> not bad seeing as there was 70KB of BF code originally
14:43:17 <elliott> last i checked lostkng was 2 megs...
14:43:28 <elliott> 12:10:13 <asiekierka> so i'm uploading the recompiled Adv (the first BF interactive fiction game)
14:50:54 <Sgeo> The favored social networking sites for stoners
14:56:35 <elliott> yes, it's hilariously awful
14:56:37 <elliott> i don't know why i know this
14:56:50 <elliott> Phantom_Hoover: http://2.bp.blogspot.com/_Om8bnRjTT6s/SsyAEgFtLzI/AAAAAAAAAxE/yeVNx4zcGyw/s400/pr_Dennis_Menace.gif
14:56:55 <Phantom_Hoover> And it was created 5 days before the true Dennis the Menace?
14:57:04 <elliott> Phantom_Hoover: LOOK HOW MENACING HE IS
14:58:41 <oerjan> wait there is a _british_ dennis the menace?
14:58:45 <fizzie> Dislike button: beamer's \pause (and other overlay-spec commands) do not work at all inside amsmath 'align' and friends, only in plain, ugly eqnarray.
14:58:46 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
15:00:03 <fizzie> Norwegians don't cackle, they just burn churches.
15:00:21 <oerjan> sheesh, i'm not varg vikernes
15:01:25 <Phantom_Hoover> I dimly recall having to battle with Human on RationalWiki to get that to stay in an article on something.
15:01:43 <fizzie> Curiously enough, Google image search for "norway church" returns only unburned churches. There must be something wrong with their algorithm.
15:01:50 <Phantom_Hoover> I have a depressing feeling the attitude was "burning churches MUST be good: religion is EVIL!"
15:02:35 -!- elliott has quit (Ping timeout: 240 seconds).
15:02:42 <fizzie> (Okay, 37th result is that Varg guy, and 61st is a newspaper clipping about church-burning.)
15:15:47 -!- Tritonio has quit (Quit: Leaving).
15:26:56 -!- copumpkin has joined.
15:28:10 -!- cheater- has quit (Ping timeout: 240 seconds).
15:29:05 -!- cheater- has joined.
15:40:10 -!- asiekierka has quit (Ping timeout: 240 seconds).
15:44:01 <kfr> Phantom_Hoover: No, actually many people involved with it were essentially theists
15:44:55 <kfr> Vikernes is an atheist but he things pre-Christian Indo-European religions are awesome nonetheless
15:44:58 <kfr> thinks* rather
15:45:59 -!- ais523 has joined.
15:47:53 <oerjan> ^ul ((::**)~^)((((:((0)(!(1)(!(2)(!(3)(!(4)(!(5)(!(6)(!(7)(!(8)(!9(_))))))))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
15:47:55 <fungot> 3, 13, 1113, 3113, 132113, 1113122113, 311311222113, 13211321322113, 1113122113121113222113, 31131122211311123113322113, 132113213221133112132123222113, 11131221131211132221232112111312111213322113, 31131122211311123113321112131221123113111231121123222113, 132113213221133112132123123112111311222112 ...out of time!
15:48:15 <ais523> is that look-and-say? In /Underload/? In /one line of IRC/?
15:48:41 <ais523> I didn't realise the lang was quite that powerful
15:48:46 <ais523> how are you representing numbers?
15:49:10 <oerjan> your "church numerals"
15:49:38 <ais523> I suppose with look-and-say, there's no reason to go beyond 3
15:50:04 <oerjan> indeed i only added up to 9 as a last change
15:50:52 <oerjan> the first string in there is the initial number
15:51:04 <ais523> now I'm trying to work out your list representation
15:51:38 <oerjan> ^ul ((::**)~^(:*)~^)((((:((0)(!(1)(!(2)(!(3)(!(4)(!(5)(!(6)(!(7)(!(8)(!9(_))))))))))(^))~*^^!S)(:a(~^)*~(()(~(~(:a~*):^))(a))~*^^)):^(()~)~**~^(:)~((a(~^)*~**)~a)~a(**~:((:)~(*)**)~a*~(^))**a(~*^^^!!^)***(~)~a(~a*^:)**a(:)**~^!!!a(~^)*~**)~a((, )S:^)**^):^
15:51:39 <fungot> 32, 1312, 11131112, 31133112, 1321232112, 111312111213122112, 311311123112111311222112, 1321133112132112311321322112, 111312212321121113122112132113121113222112, 31131122111213122112311311222112111312211311123113322112, 13211321223112111311222112132113213221123113112221133112132123222112, 111312211 ...out of time!
15:51:53 <ais523> yep, it looks like an argument list to arbitrary functions
15:52:03 <ais523> that type has a name in Perl 6, but I forget what it is
15:53:15 <fizzie> ais523: Apropos yesterday's topic, the Linux netfilter connection-tracking code (to do application-specific NAT like FTP, IRC DCC and so on) supports: UDP-Lite, Amanda (a backup protocol), FTP, H.232, IRC, NetBIOS, PPtP, SANE (a remote-access protocol specifically for scanners), SIP and TFTP.
15:53:27 <fizzie> And the SANE protocol module has been marked "experimental" for as long as I remember.
15:54:04 <ais523> I actually knew what SANE meant in that context
15:54:06 <fizzie> That's not a very big market, SANE scanner users that need to access a remote scanner from behind a NAT, I would think.
15:54:13 <fizzie> Scanner Access Now Easy, isn't it?
15:55:37 <ais523> I don't know what it /stands/ for
15:55:42 <ais523> although that's stupid enough to be plausible
15:58:10 -!- sebbu has quit (Ping timeout: 240 seconds).
16:00:05 <oerjan> the :a(~^)*~(()(~(~(:a~*):^))(a))~*^^ is an important subroutine, it is used on each initial digit in a group to set up the stack elements for handling the group data
16:00:35 <oerjan> including an equality test subroutine
16:02:03 -!- hagb4rd has joined.
16:16:20 -!- sebbu has joined.
16:29:00 <ais523> Underload is so hard to read out of context
16:29:09 <ais523> as in, you can work out what it does at a low level
16:29:22 <ais523> but it often makes no sense at a high level without knowing just what's on the stack at any given point
16:32:20 <oerjan> very well then, you can at least see my scratch file at http://oerjan.nvg.org/esoteric/las.ul
16:34:53 <ais523> hmm, really some sort of literate Underload would be helpful
16:35:12 -!- Zuu_ has changed nick to Zuu.
16:35:13 <ais523> but it'd likely have to be literate Underlambda, or at least use upp, in order to let you get the program in the right order for literate programming
16:35:39 <ais523> ooh, now I see how that digit printing works, that's ingenious
16:35:41 <oerjan> well there is at least the (...)! commenting trick
16:37:31 <oerjan> btw Talk:Monod needs deletion
16:37:35 -!- hagb4rd has quit (Ping timeout: 272 seconds).
16:37:45 -!- cheater- has quit (Ping timeout: 240 seconds).
16:38:40 -!- cheater- has joined.
16:41:18 <ais523> oerjan: indeed, that's clearly meant to be a forum post
16:43:13 <oerjan> the reference to our non-existing avatars was pretty revealing
17:12:33 -!- cal153 has quit (Ping timeout: 240 seconds).
17:17:05 <quintopia> was there a press conference already?
17:40:10 -!- Sgeo_ has joined.
17:42:32 -!- Sgeo has quit (Ping timeout: 240 seconds).
17:57:25 -!- cal153 has joined.
17:58:24 -!- oerjan has quit (Quit: Good night).
18:01:37 -!- elliott has joined.
18:04:36 <elliott> 16:04:31 <ais523> is that look-and-say? In /Underload/? In /one line of IRC/?
18:04:36 <elliott> 16:04:57 <ais523> I didn't realise the lang was quite that powerful
18:04:43 <elliott> ais523: Underload is a classic for a reason :)
18:05:09 <elliott> it's a tarpit in the best sense (any non-trivial program will use every instruction)
18:05:13 <ais523> it's just that I'm aware that simple operations can be quite tricky in Underload sometimes
18:05:22 <ais523> also, I really like that definition of "tarpit"
18:05:57 <elliott> ais523: well, it's an extension to the underlying definition ("few instructions")
18:06:11 <elliott> e.g., I wouldn't call a 100-instruction language a tarpit even if you had to use every instruction in a big program
18:06:19 <elliott> but I definitely think that languages with redundant instructions aren't tarpits
18:06:21 <ais523> that'd be a fun language in its own right, though
18:06:36 <ais523> a lang with 100 instructions, and you couldn't sensibly write a program without using all of them
18:06:39 <elliott> 16:51:56 <ais523> ooh, now I see how that digit printing works, that's ingenious
18:06:56 <elliott> http://oerjan.nvg.org/esoteric/las.ul <-- gosh, oerjan writing an actual program
18:07:28 <ais523> elliott: basically, it's a list made out of cons cells like Underload lists normally are
18:07:40 <ais523> except that the tails contain a command to pop the stack
18:07:59 <ais523> so instead of going "uncons pop" repeatedly to go "tail tail tail tail tail..."
18:08:00 <elliott> http://esolangs.org/wiki/Underload#Examples wow, I didn't realise rule 110 was so simple in underload
18:08:13 <elliott> IMO, Underload is a better esolang than Brainfuck
18:08:20 <ais523> it juts goes "uncons", and the cons cells are vaguely evil and deal with getting rid of the previous heads
18:09:08 * elliott half-suspects that Underload is actually a good language ruined (ruined being a good word here ofc)
18:09:31 <ais523> elliott: and that good language is Underlambda!
18:09:34 <elliott> ("Golf is a good program ruined", so obviously ruined is a good thing for us)
18:09:40 <ais523> although it isn't finished yet
18:09:51 <elliott> ais523: I have this strong urge to write a compiler that does the optimisations derlo does
18:10:10 -!- hagb4rd has joined.
18:11:05 <ais523> elliott: I'm not entirely convinced that all the optimisations make sense in a compiler, but go for it anyway
18:11:16 <ais523> given that derlo's already slower than the existing compiler, I'd imagine that the resulting compiler would be massively fast
18:11:30 <ais523> I could probably optimise a bit more by detecting numbers and special-casing them
18:11:35 <ais523> to make them O(1) rather than O(log n)
18:11:37 <elliott> ais523: derlo's slower than my compiler?
18:11:42 <elliott> or is there another compiler on the block?
18:11:49 <elliott> I mean the resulting code, ofc
18:12:01 <elliott> also, I thought derlo _did_ optimise numbers, that's what i was mainly planning
18:12:01 <ais523> I meant the resulting code
18:12:11 <ais523> and no, it just did things like lazy concat
18:12:17 <elliott> turn any () just consisting of "balanced" :s and *s into a number
18:12:21 <elliott> optimise operations appropriately
18:12:22 <ais523> in order to improve computational orders as much as possible
18:12:34 <ais523> also, you should detect things like !() as numbers too
18:12:52 -!- asiekierka has joined.
18:12:56 <quintopia> oh he finally finished his ul las generator...just after i went to bed last night :P
18:13:07 <ais523> also, hilarious: council want to turn off speed cameras to save money, media complain
18:13:21 <ais523> (hilarious because: council install speed cameras, media accuse them of just trying to make money)
18:13:37 <quintopia> aren't speed cameras supposed to make money? yeah. that's a fail.
18:15:00 <ais523> quintopia: no, they're supposed to slow people down
18:15:29 <quintopia> i've never seen a camera slow anyone down...
18:15:41 <quintopia> the whole purpse of speed laws is to make money
18:15:52 <ais523> quintopia: well, they cost more to run than they make
18:16:28 <ais523> no it isn't, because most people respect them
18:16:40 <ais523> and thus they hardly make any monet
18:17:04 -!- elliott has quit (Ping timeout: 255 seconds).
18:19:24 <quintopia> ah i didn't know most people respect them. i've never known them to. yeah, speed enforcement that actually works p much fails at paying for police to operate
18:19:32 <quintopia> i can understand wanting to shut them off in that case
18:20:08 <ais523> the cameras are made really visible in order that people can see them in a distance and know they should be careful about their speed
18:20:15 <Phantom_Hoover> Yeah, that kind of insane commercialism smacks of America.
18:20:19 <ais523> also, georgia the US state? or georgia the country?
18:21:03 <quintopia> well, the cameras would probably be more profitable if they hid them
18:21:56 <quintopia> i do not understand this culture of creating laws and actually making them work
18:22:06 <quintopia> laws are made to make someone money
18:23:01 <fizzie> I don't know statistics about whether they make a profit or not here, but people do slow down for the cameras. They're also not especially hidden (there's traffic signs at the start of a camera-controlled region and so), but not made especially prominent either. (And you can download unofficial crowdsourced car-navigator-point-of-interest-compatible maps of them.)
18:23:50 <fizzie> Nondescript grey here.
18:24:01 <fizzie> We have a different 'here' here.
18:24:11 <quintopia> answer only for yourself of course
18:24:15 -!- hagb4rd2 has joined.
18:26:01 -!- hagb4rd has quit (Disconnected by services).
18:26:11 -!- hagb4rd2 has changed nick to hagb4rd.
18:30:58 <quintopia> i like to include a bit of sillyness in my write-ups
18:31:15 <quintopia> the whole point of esoteric languages is not to take yourself too seriously
18:31:27 <fizzie> Espoo, next to Helsinki.
18:31:47 <quintopia> let's hang out next time in helsinki :D
18:42:31 -!- hagb4rd has quit (Ping timeout: 240 seconds).
19:02:39 <variable> <@meow> ldd gnome-kitchen-sink | wc -l
19:02:55 <ais523> I didn't realise wc /could/ overflow
19:03:04 -!- poiuy_qwert has joined.
19:03:39 <variable> ais523, I duno if it can. Twas a joke
19:07:21 <ais523> the documentation doesn't mention limits
19:07:32 <variable> Phantom_Hoover, yes | head -20000000 |wc -c
19:07:45 <variable> yes | head -200000000000 |wc -c -> running this now
19:08:18 <Ilari> Haha... Exploding truck falling into Grand Canyon... The Onion is funny.
19:09:02 <quintopia> ilari: is there a transcript of the iana press conference anywhere?
19:09:24 <olsner> what, no elliott either
19:10:05 <Ilari> variable: http://www.youtube.com/watch?v=nJdP1zK15bE
19:10:05 <quintopia> inorite? been a pretty pleasant day so far
19:10:58 -!- ais523 has quit (Read error: Connection reset by peer).
19:11:21 -!- asiekierka has quit (Remote host closed the connection).
19:14:32 <Vorpal> ah they allocated it at last
19:14:49 <Vorpal> Ilari, so... any predictions for when RIPE runs out?
19:18:08 <variable> Vorpal, they won't for variety of reasons
19:18:12 <quintopia> who cares when ripe runs out? grab your IPv4s now while they're still hot! i have 2 in romania myself
19:18:13 <Ilari> One of the official estimates (by Lagerholm) is 2012-09-08 ... But graph done by Hain has RIPE depleting in November this year...
19:18:30 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
19:18:32 <variable> there won't ever be a time when "there are no more" addresses
19:19:13 <quintopia> they will certainly come a time when the kikes of us cannot afford one however
19:19:18 <variable> Vorpal, because businesses give them back and other people are holding on to them to reallocate. The problem is when "there are not currently enough to service a businesses needs"
19:19:41 <variable> Vorpal, basically it might stay < 1000 but it won't hit 0
19:19:46 <Vorpal> variable, well sooner or later we *will* need ipv6.
19:20:00 <variable> (I'm on the IETF and NANOG mailing lists)
19:20:59 <Vorpal> variable, also I strongly dislike the huge allocations of ipv6. heck my ipv6 tunnel has a frigging /48... Something like a /60 or so would be way better, still allowing a few /64 with stateless config
19:21:01 <variable> Vorpal, the big "0" day came and went (when the final big allocations were made and now its up to the regional allocations)
19:21:18 <variable> erm - smaller than /60 have other problems
19:21:26 <variable> the routing tables get too large
19:21:51 <Ilari> => About 32 000 000 000 000
19:22:19 <Vorpal> yes indeed, but remember that is just for one tunnel
19:22:56 <Ilari> That's about 8 000 times more than there are total IPv4 addresses...
19:22:57 <Vorpal> variable, as for routing tables, that is because everyone wants multi-homed. And I seen larger allocations than /48...
19:22:57 <variable> Vorpal, take a look at Max Pierson's recent post to the NANOG
19:23:47 <Vorpal> Ilari, remember that address usage grows non-linearly
19:24:21 <variable> Vorpal, http://seclists.org/nanog/2011/Jan/1191 --> not directly on your topic - but the later responders answer it
19:24:40 <variable> ""The whole point of IPv6 is that the number of -----> prefixes <------ vastly exceeds the number of applicants that will use them.""
19:26:27 <Vorpal> variable, hopefully. But what would a normal consumer ISP customer get? A single /64?
19:26:28 <Ilari> Running out of IPv6 addresses requires at least one of 1) Totally insane allocation policies (the current ones aren't even remotely that). 2) interstellar travel and communications.
19:26:58 <Vorpal> Ilari, I'm hoping to retire on a terraformed Mars ;)
19:27:10 -!- azaq23 has joined.
19:27:29 <variable> Vorpal, I think so - but I'm not certain
19:28:08 <Ilari> 10 "logical RIR allocations" left. The Moon might need one, I think 4 would be enough for Mars...
19:28:36 <variable> Vorpal, Owen DeLong's posts on that thread explain it a lot better
19:28:47 <Ilari> And those allocations are /7s...
19:28:58 <Vorpal> variable, there should be a step in between for this use case. A consumer might want a few /64 but most won't need a full /48. Why can't they do /60 or so for this use case? It won't affect the global routing tables for consumer ISPs
19:29:15 -!- poiuy_qwert has joined.
19:29:46 <variable> Vorpal, tiny ISPs probably resell other's IP space - they don't get their own
19:29:47 <Vorpal> Ilari, but doesn't the first section (2001 or whatever it usually is) have lots of unused values
19:29:50 <Ilari> IIRC, the reason that free.fr is handing /64 instead of /60 is that CPE has problems with /60s.
19:29:52 <Vorpal> sure, some are link local or such
19:29:58 <Vorpal> but even so, most are unused
19:30:27 -!- poiuy_qwert has quit (Client Quit).
19:31:21 <Vorpal> variable, for ipv4 they definitely have their own most of the time. And I checked tables and most in Sweden have ipv6 allocations, though they don't provide it to end users.
19:31:33 <variable> Vorpal, consumer home providers are generally large and act as backbones for some cases - so they tend to get large blocks anyways. Its the tiny web hosting providers that resell
19:31:55 <Vorpal> variable, what does web hosting have to do with ISPs?
19:32:07 <Vorpal> anyway, you could possibly use that first section of ipv6, Then just tweak the wording to say 2001 or whatever is the "inner solar system" or such.
19:32:53 <Ilari> The first segments that are currently used in global unicast: 2001, 2002, 2003, 240x, 260x, 2610, 2620, 280x, 2A0x, 2C0x (85 of 8192)
19:33:10 <Vorpal> Ilari, what, that many?
19:33:13 <variable> Vorpal, I've got to go to class now :-\
19:33:15 <Vorpal> Ilari, what are they all for?
19:33:23 <Vorpal> variable, will go to sleep soon
19:34:16 <Ilari> 2001 is misc mess. 2002 is 6to4, 2003 has RIPE, 24xx is APNIC, 26xx is ARIN, 28xx is LACNIC, 2Axx is RIPE, 2Cxx is AFRINIC.
19:35:22 <Ilari> Don't ask why they stuck one of RIPE misc blocks to 2003. 2001 would had space for it (resulting that block being filled).
19:35:30 -!- nddrylliog has joined.
19:36:21 <Ilari> Or nope, that block would still have had 2001:6000::/19
19:36:28 <Vorpal> Ilari, hm, am I right in that the issue with routing tables is that the decentralised (and thus multi-homing) nature of internet makes aggregating the routes mostly futile?
19:37:01 <Ilari> At least that geographical distribution allows fair amount of aggeration...
19:37:10 <Vorpal> Ilari, hm. I think my current ipv6 is in 2001
19:37:27 <Ilari> 2001 has all sorts of mess.
19:37:29 <olsner> Vorpal: you have ipv6?
19:37:52 <olsner> ah, ok... is that fun to have? should I set one up?
19:38:07 <Ilari> Well, it has 28 allocations to RIRs..
19:39:32 <Ilari> That was apparently time before current policy of allocating /12 at a time.
19:40:35 <Vorpal> Ilari, this just messes up the routing table! And a /12 is crazy large... If it was me I would suggest a /16 to each RIR. Which should be enough for many years.
19:41:17 <Ilari> When viewed from "far away", each RIR aggerates nicely.
19:41:40 <Vorpal> Ilari, and we need another RIR you forgot. Moon: 1, Mars: 4 sure. But with the global warming on Earth it won't be long until we have ACNIC
19:42:08 <Vorpal> Ilari, isn't AC the "country code" for Antartica?
19:42:51 <Vorpal> given the global warming at the current rate it will probably be a nice spot to live in given a few hundred years
19:43:40 <Ilari> Well, current warming rate wouldn't be enough, but warming is likely to accelerate.
19:44:21 <Vorpal> Ilari, okay I adjust that line to say: current warming rate'
19:46:01 <Ilari> Latest APNIC delta: 0.03
19:46:13 <Vorpal> Ilari, what does that measure?
19:46:23 <Ilari> 3% of a block allocated in single day.
19:47:04 <Vorpal> Ilari, so... That gives us... gah can't find my calculator and I suck at mental number juggling.
19:47:08 <Ilari> AFAIK, a lot of Asia has long holidays, so one expects it to be quiet for about a week...
19:48:10 <Vorpal> Ilari, and they don't realise that it is better to invest in ipv6 now than getting more ipv4 given that it is doomed anyway?
19:50:09 <Gregor> EVERYONE WHO HAS A WII: Go play Muscle March. DO IT NOW. I want to not be the only person who has played it :P
19:50:40 <Vorpal> Gregor, I don't have a Wii, but what is that game?
19:51:18 <Gregor> Vorpal: http://www.youtube.com/watch?v=U_s7iCOj9HU
19:52:44 <Vorpal> Gregor, I don't understand the language used in that video
19:52:57 <Gregor> Vorpal: Believe me, you don't need to.
19:53:10 <Ilari> Maximum MTU IPv6 can use is only 65575. :-/
19:53:48 <Vorpal> Gregor, also that is not safe for sanity, that link
19:54:01 <Gregor> Ilari: Frankly I'm more concerned that UDP and TCP on IPv6 didn't decide to change their port bitwidth
19:54:13 <Gregor> Vorpal: ITYM "That's awesome and I want to play it NOW"
19:54:48 <Ilari> That's only defining new layer 4 protocol. Much less messy than doing layer 3 migration...
19:54:54 <Vorpal> Ilari, err, jumbo frames aren't all that useful outside very specialised networks. It makes the network respond slower to multiple clients as well. I mean given smaller frame you can interleave communication with different hosts a lot faster.
19:55:30 <Vorpal> Gregor, yeah, but in practise a host doesn't use that many open ports at once
19:55:30 <Gregor> Ilari: Yeah, but the layer 3 migration was a great time to go "oh and while we're at it ..."
19:55:34 <Ilari> AFAIK, SONET/SDH only goes to 4096. Ethernet only goes to something like 9200...
19:55:48 <Vorpal> Gregor, though an issue is of course different applications not being able to get new port numbers allocated.
19:55:50 <Gregor> Vorpal: Depends on the host. Some do, and end up having to do resource sharing just for stupid port count reasons.
19:56:23 <Vorpal> Gregor, but then, most programs don't care about officially registering a port number these days.
19:56:35 <Gregor> I'm not concerned with that, no :P
19:57:33 -!- elliott has joined.
19:57:47 <Vorpal> Gregor, a non-listening port can actually be a pair (remote-end-point,port) without issues. Meaning that you could in theory get close to the full range *per remote endpoint*
19:57:49 <Gregor> nddrylliog: TIME TO PLAY MUSCLE MARCH?
19:57:55 <Vorpal> Gregor, and if you have such systems: wtf
19:58:02 <Gregor> Vorpal: Fair enough, but nobody implements that.
19:58:57 <Vorpal> Gregor, in fact you could even do (remote-end-point,remote-port,local-port) though I think that would need to be supported on both sides. And that would be *huge* for a given remote host
19:59:13 <Vorpal> Gregor, anyway the obvious solution is to migrate to SCTP or whatever
19:59:14 <elliott> what are you talking about
19:59:33 <Vorpal> elliott, the range of TCP/UDP port numbers
19:59:38 <Ilari> SCTP has 16 bit ports.
19:59:45 <elliott> 18:32:46 <ais523> no it isn't, because most people respect them
19:59:45 <elliott> no, most people do not respect speed limits because of cameras
19:59:51 <elliott> they drive slowly when passing cameras, but then just speed up again
19:59:55 <Vorpal> Ilari, well it has multiple streams in a given connection though
20:00:27 <elliott> 18:47:45 <fizzie> Espoo, next to Helsinki.
20:00:28 <elliott> 18:48:05 <Phantom_Hoover> Espoo... *snicker*
20:00:35 <fizzie> Ilari: IPv6 packets have a special header option for 32-bit MTUs.
20:00:42 <elliott> Vorpal: ais said most people respect speed limits because of speed cameras, pretty much
20:00:43 <Vorpal> elliott, I would think it is below 50%, but still a large percentage
20:01:01 <Ilari> fizzie: Header type?
20:01:07 <fizzie> Ilari: Cf. http://tools.ietf.org/html/rfc2675 "Jumbograms"
20:01:23 <Vorpal> elliott, that just sounds wrong
20:01:45 <fizzie> Of course the whole link must support the option in question in order for that to work.
20:01:52 <Ilari> With single address, TCP (and SCTP!) would allow at most 65535 listeners at once. But the number of possible connections is 4294836225M where M is address space size.
20:02:08 <Vorpal> elliott, people don't speed because they die if they try. Though when I think about it, speeding only tends to happen during summer, a lot more rare during the winter
20:02:27 -!- impomatic has joined.
20:02:31 <fizzie> And not very many networks do have >64K MTU, ,as was mentioned.
20:03:29 <Vorpal> fizzie, indeed, don't most networks still do 1500 or less still?
20:03:58 <elliott> 19:25:43 <olsner> what, no elliott either
20:03:58 <elliott> 19:26:24 <quintopia> inorite? been a pretty pleasant day so far
20:04:04 <elliott> quintopia: be careful or I'll ... er...
20:04:07 <Vorpal> on my system, eth0: 1500, lo: 16436, sixxs: 1280
20:04:09 <elliott> ban you from #esoteric-minecraft!
20:04:45 <Vorpal> elliott, try finding a power outlet then ;P
20:05:07 <Vorpal> elliott, you should play df
20:05:18 <Vorpal> I'm currently trying to understand the military system.
20:05:51 <elliott> it sounds complicated. and scary!
20:06:01 <fizzie> Vorpal: Ethernet is a bit on the small side; AAL5 (ATM adaptation layer 5; what they use for packet-switched data over ATM, and there's quite many ATM links, I understand many DSL systems use the same protocols) default MTU is 9180 octets. (But the maximum allowed is 65535 there too.)
20:06:05 <olsner> elliott: fyi I didn't actually have anything to say to you, was just observing your absence
20:06:07 <Ilari> Vorpal: If it is GbE card, it might be possible to set MTU to over 9000!!!
20:06:13 <Vorpal> elliott, which the wiki describes as to as (in describing difference to older versions): "[...] is a new feature that is both incredibly versatile... and, initially, completely impenetrable."
20:06:33 <Vorpal> Ilari, 1) it isn't in this computer 2) I doubt the rest of the network would like that
20:06:37 <elliott> 19:43:17 <Vorpal> Ilari, I'm hoping to retire on a terraformed Mars ;)
20:06:41 <elliott> enjoy 30 minute ping to earth
20:07:12 <fizzie> Ethernet Jumbo frames are not very well standardized.
20:07:21 <Ilari> At least on this system, it is possible to set MTU larger than 1500 for eth0. Rest of the network doesn't like it tho...
20:07:50 <nddrylliog> "The theoretical capacity of a Boeing 747 filled with Blu-Ray discs is 595,520,000 Gigabits, resulting in a 37,000 Gbit/s flight from New York to Los Angeles" - http://en.wikipedia.org/wiki/Sneakernet#Theory
20:07:50 <Vorpal> <elliott> enjoy 30 minute ping to earth <-- I'm sure they will come up with a working worm hole soon!
20:07:53 <fizzie> I've tried it out on the point-to-point gigabit link between the network HD box and the router, but it didn't really have noticeable benefits; probably overhead elsewhere.
20:07:55 <elliott> 19:48:12 <Vorpal> variable, what does web hosting have to do with ISPs?
20:08:02 <elliott> ISP has a more general meaning than its day-to-day usage
20:08:06 <elliott> ISP is technically any internet service provider
20:08:13 <elliott> i.e., web hosts, email providers, are all ISPs
20:08:17 <Vorpal> elliott, well okay, but no one uses it like that.
20:08:26 <elliott> especially the kind of people who talk about IPv6
20:08:49 <Vorpal> elliott, but sure, what is a short word or abbrev for ISP in the day-to-day usage meaning then?
20:09:05 <elliott> Vorpal: internet provider?
20:09:13 <Vorpal> elliott, "IP" is already in use!
20:09:14 <elliott> An Internet service provider (ISP), also sometimes referred to as an Internet access provider (IAP), is a company that offers its customers access to the Internet.[1] The ISP connects to its customers using a data transmission technology appropriate for delivering Internet Protocol packets or frames, such as dial-up, DSL, cable modem, wireless or dedicated high-speed interconnects.
20:09:22 <Vorpal> elliott, ah yes IAP would work
20:09:26 <elliott> but i doubt anyone says /that/ :)
20:09:55 <Vorpal> elliott, well, indeed. Everyone says ISP and usually they mean that. If they use ISP in the wider meaning then I presume they use IAP!
20:10:04 <Vorpal> elliott, you don't know?
20:10:42 <Vorpal> elliott, basic example of it: connect through different IAPs for example. If one goes down, you still have access.
20:10:46 <olsner> elliott: sure you would, just to check that Vorpal isn't just pretending to know what he's talking about
20:10:52 <Vorpal> elliott, you need some routing table magic to make this work
20:11:00 <Vorpal> elliott, that is, announce the routes through *both* connections
20:11:03 <elliott> 19:57:59 <Vorpal> Ilari, and we need another RIR you forgot. Moon: 1, Mars: 4 sure. But with the global warming on Earth it won't be long until we have ACNIC
20:11:04 <elliott> Global warming is going to fuck us up well before colonising Mars is possible. :p
20:11:08 <elliott> (wait, what does AC stand for there?)
20:11:15 <Vorpal> elliott, thus it is limited to AS level
20:11:16 <elliott> <Vorpal> elliott, basic example of it: connect through different IAPs for example. If one goes down, you still have access.
20:11:23 <elliott> ok, nobody believes consumers want this, right?
20:11:29 <Vorpal> <elliott> (wait, what does AC stand for there?) <-- read 2-3 lines further
20:11:33 <elliott> the "home" part makes me suspicious
20:11:34 <Vorpal> <elliott> ok, nobody believes consumers want this, right? <-- indeed
20:11:42 <Vorpal> elliott, but a lot of businesses want it
20:11:56 <Vorpal> elliott, for a single office
20:12:17 <Vorpal> elliott, check the routing of a typical university. They will likely have several up-pipes for their AS
20:12:30 <elliott> 20:03:23 <Vorpal> Ilari, so... That gives us... gah can't find my calculator and I suck at mental number juggling.
20:12:32 <Vorpal> elliott, wait, do you know anything about BGP?
20:12:48 <elliott> I'm not an Internet-internals type
20:13:04 <elliott> my knowledge of how google.com loads ends at HTTP
20:13:21 <Vorpal> elliott, ah, the tubes can be a bit cramped at times. What with all the data going past at a respectable fraction of c.
20:13:29 <nddrylliog> what was that Google HTTP 'faster and lighter replacement' again?
20:13:45 <olsner> elliott: how can you cope with not knowing it all the way down?
20:13:59 <Vorpal> nddrylliog, they had that? For what purpose?
20:14:03 <elliott> nddrylliog: i forget but i don't think anyone adopted that
20:14:32 <Vorpal> olsner, hm do you know how optical transmitters work for fibers?
20:14:58 <Vorpal> nddrylliog, did they use it for anything? I mean, their netbook OS to talk to their own services seems a plausible use...
20:15:27 <olsner> Vorpal: gnomes and light switches
20:15:28 <elliott> olsner: because the net sucks!
20:15:37 <elliott> Google Chrome utilizes SPDY[5][6] when communicating with Google services, such as Google Search, Gmail, Chrome sync and when serving Google's ads. Google acknowledges that the use of SPDY is enabled in the communication between Chrome and Google's SSL-enabled servers.[7] SPDY sessions can be inspected in Chrome at the special URL chrome://net-internals/#events&q=type:SPDY_SESSION%20is:active .
20:16:00 <elliott> yep, gmail.com shows spdy sessions on this chrome
20:16:02 <fizzie> Vorpal: The process of how fiberoptic cables are made is also very interesting; one telecom visiting lecturer talked about it for us. I've forgotten most of the details.
20:16:04 <elliott> BIG BROTHA IS SPEEDING UP UR ADS
20:16:30 <nddrylliog> http://stackoverflow.com/questions/4065344/is-spdy-really-used
20:16:39 <nddrylliog> ah that explains why ads always load faster than the rest :)
20:16:41 <Ilari> Oh, and apparently watching a lot of your moves: You know what ajax.googleapis.com (or what it was) is?
20:17:03 <nddrylliog> Ilari: realtime search query completion?
20:17:43 <Vorpal> interesting way to think of a endpoint-fibre-endpoint system: huge opto-isolator
20:19:00 <Ilari> nddrylliog: It isn't that. It is collection of various javashit scripts... Good way to collect referer headers...
20:19:21 <elliott> Please stop calling it javashit... it's just silly.
20:20:34 <nddrylliog> seriously what's wrong with Javascript? just because its implementations are mostly horrible doesn't mean it's a bad language
20:20:47 <elliott> Oh, it's a terrible language.
20:20:52 <Ilari> Performance and security mainly.
20:20:54 <elliott> But for reasons like "terrible scoping".
20:20:59 <elliott> Security has nothing to do with JS-the-language.
20:21:13 <elliott> Anyway even if I hated JS with all my soul I'd still call it javascript. I don't say M$ either :P
20:21:17 <elliott> Plus JS performance is hardly bad.
20:21:20 <elliott> I think V8 is faster than Python.
20:21:36 <elliott> (ocf the DOM and canvas and whatever are slow as shit because the web is terribly architectured, BUT)
20:21:59 <Vorpal> <elliott> I think V8 is faster than Python. <-- a LOT of things are faster than python
20:22:10 <Ilari> Anyway, when it causes extreme loads on fast computers, its performance is bad.
20:22:10 <Vorpal> elliott, now try again with well written cython code
20:22:25 <elliott> But nobody complains about Python performance really.
20:22:28 <elliott> Ilari: No, *browsers* cause that.
20:22:36 <nddrylliog> http://shootout.alioth.debian.org/u32/benchmark.php?test=all&lang=v8&lang2=gcc
20:22:40 <elliott> Ilari: JS is used in servers too nowadays.
20:22:52 <elliott> The language isn't "slow" -- implementations are very fast nowadays.
20:22:58 <nddrylliog> nobody complains about python performance because if you need performance you're already using C modules/Cython/Psyco/Pypy/another language.
20:22:58 <elliott> It's the DOM, repainting, etc. that's slow.
20:23:01 <elliott> And those are all browser things.
20:23:05 <Vorpal> elliott, I have done it when I needed to use it in a damn lab assignment for a reversi playing AI. By switching to cython I got the ply up from ~3-4 to 7!
20:23:12 <elliott> nddrylliog: YOU FORGOT OOC
20:23:34 -!- nddrylliog has left (?).
20:23:39 <Ilari> Anyway, I have JS disabled in places because it uses too much CPU. And this is with i7...
20:24:30 <Vorpal> Ilari, I use noscript and only allow when I need it
20:24:42 <elliott> Vorpal: ooc is the thing that nddrylliog will hate us if we keep mentioning it
20:24:54 <Vorpal> elliott, oh, what is it then?
20:24:57 <Ilari> And security is even bigger reason to strictly limit what sites are allowed to execute JS. You ever have browsed high-severity browser security advisories?
20:25:12 <elliott> Vorpal: http://ooc-lang.org/ (nddrylliog standard disclaimer: he has nothing to do with the website at all)
20:25:25 <Vorpal> Ilari, I don't browse CVEs and such daily but I know what you mean
20:25:35 <Vorpal> elliott, uh why would he have anything to do with it?
20:25:42 <elliott> Vorpal: because he made the language/compiler
20:26:04 <elliott> and is stuck in CONSTANT REGRET
20:26:04 <Vorpal> elliott, so someone took over then?
20:26:13 <elliott> Vorpal: no, he just let someone else make the website
20:26:18 <elliott> A DECISION HE HAS REGRETTED EVERY DAY SINCE
20:26:24 <Ilari> Basically, the amount of browser security issues that absolutely require JS to expoit or having JS makes WAY easier is scary.
20:26:35 <Vorpal> elliott, website doesn't look too bad.
20:26:43 <elliott> Vorpal: Yes it does, it's horrible :P
20:26:48 <elliott> Vorpal: It has no information about the language, it's all marketing crap.
20:26:58 <elliott> Vorpal: Plus it has a login button... that logs in with *GitHub*.
20:27:16 <elliott> Also: We don't need no stinkin' manual, we have SCREENCASTS!
20:27:24 <Vorpal> elliott, can't he get the person to fix it and/or take over the web site?
20:27:25 <variable> <elliott> I'm not an Internet-internals type -> I am :-)
20:27:36 <elliott> Vorpal: I don't think he cares nearly enough :P
20:27:39 <Vorpal> elliott, anyway what sort of language is it? What is your opinion on the language as such
20:28:10 <elliott> It's like Java/C# had decent native-code performance and had a bastard child with Scheme and it turned out all disfigured and seriously, nddrylliog is going to kill me, stop talking about ooc.
20:28:21 <variable> <elliott> Oh, it's a terrible language. -> its not a terrible language *if you understand it and use it correctly* which 99.9% of programmers don't
20:28:30 <variable> because they learned from silly tutorial online
20:28:35 <elliott> variable: Its scoping rules are broken.
20:28:53 <Vorpal> elliott, the first bit didn't sound too bad ("decent native-code performance"), the rest however does
20:28:54 <variable> elliott, agreed. It could do with some changes - doesn't mean its absolutely horid
20:29:07 <elliott> variable: I have utmost respect for Brendan Eich, I realise he had about 10 days to create a language which looked like Java and he did the honourable thing of trying to make it as much like Scheme as possible ... but he forgot to make scoping not suck on the way there.
20:29:15 <elliott> Also, objects make painful associative arrays.
20:29:18 <Ilari> Hmm... Wonder how to create another esoteric language that seems to defy classification... :-/
20:29:32 <elliott> The language consists of pouring fire onto innocent children.
20:29:36 <elliott> The compiler is a bucket of fire.
20:30:03 <Vorpal> elliott, thing that annoys me about df: only some menus are mouse enabled. Some fully. Some only for some elements. Some not at all. Is consistency too much to ask for XD
20:30:14 <elliott> Vorpal: I doubt any REAL PLAYER uses the mouse.
20:31:07 <Vorpal> elliott, I don't unless I have a huge list and need to get an element near the middle :P
20:31:15 <Vorpal> elliott, in which case it is faster than scrolling with keyboard
20:34:58 -!- pumpkin has joined.
20:36:35 -!- copumpkin has quit (Ping timeout: 260 seconds).
20:40:17 <Ilari> F E D F I K F B K F A G L K K M L L N M M C I N H C J E C J E E H B E I A B N F 0 F 1 F o K 0 K 1 K i K o D 0 D 1 D o G 0 G 1 G i I i J o
20:44:57 <Ilari> Permuting the symbols a bit: A B C A D E A N E A F G H E E I H H J I I K D J L K M B K M B B L N B D F N J A 0 A 1 A o E 0 E 1 E i E o C 0 C 1 C o G 0 G 1 G i D i M o
20:45:33 <Vorpal> Ilari, do they have any meaning?
20:45:42 <Ilari> (that's supposed to be a cat program)
20:49:42 <Ilari> Let's just say that esolang is so heavy to execute that I wouldn't want to run 99 beers program made in it...
20:53:27 <Ilari> One kilobyte of data into that sort of cat program. Already used 2 minutes of CPU time and still not finished.
20:54:49 <Ilari> And it is not slow to load: if 16 bytes of test data is used, the finishing time is shown as 0:00.00
20:56:45 <Ilari> Start with empty string. Then extend by one character it so that it won't match given Context Free Grammar. If no such extension is possible, quit.
20:57:04 <Ilari> And repeat the extension step.
20:59:54 <Ilari> The string alphabet is 0, 1, i and o. 0 and 1 do nothing when placed. i causes byte to be read from stdin and written to end of the string using 0 and 1 (ignoring the grammar). o causes last 8 symbols to be read as byte and printed to stdout.
21:01:02 <Ilari> 446.40user 0.06system 7:27.13elapsed 99%CPU (0avgtext+0avgdata 413376maxresident)k
21:06:26 -!- Tritonio has joined.
21:07:16 <Ilari> That grammar can be written in regexp form: (i[01io]{8}o)*([01o]|i[01io]|i[01io]{8}[01i])
21:10:00 <quintopia> it sounds like you're trying to solve a pspace-complete problem and expecting it to be fast :P
21:10:10 -!- hagb4rd has joined.
21:11:07 <Ilari> Well, CFG recognition is in P.
21:14:08 <quintopia> decision is PSPACE-complete though, innit?
21:15:46 <quintopia> and what your program is doing amounts to deciding whether or not there is a string of arbitrary length that can be produced by a CFG
21:16:50 <quintopia> (in a lazy, give-up-quickly kind of way)
21:20:12 <Ilari> It repeatedly runs CFG recognition (and that's in P).
21:21:13 -!- hagb4rd has quit (Ping timeout: 240 seconds).
21:26:13 -!- hagb4rd has joined.
21:32:53 -!- hagb4rd has quit (Ping timeout: 240 seconds).
21:35:03 <Ilari> I have already optimized it a lot. Reading logs: Speedup hello world by 41 percent ... 55 percent... 86 percent... 286 percent... 118 percent... 440 percent... 38 percent...
21:47:49 -!- variable has quit (Quit: Daemon escaped from pentagram).
21:54:45 -!- MigoMipo has quit (Read error: Connection reset by peer).
21:57:20 -!- Sgeo has joined.
21:59:18 -!- augur has quit (Read error: Connection reset by peer).
21:59:41 -!- Sgeo_ has quit (Ping timeout: 240 seconds).
21:59:47 -!- augur has joined.
22:10:00 <elliott> 16:18:31 <Sgeo> Anyone use EsoAPI or PESOIX?
22:10:01 <elliott> 16:24:44 <Sgeo> Anyone use BFComp?
22:10:13 <Vorpal> elliott, what is BFComp?
22:10:29 <Vorpal> also is PESOIX a new one?
22:10:40 <elliott> Sgeo's DISSATISFACTION with it lead to PSOX.
22:15:49 -!- poiuy_qwert has joined.
22:16:00 -!- variable has joined.
22:16:15 -!- poiuy_qwert has quit (Client Quit).
22:23:06 <Phantom_Hoover> http://en.wikipedia.org/wiki/Zev_(later_Xev)_Bellringer
22:35:01 -!- pumpkin has quit (Quit: Computer has gone to sleep.).
22:36:16 <Ilari> Heh... "Death by GPS".
22:37:51 -!- variable has quit (Quit: Daemon escaped from pentagram).
22:42:35 -!- nddrylliog has joined.
22:42:50 -!- nddrylliog has left (?).
22:43:29 -!- elliott_ has joined.
22:43:38 -!- elliott has quit (Read error: Connection reset by peer).
22:44:09 <elliott_> @hoogle [Parser a] -> Parser a
22:44:11 -!- elliott_ has changed nick to elliott.
22:44:16 -!- elliott has quit (Changing host).
22:44:16 -!- elliott has joined.
22:44:18 <lambdabot> module Text.ParserCombinators.Parsec
22:45:27 -!- hagb4rd has joined.
22:45:42 <elliott> @pl choice . map (\(c,r) -> char c *> return r)
22:45:42 <lambdabot> choice . map (uncurry ((. return) . (*>) . char))
22:46:08 <lambdabot> forall a b. (a -> b -> a) -> a -> [b] -> a
22:46:22 -!- FireFly has quit (Quit: swatted to death).
22:46:24 <elliott> @pl foldl' (\poop (c,r) -> poop <|> (char c *> return r))
22:46:24 <lambdabot> foldl' ((`ap` snd) . (. fst) . (. ((. return) . (*>) . char)) . (.) . (<|>))
22:46:28 <elliott> @pl foldl' (\poop (c,r) -> poop <|> (char c *> return r)) pzero
22:46:28 <lambdabot> foldl' ((`ap` snd) . (. fst) . (. ((. return) . (*>) . char)) . (.) . (<|>)) pzero
22:46:39 <elliott> @pl foldlr (\(c,r) poop -> poop <|> (char c *> return r))
22:46:39 <lambdabot> foldlr (uncurry ((flip (<|>) .) . (. return) . (*>) . char))
22:46:46 <elliott> @pl foldlr (\(c,r) poop -> (char c *> return r) <|> poop)
22:46:46 <lambdabot> foldlr (uncurry (((<|>) .) . (. return) . (*>) . char))
22:46:50 <elliott> @pl foldr (\(c,r) poop -> (char c *> return r) <|> poop)
22:46:50 <lambdabot> foldr (uncurry (((<|>) .) . (. return) . (*>) . char))
22:54:28 -!- hagb4rd has quit (Ping timeout: 240 seconds).
23:01:17 <Phantom_Hoover> http://www.cracked.com/article_15982_5-horrifying-food-additives-youve-probably-eaten-today.html
23:03:05 <lambdabot> forall (f :: * -> *) a. (Alternative f) => f a -> f a -> f a
23:03:15 <elliott> Phantom_Hoover: Oh shush, Cracked is amusing.
23:03:19 <lambdabot> forall a b (f :: * -> *). (Functor f) => (a -> b) -> f a -> f b
23:03:33 <Phantom_Hoover> elliott, yes, but you can't churn out amusing articles all the time.
23:03:51 <elliott> @hoogle f a -> (a -> b) -> f b
23:03:52 <lambdabot> Prelude fmap :: Functor f => (a -> b) -> f a -> f b
23:03:52 <lambdabot> Control.Applicative (<$>) :: Functor f => (a -> b) -> f a -> f b
23:03:52 <lambdabot> Control.Monad fmap :: Functor f => (a -> b) -> f a -> f b
23:09:10 <elliott> @pl do x <- pSimpleUL; xs <- pUL; return (x :< xs)
23:09:10 <lambdabot> expecting letter or digit, variable, "(", "`", "!!", ".", operator or end of input
23:09:13 <elliott> @undo do x <- pSimpleUL; xs <- pUL; return (x :< xs)
23:09:13 <lambdabot> pSimpleUL >>= \ x -> pUL >>= \ xs -> return (x :< xs)
23:09:18 <elliott> @pl pSimpleUL >>= \ x -> pUL >>= \ xs -> return (x :< xs)
23:09:40 <lambdabot> forall a (m :: * -> *) b. (Monad m) => (a -> m b) -> m a -> m b
23:10:23 <Ilari> Hah... Yet another blogpost from Dr. Davis slamming wheat.
23:16:26 <Ilari> Phantom_Hoover: Collection of posts about wheat among multiple blogs: http://nigeepoo.blogspot.com/2010/03/wheat-oh-dear.html ...
23:17:22 <Phantom_Hoover> He has a BSc in engineering, so I'm assuming crackpot.
23:19:55 <Ilari> That list of blogposts about wheat in The Heart Scan Blog (Dr. Davis) is quite impressive.
23:20:36 -!- BeholdMyGlory has quit (Remote host closed the connection).
23:21:41 <Ilari> The Antony Colpo page (The Whole Grain Scam) is pretty funny. It is mainly E-Mail exchange between Colpo and some PhD idiot (presumably PhD on nutrion or somesuch).
23:24:11 -!- augur has quit (Remote host closed the connection).
23:28:04 -!- variable has joined.
23:29:58 <elliott> Googling Antony Colpo results in http://www.thegreatcholesterolcon.com/.
23:30:01 <Phantom_Hoover> (Engineers make up the majority of the PhD idiot population.)
23:30:03 <elliott> Gotta trust anyone with one of those long-email-signup pages.
23:30:43 <Ilari> Yup, Anthony Colpo wrote one of (there are multiple!) books titled "The Great Cholesterol Con".
23:30:57 <elliott> And made one of those awful spammy websites for it too!
23:31:19 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.16/20101130074636]).
23:31:32 <Ilari> Dr. Malcolm Kendrik has also written a book with that title.
23:33:04 -!- amca has joined.
23:34:15 <Ilari> While epidemilogical studies are rather crappy, just check what kind of figures China Study comes up with connection between Wheat and heart disease (and various other nasty things). Ouch.
23:34:36 -!- Sgeo_ has joined.
23:34:59 <Phantom_Hoover> "China Study" does not scream "reputable source" at me.
23:35:05 <Ilari> One of the best warnings about where too much trust on results of epidemilogical studies can lead, see Hormone Replacement Therapy.
23:35:33 <variable> a) heh - I was reading a while ago some study that showed that most studies were wrong
23:35:46 <elliott> Phantom_Hoover: Seems like http://en.wikipedia.org/wiki/The_China_Study_(book)
23:35:48 <Ilari> There is study (two actually) titled China Study. There's also book titled "The China Study" (pretty much total garbage).
23:35:49 -!- augur has joined.
23:35:52 <elliott> http://en.wikipedia.org/wiki/China-Oxford-Cornell_Study_on_Dietary,_Lifestyle_and_Disease_Mortality_Characteristics_in_65_Rural_Chinese_Counties
23:36:02 <variable> b) one study that I've been dying to do for a while is to correlate the price of tea in china with other random events :-}
23:36:40 <elliott> "What's that got to do with the price of tea in China?" "...as it turns out, absolutely nothing. Interesting!"
23:36:49 <elliott> "Perhaps I could use this as a random number generator."
23:38:07 <Ilari> And BTW, don't expect balanced discussion of The China Study (the book) on Wikipedia. There are some veganist admins there and The China Study is one of veganist bibles...
23:38:27 -!- Sgeo has quit (Ping timeout: 276 seconds).
23:38:58 <Phantom_Hoover> Ilari, you are now actually going into full-on conspiracy nutcase mode.
23:39:35 <variable> Phantom_Hoover, nearly everyone believes something completely irrational
23:39:49 <Phantom_Hoover> variable, indeed. I feel that this is why we can't have nice things.
23:39:49 <elliott> Well, sure, but not to a significant extent.
23:39:53 <elliott> (Most people, not nearly everyone.)
23:39:54 <variable> elliott, hang on a sec - there was a study on that
23:40:03 <elliott> variable: I thought most studies were wrong. Isn't there a study about that?
23:40:07 <variable> it was > 95% of people - which I consider nearly everyone
23:40:08 <elliott> Ilari: there are about a thousand wikipedia admins
23:41:18 <variable> elliott, its actually healthy to believe certain types of delusional things (like that one is above average in most areas)
23:41:50 <Ilari> Well, out of thoursand people, there are likely to be one or few veganists in that group...
23:42:11 <Phantom_Hoover> Ilari, yes, and there are likely to be a few hardcore anti-vegans.
23:42:13 <variable> (one group of things I know really well are cognitive biases and flaws)
23:42:19 <elliott> variable: But beware of Dunning-Kruger...
23:42:26 <elliott> Also, "yeah, we read Less Wrong too". :p
23:42:31 <elliott> Oh wait, I linked you to LW didn't I...
23:42:47 <pikhq> Phantom_Hoover: Read it ironically.
23:42:47 <variable> oh right - less wrong - yeah - you did
23:42:57 <pikhq> Phantom_Hoover: And then assert the earth is 6,000 years old.
23:43:09 <elliott> http://en.wikipedia.org/wiki/Talk:The_China_Study_(book)
23:43:15 <variable> elliott, I wrote a paper for university on the Dunning-Kruger effect :-}
23:43:17 <pikhq> After all, "THE BIBLE SAYS SO AND IT'S THE WORK OF GOD THEREFORE YES."
23:43:18 <elliott> lol at this joke: "I suggest we delete ANY criticism of this work immediately. Wikipedia has a duty to protect this important book/study. Conflicting studies should be expunged. Articles or quotes from conflicting Doctors need to be deleted or at least make sure you remove 'Dr.' from their name as to discredit."
23:43:26 <elliott> variable: HOW DEFORMED IS YOUR MOUTH
23:43:59 <Ilari> elliott: How you know that's a joke?
23:44:09 <elliott> Ilari: Because I was built with a sense of humour.
23:44:24 <elliott> Even if the start was believable "or at least make sure you remove 'Dr.' from their name as to discredit." gives it away as a joke.
23:44:58 <Ilari> Poe's Law does apply to veganists.
23:45:57 <elliott> Is there a reason you're saying "veganists" rather than "vegans"? ... Anyway, I know a few and they're not exactly evil or anything. :p
23:46:39 <elliott> Phantom_Hoover: sure you're aveganist, like atheists don't believe in god
23:46:43 <elliott> aveganists don't believe in vegans
23:46:50 <elliott> i guess atheists don't believe in the
23:48:15 <Sgeo_> elliott, I think I did ponder implementing it before becoming dissatisfied
23:48:17 <Ilari> Veganists are who take veganism to heights of fundamentalist religion.
23:49:00 <Phantom_Hoover> Sgeo_, shut up, me and elliott are ganging up on Ilari right now.
23:49:17 <elliott> I'm just commenting from the sidelines.
23:49:38 <Sgeo_> Is Ilari veganist or anti-veganist?
23:50:07 <Ilari> Dr. Davis is an engineer?
23:50:25 <elliott> Wheat killed my family, with its teeth.
23:50:27 <Phantom_Hoover> I'm talking about that Nigel guy. I'll stop the ad homs now.
23:50:53 <elliott> Actually if wheat had teeth that would be kinda cool except not really.
23:50:55 <Ilari> That "mutant wheat" stuff sounds more like from Dr. Davis...
23:50:57 <elliott> That's my observation for the day.
23:51:49 <Phantom_Hoover> Wild wheat shatters when its seeds have ripened, scattering them onto the earth.
23:52:07 <Phantom_Hoover> This is good if you are a wheat plant but bad if you like eating wheat seeds.
23:52:16 <elliott> `addquote <Phantom_Hoover> This is good if you are a wheat plant but bad if you like eating wheat seeds.
23:52:22 <elliott> "This is good if you are a wheat plant" is the best thing anybody has ever said.
23:52:31 <Sgeo_> It occurs to me that I know little about wheat
23:52:39 <Sgeo_> Wheaty foods come from the wheat's seeds?
23:52:50 <Sgeo_> or, just whole grain with those annoying seeds?
23:53:23 <HackEgo> 288) <Phantom_Hoover> This is good if you are a wheat plant but bad if you like eating wheat seeds.
23:53:24 <Ilari> I think there are many more hardcore vegans than hardcore anti-vegans... Especially when talking about people elevating it to fundamentalist heights.
23:54:05 <Sgeo_> I think she started turning her cheek towards me
23:56:50 <Ilari> I don't seek comments from fundamentalist vegans. Way way too painful to read anything that crazy.
23:57:11 <elliott> Animals are our friends, man.
23:59:30 <elliott> I think my bones are melting just from PH's reaction.
23:59:40 <Ilari> (In general any fundamentalist comments are usually just plain too painful to read).