00:08:45 <shachaf> does it have the equivalent of -fdefer-type-errors
00:08:53 <kmc> but I keep talking about it
00:08:59 <kmc> less clear what it means in a strict language
00:09:04 <kmc> you could do it on a per-top-level-function basis, though
00:09:12 <kmc> since we don't have type inference between top level functions, anyway
00:09:22 <kmc> shachaf: you should implement it
00:09:25 <shachaf> can't you replace a type error with a call to exit(whatever)
00:09:45 <kmc> yeah but in Haskell it will only happen if the value is actually used
00:09:52 <shachaf> or the line the type error is on
00:09:58 <shachaf> well, sure, it won't be as fine-grained
00:10:35 <shachaf> but i think i'm fine with that usually
00:10:54 <kmc> only the finest of grains will do
00:13:37 <Bike> i'm imagining this disablement being like #PRAGMA fuckthisfile
00:14:00 <kmc> it's actually #![warn(warnings)] which is... not the most clear
00:14:21 <kmc> the -Wall equivalent is deny(warnings) aka -D warnings on the command line
00:14:55 <shachaf> is it scoped or just per-file
00:15:56 <Bike> warn(warnings), lol
00:18:16 <Taneb> kmc, type errors on, eg, the wrong side of a conditional can still be deferred?
00:20:02 <kmc> shachaf: all of the warning attributes can be applied to a module or function etc.
00:20:10 <kmc> and you can change the warning handling further within that thing
00:20:22 <kmc> except there's also an attribute to make a warning fatal and forbid making it non-fatal inside
00:20:26 -!- idris-bot has joined.
00:20:49 <kmc> Taneb: yeah
00:20:52 <oerjan> ( "this is your last warning"
00:20:53 <idris-bot> "this is your last warning" : String
00:29:06 -!- mhi^ has quit (Quit: Lost terminal).
00:31:19 <Melvar> I was going to say we don’t have warnings yet, but we do … --warnpartial and --warnreach .
00:38:36 <^v> a wild oerjan
00:56:15 -!- hk3380 has quit (Ping timeout: 240 seconds).
01:07:08 -!- coppro_ has changed nick to coppro.
01:12:27 -!- olsner has quit (Ping timeout: 276 seconds).
01:14:25 <Sgeo> At work, I forgot that Java enums aren't algebraic data types
01:20:32 <Taneb> Aaaaaah it's poll day today
01:22:37 <Taneb> But who should I vote for? I don't know!
01:22:49 <Taneb> Being an adult is scary sometimes
01:24:11 <quintopia> if you can't tell, the answer is probably 'not a lawyer or activist or lobbyist'
01:24:14 <Taneb> I could vote for the Yorkshire Independence Party!
01:24:18 <Taneb> (I... actually can)
01:25:31 <Taneb> Well, "Yorkshire First"
01:27:03 <Bike> is this a racist thing
01:27:37 <Taneb> I... don't know their policies
01:27:47 <Taneb> Maybe they just don't like Lancastrians
01:28:53 <Bike> It is otherwise described as a “pragmatic” centrist party with “progressive views on economic, social and environmental issues”, with the top goal of building “a self-assured and outward-looking Yorkshire where the decisions affecting Yorkshire people are taken locally”.»
01:28:58 <Bike> man this computer is bad
01:29:17 <HackEgo> 102) <coppro> what's the data of? [...] <Sgeo> Locations in a now deceased game called Mutation <coppro> I have no problems with you being interested in online games <coppro> but the necrophilia is disturbing
01:29:34 <oerjan> Taneb: this is britain where every vote that doesn't have a chance of winning its circuit is wasted, right?
01:29:57 <Taneb> The europarl elections are different
01:31:16 <oerjan> ...what is the yorkshire independence party doing there.
01:31:41 <oerjan> but yeah i think europarl is proportional, at least partly
01:32:06 <Taneb> Europarl is proportional at the region level, Yorkshire and the Humber has 6 seats
01:32:52 <Taneb> Of the 6 from last time, 2 are no longer members of their parties
01:33:39 <oerjan> just do your part to keep ukip from becoming britain's biggest europarl party twh
01:34:40 * oerjan read about that in the newpaper.
01:36:17 <oerjan> well i _think_ it was there, anyhow.
01:36:54 <Taneb> Hexham's local newspaper is "Hexham Courant incorporating Alston Herald, Hexham Herald, Haltwhistle Herald and Haltwhistle Echo"
01:38:06 <oerjan> just call it the Hexham Chimæra and be done with it.
01:46:36 -!- olsner has joined.
01:58:22 <kmc> Taneb: anyone from the Official Monster Raving Loony Party?
02:00:12 <Taneb> I'll probably vote lib dem, there's almost as good a joke
02:01:59 <pikhq> Pity, OMRLP is definitely the best party.
02:02:23 <kmc> Taneb: how do you feel about scottish independence
02:03:21 -!- xk003 has joined.
02:05:21 -!- xk002 has quit (Ping timeout: 252 seconds).
02:06:06 <kmc> it would probably send the UK lurching even further to the right :/
02:06:31 <Taneb> ...why am I drawing the 20th iteration of the dragon fractal using Python's turtle module
02:07:06 <kmc> that sounds like a fine idea
02:08:12 <Bike> why am i drawing the 20th iteration of the dragon fractal using paper and pen
02:09:17 <Taneb> Because you have too much time and paper
02:10:24 <kmc> do you ever draw a sierpinksi triangle by hand using the randomized algorithm?
02:10:25 <kmc> it's p. fun
02:11:23 <Bike> i was more the "make fun of you for doing that" guy
02:11:47 -!- realzies has quit (Quit: realzies).
02:13:21 <pikhq> Randomized algorithm?
02:13:25 <pikhq> I haven't heard of that one.
02:14:50 <Bike> http://en.wikipedia.org/wiki/Sierpinski_triangle#Chaos_game
02:16:59 -!- xk003 has quit (Ping timeout: 252 seconds).
02:18:57 -!- tromp_ has joined.
02:19:26 -!- xk003 has joined.
02:19:31 -!- xk004 has joined.
02:19:42 <oerjan> http://en.wikipedia.org/wiki/Sierpinski_triangle#Chaos_game
02:20:29 -!- xk003 has quit (Client Quit).
02:20:34 -!- xk004 has quit (Read error: Connection reset by peer).
02:20:50 -!- xk002 has joined.
02:21:01 -!- prooftechnique has quit (Quit: leaving).
02:24:45 <kmc> touchy stupidpad
02:27:39 -!- realzies has joined.
02:31:48 <oerjan> Bike: psst http://biohaskell.org/
02:32:28 <Bike> i looked up some code for this book i'm reading and it uses a matlab feature i didn't know existed, namely, declaring variables as global within a function
02:32:36 <kmc> Bike: hilarious
02:32:46 <oerjan> Christian Höner zu Siederdissen is quite a name.
02:33:19 <Sgeo> This is a great theme song https://www.youtube.com/watch?v=CGufyFt6zQc
02:33:31 <Bike> if someone writes a haskell thing that lets me vary parameter sets to a function that takes like a million parameters and doesn't look ugly as fuck i'll use it
02:34:05 <Jafet> If they're using the biohazard symbol to represent biology, they should also use \bottom to represent haskell
02:34:22 <oerjan> Bike: CLEARLY you need lenses hth
02:35:12 <Bike> well, for example, hodgkin-huxley has like a dozen parameters. i just want something to run a model with various sets of arguments that doesn't look terrible
02:35:30 <Bike> right now i'm loading variable assignments from a file (uuuuuuugh)
02:35:51 -!- xk002 has quit (Ping timeout: 240 seconds).
02:36:03 <Bike> relatedly, are there any good ode sims in haskell, i guess
02:36:15 <elliott> Bike: python has that matlab feature
02:36:59 <Bike> python has every matlab feature
02:37:15 <Jafet> Does that mean all matlab features are bad
02:37:47 <monotone> JavaScript has that feature too. Every once in a while someone actually uses it intentionally.
02:37:50 <Bike> have you ever seen a matlab program??
02:37:53 <oerjan> no, it means some python ones are hth
02:38:12 <Bike> here you know what i'll show you what i was looking at: http://www.izhikevich.org/publications/dsn/addprob.pdf
02:38:18 <Bike> point one: code is a pdf
02:38:57 <Jafet> Haskell's horrible record system is actually a p. good parameter system
02:39:06 <Bike> point two: I = par(13)*(m.^p.*h.^q)'.*((v(:,2)-E)*ones(1,length(times)));
02:39:07 -!- irctc664 has joined.
02:39:08 <Bike> THIS IS MY LIFE
02:39:15 -!- irctc664 has quit (Client Quit).
02:39:21 <Bike> yeah that's right run
02:39:26 * oerjan slips Bike the cyanide pills
02:39:35 <Bike> nobody could love me
02:39:35 <kmc> Jafet: truth
02:41:02 -!- xk002 has joined.
02:42:51 -!- xk002 has quit (Read error: Connection reset by peer).
02:53:46 -!- BeingUntoDeath has joined.
02:54:13 -!- Sgeo_ has joined.
02:54:19 -!- nycs has joined.
02:57:14 -!- not^v has joined.
02:57:15 -!- douglass1 has joined.
03:03:05 -!- Gracenotes has quit (Ping timeout: 258 seconds).
03:03:11 -!- douglass_ has quit (Ping timeout: 258 seconds).
03:03:12 -!- Sgeo has quit (Ping timeout: 258 seconds).
03:03:12 -!- `^_^v has quit (Ping timeout: 258 seconds).
03:03:17 -!- EgoBot has joined.
03:03:40 -!- EgoBot has quit (*.net *.split).
03:04:02 -!- EgoBot has joined.
03:05:23 <oerjan> !addinterp hug sh echo 'Keep your dirty fingers off me!'
03:05:24 <EgoBot> Interpreter hug installed.
03:06:37 -!- BeingUntoDeath has quit.
03:06:58 -!- BeingUntoDeath has joined.
03:07:23 -!- Gracenotes_ has joined.
03:07:23 -!- not^v has quit (Read error: Connection reset by peer).
03:07:23 -!- not^v has joined.
03:07:24 -!- MDude has quit (Ping timeout: 252 seconds).
03:07:36 -!- MDude has joined.
03:09:06 -!- conehead has quit (*.net *.split).
03:20:21 -!- ^v has quit (Quit: Leaving).
03:32:54 -!- Sorella has quit (Quit: It is tiem!).
03:34:21 <oerjan> "As obvious as this sounds, it seems to me that large swaths of consciousness-theorizing can just be summarily rejected for trying to have their brain and eat it in precisely the above way."
03:34:31 <oerjan> http://www.scottaaronson.com/blog/?p=1799
03:35:41 -!- BeingUntoDeath has quit.
03:36:50 -!- BeingUntoDeath has joined.
03:41:44 <Bike> mathematical theory of consciousness lol
03:45:14 <Bike> "Strikingly, despite the large literature about Φ, I had a hard time finding a clear mathematical definition of it" yep lol
03:45:56 <Bike> these people need jesus + thermodynamics
03:46:20 -!- oerjan has quit (Quit: Good night).
03:47:01 <Bike> the pancreas joke is funny given that i have a paper on modeling pancreatic cells open
03:48:43 -!- BeingUntoDeath has quit.
03:49:44 -!- BeingUntoDeath has joined.
03:50:05 -!- not^v has quit (Quit: http://i.imgur.com/Akc6r.gif).
03:53:47 -!- BeingUntoDeath has quit (Client Quit).
03:54:01 -!- BeingUntoDeath has joined.
04:16:11 -!- MDude has changed nick to MDream.
04:43:42 -!- Positive07 has joined.
04:47:57 <Bike> maxin', relaxin'
04:48:23 <kmc> `relcome Positive07
04:48:25 <HackEgo> 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.)
04:49:34 <Positive07> Yeah I come from there... Im quite intrigued by esolangs and looking for ideas for a new one
04:51:36 <quintopia> well, usually people come here because they already have some glimmer of an idea and want some feedback on how to improve it
04:51:57 <quintopia> and some are willing to offer said feedback as long as its not another brainfuck clone :P
04:53:35 <Bike> http://scholar.google.co.uk/citations?sortby=pubdate&hl=en&user=suo5D8wAAAAJ&view_op=list_works roll w/it
04:54:55 <coppro> Positive07: do something involving graph grammars
04:55:40 <quintopia> grammars could be more sophisticated than eodermdrome
04:55:40 <coppro> I mention graph grammars for purely selfish reasons
04:55:55 <Positive07> Yeah have some ideas... gifs qrs and cym plus some instructions, basically you take your code and make qr codes with it then put it into cyan magenta and yellow channels and combine them., do this multiple times andmakes frames of a gif
04:56:49 <quintopia> that sounds less like a language and more like an encoding scheme
04:57:19 <Positive07> yeah it is, the language its underneath
04:59:07 <Positive07> underneath its brainfuck :D just kidding -.-
05:00:26 <Positive07> hey you are into electronics too! Thats great
05:01:23 <Bike> i am not adamatzky, i just like electrocytoplasm
05:01:47 <Bike> dude's made a career out of building electronic things out of slime molds. i respect that
05:02:23 <coppro> I'm not into electronics
05:02:26 <coppro> why does google say I am?
05:04:40 <Positive07> Hated graph grammars since I was a child... not because I'm bad at it but I just dont like to represent data like that
05:06:35 <kmc> I've reached inbox 666
05:11:53 -!- tertu has joined.
05:19:38 <kmc> well I didn't have any email when I was born
05:19:41 <kmc> I didn't even have an email address
05:19:44 <kmc> unlike kids these days
05:23:11 <quintopia> i thought you were working on reducing the number of emails in your inbox
05:23:40 <quintopia> i was impressed, since it would take me a full day or even longer to get down to 666
05:27:09 -!- Positive07 has quit (Quit: Page closed).
05:29:31 -!- Positive07 has joined.
05:50:41 -!- aloril has quit (Ping timeout: 264 seconds).
06:01:45 -!- Sgeo has joined.
06:03:28 -!- aloril has joined.
06:04:08 -!- Sgeo_ has quit (Ping timeout: 252 seconds).
06:06:16 -!- BeingUntoDeath has quit.
06:13:09 <kmc> /topic серафими многоꙮчитїи
06:15:28 <kmc> why does italic г look like a backwards 's'?
06:16:18 -!- Positive07 has quit (Ping timeout: 240 seconds).
06:22:32 -!- FreeFull has quit.
06:25:07 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39637&oldid=39633 * Sacchan * (+186)
06:26:13 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39638&oldid=39637 * Sacchan * (-27)
06:26:23 <shachaf> i don't see it for https://en.wikipedia.org/wiki/Ghe_with_upturn or https://en.wikipedia.org/wiki/Gje
06:26:41 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39639&oldid=39638 * Sacchan * (-6)
06:27:35 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39640&oldid=39639 * Sacchan * (+20)
06:27:40 <shachaf> https://bugs.launchpad.net/ubuntu-font-family/+bug/823276
06:31:46 <kmc> <span color="#900">гѓ</span> I Can't Believe It's Not HTML
06:33:13 <shachaf> kmc: do you think rustc would be a better place to spend one's time than ghc if one was to work on some compiler
06:33:18 <shachaf> perhaps that's an odd comparison to make
06:35:36 <kmc> bit underspecified
06:35:47 <kmc> it depends on your goals
06:36:30 <kmc> if you want to work on exotic type system features, no
06:37:49 <kmc> it makes sense for me because working on rustc helps me do my job :)
06:37:51 <kmc> maybe I will actually implement loadable warnings
06:38:43 <kmc> do you have any nifty use cases for that? i gave a few on https://github.com/mozilla/rust/issues/14067
06:39:23 -!- hk3380 has joined.
06:39:52 <kmc> are there C or C++ projects today which use GCC plugins to implement custom static checking?
06:40:33 <kmc> PaX uses GCC plugins to harden the Linux kernel in various ways, but I think all of them involve modifying the code
06:41:52 <shachaf> I think there are things for clang.
06:45:35 <kmc> it's so nice doing systems programming in a language with a macro system that isn't shit
06:46:42 <shachaf> E.g. https://tecnocode.co.uk/2013/12/09/clang-plugin-for-glib-and-gnome/
06:48:59 <kmc> now I remember that there's also a plugin, written in Python, to audit usage of the CPython API
06:49:18 <kmc> a plugin for GCC
07:03:14 -!- glogbackup has quit (Ping timeout: 240 seconds).
07:04:37 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39641&oldid=39640 * Sacchan * (+59)
07:05:35 -!- hk3380 has quit (Ping timeout: 265 seconds).
07:06:32 -!- clog has quit (Ping timeout: 245 seconds).
07:10:54 <kmc> http://en.wikipedia.org/wiki/Tempest_prognosticator
07:29:55 -!- Bike has quit (Ping timeout: 240 seconds).
07:29:58 <HackEgo> [wiki] [[0(nop^)]] http://esolangs.org/w/index.php?diff=39642&oldid=39636 * John Misciagno * (-102)
08:13:12 -!- tromp_ has quit (Remote host closed the connection).
08:34:59 -!- Patashu has joined.
08:40:43 -!- Patashu has quit (Disconnected by services).
08:40:43 -!- Patashu_ has joined.
08:47:19 <myname> why the hell does cabal install complains about not finding a shared object if the corresponding package is installed >_>
09:06:18 -!- clog has joined.
09:26:17 -!- Phantom_Hoover has joined.
09:26:17 -!- Phantom_Hoover has quit (Client Quit).
09:26:20 -!- Phantom__Hoover has joined.
09:28:22 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
09:38:08 -!- Slereah has joined.
09:39:31 -!- Slereah_ has quit (Ping timeout: 240 seconds).
09:40:09 -!- Phantom_Hoover has joined.
10:19:02 -!- boily has joined.
10:24:05 -!- hk3380 has joined.
10:25:55 -!- MindlessDrone has joined.
10:45:34 -!- KingOfKarlsruhe has joined.
11:01:31 -!- boily has quit (Quit: Poulet!).
11:16:44 -!- tromp_ has joined.
11:21:13 -!- tromp_ has quit (Ping timeout: 252 seconds).
11:25:50 -!- KingOfKarlsruhe has quit (Quit: ChatZilla 0.9.90.1 [Firefox 29.0/20140421221237]).
11:31:37 -!- MindlessDrone has quit (Quit: MindlessDrone).
11:37:42 -!- Slereah has quit (Ping timeout: 265 seconds).
11:39:31 -!- Slereah has joined.
11:43:40 -!- xk002 has joined.
11:53:50 -!- Tritonio has joined.
12:05:57 -!- xk002 has quit (Ping timeout: 252 seconds).
12:07:42 -!- xk002 has joined.
12:13:26 -!- oerjan has joined.
12:15:06 <mroman> "pattern syntax in expression context"
12:15:36 <mroman> You can't nest case in Haskell?
12:16:53 <mroman> I guess it's some tab/spaces mixup then
12:17:39 <mroman> My editor inserts tabs when I do 4 spaces
12:17:59 <oerjan> you can program haskell with that setting.
12:18:12 <oerjan> well, technically you can.
12:18:32 <fizzie> You can tune a filesystem, but you can't tunafish.
12:18:32 -!- yorick has joined.
12:18:37 <mroman> you can. It just does'nt work when nesting cases :D
12:18:51 <oerjan> but then you must make sure to _only_ use tabs for indentation, and never do alignment-based layout.
12:19:14 <oerjan> mroman: you know haskell _defines_ tabs as equivalent to 8 spaces, right?
12:19:33 <oerjan> (or precisely, 8 space column fits)
12:19:43 <FireFly> I've seen python code-golfs make use of that
12:19:52 <mroman> oerjan: no. But I know that it's probably a bad idea to mix tabs and spaces since indentation matters
12:21:20 <mroman> or maybe it inserts tabs when pressing enter
12:21:25 <mroman> because it has this auto indentation stuff
12:21:40 <mroman> not worth discussing it ;)
12:21:56 <mroman> I've changed it to "replace tabs with spaces" and things work fine
12:27:39 <oerjan> `addquote <Bike> dude's made a career out of building electronic things out of slime molds. i respect that
12:27:41 <HackEgo> 1196) <Bike> dude's made a career out of building electronic things out of slime molds. i respect that
12:30:34 <oerjan> <kmc> why does italic г look like a backwards 's'? <-- because cyrillic italics is insane, hth
12:31:01 <fizzie> I actually wondered for quite long why "italic r" would look like a backwards s.
12:31:38 <fizzie> And also how those wikilinks about cyrillic characters are relevant w.r.t. r.
12:32:17 <oerjan> insane enough that it's a problem for me that google translate italicizes its spell correction suggestions.
12:32:28 <Melvar> You know how traditional cursive r looks not much like print r either?
12:33:08 <oerjan> oh i'm sure the latin alphabet is just as insane. what's this lower case madness?!?!
12:33:12 -!- Sgeo has quit (Read error: Connection reset by peer).
12:33:25 <oerjan> (admittedly cyrillic borrowed it from latin)
12:34:08 <fizzie> YES I ALSO THINK LOWER CASE IS ENTIRELY UNNECESSARY AND THE SAME GOES FOR PUNCTUATION TBH
12:34:31 -!- ^v has joined.
12:34:49 <oerjan> actuallyitscapitalsthatareunnecessaryasarespacesmanywritingsystemsdofinewithoutthem
12:36:02 <mroman> spacesareunnecessarytoo
12:36:14 <mroman> japanesedoesnthavethemandtheycanreadfineiguess
12:36:34 <fizzie> If it's a unicameral writing system, how can you tell if it's lacking lower or upper case?
12:36:42 <nortti> VVESHOVLDGOBACKTOTHEORIGINALLATINALPHABET
12:37:11 <nortti> ᚨᛚᛊᛟ᛬ᚱᚢᚾᛖᛊ᛬ᚨᚱᛖ᛬ᚠᛁᚾᛖ᛬ᛏᛟᛟ
12:37:11 <Melvar> mroman: They have three distinct grapheme sets though.
12:42:22 <Melvar> Has anyone else found firefox more crashy since 29?
12:43:37 <oerjan> well 28 is perfect, 29 has _got_ to be worse.
12:44:39 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39643&oldid=39641 * Sacchan * (+2)
12:49:52 <mroman> http://codepad.org/3kHXYnYM <- is this primitive recursion?
12:51:58 <oerjan> psst never use foldl(1) without ' hth
12:55:25 <oerjan> mroman: eval (Recurse xs) (Recurse xs) ... doesn't terminate afaict, thus it's definitely not primitive.
12:56:05 -!- Patashu_ has quit (Ping timeout: 264 seconds).
12:56:23 <mroman> oerjan: You can trivially express programs that don't terminate
12:56:25 <fizzie> I guess Firefox 496 will again be fine, then?
12:56:37 <oerjan> mroman: not with primitive recursion hth
12:56:40 <mroman> i.e looping until it becomes zero like the example given in main
12:56:51 <mroman> just use a positive start number and it won't halt
12:59:12 <oerjan> mroman: can you express f 1 = 1; f x | odd x = 3*x + 1 | otherwise = x `div` 2
12:59:49 <oerjan> 's pretty non-primitive recursive (since it's unproved that it halts)
13:00:01 -!- tromp_ has joined.
13:01:37 <oerjan> or are you saying that all nontermination is trivial, that might be
13:08:30 <mroman> http://codepad.org/dWcrt7fK <- jep, that's pretty much collatz
13:08:49 <oerjan> mroman: you might rewrite your Add as Value $ foldl1' (\qa b -> case eval b r args of Value qb -> qa + qb; _ -> 0) 0 xs, that's slightly different in the case the _first_ list element does not evaluate to a Value but i doubt you really want that corner case anyhow
13:10:21 <oerjan> your original ignores the value of the second element if the first does not give a Value
13:11:30 <mroman> I probably should error instead of Value 0 anyway
13:12:28 <oerjan> are args always Values?
13:12:58 <mroman> i.e Add [Mul [Value 3, Get 0], Value 1]
13:13:03 <oerjan> because that was the only way i could see your eval not giving a Value
13:13:03 <mroman> there's an expression as an argument
13:13:34 <mroman> you mean something else
13:13:35 <oerjan> (i mean if it doesn't have to be)
13:13:50 <mroman> I wan't to add some rewrite stuff too later
13:14:04 <mroman> which means that eval can return expressions too
13:14:24 <oerjan> ok well since you can express collatz it's definitely not primitive recursion.
13:20:24 <oerjan> if you error out it would probably be simpler to separate the folding and the evaluation: Value . foldl1' (+) $ map (\x -> evalToValue x r args) xs
13:21:10 <oerjan> although that could leak if you make it monadic later.
13:21:51 <lambdabot> Monad m => (a -> b -> m a) -> a -> [b] -> m a
13:24:23 -!- nycs has changed nick to `^_^v.
13:26:23 <mroman> You can actually call any function you want
13:26:29 <mroman> by just making a case statement in your function
13:26:41 <mroman> then just add a parameter to recurse which defines what it should do
13:27:12 <mroman> I pretty much suspect so
13:28:48 <mroman> I gotta define a syntax and write some parsec stuff for it
13:56:07 <mroman> (if== $0 1 1 (if== 1 (odd $0) (r (add (mul 3 $0) 1)) (r (div $0 2))))
13:56:15 -!- tertu has quit (Ping timeout: 240 seconds).
14:01:24 <mroman> Does this qualify as an esolang :D?
14:06:08 <oerjan> probably. it's not immensely radical, mind you...
14:07:04 <oerjan> this has never prevented anyone from adding stuff to the wiki. (ok, maybe someone.)
14:07:08 -!- tertu has joined.
14:14:39 -!- nooodl has joined.
14:18:41 <mroman> oerjan: There are probably already languages like that, yeh
14:22:02 -!- Sprocklem has joined.
14:28:13 <impomatic> mroman: did you want to enter that warrior in the tournament? There are a few other new players taking part...
14:30:07 <mroman> can you give me the link again?
14:46:51 <mroman> impomatic: memory is set to dat #0,#0 after each run, right?
14:50:26 <impomatic> mroman: It's dat $0, $0 and http://corewar.co.uk/spring2014.htm
14:50:28 -!- spiette has joined.
15:02:38 <mroman> impomatic: Mine uses a distance of 37, and -35
15:02:50 <mroman> so if it misses small warriors it will most likely loose
15:25:05 -!- not^v has joined.
15:38:37 -!- Tritonio has quit (Ping timeout: 252 seconds).
15:41:59 -!- oerjan has quit (Quit: leaving).
15:42:40 <mroman> Can I somehow find out in which commit a certain directory was deleted and recover it with git?
15:43:24 <not^v> not specifically no, you will have to scrape commits manually
15:43:53 <not^v> you can make a program to use the github api to find the last time something was modified
15:46:23 <mroman> why the hell am I even using a version control system
15:46:34 <`^_^v> if you know a file in the directory, you can do git log -- directory/file
15:46:41 <mroman> it's write once revert never
15:46:58 <`^_^v> i dont think git "tracks" just the directories
15:47:16 <`^_^v> well i guess it does
15:47:21 <`^_^v> but there have to be files in it
15:47:39 <`^_^v> i think git log -- directory would work
15:48:05 <mroman> I think checkout a certain commit should do it
15:49:18 <`^_^v> to recover the actual file, you can checkout or show with the commit hash
15:49:25 -!- MDream has changed nick to MDude.
15:49:28 <Jafet> Version control was invented to remember all the stupid things you did
15:50:15 <Jafet> and to vacuum up your time in the form of "commit messages".
15:50:54 <Jafet> Don't let it have its way! Make your next commit message asdfasgafsgasfy.
15:51:43 -!- shikhin has joined.
15:52:15 -!- tertu has quit (Ping timeout: 240 seconds).
16:02:48 -!- Rikkol has joined.
16:07:52 -!- Rikkol has left.
16:17:28 -!- Bike has joined.
16:47:28 <kmc> git has a pretty effective set of tools for erasing the stupid things you did
16:55:41 <int-e> Jafet: you know Linus' tech talk on git (at google), right?
16:56:43 <int-e> Jafet: because what you just said reminded me a lot of the opening words, where he talks about managing linux as a collection of tarballs and patches ... and then says that that's a superior source code management system compared to CVS :)
16:58:58 -!- nooodl_ has joined.
17:03:24 -!- nooodl has quit (Ping timeout: 276 seconds).
17:07:00 -!- FreeFull has joined.
17:12:31 <Taneb> Are any of you people in Austria
17:16:52 -!- MoALTz has joined.
17:24:14 <Taneb> I am not very good at this "getting stressed by exams" thing
17:24:22 <Taneb> I'm sat here singing Eurovision songs
17:25:39 <int-e> because after you get burned in an exam you can always rise again like a phoenix?
17:25:55 <Taneb> Because alcohol is free!
17:26:08 <Taneb> We are the winners!
17:26:08 <int-e> you don't have to sing for that, do you ...
17:26:25 -!- Sprocklem has quit (Quit: leaving).
17:26:50 <Taneb> I wasn't actually too keen on Rise Like A Phoenix, I preferred Moustache
17:28:41 <Taneb> Although I've gone off Moustache a bit now
17:35:51 <int-e> Hmm, good use of stage, but I'd rather switch off the sound, and they should rethink their color scheme ;-)
17:36:24 <int-e> <-- will not become a Moustache fan
17:37:55 -!- not^v has quit (Ping timeout: 240 seconds).
17:38:19 -!- not^v has joined.
17:39:13 -!- shikhout has joined.
17:41:26 -!- shikhout has quit (Client Quit).
17:42:38 -!- shikhin has quit (Ping timeout: 252 seconds).
17:52:00 -!- Cyragia has joined.
18:21:34 <mroman> No standard for pseudocode syntax exists, as a program in pseudocode is not an executable program. <- I see some work for the ESOSC :P
18:23:30 <FireFly> Surely some standards organ must've tried to standardise pseudocode syntax?
18:24:04 -!- tertu has joined.
18:28:06 <olsner> I think if you really standardise pseudocode it will no longer be pseudo and becomes just "code"
18:33:45 -!- Phantom_Hoover has quit (Ping timeout: 276 seconds).
18:35:53 <Slereah> If you write a pseudocode compiler
18:39:36 -!- not^v has quit (Quit: http://i.imgur.com/Akc6r.gif).
18:49:53 <mroman> If you write a pseudocode compiler
18:49:57 <mroman> it's not pseudocode anymore
18:50:14 <mroman> all programs that compile with your compiler would have to be rewritten
18:50:25 <mroman> (compile and do what they are supposed to do)
18:52:28 <Slereah> But what if I write a compiler for that other languag D:
18:54:06 <MDude> That reminds me that I'd like to make something that works like a natural language translater, but train it to attempt to turn an informal language into something that can be compiled or interpreted.
18:55:58 <mroman> bitch please print Hello, world!
18:56:35 <MDude> Runnnig Gizoogle on BASIC code does sound like a good idea.
18:58:45 <MDude> http://gizoogle.net/index.php?search=http%3A%2F%2Fen.wikipedia.org%2Fwiki%2FBASIC&se=Gizoogle+Dis+Shiznit#Examples
18:59:07 <MDude> 10 PRINT "Yo muthafucka, World!"
19:00:15 <MDude> Oddly, the code itself ins't really changed.
19:00:58 <MDude> I guess because the keywords themselves are too terse to be picked up as translatable phrases.
19:03:06 <Bike> https://twitter.com/0xabad1dea/status/469541728828280832/photo/1/large programming
19:05:00 <Slereah> You know who else did that
19:05:08 <Slereah> And their Plain English Compiler
19:05:25 <FireFly> How did I not know about gizoogle before?
19:05:54 <MDude> DOn't think I've read about the Plain English Compiler, but I think that sounds like something not made as a joke.
19:06:53 <FireFly> I bet it's an implementation of http://en.wikipedia.org/wiki/English_(programming_language)
19:07:07 <Slereah> MDude : It is terrifyingly not a joke
19:07:16 <FireFly> oh wait, no, it's that silly other thing
19:07:22 <Slereah> http://www.theonion.com/articles/family-saved-by-threeway-inflatable-goat,4579/
19:07:31 <Bike> no, that is the correct link.
19:07:31 <Slereah> https://web.archive.org/web/20140108064338/http://osmosian.com/
19:07:44 <MDude> I was thinking of just having something that tries to avoid making obviously bad decisions, but doesn't particularly care what your actual intent is.
19:07:46 <FireFly> http://www.osmosian.com/ works too
19:07:57 <Slereah> The older version had more content though
19:08:18 <MDude> Wonder why it got cut.
19:08:33 <Slereah> Because he wants to sell it, no doubt
19:08:39 <Slereah> The old site had the compiler for free!
19:09:59 <FireFly> It doesn't actually work, does it?
19:10:14 <Slereah> it's a programming language
19:10:25 <Slereah> I can't vouch for how well it works
19:10:53 <FireFly> Have you tried compiling their sample draw-anything-you-can-name-in-the-style-of-Claude-Monet program?
19:11:11 <Slereah> But all that program does is
19:11:23 <Slereah> It googles "Monet" and put up a picture
19:11:42 <Slereah> In the hope that it's not porn, perhaps
19:11:56 <FireFly> What if I want porn in the style of Claude Monet?
19:29:27 -!- hk3380 has quit (Ping timeout: 245 seconds).
19:31:51 -!- xk002 has quit (Ping timeout: 240 seconds).
19:32:35 -!- hk3380 has joined.
19:37:27 -!- Bike has quit (Ping timeout: 276 seconds).
19:40:33 -!- Slereah_ has joined.
19:43:56 -!- Slereah has quit (Ping timeout: 265 seconds).
19:50:07 <MDude> I like how the old version of the site has a page of "What our customers could be saying" (if we had any).
19:50:46 <kmc> the unconscious part of my mind that plays 2048 is learning new move sequences that the conscious mind is unaware of
19:50:49 <kmc> the ego is a joke
19:54:24 <Slereah_> Jesus would have said "Had I known about it, I wouldn't have died for mankind!"
20:00:15 -!- AnotherTest has joined.
20:04:28 -!- mhi^ has joined.
20:10:26 -!- conehead has joined.
20:13:48 -!- Cyragia has quit (Ping timeout: 240 seconds).
20:22:30 -!- edwardk has joined.
20:23:42 <MDude> http://www.gizoogle.net/index.php?search=http%3A%2F%2Fwww.w3schools.com%2Fjs%2Fjs_intro.asp&se=Gizoogle+Dis+Shiznit
20:23:45 <MDude> "ECMA-262 is tha straight-up legit name. ECMAScript 5 (version 1.8.5 - July 2010) is tha sickest fuckin standard. "
20:24:50 <MDude> I'm still wondering what IRP is.
20:25:00 -!- Bike has joined.
20:31:07 <FireFly> There's also a page on the wiki
20:36:17 -!- AnotherTest has quit (Ping timeout: 264 seconds).
20:50:02 -!- conehead has quit (Quit: Textual IRC Client: www.textualapp.com).
20:51:23 <elliott> how many times has it been the end of ipv4 now?
20:51:55 <elliott> "We’ve been talking for many years about IPv4 addresses running out. Now it’s happening. [...] Yes, there are enough IPv4 addresses in the overall system right now that we’re not running out of addresses TODAY … but we are basically OUT at the top-level." come on, I heard the exact same thing years ago! hurry up already
20:57:59 <Bike> climate change is still going to kill us all too. just have some patience
21:02:57 -!- Patashu has joined.
21:10:37 <FreeFull> I want ISPs here to provide IPv6 ):
21:12:47 -!- madbr has joined.
21:13:11 <madbr> hm, I wonder how much data you could squeeze into a huge 2d barcode
21:14:58 <FreeFull> You have to be error resistant
21:15:20 <madbr> something you could successfully scan with a cheap CCD (cell phone cam)
21:15:34 <FreeFull> http://ronja.twibright.com/optar/ This works for black-and-white
21:15:43 -!- Patashu_ has joined.
21:15:44 -!- Patashu has quit (Disconnected by services).
21:15:56 <madbr> given optimal camera position and a barcode printed with a good process (possibly in color even)
21:15:57 <FreeFull> optar probably requires a scanner for any large amount of data though
21:16:16 <madbr> "Optar fits 200kB on an A4 page"
21:19:25 <FreeFull> You can probably fit more if you're smarter
21:20:14 <madbr> yeah but it probably has to be resilient to being photographied at unfavorable angles
21:20:23 <madbr> like having some perspective in the shot
21:20:34 <Bike> probably want some error correction, too.
21:20:46 <madbr> yeah it will need heavy error correction
21:21:06 <madbr> on the other hand it can probably be in color
21:21:48 <Bike> that makes things dependent on the ambient lighting, in addition to positioning.
21:22:24 <madbr> but ambient lighting effects will probably have a low frequency
21:22:59 <madbr> ie will be roughly the same for a whole block of, say, 32x32 pixels
21:24:07 <madbr> also, the camera can probably have a flash
21:24:56 <madbr> which means that if most of the light contribution comes from the flash it should have roughly consistent color temperature
21:25:28 <madbr> otoh you'd have to compensate for stuff like ink gamut (damn you cmyk)
21:26:13 <madbr> or you could use color as a secondary data channel
21:26:21 <madbr> for synchronization and error correction
21:26:38 -!- Patashu_ has quit (Ping timeout: 240 seconds).
21:27:27 <madbr> put the real data in the green channel, use red and blue as synchronization and error correction data
21:27:55 -!- xk002 has joined.
21:28:46 <mroman> How many bits is that per cm^2?
21:29:24 -!- nooodl has joined.
21:29:29 <mroman> > (21.0 * 29.7) / (200*1024*8)
21:29:38 <madbr> about 25x25 per cm²
21:30:09 <mroman> > ((21.0 * 29.7) / (200*1024*8))**(-1)
21:30:50 <madbr> it could probably use very high dpi commercial printing
21:30:52 <mroman> but bits / cm^2 looks good
21:31:12 <mroman> that's pretty good I think
21:31:27 <int-e> > 21.0 * 29.7 * 2^4
21:31:38 <elliott> printers don't print in RGB though...
21:31:49 <madbr> true but you can correct for that
21:32:48 -!- augur has quit (Remote host closed the connection).
21:33:04 <mroman> > ((21.0 * 29.7) / (100*1024*8))**(-1)
21:33:09 -!- nooodl_ has quit (Ping timeout: 276 seconds).
21:33:14 <madbr> use local maxima as white, local minima as black, then find a local 100% cyan pixel, use that as cyan maxima, repeat for yellow, magenta, red, green and blue
21:33:42 <madbr> build some kind of matrix to reverse the effect of the color shift
21:35:22 <madbr> or you could stay b&w and increase dpi instead, but then you need more printing dpi and a better camera
21:36:27 <madbr> there's also the issue that the user might incorrectly orient the camera and miss a part of the barcode
21:36:29 <int-e> 25.6 bits per square mm; 0.2mmx0.2mm pixels (since apparently they don't include error correction in their numbers)
21:37:33 <madbr> makes sense, it's printed at 200dpi (0.127mm)
21:38:27 <fizzie> There are a couple of color 2D barcode standards.
21:39:04 <fizzie> Like the Microsoft Tag.
21:39:24 <madbr> yeah but they're designed to fit in an url
21:39:43 <int-e> I should read more carefully. "That corresponds to 200kB per page when taking overhead into account."
21:41:57 <fizzie> HCCB's goal for the colors is to improve density compared to black-and-white, though.
21:42:14 <fizzie> "Currently laboratory tests have yielded using eight colors, 2,000 binary bytes, or 3,500 alphabetical characters per square inch in its highest density form using a 600dpi business card scanner."
21:43:14 <fizzie> 2000 bytes per square inch is... 2480 bits per square centimetre, I guess.
21:46:55 <madbr> considering a sheet of paper has something like 80~ usable square inches if you count margins (8½ x 11) that's something like 160kb for the whole page
21:48:13 <fizzie> > [8.5*2.54*11*2.54, 21*29.7]
21:48:19 <fizzie> Our papers are bigger.
21:48:47 <FireFly> I don't think that's what "printed copy" means
21:50:56 -!- oerjan has joined.
21:52:10 <madbr> also kinda wonder how heavy to decode these things are
21:52:14 -!- boily has joined.
21:58:30 -!- b_jonas has quit (Ping timeout: 276 seconds).
22:02:57 <madbr> though I guess if your cpu runs at at least a couple hundred mhz (arm SOC etc) it should be acceptable
22:05:49 <int-e> oh, this one is good. "Reducing the space necessary to keep accounting records that are mandatory to be kept on paper"
22:07:51 <madbr> would it make sense to print at very high dpi in color, and then apply a very aggressive error correction algo so that even if over half of the data or more is lost, you still get the correct output?<
22:09:16 <MDude> Print it as a hologram on a sheet of acrylic.
22:10:09 <madbr> you can't scan that with a CCD camera :D
22:10:32 <MDude> Not in a single pass.
22:10:44 -!- b_jonas has joined.
22:11:40 <MDude> The scanning program could ask you to look at it form different angles until the file is read, though.
22:12:13 <madbr> it has to be more practical than just putting a QR code as an URL and getting the data off of wireless internet
22:12:34 <madbr> also it would be easier to just print multiple pages of data
22:14:08 <int-e> I wonder whether the QR-style synchronization pattern (concentric squares) is superior to the checkerboard pattern or not ...
22:14:31 <MDude> Just get a high DPI printer and print it normally several times over.
22:14:38 <int-e> (and I'm afraid that the answer will depend on the printer)
22:15:07 <MDude> Then when you scan it, take the bitmaps of the different versions and do a majority vote per pixel.
22:16:55 <MDude> Plus overhead for getting it lined up.
22:17:06 <oerjan> <Bike> no, that is the correct link. <-- absolutely
22:17:09 <madbr> another issue is, what if the picture is taken too far away, then the data will be too small and won't scan okay
22:17:47 <MDude> Give you user a telescoping camera lens.
22:18:35 <MDude> It's not like you'd expect to be reading office document from far away to begin with.
22:20:18 <madbr> well, it can't be further away than about a meter
22:21:29 <madbr> but it can probably be like 10cm away
22:21:58 <madbr> there's probably a 2x range at least
22:28:12 -!- Bike has quit (Ping timeout: 245 seconds).
22:29:42 <madbr> maybe it would be easier to use some kind of frequency domain encoding?
22:30:46 <madbr> considering it's going to be somewhat smudged once snapped by the camera
22:35:54 -!- tertu has quit (Ping timeout: 258 seconds).
22:35:58 -!- Phantom_Hoover has joined.
22:39:00 <madbr> maybe you could overlay the data 4 times (in a 2x2 grid) at high DPI, and then once at lower DPI
22:39:29 <madbr> so that if the shot is too close and part of the data is missing, you can still recover it
22:39:52 <madbr> and if it's too far, then you have the whole frame and you can use the low DPI data
22:44:40 -!- Sorella has joined.
22:53:19 -!- mhi^ has quit (Quit: Lost terminal).
22:58:30 -!- Bike has joined.
23:20:43 -!- douglass1 has quit (Ping timeout: 252 seconds).
23:29:52 -!- xk002 has quit (Ping timeout: 245 seconds).
23:47:23 <kmc> https://gist.github.com/huonw/be05427dc80e44f1a594 a Rust compiler plugin to randomize the order of struct fields, for exploit hardening
23:53:53 <quintopia> Rust is still on my steam wishlist, waiting to be discounted
23:59:13 <kmc> quintopia: https://github.com/mozilla/rust/issues/12723