00:00:15 <oerjan> so the differences are probably not even considered a bug
00:00:19 -!- copumpkin has quit (Quit: Textual IRC Client: www.textualapp.com).
00:02:40 <oerjan> all my attempts at googling rectangular blocks of text seem to give pages talking about putting text inside a rectangle, which may still wrap arbitrarily there.
00:04:00 <oerjan> or about wrapping text _around_ a rectangle.
00:05:09 <monotone> You may as well just manually break the lines, unless that's prohibited in this language...
00:06:38 <oerjan> monotone: all my recent css changes were to programs that cannot have whitespace in them
00:06:53 <oerjan> not without major redesign, anyway
00:08:18 -!- augur has quit (Read error: Connection reset by peer).
00:08:34 <oerjan> i wonder if text-wrap:suppress would have worked if any browsers supported it
00:09:19 <oerjan> the danger is that the browser would still prefer normal break points.
00:09:20 -!- augur has joined.
00:14:00 <oerjan> dammit setting word-break still makes it prefer the usual break points.
00:27:34 -!- augur has quit (Remote host closed the connection).
00:34:36 <oerjan> Bike: can you look at http://esolangs.org/wiki/Hello_world_program_in_esoteric_languages#dupdog again?
00:35:25 <oerjan> i managed to find http://jsbin.com/bulletproof-responsive-pre/2/edit
00:36:08 <oerjan> and adding the IE line from their made it look good to me
00:36:59 <oerjan> elliott: it the \9 something fiendish?
00:37:17 <elliott> yeah, I think it's exploiting a parser bug
00:37:21 <elliott> or at least a parser disagreement
00:37:51 <elliott> lots of evil like this circa the early 2000s: http://tantek.com/CSS/Examples/boxmodelhack.html
00:37:53 <oerjan> anyway, it's the first thing i found that seems to even understand my problem :P
00:38:23 <kmc> oh that guy (Tantek Çelik) sits next to me at Mozilla SF
00:39:02 <kmc> i showed him some videos of trains.
00:39:20 <kmc> the owner of the domain elliott just linked to
00:41:22 <oerjan> elliott: ok it's a hack but does the result look ok?
00:42:22 <Bike> sorry i'm too busy dealing with printers to help
00:42:34 <oerjan> the attack of the zombie printers
00:42:40 <Bike> 27 jobs in queue and the first is from an hour but hell if i can fix anything!!
00:43:09 <Bike> i think i'll just print in another lab that doesn't suspect anything
00:43:14 <oerjan> anyone else with firefox?
00:43:33 <elliott> oerjan: can I interest you in a firefox installer? :P
00:44:35 <oerjan> now to make pizza, i think
00:44:51 <Bike> i see oerjan is an experienced web programmer
00:48:09 <kmc> http://www.jerkcity.com/jerkcity896.html
00:52:07 <oerjan> obviously not, there's no way the two lines are compatible without some serious hacking.
00:56:06 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
00:57:40 <oerjan> wait scratch that, the solution is obvious: just bring your children and register them as web developers. that automatically implies the rest.
01:00:53 -!- fungot has quit (Ping timeout: 248 seconds).
01:10:47 <Bike> https://i.cloudup.com/YTMd2VkRWl-2000x2000.jpeg Fox explains github
01:14:07 <oerjan> elliott: i put it in a class definition, since it's pretty long and at least then we can replace it if we find something better.
01:15:32 -!- nisstyre has quit (Read error: Connection reset by peer).
01:22:37 <Roujo> The rest is twitch-worthy, but more or less right =P
01:22:44 <Roujo> But the "pull request"... Nope
01:22:53 <Bike> gotta request a pull on that reposotory
01:23:30 <Roujo> What, is Github going public?
01:24:03 <Bike> https://github.com/blog/53-github-services-ipo maybe?
01:26:38 <elliott> Bike: that's a blog post from 2008 and it says the title is a lie in the first sentence... :p
01:27:40 <Bike> elliott: the fox image is undated!!
01:37:52 -!- Bike has quit (Ping timeout: 256 seconds).
01:38:54 -!- nisstyre has joined.
01:40:30 -!- audioPhil has quit (Read error: Operation timed out).
01:45:10 <oerjan> elliott: the atrocities have now been properly documented.
01:53:15 -!- Bike has joined.
01:58:39 <elliott> oerjan: you're good at the admin thing
01:58:44 <elliott> doing all this work for me.
02:01:39 <Bike> well, this is a new and exciting printer problem, it interprets everything i print as one blank page.
02:01:42 <Bike> fuck everything.
02:02:05 <oerjan> why would you want to fuck a blank page
02:02:33 <Bike> because it's there
02:02:43 <oerjan> paper cuts are nasty, man
02:03:54 <oerjan> in the mean time, you might take a hard reload on http://esolangs.org/wiki/Hello_world_program_in_esoteric_languages#dupdog to confirm the latest hack works in firefox too.
02:03:56 <Roujo> Not when your dick's made out of metal
02:04:02 <Roujo> Although I should really get that checked
02:04:21 <oerjan> (it needs to reload common.css)
02:05:39 <oerjan> (it _should_ work, since i stole it from elsewhere.)
02:06:37 <oerjan> (it might not work in less common browsers since i only stole one line of it.)
02:08:02 <oerjan> in particular i did not steal the line with a webkit comment.
02:09:02 -!- Bike has quit (Ping timeout: 256 seconds).
02:10:39 -!- Bike has joined.
02:17:24 <Roujo> `reload common.css
02:17:25 <HackEgo> Reload of common.css complete.
02:17:43 <HackEgo> echo Reload of $1 complete.
02:18:05 <Roujo> Yes, it's primitive
02:18:08 <oerjan> Roujo: i'm afraid that will only work for HackEgo's browser.
02:18:16 <Roujo> That's all we need
02:18:27 <Roujo> We just need to install lynx there and browse via him, over IRC
02:18:38 <HackEgo> install: missing destination file operand after `lynx' \ Try `install --help' for more information.
02:18:39 <HackEgo> bash: version: command not found
02:18:50 <Roujo> `run install lynx lynx
02:18:52 <HackEgo> install: cannot stat `lynx': No such file or directory
02:19:02 <HackEgo> --2013-09-13 02:19:02-- http://hats/ \ Connecting to 127.0.0.1:3128... connected. \ Proxy request sent, awaiting response... 403 Forbidden \ 2013-09-13 02:19:02 ERROR 403: Forbidden.
02:19:14 <Roujo> `run wget http://psychose.ca
02:19:17 <HackEgo> --2013-09-13 02:19:15-- http://psychose.ca/ \ Connecting to 127.0.0.1:3128... connected. \ Proxy request sent, awaiting response... 403 Forbidden \ 2013-09-13 02:19:16 ERROR 403: Forbidden.
02:19:26 <Roujo> It's blocked altogether
02:19:44 <oerjan> hint: HackEgo has a whitelist for web access.
02:20:00 <oerjan> hint 2: `fetch works around it.
02:20:27 <oerjan> `fetch http://psychose.ca
02:20:28 <HackEgo> http://psychose.ca: Scheme missing.
02:20:36 <oerjan> `fetch http://psychose.ca
02:20:38 <HackEgo> 2013-09-13 02:20:37 URL:http://psychose.ca/ [78/78] -> "index.html" [1]
02:20:48 <HackEgo> <html> \ <head> \ <title>Psychose</title> \ </head> \ \ <body>Yayyyy!</body> \ </html>
02:21:28 <oerjan> the catch is that `fetch is not useable from other commands.
02:30:16 <coppro> http://imgur.com/7tImjmw
02:33:24 -!- mnoqy has quit (Quit: hello).
02:34:15 <oerjan> i was about to ask fungot but he's gone again.
02:34:39 <oerjan> which raises the question: has fungot ever been seen together with that thing
03:11:15 -!- tswett has joined.
03:11:57 <tswett> So I pondered how this Hylisk program would translate into Haskell: "module Main where main :: World -> World; main = \x -> x"
03:12:17 <tswett> I concluded that "\x", "->", and "x" would translate to "return ()", ">>", and "return ()", respectively.
03:12:27 <tswett> Thus, "module Main where main :: IO (); main = return () >> return ()"
03:14:24 <tswett> If "\x -> x" were to be translated into a Haskell type a -> a instead of a Haskell type IO a, it would, of course, translate as "id . id"
03:15:37 <tswett> Yeah, man, dang, amirite.
03:41:54 <myndzi> sometimes irc 'help' channels really make me rage
03:42:20 <myndzi> gotta love it when you get a guy who's doing everything he can to show he's smarter than you instead of actually doing something simple like pasting a link
03:44:11 <oerjan> you'll just have to change to doing everything in haskell
03:44:37 <oerjan> ok they'll still show they're smarter than you, but they'll _also_ try to help.
03:46:46 <tswett> "Hey, I'm having trouble working through Linux From Scratch." "Do it in Haskell."
03:56:54 -!- Bike has quit (Ping timeout: 264 seconds).
04:00:10 -!- conehead has quit (Quit: Computer has gone to sleep.).
04:08:34 -!- Bike has joined.
04:15:25 <Sgeo> This reminds me, I should do Linux From Scratch
04:16:08 <kmc> does LFS give you a modern Linux desktop or is it @90s_Linux_User all the way
04:16:59 <Sgeo> Doesn't matter, I'm not planning on using it as a 'real' system
04:17:23 <kmc> it matters as far as whether it teaches you "how things really work"
04:18:06 -!- oklopol has quit (Read error: Connection reset by peer).
04:18:23 -!- oklopol has joined.
04:20:10 -!- Ghoul_ has quit (Ping timeout: 261 seconds).
04:20:53 <pikhq> kmc: LFS proper gives you a fairly sane build environment *for* a modern Linux desktop that just happens to be missing nearly every dependency.
04:21:07 <pikhq> BLFS describes how to build stuff you'd care about, like, say, XFCE.
04:21:50 <Sgeo> pikhq: do you still like Tcl?
04:22:04 * Sgeo is currently on a Tcl kick
04:22:26 -!- myndzi has quit (Excess Flood).
04:22:40 <pikhq> I consider it a fairly good example of how to do language minimalism without making something utterly impractical.
04:23:37 <pikhq> I mean really, you can do a Tcl implementation in like 500 lines of not-crazy C.
04:24:01 <Sgeo> Can a Tcl be done in a little bit of not-crazy LSL?
04:24:05 -!- myndzi has joined.
04:24:05 <Sgeo> I should attempt to do that
04:24:37 <Sgeo> Although, this 500 not-crazy C implementation, might not be as efficient as the official implementation?
04:24:51 <Sgeo> The official implementation doesn't internally do EIAS
04:25:34 <pikhq> The official implementation is a bytecode compiler, and does more "everything can be a string" than EIAS.
04:25:58 <Sgeo> Can a suspended coroutine be a string?
04:26:23 <Sgeo> I'm imaginine everything in the Tcl universe being serializable... except suspended coroutines
04:36:37 <oerjan> just keep a serializable continuation.
04:36:44 -!- Ghoul_ has joined.
04:40:43 <kmc> tcl hhas coroutines?
04:43:33 <pikhq> Not to my knowledge.
04:44:07 <pikhq> Though you could probably hack it in via stack inspection. :)
04:44:22 <pikhq> Shame the call stack isn't actually first class.
04:49:21 -!- SingingBoyo has joined.
04:51:29 -!- SingingBoyo has quit (Client Quit).
04:51:54 <Sgeo> kmc: pikhq: new to 8.6: http://www.tcl.tk/man/tcl8.6/TclCmd/coroutine.htm
04:56:55 <Sgeo> Also, there's a built-in OO system in 8.6
04:57:19 <Sgeo> Intended more to be a framework on which to make OO systems, but usable independently in and of itself
04:57:37 <Bike> how would the call stack being first class work
04:58:15 <Sgeo> In Tcl specifically, or in general?
04:58:22 <pikhq> Sgeo: That was 8.5.
04:58:28 <pikhq> I think 8.6 added to it though.
04:58:35 <Bike> in general, i guess
04:58:55 <Sgeo> http://wiki.tcl.tk/18152
04:59:08 <Sgeo> Bike: not sure how, but Smalltalk does it
04:59:35 <pikhq> Sgeo: Ah, yes, a wrapper for the constructs 8.5 had.
04:59:45 <Sgeo> pikhq: I think there was an independent package for 8.5 and it became part of core in 8.6
04:59:47 <Bike> i mean the data structure's simple to imagine. a stack made up of frames.
05:02:52 -!- SingingBoyo has joined.
05:04:42 -!- Bike has quit (Ping timeout: 264 seconds).
05:08:08 -!- carado has joined.
05:08:53 -!- jconn has quit (Ping timeout: 248 seconds).
05:15:40 -!- Bike has joined.
05:49:31 -!- FreeFull has quit.
06:04:07 -!- tswett has left ("http://quassel-irc.org - Chat comfortably. Anywhere.").
06:06:09 -!- jconn has joined.
06:07:19 -!- douglass has quit (Read error: Operation timed out).
06:28:04 -!- douglass has joined.
06:48:50 -!- CADD has joined.
06:49:04 <HackEgo> slist: Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
06:50:14 -!- Bike has quit (Ping timeout: 240 seconds).
06:57:28 -!- Bike has joined.
07:01:17 <HackEgo> olist 918: shachaf oerjan Sgeo FireFly
07:16:44 -!- douglass has quit (Ping timeout: 256 seconds).
07:17:04 -!- Bike has quit (Ping timeout: 246 seconds).
07:19:10 -!- carado has quit (Ping timeout: 246 seconds).
07:24:21 -!- Bike has joined.
07:32:02 -!- Bike has quit (Ping timeout: 256 seconds).
07:37:23 -!- Bike has joined.
07:42:17 -!- Bike has quit (Ping timeout: 256 seconds).
07:48:37 -!- Bike has joined.
07:52:27 <oerjan> the bikes keep rolling past
07:56:58 -!- Bike has quit (Ping timeout: 246 seconds).
08:04:08 -!- Bike has joined.
08:11:35 -!- Bike has quit (Ping timeout: 260 seconds).
08:32:21 <fizzie> http://sprunge.us/Pcjf?c best macro?
08:38:22 <oerjan> !c printf("%d", 01001);
08:39:11 <lifthrasiir> fizzie: can it just use 16 macros B_0000..B_1111 and make B4(x) expand to the expansion of B##x?
08:39:32 <oerjan> !c print("%d",(((01001)&1)|(((01001)>>2)&2)|(((01001)>>4)&4)|(((01001)>>6)&8)));
08:39:43 <oerjan> !c printf("%d",(((01001)&1)|(((01001)>>2)&2)|(((01001)>>4)&4)|(((01001)>>6)&8)));
08:41:15 <fizzie> Deewiant: It's the standard set on that bot.
08:41:29 <fizzie> !c printf("%d", 0b1001); /* the power of GCC extensions */
08:42:00 <Deewiant> geordi or something like that?
08:44:24 <fizzie> And sure, it's easy enough to use 16 macros for B4 -- but the octal approach can go up to a B22.
08:44:52 <fizzie> (01111111111111111111111 is less than 2^64.)
08:45:24 -!- Bike has joined.
08:50:17 -!- Bike has quit (Ping timeout: 256 seconds).
08:50:39 <fizzie> TIL: a cow is more likely to stand up in the next 15 minutes the longer it has been lying down, but not any more likely to lie down in the next 15 minutes the longer it has been standing up.
08:51:00 <fizzie> (2013 Ig Nobel Prices were given out recently.)
08:51:44 <fizzie> (Also learned: our university library agreements include the Applied Animal Behaviour Science journal.)
08:54:43 <oerjan> so cows stand up exponentially
08:55:29 <oerjan> fizzie: is that from an elsevier bundle?
08:55:42 <fizzie> It's an Elsevier journal, anyway.
08:56:05 <oerjan> also, can you guess what is currently missing...
08:56:38 -!- fungot has joined.
08:56:41 -!- Bike has joined.
08:57:27 <oerjan> probably bike, he's very on and off today
09:08:18 -!- Bike has quit (Ping timeout: 264 seconds).
09:10:03 -!- Bike has joined.
09:15:17 -!- MindlessDrone has joined.
09:17:24 -!- MindlessDrone has quit (Client Quit).
09:20:16 -!- SingingBoyo has quit (Ping timeout: 256 seconds).
09:23:26 -!- Bike has quit (Ping timeout: 240 seconds).
09:25:11 -!- CADD has quit (Ping timeout: 246 seconds).
09:44:29 <fizzie> http://sprunge.us/VeGL "lies, more lies, and documentation."
09:47:10 -!- oerjan has quit (Quit: Later).
10:06:12 -!- MindlessDrone has joined.
10:13:31 -!- audioPhil has joined.
10:13:31 -!- audioPhil has quit (Changing host).
10:13:31 -!- audioPhil has joined.
10:29:35 -!- Bike has joined.
10:41:18 -!- Bike has quit (Ping timeout: 256 seconds).
10:43:11 -!- Bike has joined.
10:58:26 -!- epicmonkey has joined.
11:21:26 <Jafet> fizzie: it doesn't say that it doesn't turn on other optimizations
11:25:25 -!- Bike has quit (Ping timeout: 263 seconds).
11:29:47 -!- yorick has joined.
11:31:29 -!- Sgeo has quit (Read error: Connection reset by peer).
11:39:54 -!- nisstyre has quit (Remote host closed the connection).
11:48:40 -!- Bike has joined.
12:02:27 -!- Phantom_Hoover has joined.
12:17:07 -!- boily has joined.
12:40:38 -!- metasepia has joined.
12:56:13 <boily> Roujo: http://pastebin.ca/2449462
13:24:18 -!- mnoqy has joined.
13:25:27 <Gregor> fizzie: There is a flag to get gcc to give you all the flags that -Owhatever expands into.
13:27:34 <boily> hm. there is such a thing as Objective-C++.
13:27:53 <Gregor> Yes, it's for gluing objc to C++
13:28:42 <boily> oh. I was confused for a moment because of the clashing object models.
13:29:34 <Gregor> It basically exists so that Safari can be written using an Objective-C UI and C++ VM ;)
13:30:55 <boily> now I'm disgusted.
13:48:31 <Gregor> Next up, Object-PL/I++
13:48:42 <Gregor> Nowait, Objective-Fortran++
13:50:27 <boily> I think Object-PL/I++ is the most terrifying.
13:51:13 <Gregor> Objective-COBOL++Script.NET?
13:52:36 * boily runs away from Gregor. “AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAH!”
13:52:58 -!- conehead has joined.
14:02:12 -!- Taneb has joined.
14:04:10 <boily> the Igs are in! http://www.improbable.com/ig/winners/#ig2013
14:07:54 <Bike> Beauty Is in the Eye of the Beer Holder
14:09:19 <Bike> ok i'm pretty honestly impressed by the archaelogy one
14:10:04 -!- conehead has quit (Quit: Computer has gone to sleep.).
14:10:20 <Bike> huh, i see one of the 2012 winners on twitter all the time. didn't know they did the eiffel study
14:12:32 <boily> parboiling a dead shrew. what the fungot where they thinking.
14:12:32 <fungot> boily: must be canadia
14:12:46 <boily> well, in fact, it is.
14:14:15 <Bike> that's just it, it's actually a pretty good idea
14:14:23 <Bike> i mean how else would you tell which bones would survive?
14:17:02 -!- rodgort has quit (Ping timeout: 240 seconds).
14:18:56 -!- rodgort has joined.
14:25:06 <boily> Bike: that's besides the main problem of why would you voluntarily eat a parboiled dead shrew.
14:25:50 <Taneb> boily, guess what I had to drink with my lunch
14:27:29 -!- AnotherTest has joined.
14:27:46 <boily> Taneb: I'm curious. or in that case, I think you are the one who was curioused.
14:28:03 <lambdabot> Local time for Taneb is Fri Sep 13 15:28:06
14:28:20 <Taneb> I think your thought patterns are along the right line, as indeed I was curious
14:28:31 <Taneb> And my drink satisfied my curiosity
14:30:16 <boily> Taneb: an argentinian coworker supplies me with mate tea-bags while we wait for another argentinian coworker to replenish the main mate stock.
14:30:36 <Taneb> boily, getting colder
14:31:19 <boily> it was only a tangential beverage. I still support the Hexhamite Botanical Infusion Hypothesis.
14:39:48 <Roujo> @ask boily What would you recommend to me re: learning Japanese?
14:51:15 <boily> I'm still here. only having weird python errors.
14:52:15 <boily> Roujo: I'd say the “Minna no Nihongo” textbooks, because they are popular, and we went through them in uni.
14:52:51 <Roujo> @tell boily I know you're here, I just wanted to ask you something so I @asked it
14:58:13 -!- Fiora has quit (Ping timeout: 248 seconds).
14:59:03 -!- Fiora has joined.
15:08:20 <boily> @tell Roujo you vile bot abuser.
15:08:45 <lambdabot> boily said 24s ago: you vile bot abuser.
15:09:09 <Roujo> boily: Also, thanks fot the tip, I'll look into those ^^
15:10:01 <boily> Roujo: as I said yesterday, I can lend you mine.
15:10:30 <kmc> fungot: sing me a song
15:10:31 <fungot> kmc: i never found the original paper on syntactic closures? syntax-case ( obviously not... :)
15:10:31 <Roujo> Ah, I missed that =P
15:10:47 <Roujo> Did I mess it up again?
15:10:48 <kmc> ^style hats
15:10:55 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
15:11:03 <kmc> ^style europarl
15:11:03 <fungot> Selected style: europarl (European Parliament speeches during approx. 1996-2006)
15:11:07 <kmc> fungot: sing me a song of eurocrats
15:11:10 <fungot> kmc: mr president, this report reflects the policy consensus between the partners, a process that began in 1989 with the first phase of the pig cycle. one thing is certain: we should insist that other contributors fulfil their commitments. otherwise the gap between levels of prosperity in the region? to my way of thinking, mrs ghilardotti, particularly, within the framework of protectionism. moreover, this draft, in its opinion
15:11:23 <Roujo> Is it possible for us to add themes to the list?
15:11:37 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
15:13:54 <boily> fizzie: I think Roujo wants you to add a new fungostyle.
15:14:31 <Roujo> Yeah, I think the RFCs might be pretty cool to markov from ^^
15:14:43 <fungot> Selected style: fungot (What I've said myself)
15:14:50 <Roujo> This might get redundant
15:15:02 <boily> fungot: do you redund with a runcible spoon?
15:15:03 <fungot> boily: just to help an fnord archive)
15:22:57 <fizzie> There are some instructions about the style-making machinery.
15:23:25 <fizzie> The RFCs do not sound like a bad idea.
15:24:56 <fizzie> I'm more inclined to add things if someone else does the necessary preprocessing into the "one full IRC comment per line, no ASCII art and page numbers and such" format, though.
15:32:01 <Roujo> Sure, that was my original question:
15:32:08 <Roujo> <+Roujo> Is it possible for us to add themes to the list?
15:32:25 <Roujo> I was ready to do the work needed, I just wanted to know if I had the rights to do it
15:32:37 -!- oerjan has joined.
15:32:47 <fizzie> Well, no. It needs the editing of some files.
15:33:02 <Roujo> What do you mean by "one full IRC comment per line"?
15:33:26 <fizzie> I mean, one example of the sort of a thing it should generate as a single comment.
15:33:45 <fizzie> The "can stop here" flags are set based on that.
15:34:13 <fizzie> (The comment length model is particularly cruddy.)
15:34:20 <Roujo> Okay, so it's not a pure markov thing
15:34:47 <fizzie> It's more or less the same, just the viewpoint is slightly different.
15:35:08 <oerjan> an ngram model is just a markov model with a sliding window state
15:35:17 <Roujo> I have sliding windows at home
15:36:00 <oerjan> `run echo "An ngram model is just a markov model with a sliding window state" >wisdom/'ngram model'
15:36:19 <oerjan> hm should markov be capitalized
15:36:40 <oerjan> `run echo "An ngram model is just a Markov model with a sliding window state" >wisdom/'ngram model'
15:36:54 <fizzie> Roujo: http://lib.tkk.fi/Diss/2007/isbn9789512288946/article4.pdf -- that's what it uses.
15:37:29 <oerjan> i suppose this would be technically falls if the ngram model had n unbounded. but i don't think fizzie does that.
15:37:52 <oerjan> i cannot even blame autocorrection.
15:38:06 <fizzie> oerjan: I don't. But it is a variable-length ngram model.
15:39:14 <fizzie> (Still, that's mostly a representation detail.)
15:39:56 <oerjan> i suppose you could have variable width windows.
15:40:27 <oerjan> obviously the width won't increase by more than 1 each step.
15:40:59 <fizzie> Roujo: If you *really* want to "do the work", you can go all the way to the actual model files used by fungot. But that requires getting some software.
15:41:00 <fungot> fizzie: so, let's say i call them mindless games. if we hit every stupid person, any person going to the theater of others. god keep. i will walke my selfe, richard. then, heaven, i love thee well; and, by my friend, c
15:41:27 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot* homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
15:42:14 <oerjan> i wonder if any of that was not shakespeare.
15:42:39 <fizzie> I don't think "hit every stupid person" was necessarily Shakespeare.
15:44:07 <Roujo> I'm mostly wondering what an actualt "prepared" document looks like. Stripping the page number, table of contents and such is easy enough, but I'm not sure I see what the lines are supposed to look like
15:44:49 <Roujo> Do you still have one of those from the other styles for me to take a look at? =)
15:46:16 <oerjan> for irc it's simple enough i assume, just use the part after : in the PRIVMSG
15:47:00 <oerjan> or after > in our usual style logs
15:47:27 <Roujo> So... just sentences?
15:47:39 <oerjan> for a book there'd be the question of how big blocks to use. maybe entire paragraphs.
15:47:52 <oerjan> you'd want it to be able to continue after a period, after all.
15:48:59 <oerjan> there's nothing really preventing you joining the whole book into line (well, maybe there's a length limit) but that would probably give some unnatural jumps.
15:49:08 <oerjan> across chapters and stuff.
15:49:37 <fizzie> Roujo: Well, I mean, it's really a matter of taste.
15:49:57 <oerjan> basically any point in the book where you want the model to be able to go from one word to the next, you want the two words on the same line.
15:50:03 <fizzie> What kind of things would you like the bot to output as a single reply, pretty much.
15:50:25 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot* homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
15:50:27 <fizzie> I've used sentences for some styles.
15:50:32 <fungot> Selected style: europarl (European Parliament speeches during approx. 1996-2006)
15:50:42 <oerjan> fungot: i bet not for this one
15:50:43 <fungot> oerjan: mr president, people outside this house. in the column on the right were disappointed that these were not in a position to consider that although the minutes mentioned specifically the epp group will therefore not support any initiative which stresses the importance of the human race in general and the prospects of enlargement and the report, the european union
15:51:12 <oerjan> fungot: that had disturbingly few periods.
15:51:14 <fungot> oerjan: madam president, i also believe that the introduction of profit sharing through the exchange of information, also exercise control of this house surely has so little room for manoeuvre. what is worse, it has been under intense discussion for days if not weeks. that confirms how right parliament was to start both the accession and negotiation procedures. as a general question. could you not arrange for people to work in
15:51:24 <fizzie> oerjan: It's not exactly strict, because it always has the zero-length context to fall back on.
15:52:19 <fizzie> The stopping is not strictly from the ngram model, it doesn't stop when the sentence-end token happens to be generated (like you might expect).
15:52:23 <oerjan> oh, so it can use that other than at the beginning?
15:53:00 <fizzie> oerjan: More than that: it can't use that at the beginning, because the context at beginning is the sentence-start token, and that always has some possible ways to continue.
15:53:43 <oerjan> um, but that means the zero length context would be even less continuity than starting a new sentence.
15:54:07 <fizzie> (It's just the overall word frequencies.)
15:54:15 <fizzie> But technically if it ends up generating a word that has never been followed by anything, but it decides not to stop there, I think it'll use that.
15:55:14 <oerjan> it feels like it would be more natural to use the sentence-start token for it.
15:55:30 <fizzie> Yes, but I don't think it does.
15:55:40 <fizzie> This is all from memory, it might work slightly differently.
15:55:54 -!- nooodl has joined.
15:56:06 -!- augur has joined.
15:56:14 <fizzie> It's also possible that it actually contains the sentence-end tokens and always stops if one of those is generated, and the extra ad-hoc rules were just an extra on top of that.
15:57:04 <oerjan> fungot: i guess you talk well enough, anyhow.
15:57:05 <fungot> oerjan: mr president, during your final six months as decided at the copenhagen summit in two minutes all i can say that the reason for this is the only possible decision. allow me, on the early warning system was set up in nairobi.
15:57:52 <oerjan> fungot: that does not sound like very good time management.
15:57:54 <fungot> oerjan: mr president, naturally i will be happy to keep parliament fully informed to the greatest possible interest in ensuring that we have failed to keep its position for the future, when it comes to this parliament: is the crisis of which the discharge has been shown with regard to preparations which are not their own personal interests. if a man campaigns for the withdrawal of troops and hardware to be deployed and allowing
15:57:57 <fizzie> As for Europarl, the units for that were single statements made by one speaker.
15:58:03 <fizzie> (They are quite long.)
15:58:18 <fizzie> So there's continuity over sentences there.
15:58:43 <fizzie> And there are some styles where the preprocessing has been quite rudimentary, and there are leftover line breaks, and it sometimes stops right in the middle.
15:58:46 <fungot> Selected style: c64 (C64 programming material)
15:58:49 <fizzie> fungot: Was this one of those?
15:58:50 <fungot> fizzie: memtop 65061, 65433 ( decimal). if you know that it is set to 0
15:59:04 <fizzie> fungot: ...if I know that, then what?
15:59:04 <fungot> fizzie: searching for game timers, and do not count when it is necessary to select the keyboard.
15:59:52 <fizzie> Still, extracting sentences is one reasonable thing to do.
16:00:57 <oerjan> <Gregor> Next up, Object-PL/I++ <-- i'm thinking of Objectivist-B. you have to start at B, because A is A.
16:01:57 <oerjan> (objectivist-C is of course well known already.)
16:04:06 <fizzie> Yeah, the stopping rule -- assuming the Perl prototype matches fungot -- is, in fact, "for each point where $prefix->{canstop}, stop if rand(20) < (number of generated words)/5 + 1". So it depends partly on the model ('canstop' is set if the sentence-end token is possible in that context), but the probability for generating it is explicitly set based on the current length instead of using its frequency.
16:04:07 <fungot> fizzie: once you've made a sprite onto the stack entries made by for, loading, and change the number of the number is written to with a secondary address of the logical file
16:06:07 <fizzie> In retrospect I really should've used either just the sentence token (and accepted the somewhat bad length distribution modeling), or used a proper separate length model (say, one Gaussian) with parameters set from the data.
16:06:42 <fizzie> Still, "it works", as an engineer would say.
16:06:53 <oerjan> fungot: are you any good at generating gaussians?
16:06:53 <fungot> oerjan: purpose: set ieee bus. remember, that is filled. it is an arbitrary starting point similar to def fn. def fn a(x)=x+7 with a secondary command, it can be reset to 0
16:07:24 <fizzie> I don't think that's how you generate Gaussian random numbers.
16:10:41 * oerjan vaguely recalls that they were simplest to generate in pairs.
16:11:12 <fizzie> The good old Box-Muller.
16:11:34 <fizzie> Both do make a pair of norma deviates from a pair of uniformly distributed numbers.
16:11:53 <fizzie> Another fungot deficiency: it always selects the longest context with words, instead of using the backoff weights. (That's one reason for those loops.)
16:11:53 <fungot> fizzie: by looking at the time, the 6566/ 6567 fetches data from.
16:12:49 <oerjan> istr you removed one of the reasons
16:13:12 <fizzie> Yes, there was also a bug, the details of which I forgot.
16:13:36 <oerjan> something about it only looking at a limited subset of the options, iirc
16:14:13 <fizzie> Yes, the number of ? iterations (each iteration generates 2 bits) was too low for the varikn-generated models.
16:15:21 <fizzie> (The frequencies I get out from varikn are just numbers in the [0, 1] range, which I multiplied with some big constant K because the bot works with integers -- the old styles had actual word counts there, and the numbers were generally smaller.)
16:18:30 <oerjan> https://en.wikipedia.org/wiki/Ziggurat_algorithm
16:21:21 -!- Koen_ has joined.
16:21:57 -!- conehead has joined.
16:22:52 <fizzie> Roujo: Getting back on track, http://git.zem.fi/fungot/blob/HEAD:/varikn/readme.txt has the instructions if you want to actually make a model (in which case you get to do the parameter-tweaking); or you can skip step 0 and stop at step 1, because the rest is really quite mechanistic.
16:22:52 <fungot> fizzie: large amounts of data is permanently read into these three bits are used to read the output of the sprite's 21-line length.
16:23:08 <fizzie> (If you do want to fiddle with it, the -D and -E parameters in those instructions are "wrong".)
16:23:25 <Roujo> fizzie: Alright, thanks ^^
16:24:33 <fizzie> ...I forgot that step #7 was still there...
16:32:43 <Koen_> oerjan: did you actually modify the dupdog program or is the diff thingy just confused?
16:35:24 <oerjan> Koen_: i took away all the newlines
16:35:45 <Koen_> that's what I thought
16:36:01 <oerjan> and eventually added a pre with the new class
16:36:31 <Koen_> I just wanted to make sure this wasn't "oh hey btw here is an optimization"
16:36:48 <oerjan> Koen_: the idea was to make the formatting by css instead, so that you can cut and paste it correctly.
16:37:26 <oerjan> since the actual program definitely should not have newlines in it.
16:41:07 -!- Koen_ has quit (Read error: Connection reset by peer).
16:41:11 -!- Koen__ has joined.
16:48:07 -!- douglass has joined.
16:52:50 <Roujo> "Bank robbery is punishable by 20 years in federal prison"
16:53:24 <boily> Roujo: did you listen to the whole thing?
16:53:35 <Roujo> I've just gotten to that point
16:54:07 -!- 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?").
16:54:27 -!- Koen_ has joined.
16:54:34 <Roujo> http://www.youtube.com/watch?v=n7JAns3PsB0
16:54:36 <boily> Phantom_Hoover: the whole shebang → http://youtu.be/MIDuZq7RVAM?list=PLTUlTwlsdlFQhHsAoG7sCxumrigHW-qJk
16:57:25 <boily> I have also the puddle version, but it's OVER 9000 decamyriabytes!
16:58:11 -!- FreeFull has joined.
16:58:17 * Roujo smacks boily around with a calisse de grosse truite
16:59:02 <FreeFull> What is calisse de grosse truite?
16:59:17 <Roujo> "Fucking big trout" =P
16:59:22 <boily> Roujo: MWAH AH AH! moéssi I can make bilingual mauvais jeux de mots!
16:59:23 <HackEgo> FreeFull: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
16:59:25 -!- augur has quit (Remote host closed the connection).
16:59:53 -!- augur has joined.
17:00:05 <FreeFull> Well, I też can mówić w dwóch językach.
17:01:11 <boily> my tongue. it is not happy.
17:01:23 <boily> Roujo: puddle. flaque. make the connection.
17:01:35 <Roujo> boily: I know. Believe me, I know.
17:03:26 -!- Bike has quit (Ping timeout: 240 seconds).
17:03:57 <Roujo> `thanks hghbhmmmmhghh
17:03:58 <HackEgo> Thanks, hghbhmmmmhghh. Tghbhmmmmhghh.
17:04:26 -!- augur has quit (Ping timeout: 264 seconds).
17:05:26 * oerjan swats Roujo and boily for making him listen to this painful thing -----###
17:05:37 * kmc hugs douglass
17:05:39 -!- Bike has joined.
17:05:51 * boily hides behind Bike to dodge oerjan's swat.
17:06:16 * Roujo build a shed around Bike
17:06:19 <elliott> oerjan: if you think that's painful...
17:06:30 * Roujo then debates about the most suitable color for the shed
17:06:39 <Bike> boily: well it's a job where you go to syria to look at centuries-old garbage.
17:06:40 <Bike> it's my favorite color
17:06:47 <kmc> orange! that's right!
17:06:48 <Bike> not even a question tbh......
17:07:15 <oerjan> elliott: oh i'm sure there are far worse things.
17:07:41 <boily> my shirt is orange.
17:08:19 <boily> we're bad at bikeshedding. we need something more... disagreement-prone.
17:09:40 <oerjan> elliott: but there were these words they mentioned that i tried to listen until i found, but it was impossible to hear the words said _and_ having the counting song be at a bearable volume at the same time.
17:09:47 <Roujo> `run cat cat | cat
17:10:49 <HackEgo> cat: run: No such file or directory
17:11:45 <Bike> "| cat" is kind of awesomely pointless
17:12:29 <Roujo> `run cat cat | echo
17:12:40 <Koen_> what! it should echo a newline
17:12:43 <Roujo> `run cat cat | cat | cat | cat | cat
17:12:57 <oerjan> i think that worked perfectly
17:13:24 <oerjan> Koen_: HackEgo strips final newlines.
17:14:59 <oerjan> ^def ul (That's interesting!)S
17:15:14 <boily> \input insert \x0a
17:15:32 <oerjan> ^ what do you think about this?
17:15:42 <fungot> echo reverb rev rot13 rev2 fib wc ul cho choo pow2 source help hw srmlebac uenlsbcmra scramble unscramble asc ord prefixes tmp test celebrate wiki chr ha rainbow rainbow2 welcome me tell eval elikoski list ping def a
17:16:38 <fungot> (Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot)S
17:16:50 <fungot> (That Pong alone cannot stop!)S
17:17:37 -!- Taneb has quit (Quit: Leaving).
17:17:52 <kmc> we have found a bug arising from the fact that Rust macros are not hygenic wrt named loop break labels
17:18:18 <Bike> are rust macros just c style or what
17:18:57 <oerjan> is rust trying to have scheme style hygienic macros
17:19:02 <Roujo> Fun fact. "`run echo rofl" != "`run `which echo` rofl"
17:19:05 <Roujo> And I have no idea why
17:19:28 <Roujo> `run `which echo` rofl
17:19:49 <kmc> Bike: no they are closer to Scheme macros although only recently partially hygenic http://static.rust-lang.org/doc/tutorial-macros.html
17:19:49 <Bike> `run `which cat` cat
17:19:52 <oerjan> Roujo: echo is normally intercepted by the shell for efficiency.
17:20:06 <Bike> kmc: oh, nice.
17:20:42 <Bike> well i don't get this syntax at all but whatever
17:20:47 <kmc> i kinda don't like these special purpose macro languages. just lemme run code at compile time and spit out an AST!
17:21:39 <Bike> you can't expose compiler things like the AST at compile time! the world would go mad
17:21:45 <kmc> but it works out ok so far
17:22:03 <elliott> ASTs are such a poor interface though
17:22:10 <Bike> lisp more like shitsp
17:22:12 <elliott> (not that I really have a better suggestion)
17:22:22 <Bike> what's wrong with asts
17:22:35 <kmc> Phantom_Hoover: Rust is a systems language that provides explicit control over allocation, but in a way where the compiler can prove memory safety
17:23:21 <kmc> so it has a few novel features supporting that, and the rest is just like "let's design a systems language for 2013 not 1970" so things like pattern matching, macros, real module system, etc
17:24:15 <elliott> did Phantom_Hoover just successfully troll kmc? :p
17:24:37 <Bike> troll? i thought he just asked what rust was and kmc answered
17:24:50 <kmc> it's not like it took me long to compose that answer ;P
17:24:58 <kmc> since people keep asking me that
17:25:13 <elliott> I thought it was in response to
17:25:14 <elliott> 18:21:51 <Phantom_Hoover> ah but that's why you use lisp
17:25:14 <elliott> 18:22:01 <Phantom_Hoover> because it doesn't have syntax, obviously
17:25:15 <boily> kmc probably has a macro to explain rust.
17:25:18 <kmc> if you really want to troll ask "how is this better than Go?"
17:25:24 <Bike> that's too obvious\
17:25:44 <kmc> Go is a completely different language for a completely different niche, and yet people can't stop comparing Rust to it for some reason
17:25:52 <Bike> kmc trolls himself
17:26:30 <Phantom_Hoover> kmc, and if you added monads it would basically be haskell
17:26:32 <boily> oerjan: naturellement, mais pourquoi en ce moment-ci?
17:26:48 <Bike> ah, now there's a half-decent troll
17:26:51 <kmc> Phantom_Hoover: to Go or Rust?
17:27:04 -!- conehead has quit (Quit: Computer has gone to sleep.).
17:27:24 <elliott> recipe for Haskell: add monads to water and season according to taste
17:27:50 <boily> one day, the majority of languages will have monads in them.
17:27:59 <Bike> a beautiful future
17:28:08 <nooodl> what's the recipe for the strathclyde haskell enhancement
17:28:16 <Bike> random thought: when did "assembly" become a monolithic thing
17:28:25 <nooodl> is it just a bunch of seasoning
17:28:26 <Bike> since it's, not.
17:29:11 <Fiora> Bike: maybe it's like, it's a monolithic thing to people who are kind of outside of it?
17:29:19 <Fiora> and don't really understand it
17:29:21 <Bike> Fiora: well yes, but why
17:29:23 <Roujo> Instant Haskell: Just Add Water! (Ingredients: Pure Monad Powder)
17:29:25 <Bike> or when rather
17:29:32 <nooodl> Phantom_Hoover: is that a recipe or a personal thing
17:29:35 <Fiora> Bike: maybe it's like, how "functional languages" are a monolithic thing to some peole
17:29:45 <Fiora> like they're al lisp or something I don't know (?)
17:29:52 <Bike> well i mean that's taken as a group
17:29:59 <Bike> they're probably vaguely aware that ML and Lisp have differences
17:30:09 <elliott> I think lisp suffers from that itself
17:30:10 <Bike> but people talk about "assembly" as if it as one language
17:30:18 <Bike> suffers from what, being seen as monolithic?
17:30:19 <elliott> like people lump all the languages with lots of parens into one thing
17:30:24 <Bike> yes it's silly
17:30:42 <oerjan> boily: because Bike asked what's wrong with it
17:30:44 <Bike> the whole "lisp is god" thing doesn't help, but whatever
17:30:46 <elliott> it might just be a thing people do in general, "C/C++" and so on...
17:30:48 * boily gives a stainless bloodless machine-washable ceramic stabbing salt-shaker with nice floral motifs
17:30:56 <kmc> yeah the space of languages looks really different depending on where you are
17:31:02 <elliott> I guess assembly kind of has it the worst given that it's not even really a language family so much as a type of language
17:31:17 <kmc> people will argue forever about the differences between Python and Ruby and JavaScript and they all seem kinda the same to me
17:31:21 <Bike> at least c++ is "almost" a superset of C. scheme and common lisp and elisp are pretty damn different
17:31:33 <Bike> what's a little dynamic scope between friends, i guess
17:31:43 * boily gives the aforementioned salt-shake to noodl. you know, verb target and stuff...
17:31:51 <kmc> Bike: the whole point of "Lisp" is that you get to cherry-pick the language attributes you like from Common Lisp, Scheme, Clojure, etc. to support whataver point you're trying to make
17:31:52 <Bike> i just wish we could compile any source into a binary module and link wildly with other so compiled modules.
17:32:04 <kmc> Bike: hey wanna add C++ FFI to Rust?
17:32:08 <kmc> that would really help us out
17:32:13 <kmc> sincerely, servo team
17:32:24 <Bike> i know a guy who's trying to make a CL implementation with a C++ ffi, presumably because he's crazy
17:32:29 <kmc> presumablyp
17:32:43 <kmc> I don't think you can do C++ FFI in general without outputting lots of C++ code and compiling it
17:32:49 <Bike> yeah that's what he's doing >_<
17:32:56 <kmc> well then at least he's on the right track!!
17:33:06 <Bike> i mean, i just feel like this whole language separation thing is kind of wrong
17:33:25 <kmc> yeah, unfortuantely languages tend to have type systems / memory models that are hard to reconcile
17:33:38 <kmc> which is why we resort to lowest common denominator: C types or JSON or such
17:33:42 <Bike> you could have a binary object file that defines calling convention and so on, and have a universal symbol something
17:34:01 <kmc> plus, like, stuffing a bunch of language runtimes into the same process can be tricky
17:34:39 <Bike> i guess i don't know enough about that
17:34:46 <Bike> don't even know many languages where the runtime is a .so
17:34:47 <elliott> trying to force every language to play together nicely kinda just homogenises the languages you can use
17:35:29 <elliott> you can sorta go for only really supporting one kind of language well but having really good tight integration between everything, or supporting a wider variety of languages but everything is kinda taped together haphazardly, unix is more the latter
17:35:56 <Bike> maybe i'm just annoyed at rubygems and whatever working by downloading source.
17:36:38 <Bike> oh well. not my field. fuck computers.
17:38:13 <Bike> because i have an adamatzky fetish
17:39:15 <elliott> Phantom_Hoover: Bike wants all of the pain of CS without any of the money (;_;)
17:39:29 <elliott> that's a ;_; in parens!! not a ;_; on a face
17:39:42 <Roujo> Could be both, really
17:40:01 <elliott> how much do neuroethologists get paid... what do neuroethologists even do??
17:40:15 <Bike> well pretty soon i'm going to be a lab monkey
17:40:31 <kmc> oh, did you get one of the research jobs?
17:40:32 <Bike> all the money, and none of the pain because none of my supervisors and employers care about how bad my code is if it works
17:40:50 <Bike> kmc: interview today, but even if i don't get that one i'm pretty set on getting /a/ lab job
17:40:55 <kmc> cool, good luck
17:41:32 <Bike> it will probably involve simulating the activity of muscle proteins, but that's just a guess
17:48:27 <boily> ~duck neuroethology
17:48:28 <metasepia> Neuroethology (from Greek - neuron "nerve" and & - ethos "habit or custom") is the evolutionary and comparative approach to the study of animal behavior and its underlying mechanistic control by the nervous system.
17:49:02 <boily> Greek language is weird. how in fungot can you utter a ½...
17:49:02 <fungot> boily: there are several filter controls to set the system clock.
17:49:45 <boily> ~duck neurotheology
17:49:45 <metasepia> Neurotheology, also known as spiritual neuroscience, attempts to explain religious experience and behaviour in neuroscientific terms.
17:50:04 <Bike> avoid neuroetheology, hth.
17:50:18 <Bike> wow good misspelling.
17:51:07 <Bike> well there is neurology, if you like.
17:51:15 <boily> Bike: I was curious if neurotheology existed. it does. I am disturbed.
17:51:26 <Bike> it's not that great.
17:51:39 <kmc> it seems like a totally good and reasonable thing to study......................
17:51:42 <kmc> if a kind of dumb name
17:51:44 <boily> ~duck neurogeriatology
17:51:53 <Roujo> ~duck neuroscatology
17:52:02 <boily> ~duck neuropathology
17:52:02 <metasepia> Neuropathology is the study of disease of nervous system tissue, usually in the form of either small surgical biopsies or whole autopsy brains.
17:52:20 <Bike> kmc: most of what i know aboutit involves stupid arguments about how shocking one part of the brain making people more religious means something something dawkins fanboyism
17:52:28 <kmc> I don't care about any of that
17:52:39 <kmc> i'd like a rationalistic view of the weird shit I've experienced on drugz
17:52:53 <kmc> I don't think it'll happen any time soon, though
17:53:02 <kmc> it's probably Hard Problem complete
17:53:04 <Roujo> ~duck neurosociology
17:53:04 <metasepia> Neurosociology is the application of neurobiology to the study of society.
17:53:23 <Roujo> The more we know =P
17:53:23 <Bike> is this surprising people
17:53:33 <Bike> "neuro-" is like "computational" in terms of sticking it in front of everything
17:53:42 <Roujo> s/is this/are these/
17:53:51 <Roujo> Answer: Maybe, I haven't met them yet
17:54:02 <elliott> ~duck computational neurology
17:54:09 <Bike> neurology is doctoring
17:54:12 <oerjan> <boily> Greek language is weird. how in fungot can you utter a ½... <-- you still haven't fixed that unicode bug?
17:54:12 <fungot> oerjan: preparatory routines: talk, a special reversed character appear in the
17:54:13 <Roujo> ~duck neurocomputer
17:54:15 <Bike> ~duck compoutational neuroscience
17:54:18 <Roujo> ~duck neurocomputing
17:54:38 <Bike> kmc: but yeah most of this stuff seems like trying to understand the principles of word processor design by examining logic gates.
17:54:42 <Roujo> ~duck neuroprogramming
17:54:58 <Bike> Roujo: http://en.wikipedia.org/wiki/Neuromorphic_engineering
17:56:26 <boily> oerjan: *LA LA LA* can't hear you through the encoding problems *LA LA LA*
17:56:58 <metasepia> A trademark used for a canned meat product consisting primarily of chopped pork pressed into a loaf.
18:00:03 -!- Bike has quit (Quit: Reconnecting).
18:00:15 -!- Bike has joined.
18:01:23 <Bike> soon, there will be no psychology topic i cannot gesticulate angrily about. MWA HA HA HA HA
18:01:43 <kmc> keep your gesticles in your pants, tia
18:01:55 <Roujo> I don't see how that's going to fix your connection, but eh
18:02:06 <kmc> perhaps bonghits will fix your connection
18:02:14 <Bike> it's a thought
18:02:22 <Roujo> Bike: Most things are
18:03:15 <Bike> wow wrong meme
18:03:19 <Bike> to the corner with you
18:06:40 -!- augur has joined.
18:10:54 <Roujo> I realized a couple of days back that every single IRC channel can be read as an hashtag
18:11:43 <Koen_> what about ##doublehashtagschannels
18:12:06 <mnoqy> you can make hashtags about hashtags can't you
18:12:13 <Bike> «"Burdens are the foundations of ease, & bitter things the forerunners of pleasure" #Rumi #Spiritual #Quote #Esoteric #Mystic #Sufi #Wisdom»
18:12:48 <Roujo> And it gets even better when you think of the # channels
18:12:55 <Bike> #Esoteric : If you keep presenting yourself as a slave, you are a slave because you obey
18:13:05 <Roujo> Because apparently, # is a valid channel name
18:13:24 <Roujo> # on freenode was crap last I went there, though
18:13:45 <Bike> Should We Consider the Symbolism of Monster Energy Drinks as Being Satanic?
18:14:02 <Koen_> is # a requirement of the protocol or of the servers? or just common practice?
18:14:10 <Bike> it means something specific
18:14:14 <Bike> there are also &channels and stuff
18:14:17 <Bike> nobody cares though
18:14:42 <Koen_> what if I held a server on my computer? what if I cared?
18:14:44 <Roujo> #channels are network-wide, &channels are server-specific
18:14:54 <Koen_> there's emphasis on this "I" as I put in in caps
18:14:54 <Roujo> But yeah, they're required by the RFC, AFAIR
18:14:57 <Bike> i just said you don't care koen
18:15:36 <Koen_> what if my name is nobody
18:15:38 <Bike> The Hidden (And Not So Hidden) Messages in Stanley Kubrick's "Eyes Wide Shut"
18:16:38 <Bike> https://twitter.com/Illumine_Nation good
18:16:46 -!- Taneb has joined.
18:26:22 <metasepia> A chemist who discovers a laundry additive that rhymes with "bright."
18:28:06 <boily> greco-roman extreme ironing.
18:44:32 -!- variable has changed nick to trout.
18:51:47 <boily> Roujo: apparently, # is indeed a valid channel name. but is it «#», or «##»?
18:52:06 <Roujo> ## is yet another channel
18:53:23 <boily> uhm. you, like, have, like, joined them all, like.
18:54:26 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
18:55:29 <Roujo> I used to hang in #
18:55:56 -!- Phantom_Hoover has joined.
18:56:30 <Roujo> But yeah, I do tend to join # on every network I'm on
18:58:25 <Roujo> It's a thing on most networks
18:58:35 <Roujo> But it's banned on some =P
18:58:55 <Roujo> I think Coldfront just says "This channel may not be used"
19:02:48 -!- Bike has quit (Ping timeout: 256 seconds).
19:03:24 <boily> Roujo: trying out the various punctuational channels. I got autokicked from «_».
19:04:31 <boily> and #% is invite only.
19:07:17 <boily> Roujo: say, try to join #¨.
19:08:06 <Roujo> Illegal channel name =P
19:08:32 <boily> someone else who has a client that Accepts the Greatness of the Diæ̈resis?
19:08:45 <Roujo> Nah, it's the server who doesn't want to
19:08:50 <Roujo> [kornbluth.freenode.net][479] Roujo #¨ :Illegal channel name
19:10:22 <boily> oh. and if you just «/join ¨»?
19:14:29 -!- Bike has joined.
19:15:29 <Roujo> boily: That aliases to "join #¨"
19:15:41 <Roujo> If I "/quote join ¨"...
19:15:51 <Roujo> [kornbluth.freenode.net][479] Roujo ¨ :Illegal channel name
19:19:50 <Roujo> http://www.youtube.com/watch?v=ms-IFYNrjhU
19:20:30 -!- AnotherTest has quit (Ping timeout: 264 seconds).
19:24:19 <ion> roujo: wat
19:27:42 <boily> ion: listen to the thing. discuss with oerjan.
19:29:08 <Taneb> Can someone give me a sensible terminal size to stream Dwarf Fortress with?
19:30:41 <Koen_> is dwarf fortress an ascii motion picture?
19:30:55 <Taneb> Koen_, it's an ANSI universe simulator
19:30:59 <boily> Taneb: 80 × 25 is not good?
19:31:09 <Taneb> boily, it probably is
19:31:38 <boily> Koen_: it's a unicodified CP437 universe simulator, down to sweat drops and eyeball nerves.
19:32:34 <Taneb> I'm streaming Dwarf Fortress, telnet termcast.org
19:47:50 -!- oerjan has quit (Quit: leaving).
19:49:00 <Koen_> have we got a stack-based language where there is one data stack, and the program is a list of instructions, with the first instruction dropped before it's executed? and some instructions allow to push a "next instruction" before the rest of the program in function of what's on top of the data stack?
19:49:20 <Bike> underload but slightly suckier
19:49:58 <Koen_> and for so long I have resisted going into how underload works
19:50:29 <Bike> it's pretty easy
19:51:03 <Bike> "it even forms a monoid"
19:51:44 <boily> since when underload is monoidal?
19:51:56 <Bike> since i said so just now keep up motherfucker
19:52:10 <Roujo> That's what I got from monads. They're like single gonads, really.
19:52:14 -!- epicmonkey has quit (Ping timeout: 240 seconds).
19:53:42 * boily replaces Bike with a small semigroup
19:58:34 -!- conehead has joined.
19:59:58 -!- conehead has quit (Remote host closed the connection).
20:00:12 -!- conehead has joined.
20:29:50 -!- Bike has quit (Ping timeout: 240 seconds).
20:33:47 <elliott> Taneb: someone wants to see your dining room
20:40:49 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:54:57 -!- Koen_ has quit (Quit: Koen_).
21:08:26 -!- mnoqy_ has joined.
21:09:10 -!- mnoqy has quit (Disconnected by services).
21:09:12 -!- mnoqy_ has changed nick to mnoqy.
21:10:34 <boily> mnoqy: will you remonqy eventually, or are you mnoqy for the foreseeable future?
21:11:14 <mnoqy> it's terminal i'm afraid
21:18:51 -!- douglass has quit (Ping timeout: 256 seconds).
21:38:44 -!- Taneb has quit (Quit: Leaving).
21:52:04 -!- Bike has joined.
22:03:33 <Phantom_Hoover> MEANWHILE IN /R/BITCOIN: http://www.reddit.com/r/Bitcoin/comments/1mbo1y/my_parents_dont_believe_in_btc_how_can_i_convince/cc7o88v
22:07:13 <boily> WORK DOOOOONE! YEEEEES!
22:07:24 <boily> I'm fungotly hungry.
22:07:24 <fungot> boily: to scroll the following:
22:07:35 -!- boily has quit (Quit: foodening).
22:07:40 -!- metasepia has quit (Remote host closed the connection).
22:07:44 <Bike> that's a lot of scrolling
22:08:20 <Bike> so, does anyone know about that thing where you use commodity GPUs to do subsupercomputing? fiora?
22:09:51 <kmc> subsupercomputing?
22:10:10 <Bike> supercomputing on the cheap, basically.
22:10:19 <kmc> I know a few small things
22:10:24 <Bike> i guess it's just opencl stuff. but i don't know any of that.
22:10:33 <kmc> GPGPU is a general term used here
22:10:36 <Bike> thanks phantom, that was the joke
22:10:45 <olsner> it's probably not very "super" if it's also "cheap"
22:10:47 <Bike> "general-purpose graphical processing unit"?
22:11:21 <Bike> i think you (fiora) linked a paper once about breaking crypto using commodity GPUs or something
22:11:25 <Bike> i don't know. kinda high right now
22:11:41 <Fiora> um. I know a little but like what about it
22:11:57 <Fiora> @_@ that's kind of vague
22:11:59 <Bike> i don't know >_>
22:12:07 <Bike> i am the eternal noob.
22:12:30 <Fiora> like um what do you want to know >_<
22:12:32 <olsner> fungot: did anyone say anything interesting since this morning?
22:12:33 <fungot> olsner: 120 print nam tab(25) " total" 120 for t=0 to 1 depends on the tape at a time. the book. if corresponding bits of color ram. the
22:13:56 <olsner> fungot: I'll just take that as a "maybe"... write funnier things?
22:13:56 <fungot> olsner: if the number at the next available location in the middle of variable storage area. therefore, this register. the
22:17:36 -!- SingingBoyo has joined.
22:25:46 <Bike> maybe i should just jack a gpu and write fibonacci in opencl or whatever shit
22:25:46 <Bike> what's "hello world" for that
22:29:42 <kmc> `unicode SNOWMAN
22:31:05 <Phantom_Hoover> i'm noticing progressively more advertising getting through adblock+...
22:33:04 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name 'POO'"
22:33:15 <Bike> `unicode PILE OF POO
22:33:17 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name 'PILE OF POO'"
22:33:18 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name 'PILE'"
22:33:31 <Bike> hm, i thought that was it for sure
22:33:39 <Bike> oh, it is... is it not in the db?
22:33:51 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name 'U+1F4A9'"
22:33:58 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name '1F4A9'"
22:33:58 <Bike> `unicode just takes a char name
22:34:01 <Bike> not a codepoint
22:34:01 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unicode", line 5, in <module> \ print u"".join(map(unicodedata.lookup, sys.argv[1:])).encode("utf-8") \ KeyError: "undefined character name 'just takes a char name'"
22:34:06 <Bike> look shut up HackEgo
22:34:47 <Bike> how would it be exploitable
22:36:08 <kmc> hey did you all see http://blog.sucuri.net/2013/09/ask-sucuri-non-alphanumeric-backdoors.html non-alphanumeric php backdoor shell
22:36:23 <kmc> this is esoprogramming for sure
22:36:32 <kmc> I keep saying that exploits are weaponized esoprogramming...
22:36:53 <FreeFull> Bike: Since you wanted it so much... )) 💨💩
22:39:21 <olsner> kmc: it involves PHP... do I really want to see?
22:39:29 -!- epicmonkey has joined.
22:39:49 <kmc> @$_[]=@!+_; $__=@${_}>>$_;$_[]=$__;$_[]=@_;$_[((++$__)+($__++ ))].=$_; $_[]=++$__; $_[]=$_[--$__][$__>>$__];$_[$__].=(($__+$__)+ $_[$__-$__]).($__+$__+$__)+$_[$__-$__]; $_[$__+$__] =($_[$__][$__>>$__]).($_[$__][$__]^$_[$__][($__<<$__)-$__] ); $_[$__+$__] .=($_[$__][($__<<$__)-($__/$__)])^($_[$__][$__] );
22:39:56 <kmc> there now you have seen it whether you want to or not
22:40:14 <olsner> ah, luckily Haskell is not susceptible to the same backdoors
22:40:18 <Bike> it's code to run a shell (takes arbitrary calls to do from GET)
22:40:21 <kmc> http://i.imgur.com/IZrScxv.jpg
22:40:34 <elliott> kmc are you hacking our aim??
22:40:38 <kmc> little bitty bit
22:40:51 <elliott> no it's definitely hecking
22:41:00 <Bike> why would you heckle an aim
22:41:06 <Bike> aim heckles itself
22:41:10 <olsner> huh. I always thought it was heckling
22:41:57 <FreeFull> kmc: Someone probably reared that train
22:42:26 -!- epicmonkey has quit (Read error: Operation timed out).
22:42:39 <kmc> FreeFull: lolololol
22:45:00 <olsner> I guess a suitably limited version of PHP could make a nice esolang
22:45:52 * kmc nests iframes 50 deep
22:45:59 <olsner> hmm, and where except make is @ a prefix to suspend output?
22:46:15 <elliott> in PHP it suspends errors, I think...
22:46:39 <olsner> kmc: I think most (non-IE) browsers limit iframes to about 9 deep
22:47:06 <kmc> olsner: DOS batch files I think
22:47:21 <olsner> I've tried infintely nested iframes and everyone was fine except oerjan whose browser crashed
22:47:47 -!- augur has quit (Read error: Connection reset by peer).
22:48:01 -!- augur has joined.
22:48:31 <olsner> so what's the origin of @ as a silencing/error-suppressing prefix?
22:49:02 <kmc> the shape represents your errors spiraling into a black hole
22:56:38 -!- SingingBoyo has quit (Ping timeout: 240 seconds).
23:01:18 -!- nooodl has quit (Ping timeout: 264 seconds).
23:08:03 -!- zzo38 has joined.
23:12:40 <kmc> we'll have to agree to disagree
23:12:43 <kmc> hey zzo38, how's it going
23:13:13 <elliott> you use @ to suppress errors because @ is perfect and never has errors!
23:14:15 <elliott> don't give me that look... that... smell
23:15:30 <zzo38> Just wait a few minute please
23:15:33 <zzo38> I will answer in a bit later
23:15:40 <lambdabot> boily said 4d 1h 46m 36s ago: ♪moof♪ you have quotes!
23:15:59 <zzo38> boily: What quotes is that?
23:16:00 -!- Bike has quit (Ping timeout: 260 seconds).
23:16:15 <mnoqy> boily added your quotes to the pdf in the topic
23:22:18 <kmc> http://en.wikipedia.org/wiki/File:NancyReaganMrTChristmas1983.jpg
23:22:19 <kmc> http://upload.wikimedia.org/wikipedia/commons/4/41/Elvis-nixon.jpg
23:25:02 <elliott> NancyReagenMrTChristmas1983.jpg is definitely up there in terms of good filenames
23:31:30 -!- Bike has joined.
23:33:12 -!- yorick has quit (Remote host closed the connection).
23:41:05 -!- augur has quit (Remote host closed the connection).
23:43:57 <zzo38> I have made some levels of "Attribute Zone" game in QBASIC (I don't have it here right now though); later I intend to put in Famicom, since the game mechanics are based on the limitations of the Famicom PPU.
23:45:27 <kmc> good premise for a game
23:46:46 <zzo38> I don't know if there are other games like that.
23:47:19 <Bike> the platonic spirit of platformer characters must get through the platonic platformer game
23:47:56 <Bike> terrifying imo
23:48:29 <zzo38> Bike: I do not understand you.
23:49:10 <Bike> me neither,man. me neither.
23:49:56 <zzo38> I don't know what a platonic platformer game is either.
23:50:12 <Bike> y'know, world of forms
23:51:11 -!- copumpkin has joined.
23:52:26 <zzo38> "Attribute Zone" is not platformer game though.
23:58:36 -!- Bike has quit (Ping timeout: 256 seconds).