←2013-11-09 2013-11-10 2013-11-11→ ↑2013 ↑all
00:02:43 <oerjan> `unidecode ⟩
00:02:45 <HackEgo> ​[U+27E9 MATHEMATICAL RIGHT ANGLE BRACKET]
00:03:00 <fizzie> oerjan: MAYBE.
00:04:14 <ais523\unfoog> oerjan: there's an annotation in Unicode saying that that's the correct one to use for a ket
00:06:21 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds).
00:08:34 <int-e> > forever ""
00:08:35 <lambdabot> []
00:08:44 <oerjan> ais523\unfoog: aha
00:09:30 <oerjan> > execWriter . forever $ tell "hi! "
00:09:31 <lambdabot> "hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi...
00:10:00 <ais523\unfoog> `yes hi
00:10:01 <HackEgo> hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \ hi \
00:11:22 <int-e> @unlambda ``ci`c`. `.!`.i`.hi
00:11:22 <lambdabot> hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi! hi!...
00:12:12 <oerjan> > fst $ (runRWS . forever $ do x <- get; tell [x]; put (x+1)) undefined 0
00:12:14 <lambdabot> Couldn't match expected type `(a0, b0)'
00:12:14 <lambdabot> with actual type `(a1,...
00:12:23 <oerjan> :t runRWS
00:12:24 <lambdabot> RWS r w s a -> r -> s -> (a, s, w)
00:12:31 <oerjan> oh
00:13:09 <oerjan> > view _3 $ (runRWS . forever $ do x <- get; tell [x]; put (x+1)) undefined 0
00:13:11 <lambdabot> [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,...
00:15:25 <oerjan> > view _3 $ (runRWS . forever $ do x <- get; tell [x]; modify . (+) =<< ask) 2 1
00:15:27 <lambdabot> [1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,5...
00:16:53 <ais523\unfoog> `seq 1 2 10000
00:16:54 <HackEgo> seq: invalid floating point argument: 1 2 10000 \ Try `seq --help' for more information.
00:16:58 <ais523\unfoog> err
00:17:01 <oerjan> > (0$0&)
00:17:01 <ais523\unfoog> `run seq 1 2 10000
00:17:02 <lambdabot> The operator `Control.Lens.Combinators.&' [infixl 1] of a section
00:17:02 <lambdabot> must...
00:17:03 <HackEgo> 1 \ 3 \ 5 \ 7 \ 9 \ 11 \ 13 \ 15 \ 17 \ 19 \ 21 \ 23 \ 25 \ 27 \ 29 \ 31 \ 33 \ 35 \ 37 \ 39 \ 41 \ 43 \ 45 \ 47 \ 49 \ 51 \ 53 \ 55 \ 57 \ 59 \ 61 \ 63 \ 65 \ 67 \ 69 \ 71 \ 73 \ 75 \ 77 \ 79 \ 81 \ 83 \ 85 \ 87 \ 89 \ 91 \ 93 \ 95 \ 97 \ 99 \ 101 \ 103 \ 105 \ 107 \ 109 \ 111 \ 113 \ 115 \ 117 \ 119 \ 121 \ 123 \ 125 \ 127 \ 129 \ 131 \ 133 \ 135
00:17:37 <oerjan> > [1,3..]
00:17:39 <lambdabot> [1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,5...
00:18:25 <oerjan> I WIN
00:18:45 <oerjan> (i wasn't trying to golf, anyway)
00:18:47 -!- nooodl has joined.
00:18:59 <ais523\unfoog> nor was I
00:22:09 * oerjan briefly wonders how many of the "quantum" esolangs actually are related to quantum computing rather than the "take several paths at a time (and discard the ones whose result you don't like)" misunderstanding of it
00:22:37 <oerjan> i vaguely _think_ i may have seen one.
00:23:02 -!- Sprocklem has quit (Ping timeout: 264 seconds).
00:23:05 <ais523\unfoog> nondeterministic esolangs are still interesting
00:23:12 <zzo38> Well, a few have Hadamard transformers I think
00:23:17 <ais523\unfoog> you just shouldn't call them quantum
00:23:31 <zzo38> ais523: Yes I think you are correct
00:23:34 <oerjan> right (to both of you)
00:23:43 -!- Phantom_Hoover has joined.
00:24:37 <oerjan> there's however a somewhat interesting complexity class that you get by combining both those ideas, "quantum computing with superselection"
00:25:21 <oerjan> it's not known to be equal to either of the subsets, but fits inside PSPACE. it's in the graph at the top of aaronson's blog.
00:27:07 <oerjan> PostBQP it was
00:28:05 <oerjan> oh it was postselection, not superselection
00:29:55 <oerjan> superselection also means something, not exactly sure what it's for
00:31:31 <oerjan> oh and it's equal to PP, a complexity class defined without any quantum
00:32:04 <Bike> it's cool how seemingly half of aaronson's job is telling peole quantum computers suck
00:32:57 -!- Phantom_Hoover has quit (Ping timeout: 272 seconds).
00:34:19 -!- Phantom_Hoover has joined.
00:39:47 <ais523\unfoog> for my Physics A-level project, I wrote something about how quantum computers work
00:40:00 <ais523\unfoog> and implemented a simulator capable of handing Shor's algorithm on 8-bit integers
00:40:09 <ais523\unfoog> calculating a probability distribution for the result
00:40:16 <Bike> neat
00:40:19 <ais523\unfoog> it had like a 30-40% chance of getting the correct answer, IIRC
00:40:31 <ais523\unfoog> quantum computer use in practice is, you run the experiment, then you check the result
00:40:42 <ais523\unfoog> and try again if it was wrong
00:45:58 <APott> So does anyone know anything about how virtual machines are implemented?
00:46:05 <APott> As in something like the JVM
00:46:39 <Bike> well yeah, they're like any language interpreter, except you have a billion dollars so you can make them fast as shit
00:47:44 <APott> lol
00:48:09 <APott> But I read about various virtual machine implementations
00:48:20 <APott> Like Google's V8 Javascript engine
00:48:34 <APott> they say they go straight from AST to Machine code
00:48:47 <Bike> that's a (just in time) compiler.
00:48:48 <APott> this is why it's so fast, how do they execute machine code?
00:48:51 <APott> Yes
00:49:04 <Bike> how do they execute machine code? well, they have these things called computer processors,
00:49:16 <APott> I mean programatically
00:49:41 <APott> I've read they basically get all the instructions in byte format or whatever then create a function pointer to the bytes in memory
00:49:51 <APott> then they can just call it like a function
00:50:01 <Bike> Something like that.
00:51:16 <ais523\unfoog> Gregor did something like that for the IOCCC
00:51:36 <Bike> i mean, obviously that's not conforming C or whatever, but you still get the idea.
00:51:45 <APott> wow
00:51:47 <APott> yeah
00:52:24 <Bike> that's von neumann architecture for ya.
00:53:00 <APott> lol
00:56:34 <Bike> did i tell a joke by accident...
00:57:01 <APott> no it's just I don't know as much as you so I said lol as a response
00:57:15 <APott> as i began reading a wikipedia article
00:57:32 <Bike> i've never really understood that. i thought 'lol' meant laughing
00:58:04 <APott> because before I looked it up I thought it was a joke since I didn't know what It meant
00:58:14 <APott> and it was kinda stated like one
00:58:58 <Bike> no that's pretty much just how i write
00:59:29 <APott> That's fine.
00:59:48 <Bike> i am The Living Joke
00:59:57 <APott> lol xD
01:02:15 <APott> are there any popular languages that interpret the ast
01:02:16 <APott> ?
01:03:01 <int-e> @unlambda ```s``s.1`k``s`k`s`k.1`s`k.,i``s``s`ks``s`kk``s``si`ki`ki``s``s`ks``s`k`si``s`kk``s``s`ks``s`k`s`ks``s`k`s`kk``s``s`ksk`k`s`k.1``s``s`ksk`k`s`k.0`ki
01:03:01 <lambdabot> 1,10,11,100,101,110,111,1000,1001,1010,1011,1100,1101,1110,1111,10000,10001...
01:03:21 <int-e> I forgot how icky that programming language (Unlambda) really is :)
01:03:27 <Bike> APott: what do you mean
01:04:23 <APott> I'm just curious
01:04:35 <APott> Like as I mentioned earlier with V8
01:04:47 <pikhq> Sure. Ruby.
01:04:59 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
01:05:00 <APott> It translates the ast to machine code, I'm assuming other languages do that but are any popular that interpret ast?
01:05:04 <APott> oic, ruby
01:05:13 <Bike> that's how most interpreters work...
01:05:21 <Bike> you have a "parse" step and then an "evaluation" step.
01:05:25 <APott> I've never written in it before. I don't even know what it's mainly used for
01:05:31 <APott> oic
01:05:36 <APott> yeah
01:05:51 <Bike> doing them both together would be kind of silly ("we call it Perl")
01:05:57 <int-e> bytecode interpreters used to be popular.
01:06:05 <APott> Just what everything i've read has made me believe that's not enough or suitable
01:06:10 <Bike> what?
01:06:14 <pikhq> int-e: Still are.
01:06:22 <int-e> parse -> bytecode -> interpret that, rather than an AST
01:06:33 <pikhq> Bytecode interpreters are profoundly faster than AST interpreters.
01:06:36 <Bike> yeah that... hapens all the time, int-e
01:06:58 <Bike> i have a language i should write a bytecode thing for but effffffort
01:06:58 <pikhq> Ruby's the only mainstream language that really *used* a direct AST interpreter.
01:07:16 <APott> Okay, I didn't know that, thanks
01:07:38 <APott> Why would a bytecode interpreter be faster since the AST has to be translated to bytecode which is then interpreted?
01:07:39 <zzo38> Maybe you might want to interpret some kinds of macros while parsing AST
01:07:57 <Bike> APott: the idea is that the ast->bytecode step takes some effort but you only have to do it onc.
01:08:00 <int-e> APott: better locality, smaller memory footprint --> better for caches
01:08:00 <Bike> e
01:08:09 <APott> oh okay
01:08:23 <Bike> that too yeah.
01:11:41 <pikhq> It's also quite a bit quicker to branch on.
01:12:03 <pikhq> Especially if you do somewhat sophisticated tricks to help the CPU's branch predictor along.
01:12:29 <Bike> fizzie: this shit reminds me, how much can i rely on matlab's jit? also if you want me to stop reminding you matlab exists that's ok
01:13:02 <pikhq> Bytecode's also nicer in terms of future development because you can generally add a JIT later.
01:13:16 <APott> yeah
01:13:32 <zzo38> pikhq: Especially if it is harvard bytecode, that is.
01:14:05 <Bike> i've been wondering if i should bother rearranging things to index into a matrix instead of branching but it seems like such a pain
01:14:12 <pikhq> zzo38: Which you almost always do for interpreter bytecodes. :)
01:14:53 <Bike> ... i wonder if even just using switch/case instead of elseifelseifelseif would speed it up. maybe the jit does jump tables
01:15:42 <Bike> or i could just give up and start rewriting the thing in scipy.
01:17:50 -!- tswett_q has changed nick to tswett.
01:19:13 <oerjan> APott: calling machine code in memory can be basically as you heard, except that nowadays you also need to change the security settings of the memory. by default you cannot write to and execute the same part of memory (this is to make attacks harder by not making it too easy for attackers to execute anything they want.)
01:20:18 <APott> oic
01:20:39 <oerjan> (also helps against some bugs, i guess)
01:20:56 <Bike> some terrifying bugs
01:21:06 <APott> Will just raw instructions execute or will there be some sort of other step before it's executed?
01:21:08 <zzo38> pikhq: A few aren't; a few partially are. In Z-machine, it is possible to execute instructions from RAM, although usually it isn't; there is the high ROM area which can contain only instructions and packed text strings. JUMP with a variable operand might still make compiling it a bit more difficult, though.
01:21:54 <Bike> APott: well you set up the argument stack and all, is that what you mean
01:22:04 <Bike> if there are arguments.
01:22:09 <zzo38> (In addition, there is the restriction of where routines and text strings can start, so you don't have to check all of them; also, routines always begin with a byte having the high four bits cleared, so if any of them are set you can skip that one.)
01:22:28 <APott> Bike: Yeah that makes sense
01:23:08 <pikhq> zzo38: Hence "almost". :)
01:23:33 <oerjan> APott: well it's as raw as you get to put into memory. although nowadays computer processors transform it to even rawer "microcode" internally.
01:23:51 <APott> yeah
01:23:53 <Bike> rawr
01:23:53 <zzo38> I am in the process of writing a program to convert OASYS codes (which is harvard) into Z-machine codes; currently no optimizations are done.
01:23:54 <oerjan> APott: but it's the same format as most programs use.
01:24:39 <zzo38> (OASYS also lacks variable jumps; this also makes it easier to compile.)
01:24:59 <APott> I've completely lost track of what you're talking about
01:25:37 <Bike> none of this should be new to you, since "load a program into memory and run it" is the same thing that happens when you run a program from your desktop :)
01:26:04 <APott> I was talking to zzo38
01:26:15 <Bike> `quote zzo38
01:26:17 <HackEgo> 27) <zzo38> I am not on the moon. \ 116) <zzo38> Some people are reasonable, some people who are not reasonable insist on changing things so therefore progress depends on not reasonablepeple \ 150) <zzo38> catseye: Please wake up. Not recorded for this timezone. The big spider is not your dream \ 174) <pikhq> zzo38: A better definition would probab
01:26:43 <zzo38> `quote 174
01:26:44 <HackEgo> 174) <pikhq> zzo38: A better definition would probably fix Avogadro's number. <Sgeo> It's broken?
01:27:03 <APott> uhh
01:30:13 <Bike> i woulda thought free vacuum permitivitty or something would be more 'broken'.
01:32:25 -!- ais523\unfoog has quit.
01:37:04 -!- yorick has quit (Remote host closed the connection).
01:49:40 <APott> So are any of you asian?
01:49:40 <pikhq> "Fix" in this sense meaning "set it to a fixed number, rather than an experimental quantity"
01:49:48 <pikhq> No, but I speak Japanese.
01:49:56 <APott> cool
01:50:07 <Bike> where did that come from
01:50:33 <APott> If you knew me irl I have an ongoing inside joke about asians
01:50:43 <Bike> is it racist
01:50:46 <APott> I always joke about asians but I have nothing against them, they're awesome
01:50:53 <APott> http://www.bearnerd.com.br/wp-content/uploads/2013/08/Asians-meme.jpg
01:51:06 <APott> it is not racist
01:53:18 <elliott> are you sure
01:56:56 <APott> 0100% positive I'm not racist
01:57:01 <APott> 100%*
01:57:08 <APott> I talk to asians irl on a regular basis
01:57:13 <APott> there is no problem
01:57:24 <Bike> like is this the "asians are all super smart math nerds" sort of crud
01:57:37 <APott> not really no
01:57:51 <APott> just kinda like that picture i linked. Just--asians.
01:58:04 -!- Koen_ has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?").
01:58:05 <APott> Anyway lets move on,
01:58:14 <Bike> baffling
01:58:42 <APott> lol that message is awesome whatever you would call that.
02:04:35 <zzo38> APott: Do you mean the quit message?
02:05:58 <APott> indeed
02:06:35 <elliott> APott: is your criterion for whether or not you are racist actually "I talk to asian people"
02:06:55 <APott> dude
02:07:00 <APott> I'm not racist
02:07:01 <APott> at all
02:07:06 <oerjan> as a general rule, if you have to explain why you're not racist you are already on thin ice.
02:07:27 <oerjan> and it gets thinner the stronger you argue.
02:07:28 <zzo38> Because I speak a language, I am racist too, although I try not to be too racist.
02:07:29 <APott> thin ice doesn't mean i'm racist
02:07:43 <APott> Definition of racism: the belief that all members of each race possess characteristics or abilities specific to that race, esp. so as to distinguish it as inferior or superior to another race or races.
02:08:14 <APott> I believe none of that
02:08:28 <oerjan> i was starting to write a response but then realized i was trolling.
02:08:36 <APott> ha lol
02:09:15 <zzo38> Do you know any Washizu mahjong players?
02:09:28 <APott> nope I don't even know what that is
02:09:31 -!- tertu has quit (Ping timeout: 252 seconds).
02:10:05 <zzo38> Do you know any mahjong game?
02:10:37 <APott> *opens google*
02:11:36 <APott> I suppose I've seen something like that before
02:11:46 <nooodl> i've only ever played mahjong solitaire. i kinda want to play actual mahjong once
02:13:34 <zzo38> nooodl: Japanese mahjong? There are many different kinds; Japanese mahjong is popular in many places. I play Japanese mahjong.
02:13:48 <nooodl> yeah, japanese mahjong
02:13:59 <zzo38> nooodl: Do you know anything about the rules?
02:15:04 <nooodl> i don't. all i know is it's similar to card games like bridge
02:15:35 <quintopia> do you like tichu
02:15:37 <zzo38> There are some similarities to some card games, such as bridge and rummy
02:15:48 <quintopia> zzo38: would you like to have a tichu night
02:16:09 -!- Taneb has joined.
02:16:09 <quintopia> you and me and two of your friends
02:16:14 <Taneb> Hi
02:16:21 <quintopia> sup
02:17:53 <oerjan> inf
02:18:15 <zzo38> quintopia: I do not know what that means.
02:18:32 <quintopia> zzo38: it's a good game you'd like it.
02:21:16 <zzo38> That doesn't explain anything.
02:23:15 <oerjan> <Taneb> *shudder* that was Lewis Carrol <-- um pretty sure it was dr seuss?
02:23:38 <APott> wow that's old but that confused me to
02:23:48 <Taneb> It was a line from Humpty Dumpty in Through the Looking Glass
02:23:53 <oerjan> possibly carroll wrote it too
02:24:06 -!- ^v has quit (Remote host closed the connection).
02:24:13 <APott> I don't even know the line that was spoken, what was it again?
02:24:51 <APott> http://www.goodreads.com/quotes/44212-i-meant-what-i-said-and-i-said-what-i
02:24:58 <APott> I said what I meant and meant what I said?
02:25:21 <APott> I meant what I said and I said what I meant*
02:25:24 <oerjan> <Bike> i meant what i said and i said waht i meant
02:25:39 <APott> yeah
02:25:42 <oerjan> Taneb: i think you may be confusing it with another quote
02:25:57 <APott> That was from a Dr. Suess story I'm pretty sure
02:26:03 <shachaf> hion
02:26:09 <APott> I suppose that could have beena coincidence or he borrowed it
02:26:15 <APott> depends what you're referring to
02:26:19 <oerjan> 'If I'd meant that, I'd have said it,' said Humpty Dumpty.
02:26:30 <APott> yes but that's not what was said
02:26:31 <Bike> if it helps i was actually quoting a porno
02:26:37 <Taneb> oerjan, hmm, possibly
02:26:45 <APott> lolo
02:26:49 <elliott> it definitely wasn't Dr. Suess.
02:27:07 <oerjan> elliott: well we have to adjust for the fact it was "waht"
02:27:15 <APott> but that was in a Dr. Suess story, specifically the onewith the elephant
02:27:18 <shachaf> oerjan: But what did Humpty Dumpty mean by that?
02:27:29 <APott> horton or whatever it's name was
02:28:02 <APott> published in 1940
02:28:57 <APott> http://en.wikipedia.org/wiki/Horton_Hatches_the_Egg#Analysis
02:29:04 <APott> near the bottom of the paragraph
02:29:13 <APott> I don't even know why we're talking about this lol
02:29:52 <APott> second paragraph
02:30:03 <oerjan> because this is a channel for incorrigible nitpickers
02:30:16 <APott> lol
02:31:37 <Taneb> "When I use a word," Humpty Dumpty said in rather a scornful tone, "it means just what I choose it to mean——neither more nor less."
02:31:38 <Taneb> That was what I was thinking of
02:31:38 <Taneb> Sorry about this
02:32:14 <APott> Alright it's fine
02:32:28 <APott> same message completely different words.
02:32:57 <oerjan> I DON'T THINK IT WAS THE SAME MESSAGE
02:33:02 <oerjan> *cough*
02:33:11 <Bike> it wasn't a very good porno
02:33:31 <Taneb> Bike, it's probably precisely as good as every porno I have ever seen
02:33:57 <APott> close to the same--lol
02:34:18 * oerjan is cleverly assuming Taneb has seen zero.
02:34:29 <Taneb> oerjan, correct
02:34:30 <Taneb> "Then you should say what you mean," the March Hare went on.
02:34:30 <Taneb> "I do," Alice hastily replied; "at least—at least I mean what I say—that's the same thing, you know."
02:34:52 <Taneb> Another near-miss Carroll quote
02:35:10 <APott> but that's not what was said, i think I won here
02:35:50 <APott> this si the stupidest conversation ever
02:35:56 <APott> is* lets end it
02:36:09 <Taneb> I've had stupider
02:36:29 <APott> one of the stupidest conversations ever*
02:36:49 <APott> bison poop
02:36:54 * oerjan hands APott the Useless Argument Champion Cup, in genuine pewter.
02:37:02 <APott> :D
02:37:40 <APott> So what OS's you guys running?
02:37:51 <pikhq> Linux deagol 3.10-3-amd64 #1 SMP Debian 3.10.11-1 (2013-09-10) x86_64 GNU/Linux
02:37:57 <Bike> be
02:38:17 <APott> I'm running Ubuntu Linux 13.04
02:38:28 <APott> 64 bit
02:38:33 <Taneb> Linux Mint, but not for the long term probably
02:38:44 <oerjan> i keep annoying everyone else by staying on Windows despite not liking it much (8.1 at the moment)
02:38:55 <APott> I won't comment on that.
02:39:16 <Bike> there's a guy in my twitter feed who can't figure out how to turn off his windows machine after the 8.1 upgrae
02:39:20 <APott> Linux Mint with what GUI? Last time I used it with Cinnamon
02:39:29 <APott> lol xD
02:39:42 <Taneb> Yeah, Cinnamon
02:39:55 <APott> Cinnamon was pretty nice, needs some work though
02:39:57 <zzo38> Bike: Does the SHUTDOWN command still work?
02:40:11 <Bike> i didn't ask
02:41:01 <zzo38> The menus and stuff are all different, but most of the same keyboard shortcuts continue to work (even to access things that aren't available on any menu anymore), and cmd.exe is still available, at least in Windows 8.0
02:41:40 -!- shikhin has joined.
02:42:16 <Bike> i don't just, i don't like ask near-strangers "have you tried command prompt"
02:42:24 <APott> :D
02:42:42 <APott> I have 8.0 on a separate partition just in case of an emergency
02:43:03 <APott> because sometimes you can fuck up linux quite easily
02:43:33 <APott> what...
02:43:40 <trout> APott: FreeBSD
02:44:11 <APott> That's what you run?
02:44:12 <trout> using openbox+lxpanel but looking to change
02:44:14 <trout> APott: yes
02:44:19 <APott> I've never tried it
02:44:25 <APott> any good, long term use?
02:44:26 <zzo38> Linux, Windows, a lot of systems are complicated and can be mixed up easily; DOS is a simple system that boots very fast and runs very fast too.
02:44:32 <trout> APott: absolutely
02:44:39 <APott> interesting
02:44:44 <Taneb> zzo38, wasn't DOS considered quite bloated when it was released?
02:44:53 <trout> APott: if you are going to try it, use pcbsd
02:44:55 <Bike> finally, a workable solution to the problem of computers. dos.
02:45:09 * trout peddles Bike
02:45:23 <APott> zzo38: This is true but sometimes some simple hardware/driver mix ups can be an issue, on windows things are a bit more stable
02:45:26 <Bike> pedals?
02:45:26 <quintopia> "I meant what I said what I meant; an elephant's faithful one hundred percent" --oh, childhood, thanks for reminding me :)
02:45:35 <Bike> or are you trying to sell me...........................
02:45:38 <APott> yeah ^^ :D
02:45:44 <zzo38> Taneb: Yes, but that is just because many people said that; it does contain some features it doesn't need but it is still pretty good compared to others
02:45:44 <trout> Bike: nah, trying to ride on the Bike
02:45:53 <quintopia> (locked up my computer egojsouting a long battle. would have replied sooner.)
02:46:00 <APott> my bikes back tire is flat
02:46:14 <APott> bike's*
02:46:28 <Bike> well if you're riding me you would use the pedals, not the peddles
02:46:31 -!- Taneb has quit (Quit: it is time for taneb to sleep).
02:46:43 <Bike> i don't mean to be rude here, you're the operator, but this is pretty basic
02:46:51 <APott> yeah lol.
02:47:26 <APott> So I found these old Beatles vinyls on E-Bay and I think I'm going to bid on them
02:47:38 <APott> The seller placed the starting price quite low
02:48:04 <zzo38> Do you have a record player though? (If not, it isn't that complicated anyways)
02:48:20 <Bike> well beatles vinyl isn't rare
02:48:56 <zzo38> Do you like Pokemon card (with the old-style rules)?
02:49:20 <APott> lol wut
02:49:28 <APott> I do not but am going to invest in one
02:49:56 <APott> I'm interested in starting a collection of them, since me grandfather gave me a good amount of them I thought why not?
02:51:08 <Bike> https://github.com/search?q=exec+sudo+%24_GET&type=Code
02:52:28 <APott> What about it?
02:52:43 <Bike> it's beautiful
02:52:51 <APott> a lot of bad practice
02:52:54 <zzo38> If "exec sudo $_GET" is a PHP code, it is a terrible code
02:52:55 <APott> but okay
02:53:15 <zzo38> (I haven't actually followed the link so I don't know how many results and that stuff)
02:53:34 <Bike> echo shell_exec("sudo ./smbkill ".$_GET['kill']." 2>&1");, what could possibl go wrong
02:53:41 <APott> a lot of shit
02:53:50 <zzo38> Bike: A lot of stuff
02:53:50 <APott> this is the worst php practice I've ever seen
02:55:13 <quintopia> i agree with Bike
02:55:16 <quintopia> it's amazing
02:55:30 <APott> That so many people are writing horrible code?
02:55:32 <APott> yes
02:55:51 <APott> not necessarily horrible but extremely insecure
02:55:51 <Bike> not sure if this is quite as funny as people putting private pgp keys in github dotfiles, but it's up there
02:56:03 <APott> yeah lol
02:56:34 <zzo38> A lot of PHP programmers write terrible programs though
02:56:46 <APott> The language promotes bad practices
02:57:09 <zzo38> Well, PHP isn't a very good programming language anyways, due to a number of problems (even ignoring bad practices)
02:57:15 <APott> I know first hand because it was the first language I learned and when I began learning others I realised how bad the stuff I was doing was
02:57:24 <APott> I need a comma in there lol
02:57:52 <shachaf> zzo38: What is a very good programming language?
02:58:21 <zzo38> shachaf: I don't really know but many are OK; they can be good for different purposes.
02:58:46 <APott> PHP being good for web scripting
02:59:00 <APott> I think it does it's job fine it just has it's issues
02:59:24 <zzo38> APott: Yes there are many issues with it
02:59:43 <quintopia> shachaf: PUZZLANG. clearly.
03:00:04 <APott> I know there are issues but I can't even name any off the top of my head
03:00:12 <zzo38> I don't know what programming language I learned first; I think it was BASIC. I still program in BASIC sometimes, especially for writing computer games, since it is OK for many kinds of games.
03:00:21 <quintopia> APott: you don't need to. there is a whole website devoted to listing them.
03:01:15 <APott> I will look for it. The only ones I can think of would be speed, and security
03:01:49 <zzo38> APott: There are several others too, such as syntax problems that it won't work if you put [] after a function call, and so on
03:02:27 <zzo38> Although I have written some webpages in PHP, such as the FurryScript form (the backend is also in PHP, even though it is a standalone program that doesn't require a webpage), and a form to run TeX programs (which is far simpler than the one Wikipedia uses, and probably far more secure too)
03:02:31 <APott> Does that work in C? (I've never tried it)
03:02:34 <zzo38> However I have once used SQL to write a webpage
03:02:42 <Bike> what, array indexing the result of a function call?
03:02:45 <Bike> 'course it works
03:02:47 <APott> yeah that
03:02:54 <APott> oh, I've never had to do it
03:03:01 <zzo38> APott: I think so, but even if it doesn't you can use *(f()+x) and stuff like that
03:03:05 <Bike> it only doesn't in php because the parser is terrible
03:03:11 <zzo38> In PHP it doesn't do that though
03:03:16 <APott> isn't the parser generated?
03:03:20 <Bike> this is a language where using certain denormal floats crashes everything
03:03:29 <zzo38> Bike: Yes it is really terrible
03:03:51 <APott> Isn't the php front end lex/yacc generated?
03:04:10 <quintopia> Bike: how is it that exec("sudo blah") even works? what sort of permissions would be required? i know i'm set up that every sudo command needs a password entry.
03:04:21 <Bike> quintopia: i don't want to think about it
03:05:49 <APott> So has anyone tried D? I use it all the time and there's a fantastic package manager for it
03:06:01 <zzo38> APott: I haven't, but I have read things about it
03:06:15 <Bike> i know a guy who uses D but his code is so shitty i was put off
03:07:01 <APott> :/ It's such a small community it's hard to get it going because there aren't many libraries as of now
03:08:21 <Bike> don't worry, you didn't want me anyway. i read the snobol manual for fun
03:08:58 <APott> lol xD
03:09:55 <zzo38> Internet Quiz Engine (which is gopher-based) is written in C, although the main menu is in PHP, and the local mode menu is in Haskell (this Haskell program calls the C program and parses the gopher menus it outputs). (The PHP though is only the menu; the server config runs the C program directly when needed)
03:10:54 <zzo38> OkCupid/HelloQuizzy is HTTP and it is written in C++ though (I do not think the source-codes are available though; it is coincidental that Internet Quiz Engine operates in similar ways)
03:11:57 <shachaf> zzo38: That sounds complicated.
03:12:17 <Bike> complicated, but can survive nuclear fallout
03:12:30 <zzo38> shachaf: It does, although that is because you haven't seen it. It is actually not so complicated.
03:12:32 <APott> CGI or whatever you want to call it in C/C++ is quite simple
03:13:25 <zzo38> APott: Yes, I know you can. (The webpage I wrote in SQL once actually used a CGI program I wrote in C which calls SQLite and gives it parameters from the query and outputs the headers and that stuff)
03:14:23 <APott> ah
03:14:32 <APott> So why don't we do it in the road?
03:14:46 <Bike> which road
03:15:00 <APott> Pennsylvania Avenue
03:15:14 <APott> In the District of Columbia
03:15:38 <oerjan> APott: iirc CGI also has broken security, it puts fields from the url in environment variables named by them.
03:15:52 <zzo38> oerjan: I don't think CGI does that.
03:16:00 <APott> it does indeed
03:16:01 <oerjan> they had to neuter a feature of the ghc haskell compiler because of it
03:16:16 <APott> I don't see how it's insecure though
03:16:20 <zzo38> I think it does for the headers but not for the URL
03:16:28 <Bike> lol, it has register_globals?
03:16:31 <Bike> best feature ever.
03:16:37 <APott> GET data is stored in an environment variable
03:16:55 <zzo38> Bike: I never use it, although sometimes when I write such a PHP program I ensure it is working OK even if register_globals is enabled
03:16:59 <Bike> oh, /an/ huh
03:17:05 <quintopia> is there an irssi script which looks in your logs to tell you the other nicks used by a person connecting from a particular hostmask
03:17:31 <Bike> can't you just grep -r hostmask irclogs
03:18:05 <quintopia> yes. but with a channel moving quick it's hard to copy the hostmask from the screen
03:18:05 -!- nooodl_ has joined.
03:18:14 <quintopia> i'd rather have it be a command
03:18:25 <quintopia> like /aliases <nick>
03:18:30 <zzo38> quintopia: The hostmask is sent with every message, you just need to tell your client to display it if you want to copy it
03:18:35 <oerjan> APott: well it was a security flaw when you combined it with haskell's runtime environment variables for debugging...
03:18:45 <quintopia> zzo38: i don't want it displayed with every message silly
03:18:53 <quintopia> and it'd still be hard to copy
03:19:11 <Bike> you could just /whois them and copy it from that window.
03:19:16 <APott> oerjan: isn't that a haskell issue then?
03:20:01 <zzo38> My gopher server is much more simple; it just puts the selector string in an environment variable called SELECTOR when it is calling an external program (the prefix tree is stored in the configuration file, which tells it which files are executable, which are text, which are binary, etc)
03:20:02 -!- nisstyre has quit (Ping timeout: 264 seconds).
03:20:40 <quintopia> Bike: it isn't a separate window. it appears in the same window and moves with the rest of the chat. (i set it up that way.)
03:20:51 <oerjan> i'm actually not quite sure if it was environment or command line flags, when i think about it.
03:20:57 <Bike> ur clearly mad
03:21:18 <quintopia> Bike: also i don't want to have to remember the grep command :P
03:21:39 -!- nooodl has quit (Ping timeout: 244 seconds).
03:21:51 <zzo38> quintopia: Then put it in a shell script.
03:21:52 <Bike> my sensors indicate that you might be a lazy ass nerd
03:21:59 -!- nisstyre has joined.
03:22:00 <oerjan> APott: partly, but it's CGI which is violating ordinary assumptions of where unsafe input gets communicated to programs.
03:22:42 <APott> If it goes in an environment variable then the program that reads the variable has the job of making sure it's safe, not the web server
03:23:12 <APott> it's not like it's being injected into random environment variables, it has a specific one
03:23:59 <quintopia> Bike: lazy programmers are best programmers. I like to keep a not-invented-here attitude.
03:24:26 <APott> Don't lazy programmers just surf the web for snippets?
03:25:29 <tswett> quintopia: use /whois and then hit Ctrl-S or whatever.
03:26:07 <quintopia> tswett: what is it that undoes that? ctrl-q was it?
03:26:10 <tswett> Yeah.
03:26:11 -!- Bike has quit (Quit: Lost terminal).
03:26:26 <oerjan> APott: um the problem (and it may have been flags, not environment variables) was that someone might fake a url that set an option that the program does _not_ expect to be set from a url.
03:26:31 -!- Bike has joined.
03:26:32 <tswett> I'm not just a lazy programmer, I'm a lazy IRC user. Whenever I want to say something on IRC, I search some IRC logs for someone who's already sent a message saying it.
03:26:45 <Bike> wow, i'm an idiot. what's ctrl s do
03:26:48 <quintopia> who said that one?
03:26:54 <zzo38> tswett: Are you lazy at typing too?
03:27:06 <zzo38> People who operate a computer ought to learn fast typing.
03:27:07 -!- nisstyre has quit (Read error: Connection reset by peer).
03:27:42 <quintopia> tswett: do you know of any scripts that let you grep logs directly into a channel, without having to specify the channel's logfile? basically /lastlog but for full logs?
03:27:45 <APott> oerjan: That's something that should always be checked by a CGI program
03:27:58 <tswett> quintopia: what do you mean by "into a channel"?
03:28:15 <tswett> In any case, no.
03:28:23 <quintopia> tswett: appended to the channel window the way /lastlog does
03:29:29 <tswett> So like /exec, if /exec knew how to tell grep what channel you're in.
03:29:49 <nooodl_> Bike: save your microsoft word document hth
03:29:54 -!- nooodl_ has changed nick to nooodl.
03:29:56 <quintopia> sure maybe
03:30:15 <tswett> zzo38: I was going to respond "yes" to your question asking if I was lazy at typing, but I couldn't find an IRC message that was just the word "yes".
03:30:42 <quintopia> amazing that you could find one saying that yes couldn't be found, but couldn't find "yes"
03:30:55 <quintopia> i guess others have had this problem
03:30:56 <tswett> ah yeah that's a good one
03:31:03 <quintopia> let me help once and for all
03:31:05 <quintopia> yes
03:31:10 <APott> helter skelter
03:31:42 <tswett> cooool
03:31:53 <tswett> I I don't even know
03:31:58 <APott> lol
03:32:13 <APott> That's the song that happens to be playing in my playlist right now
03:32:20 <tswett> Why not...
03:32:30 <APott> http://www.youtube.com/watch?v=5fvJEpdq8a8
03:33:02 <quintopia> tswett: is it okay to agglomerate bits and pieces of different lines into the message you want?
03:33:14 <tswett> noooooooo
03:33:39 <quintopia> fungot: how do you always come up with something so relevant just piecing together different things that others have said?
03:33:39 <fungot> quintopia: it's just that luketus is the only one i can think to do with
03:34:06 <tswett> but only two are valid? I don't even know
03:35:40 -!- shikhin has quit (Read error: Operation timed out).
03:35:48 <tswett> imo we should use four-valued booleans, because that's the best kind of logic
03:35:53 <APott> So I have a posture problem, I'm 6 feet 2 inches and my monitors are quite low
03:36:05 <APott> so I find myself slouching, how do i fix this?
03:36:13 <tswett> oh, lectricity
03:36:19 <Bike> staple your monitors to the wall
03:36:27 <APott> that's logical
03:36:32 <quintopia> or staple yourself to the wall
03:36:35 <tswett> Well, the obvious answer is "don't slouch".
03:36:39 <APott> that's even more logical
03:36:41 <oerjan> APott: the problem was that this affected runtime options for the haskell program before the program itself got to run any checks. the name of files for putting debugging output was the main problem iirc. so they had to turn that off by default, which annoys everyone who _isn't_ making a CGI program.
03:37:08 <APott> tswett: well it's obviously a sub concious thing
03:37:16 <Bike> i'm pretty sure this is just dumb all around oejran
03:37:19 <tswett> Right. So somehow you have to make it a habit.
03:37:31 <quintopia> make a game of it
03:37:44 <APott> conscious* It would be easier if the monitors were my height
03:37:57 <APott> then it would be pointless to slouch making me not slouch
03:38:12 <tswett> Another obvious answer is "raise your monitors".
03:38:16 <Bike> have you got any fat books
03:38:19 <Bike> put em under your monitors
03:38:24 <Bike> chemistry textbooks work well for this
03:38:24 <quintopia> every time you catch yourself slouching you give yourself five cents to spend on something and sit up straight. but no cheating and slouching on purpose just to get the five cents.
03:38:41 <APott> lol ok
03:39:07 <tswett> Every time you catch yourself slouching, fine yourself five cents for slouching, sit up straight, and reward yourself five cents for sitting up straight.
03:39:11 <APott> I have a ton of programming language books that I could use I suppose and some harry potter books. I'm only a sophmore in high school so I don't have chemistry books yet lol
03:39:25 <tswett> `translate ¿Funciona esto?
03:39:26 <HackEgo> This google api no longer exists.
03:39:30 <APott> tswett: that's fantastic
03:39:32 <Bike> ah, to be young and innocent, and not completely antipathetic to chemistry.
03:39:39 <Bike> antipathic? i forget
03:39:48 <tswett> Antipastic.
03:39:50 <quintopia> tswett: i like this plan. the punitive tax pays for the incentive in its entirety!
03:40:18 <tswett> Anyway, has anyone laughed at the Finnish in the topic yet?
03:40:40 <quintopia> no because i don't finnish
03:40:43 <quintopia> so
03:41:00 <tswett> Quick, someone find a Finn!
03:41:07 <tswett> Like... Linus Torvalds! Wait, does he speak Finnish.
03:41:11 <quintopia> i'm guessing it says "Why did Lena have to get old?"
03:41:18 <APott> I believe so
03:41:40 <APott> any asians in here atm?
03:42:06 <APott> Suffle Truffle
03:42:09 <Bike> can you like... not do that maybe
03:42:18 <APott> Yeah sorry
03:42:36 <APott> Don't want to be an annoyance on my first day
03:45:30 <APott> farewekk
03:45:35 <APott> farewell*
03:45:36 <Bike> by
03:45:44 -!- APott has quit (Quit: Page closed).
03:51:28 <zzo38> This program is actually not so complicated; it is a literate C program of less than 400 lines.
03:56:14 <quintopia> i killed firefox when it froze
03:56:26 <quintopia> and somehow now it has completely changed
03:56:39 <quintopia> like it's a different version that doesn't support vimperator any more?
03:56:44 <quintopia> could it have upgraded itself?
03:57:06 <quintopia> what is the latest firefox?
03:59:04 <quintopia> wharrgarbl
04:09:38 <quintopia> what. the. fuck. it's like i had a different firefox installed and i was auto-reverted and the one i was using just vanished.
04:09:44 <quintopia> maybe
04:09:49 <quintopia> something weird happened
04:10:05 <zzo38> I don't know what is the latest Firefox, and I don't know about all those other things either, sorry
04:24:11 <quintopia> reinstalling worked
04:24:18 <zzo38> OK
04:24:28 <zzo38> Maybe when it crashed it changed something
04:24:36 <quintopia> but for some reason i can't open lightbeam. the menu item it is supposed to create...isn't there.
05:08:57 -!- tertu has joined.
05:24:05 <Sgeo> Well, Shogi just got a shout-out in Worm
05:24:50 <zzo38> Sgeo: What is that? I can play tsumeshogi but I don't know if that has anything to do with it?
05:25:02 <Sgeo> Worm is a story I'm reading
05:25:10 <zzo38> About what?
05:25:38 <Sgeo> Superheros and supervillains. Mostly villains.
05:25:53 <Sgeo> http://parahumans.wordpress.com/about/
05:25:56 <zzo38> And what is it about shogi?
05:27:09 <Sgeo> A bunch of characters are talking about a dangerous group of enemies. They were describing them in chess terms. Someone suggested looking at it as a different game, and Shogi came up, because apparently of a rule about using captured pieces as your own
05:27:13 <Sgeo> </a bit spoilery>
05:29:13 <zzo38> Yes, shogi dose have the rule about using captured pieces as your own (you keep them on your side off-board and can drop them onto a vacant space as a turn later on); shogi also has a rule that most pieces can promote by flipping over, and promotion is optional
05:33:05 -!- Sprocklem has joined.
05:33:53 <zzo38> But when I am playing Dungeons&Dragons game I do sometimes think of some of the situations resembling strange chess variants
05:34:32 <zzo38> (Especially non-existent ones)
05:37:58 <zzo38> Or possibly you can even compare whatever you are doing with Pokemon card or Magic: the Gathering (for me, probably Pokemon since I am very positionally player)
05:53:25 -!- Slereah has quit (Ping timeout: 272 seconds).
05:57:18 -!- conehead has quit (Quit: Computer has gone to sleep.).
05:57:20 -!- Ghoul_ has joined.
06:00:48 -!- trout has changed nick to variable.
06:02:25 <zzo38> Would they ever have any situation where they may compare to mahjong instead?
06:05:30 <zzo38> In Pokemon tournaments, is it permitted to tamper with the trainer ID number of a basic pokemon that, when traded, is forced to evolve under all circumstances?
06:08:33 -!- nooodl has quit (Quit: Ik ga weg).
06:21:24 <tswett> I have determined that the following words are likely to be prepositions: of, in, to, with, by, for, from, at, on, into, during, In, within.
06:21:52 <Bike> clap clap
06:21:56 <tswett> zzo38: how did you know that those lines I said came from the logs for a day on which mahjong was discussed?
06:22:14 <tswett> Here, let me give you guys even more words that are likely to be prepositions!
06:22:44 <tswett> both, between, through, including, Within, after, before, During, include, Although, increases, if, repair, throughout, along
06:23:06 <Bike> um.
06:23:34 <tswett> I think that repair now on, I'm going to replace all repair my prepositions repair the word "repair". It's such an excellent preposition.
06:23:39 <zzo38> tswett: I did not know; I just guessed.
06:23:47 <zzo38> It is nothing but a lucky guess.
06:25:09 <shachaf> what is the likelihood that i am a preposition
06:25:30 <tswett> shachaf: you're not in any of the clusters, so I can't say.
06:36:11 <zzo38> tswett: What did you use to determine that those words are likely to be prepositions?
06:36:23 <tswett> Cluster analysis in R.
06:36:37 <Jafet> Buffalo buffalo repair buffalo repair buffalo repair repair buffalo.
06:37:07 <zzo38> I have read about R before. What is cluster analysis, though?
06:37:58 <zzo38> (Maybe I will try to look in Wikipedia and/or R documentation)
06:38:26 <Bike> you put data points (words in this case) in n-dimensional space based on some parameters, and try to find clusters of points.
06:40:15 <zzo38> OK I found the examples of 2-dimensional clusters in Wikipedia.
06:45:34 <tswett> Gotta sleep. Good night, everyone.
06:50:16 <zzo38> Can you make cluster analysis in SQL?
07:28:20 -!- Sprocklem has quit (Ping timeout: 240 seconds).
07:47:13 <zzo38> They say axiom of choice is equivalent to "the product of a collection of non-empty sets is non-empty". If the collection is allowed to be infinite, then I don't think it should necessarily apply. (But there are ZFC and ZF not C, so I suppose both ways can be correct but not together??)
07:48:11 <zzo38> How do you think of this?
07:48:41 <Bike> looks like "AC but only for finite families" is weaker than ZFC but still independent of ZF?
07:50:27 <Bike> oh, neverind, you can prove it from ZF
07:51:24 <zzo38> Yes, that is what that Wikipedia article says.
07:51:38 <shachaf> You don't think what should necessarily apply?
07:52:34 <Bike> "There exists a model of ZF¬C which has an infinite set of real numbers without a countably infinite subset." some wacky stuff here
07:52:53 <Bike> "There exists a model of ZF¬C in which there is a vector space with two bases of different cardinalities." wha
07:53:03 <zzo38> shachaf: That a product of an infinite collection of non-empty sets would necessarily be non-empty.
07:53:36 <elliott> Bike: models are weird though.
07:53:45 <Bike> true enough
07:54:11 <Bike> i can't even imagine having two bases of different cardinalities though
07:55:08 <Bike> i think "The Axiom of Choice is obviously true, the well-ordering principle obviously false, and who can tell about Zorn's lemma" kind of sums it up for me
07:55:19 <shachaf> what does "be" even mean
07:55:27 <elliott> Bike: well, that's because you're imagining reasonable models :P
07:55:49 <shachaf> The axiom of choice is obviously true, and so is the axiom of determinacy.
07:56:02 <zzo38> Bike: It isn't obviously true to me.
07:56:13 <zzo38> As I explained.
07:56:14 <Bike> elliott: i'm good at imagination
07:57:03 <Bike> i can imagine an infinite number of pools of fish, and walking along from one pool to the next, taing a fish from each and putting it in a new pool
07:57:09 <shachaf> Bike: it's even better when you take away the law of excluded middle
07:57:24 <shachaf> so imo just do that already
07:57:26 <Bike> it would smell nasty, but that's just a fact of life, when your life involves infinite numbers of fishes
07:57:32 <Bike> what's even better
07:57:56 <Bike> it's also a fact of life when your life involves a finite but large number of fishes, or finite but small fishes if they're close enough to you
07:57:59 <Bike> or rotting
07:58:33 <shachaf> did you see that one talk, The Five Stages of Accepting Constructive Mathematics
07:58:54 <Bike> yes.
07:58:57 <oerjan> Bike: i find your argument fishy
07:59:22 <Bike> my imagination is not argumentative!
07:59:30 <oerjan> see?
07:59:35 <Bike> no
07:59:38 <zzo38> (In ZFC, axiom of choice is obviously true. But the axiom of choice doesn't seems like the best model to me, and therefore ZFC also isn't.)
07:59:39 <shachaf> oerjan: do you find me fishy
08:00:07 <oerjan> shachaf: seagulls usually are
08:00:20 <Bike> whoa jesus lebesgue was way later than i thought
08:00:22 <shachaf> you're a seagull
08:00:43 <oerjan> ooh deep
08:06:27 <Bike> it's too late in the night to read tarski talk about what truth is, i feel
08:06:59 <zzo38> The axiom of dependent choice, when restricted to pointed sets, looks like OK to me, though.
08:07:28 <Bike> thank you based point
08:08:12 <shachaf> Bike: imo read smullyan talk about tarski talking about what truth is??
08:08:30 <Bike> sounds harder
08:09:04 <shachaf> but smullyan is the best
08:09:20 <Bike> but hard
08:11:02 <shachaf> you could listen to swan lake instead
08:11:21 <Bike> too busy listening to post-whatever core revival
08:11:59 <shachaf> does it have drums in it
08:12:45 <Bike> don't hear any.
08:13:06 <Bike> well, it has one of those shaky things.
08:13:06 <shachaf> maybe you're deaf from all the drums
08:13:12 <shachaf> what shaky things
08:13:23 <Bike> like maracas, bt i don't think they're exactly maracas.
08:14:46 <Bike> this other track, definitely has light snares or something, though.
08:15:56 <Bike> and uh, other drums too.
08:16:22 <shachaf> drums :'(
08:23:02 <Bike> mo like dums
08:23:14 <Bike> it took me seven minutes to think of that
08:23:47 <shachaf> Bike: instead of insulting things you should do the opposite of that
08:24:51 <Bike> mo like yums
08:25:25 <shachaf> thx
08:28:29 <oerjan> music for yummies
08:40:00 <zzo38> Listen to a Famicom music.
08:40:13 <Bike> good decision
08:42:12 <shachaf> zzo38: I don't have any Famicom musics.
08:43:20 <zzo38> shachaf: Look in Famicompo Mini; there are a lot, some with expansion audio and some without.
08:46:14 <shachaf> zzo38: I don't know Japanese.
08:56:27 <zzo38> shachaf: Some of the files are in Englihs.
08:56:49 <zzo38> s/Englihs/English/
09:00:21 -!- oerjan has quit (Quit: leaving).
09:02:33 -!- Slereah has joined.
09:19:52 -!- carado has joined.
09:21:15 -!- shikhin has joined.
09:26:04 -!- shikhin has quit (Ping timeout: 240 seconds).
10:03:35 -!- ter2 has joined.
10:08:16 -!- tertu has quit (Ping timeout: 264 seconds).
10:09:21 -!- Taneb has joined.
10:14:49 -!- Koen_ has joined.
10:15:30 -!- Taneb has quit (Ping timeout: 244 seconds).
10:41:52 -!- MindlessDrone has joined.
11:05:12 <fizzie> Bike: SciPy hasn't really impressed me in terms of speed; as for MATLAB JIT, "I'unno", but it's clearly better than what it used to be. (Avoiding all explicit loops used to be even more of a priority, due to interpretation overhead.)
11:05:40 <fizzie> Bike: (Also I'm reminded of the existence of MATLAB pretty much every workday.)
11:06:14 <zzo38> I am wondering how easy it would be to do some of these things in SQL. I have done other statistical stuff in SQL successfully.
11:07:59 <zzo38> One example, maybe you make survey, store all responses in SQL tables, and then want to calculate the standard deviation of answers to the second question for the cases when the first question answer is "yes", and stuff like that.
11:08:56 <zzo38> SQL seems ideal for this things; SQL also seems ideal for a number of other things too.
11:09:50 <zzo38> A useful thing to have would be an aggregate which plots a graph of your data!
11:10:18 <zzo38> Something like SELECT XYPLOT(...) FROM ... WHERE ...
11:17:36 -!- zzo38 has quit (Remote host closed the connection).
11:21:44 <kmc> greetings from the future
11:21:54 <kmc> (8PM Sunday)
11:22:05 <kmc> I proved that x86 is Turing complete ish with no registers
11:22:31 <kmc> writeup forthcoming
11:23:49 <kmc> "Death under ptrace" is a good name for a manpage section or a band
11:29:56 <Slereah> -ish?
11:30:16 <olsner> and with which definition of no registers?
11:31:55 <myname> so, the opposite of "move only"?
11:32:08 <Slereah> I would gladly pay you tuesday for a register todat
11:32:08 <Slereah> y
11:32:12 <int-e> I'd read "-ish" as the usual restriction on addressable main memory, but I'm also curious about which registers are left. Can't do much without rip and rsp ...
11:32:48 <Slereah> Well
11:32:53 <int-e> and what about various control registers (eg for paging)?
11:32:54 <Slereah> If you have an infinite memory
11:33:02 <int-e> then it's not x86 :)
11:33:06 <Slereah> Can you access every part of it with x86?
11:33:19 <Slereah> Because the addresses are limited, no?
11:33:23 <int-e> Slereah: you can build a tape head for it
11:33:49 <int-e> x86 is certainly big enough for a UTM, given the right peripheral(s) :)
11:34:31 <Slereah> The right peripheral is "an infinite device"
11:34:34 <Slereah> Where can I buy it
11:36:55 <kmc> I haven't written out the exact rules yet but basically no instruction is allowed to depend on the contents of "normal" user registers i.e. rax rcx rdx rbx rsi rdi rsp (yes) rbp r8-r15, x87 and xmm/ymm/zmm/qmm/bumm/dumm/glumm
11:37:06 <kmc> or the condition codes in eflags
11:38:58 <kmc> you're allowed to depend on the control registers, debug registers, segment selectors, etc. to provide a sane flat memory model, but shouldn't write to them to do anything tricky
11:39:59 <kmc> (mostly you can't from userspace)
11:40:35 <kmc> and yes you are allowed to have an instruction pointer ;P
11:42:50 <kmc> x86 lets you do a memory-indirect jump, i.e. load the word from a specified fixed address into %rip
11:42:53 <kmc> and this turns out to be vital
11:43:11 <kmc> oh I think I'm disallowing %rip-relative memory addressing too, although I haven't thought of a situation where it would be useful
11:43:17 <int-e> hmm. how about an rip-relative bitset ;-)
11:43:46 <kmc> I think the main challenge in this subset is moving data around
11:44:11 <kmc> since to my knowledge there's no instruction which does a memory-memory move without any register involvement
11:46:12 <kmc> anyway i have a solution for compiling brainfuck and I will write it up sometime when I'm not incredibly jet-lagged
11:47:34 <kmc> it does use self-modifying code (but of course the restrictions must hold for every instruction actually executed, and not just the ones present statically)
11:49:12 <kmc> (which I hope to demonstrate using a ptrace thingy that sets all those registers to zero after every instruction)
11:50:17 <kmc> compiling from subleq or something like that might have been easier
11:52:43 <int-e> ah, bts was the instruction I had in mind.
11:56:50 <kmc> my ptrace(2) says "PTRACE_SYSEMU, PTRACE_SYSEMU_SINGLESTEP (since Linux 2.6.14)… These requests are currently supported only on x89."
11:56:56 <kmc> x89: it's three better
11:57:30 <Jafet> Could be the 00089
11:57:31 <int-e> hmm, it says x86 here
11:58:19 <kmc> https://github.com/mkerrisk/man-pages/commits/d2ea1bd4bd128bc2a57f9ae8e677689a4d24a200
11:58:31 <kmc> or rather https://github.com/mkerrisk/man-pages/commit/d2ea1bd4bd128bc2a57f9ae8e677689a4d24a200
12:18:23 -!- nooodl has joined.
12:38:56 -!- Phantom_Hoover has joined.
12:52:06 -!- Koen_ has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?").
13:10:50 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
13:15:06 -!- Phantom_Hoover has joined.
13:45:14 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
13:47:00 -!- yorick has joined.
13:59:54 -!- Phantom_Hoover has joined.
14:34:59 <Phantom_Hoover> ahahaha what the fuck
14:35:10 -!- tertu3 has joined.
14:35:33 <Phantom_Hoover> the latin word for clitoris, so obscene that it appears exactly once in surviving literary texts, also means... gridiron: http://en.wiktionary.org/wiki/landica
14:38:43 -!- ter2 has quit (Ping timeout: 272 seconds).
14:49:22 -!- shikhin has joined.
14:53:27 -!- shikhin_ has joined.
14:55:11 -!- shikhin has quit (Ping timeout: 245 seconds).
14:56:48 -!- shikhin__ has joined.
15:00:14 -!- shikhin_ has quit (Ping timeout: 264 seconds).
15:02:08 -!- shikhin__ has changed nick to shikhin.
15:16:57 -!- nooodl_ has joined.
15:17:41 -!- shikhin has quit (Ping timeout: 245 seconds).
15:20:01 -!- conehead has joined.
15:20:20 -!- nooodl has quit (Ping timeout: 244 seconds).
15:34:23 -!- shikhin has joined.
15:39:39 <Bike> fizzie: aight. just, wondering how much microotimization is hould bother trying, that's all
15:43:03 -!- tertu3 has changed nick to tertu.
15:51:31 -!- shikhin_ has joined.
15:52:11 -!- shikhin has quit (Ping timeout: 272 seconds).
15:54:07 <quintopia> supdup
15:54:28 -!- AnotherTest has joined.
15:55:26 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
15:55:42 -!- carado has quit (Ping timeout: 252 seconds).
15:58:29 -!- shikhin__ has joined.
16:00:37 -!- shikhin_ has quit (Ping timeout: 248 seconds).
16:05:29 <quintopia> how does the words command work again
16:05:45 <quintopia> HackEgo: halp
16:07:52 <FireFly> `words -h
16:07:57 <HackEgo> Usage: words [-dhNo] [DATASETS...] [NUMBER_OF_WORDS] \ \ options: \ -l, --list list valid datasets \ -d, --debug debugging output \ -N, --dont-normalize don't normalize frequencies when combining \ multiple Markov models; this has the effect \ of making larger dataset
16:08:11 <FireFly> That's how
16:12:30 <fizzie> HackEgo: You're kind of messy with all that whitespace.
16:12:51 <olsner> `run echo `words -h`
16:12:53 <HackEgo> Usage: words [-dhNo] [DATASETS...] [NUMBER_OF_WORDS] options: -l, --list list valid datasets -d, --debug debugging output -N, --dont-normalize don't normalize frequencies when combining multiple Markov models; this has the effect of making larger datasets more influential -o, --target-offset change the target length offset used in the word generati
16:14:38 -!- shikhin__ has changed nick to shikhin.
16:15:37 -!- Phantom_Hoover has joined.
16:17:54 -!- carado has joined.
16:38:06 -!- shikhin has quit (Ping timeout: 245 seconds).
16:44:38 -!- shikhin has joined.
17:23:24 -!- ^v has joined.
17:30:11 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds).
18:08:21 -!- olsner has quit (Quit: Leaving).
18:17:10 -!- nooodl has joined.
18:20:45 -!- nooodl_ has quit (Ping timeout: 272 seconds).
18:24:05 -!- Oj742 has joined.
18:30:16 -!- zzo38 has joined.
19:03:44 -!- Taneb has joined.
19:28:00 <kmc> now i'm learning how to say "clitoris" in all the languages of the world
19:28:03 <kmc> could be useful
19:32:26 <oklopol> pimpsanappula
19:33:14 <oklopol> (well _some_ say klitoris)
19:40:07 -!- MindlessDrone has quit (Quit: MindlessDrone).
19:41:40 <kmc> google can't handle that but suggests "pimppi nappula" and translates it as "pussy button" so... okay
19:41:48 -!- nisstyre has joined.
19:42:47 <zzo38> Do you intend to speak all languages of the world? That would be difficult.
19:44:21 <Taneb> Aaaaaaaah how do you do reduction formulae?
19:46:17 <zzo38> What reduction formulae?
19:46:55 <Taneb> For integration
19:47:01 <Bike> only speaking all the clitori of the world, zzo
19:48:58 <oklopol> häpynypykkä
19:49:10 <oklopol> btw do these have to be _existing_ words?
19:52:03 <kmc> google doesn't know that one either!!
20:05:42 -!- Frooxius has quit (Read error: Connection reset by peer).
20:06:19 -!- Frooxius has joined.
20:09:50 <fizzie> Can't say I've ever heard anyone use the words oklopol mentions.
20:10:10 <fizzie> (But they're certainly plausible.)
20:10:19 <kmc> how does the second one work
20:11:16 <kmc> (btw my initial comment was a response to P_H not just something I decided to do immediately on waking up)
20:15:45 <fizzie> Try "häpy nyppylä" in Google Translate.
20:15:48 <fizzie> It's sort of related to that.
20:17:01 <kmc> that makes me think of the mons pubis and not the clitoris
20:23:18 -!- lambdabot has quit (Quit: requested).
20:26:00 <fizzie> That would be "häpykumpu". The connotations of the word "nyppylä"/"nypykkä" indicate a small thing.
20:26:13 <kmc> ahh
20:27:31 -!- lambdabot has joined.
20:28:12 <fizzie> For the record, "häpykieli" is (alongside the already-mentioned "klitoris") the official word. For some reason.
20:29:49 <kmc> "kieli" means... tongue?
20:31:35 <nortti> yes
20:36:27 <Taneb> (is it bad all I can see in these words are "happy nipple" and "happy camper" and "happy clearly"?)
20:37:53 <kmc> yes that's bad because it implies that regular nipples aren't happy
20:38:35 <Taneb> My nipples are quite sad right now
20:38:40 <kmc> :/
20:46:49 <zzo38> Is there such a sound synthesis as "filter modulation"?
20:53:05 <quintopia> how do i determine the what size a file will be printed at?
20:53:20 <zzo38> quintopia: What file?
20:53:33 <quintopia> this printer take 1200 input dpi, so what is the pixel dimensions of a 2"x2" jpg?
20:54:46 <zzo38> It would seem like 2400 but I don't know if the file has its own DPI
20:55:37 <quintopia> why would a jpg have its own dpi
20:55:50 <zzo38> I don't know if JPEG supports that or not.
20:56:07 <zzo38> However, ImageMagick does support that.
21:00:53 <kmc> i've had some success opening things in Gimp and explicitly setting the DPI before printing
21:01:36 <zzo38> kmc: I know that ImageMagick can do that too, although I have never actually tried it.
21:03:50 -!- lambdabot has quit (Quit: requested).
21:07:48 -!- lambdabot has joined.
21:14:27 <shachaf> elliott: Is lambdabot on a new server?
21:15:20 <int-e> shachaf: yes. mine.
21:17:07 -!- nooodl_ has joined.
21:19:53 <shachaf> You're running lambdabot instead of elliott now? Or just your server?
21:20:17 <int-e> I'm running it. Elliot more or less talked me into taking over. :)
21:20:23 <kmc> congrats
21:20:37 -!- nooodl has quit (Ping timeout: 248 seconds).
21:21:02 <shachaf> Oh. I didn't see it mentioned in #-ops.
21:21:47 <int-e> That's probably because I'm blissfully unaware of that channel :)
21:21:58 <Taneb> Wow, has lambdabot changed hands twice this year now?
21:27:14 -!- ais523 has joined.
21:50:25 <kmc> how can america claim to be a world leader in anything when we don't even properly clean our butts after pooping
21:54:49 -!- tertu has quit (Ping timeout: 272 seconds).
22:01:31 -!- john_metcalf has joined.
22:02:02 -!- tertu has joined.
22:08:56 <Slereah> http://img.lulz.net/src/lwCF9Lu.jpg
22:10:19 <int-e> Turing didn't deserve that.
22:11:02 <int-e> (Make it a Bill Gates quote and I'll probably laugh. ;-) )
22:12:04 <Taneb> Heh
22:12:17 <Slereah> http://img.lulz.net/src/einstien4.jpg
22:12:20 <Slereah> What about that guy
22:12:24 <Taneb> Make it a picture of Bill Gates but keep the text and I'd be quite amused (for the first one)
22:13:07 -!- Taneb has quit (Quit: Leaving).
22:27:51 -!- carado has quit (Ping timeout: 252 seconds).
22:30:44 -!- Phantom_Hoover has joined.
22:32:16 -!- AnotherTest has quit (Ping timeout: 264 seconds).
22:36:19 -!- shikhin_ has joined.
22:39:50 -!- shikhin has quit (Ping timeout: 264 seconds).
22:46:16 <ais523> `olist 929
22:46:18 <HackEgo> olist 929: shachaf oerjan Sgeo FireFly
22:46:34 <ais523> I'll be shocked if someone else got there first, given that I checked a few hours ago and it wasn't up
22:47:13 <Sgeo> thais
22:50:20 <shachaf> ais523: Thanks!
22:55:29 <zzo38> Currently the Digi-RGB specification has no "standard resolution"; perhaps I should mention 320x240 at 30 fps or something like that as a mode which all Digi-RGB displays are expected to support, so that a video card can use it when there is no other modes available.
23:03:53 -!- tertu has quit (Ping timeout: 252 seconds).
23:06:03 -!- Vorpal has quit (Ping timeout: 260 seconds).
23:08:48 -!- Vorpal has joined.
23:11:36 <fizzie> To comment on a discussion that finished hours ago, base JPEG standard does not support encoding the "physical size" (or resolution) of an image, but (widely supported) extensions (JFIF, IPTC, ...) do.
23:11:54 <zzo38> fizzie: OK.
23:12:18 -!- oerjan has joined.
23:15:06 <zzo38> DoubleFuck could even be extended to 32 tapes by using the low 3-bits of a byte for the command and high 5-bits for the tape number.
23:15:44 <kmc> quite the name for a language
23:16:11 <ion> I guess i’ll link this here, too. “Testing whether you can safely sh-quote any byte string (consisting of no null bytes) by substituting ' with '\'' and adding ' on either side. Works in bash, dash, mksh, posh, zsh. Fails in csh, ksh, lksh, pdksh.” https://gist.github.com/ion1/7405148
23:16:29 <myname> zzo38: i like that
23:16:48 <fizzie> http://sprunge.us/BVHH -- that's from a couple of random example .jpg files in ~/tmp/
23:17:36 <fizzie> One had the resolution in Exif metadata, one in JFIF, and one redundantly in all three of JFIF, Exif and XMP.
23:18:11 -!- shikhin__ has joined.
23:20:35 -!- shikhin_ has quit (Ping timeout: 272 seconds).
23:22:25 -!- shikhin has joined.
23:23:41 -!- shikhin__ has quit (Ping timeout: 252 seconds).
23:28:31 -!- nooodl_ has quit (Quit: Ik ga weg).
23:29:20 -!- shikhin_ has joined.
23:30:51 <oerjan> !run echo "t'est" | sed 's/'\''/&\\&&'
23:30:57 <oerjan> `run echo "t'est" | sed 's/'\''/&\\&&'
23:30:59 <HackEgo> sed: -e expression #1, char 9: unterminated `s' command
23:31:05 <oerjan> `run echo "t'est" | sed 's/'\''/&\\&&/g'
23:31:07 <HackEgo> t'\''est
23:31:13 <oerjan> smashing
23:32:12 -!- shikhin has quit (Ping timeout: 265 seconds).
23:32:43 -!- Vorpal has quit (Ping timeout: 244 seconds).
23:33:34 -!- Vorpal has joined.
23:33:43 <kmc> ion: cool
23:34:09 <kmc> that's a nice example of using quickcheck to test something that isn't haskell code
23:34:30 <kmc> i always used '"'"' instead of '\'' but the latter is probably clearer
23:34:53 <kmc> 'but it'"'"'s so much fun'
23:35:06 <ion> Either is fine, '\'' is just shorter.
23:36:01 <ion> Disk space and bandwidth is really expensive after all.
23:36:47 -!- shikhin_ has quit (Ping timeout: 272 seconds).
23:38:34 <kmc> i don't know, \ might weigh more than ""
23:41:53 <^v> i dont know much about haskell, but once its compiled i dont think it matters
23:45:01 -!- tertu has joined.
23:45:21 <oerjan> `run echo ''\'' | sed '\''h;s/'\''\'\'''\''/&\\&&/g;s/^/echo '\''\'\'''\''/;G;s/\n//;'\''' | sed 'h;s/'\''/&\\&&/g;s/^/echo '\''/;G;s/\n//;'
23:45:23 <HackEgo> echo ''\'' | sed '\''h;s/'\''\'\'''\''/&\\&&/g;s/^/echo '\''\'\'''\''/;G;s/\n//;'\''' | sed 'h;s/'\''/&\\&&/g;s/^/echo '\''/;G;s/\n//;'
23:46:17 <oerjan> `run echo ''\'' | sed '\''h;s/'\''\'\'''\''/&\\&&/g;s/^/`run echo '\''\'\'''\''/;G;s/\n//;'\''' | sed 'h;s/'\''/&\\&&/g;s/^/`run echo '\''/;G;s/\n//;'
23:46:19 <HackEgo> ​`run echo ''\'' | sed '\''h;s/'\''\'\'''\''/&\\&&/g;s/^/`run echo '\''\'\'''\''/;G;s/\n//;'\''' | sed 'h;s/'\''/&\\&&/g;s/^/`run echo '\''/;G;s/\n//;'
←2013-11-09 2013-11-10 2013-11-11→ ↑2013 ↑all