00:01:47 -!- nooga has quit (Ping timeout: 244 seconds).
00:32:08 -!- augur has quit (Remote host closed the connection).
00:32:57 -!- augur has joined.
00:34:00 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
00:49:28 -!- DHeadshot has joined.
01:14:45 -!- DHeadshot has quit (Read error: Connection reset by peer).
01:14:51 -!- DH____ has joined.
01:22:25 <Arc_Koen> "If the queue is empty, any instruction that would normally use a value from the queue instead uses 0."
01:31:56 <Arc_Koen> after a few corrections the hello world program finally outputs "he???????e>?"
01:34:32 <Arc_Koen> and after a few more corrections "hell?η?wE?"
01:34:54 <Arc_Koen> also the number of characters is now wrong, which probably means I should go to sleep now
01:35:54 <Arc_Koen> (for the record I have no idea what the heck this η thing is doing there - my shell has never supported anything else than ascii 0-127 before
01:39:32 -!- DH____ has quit (Read error: Connection reset by peer).
01:39:37 -!- DHeadshot has joined.
01:50:05 <kmc> that's more a matter of the terminal than the shell
01:51:04 <kmc> have you actually tried non-ascii characters before?
01:51:53 -!- Jafet has joined.
01:56:10 <kmc> it's not really surprising that it would work out of the box on a modern OS
01:56:26 <kmc> amazingly, some people speak languages other than english, and want to use computers
01:56:39 <kmc> i know, right?
01:56:51 <elliott> kmc: Some people just speak English.
01:57:05 <elliott> kmc: it didn't suddenly lose accents and so on just because ASCII defined an arbitrary subset in the 60s
01:57:16 <elliott> not to mention punctuation
01:57:43 <kmc> i don't understand what point you are making
01:59:13 <elliott> That you could be more confusing about your sarcasm!
01:59:20 <elliott> (I wasn't disagreeing at all.)
01:59:34 <elliott> (Just saying that you don't need to go to foreign languages to see ASCII's inadequacy.)
01:59:57 <kmc> english does have some diacritic marks, yes
02:00:10 <kmc> and confounding hyphen with minus with en dash with em dash is annoying
02:00:29 <elliott> Quotes, too, though nobody cares about that any more.
02:00:41 <kmc> wikipedia doesn't use fancy quotes?
02:00:46 <kmc> i know someone who uses them in IM
02:00:49 <elliott> Specifically recommended against in their MOS.
02:00:59 <elliott> yes, random nerds who do it to be pedantic don't count :p
02:00:59 <kmc> on what grounds
02:01:18 <elliott> kmc: far simpler to input, basically universal, work just fine
02:01:18 -!- Jafet has quit (Ping timeout: 255 seconds).
02:01:26 <elliott> and I think causing fewer display issues with crappy fonts etc.
02:01:44 <elliott> it's not a bad justification as far as input goes... random editors would hardly bother with that
02:01:45 <kmc> i use fancy quotes in latex and pretty much nowhere else
02:01:56 <elliott> anyway I am going to sleep now
02:02:13 <kmc> high-class online publications still use them
02:02:28 <kmc> newspapers etc
02:03:38 <kmc> but then again The New Yorker still writes coöperate and such
02:03:43 <kmc> The Sha Chafer
02:05:44 <kmc> but hungarian is cooler: it has o, ó, ö, and ő
02:06:13 -!- elliott has quit (Ping timeout: 246 seconds).
02:08:41 <Arc_Koen> and german uses some weird fancy-but-reversed quotes
02:09:04 <Arc_Koen> like they confused the beginquote and endquote
02:12:32 -!- Arc_Koen has quit (Quit: Arc_Koen).
02:19:31 -!- TodPunk has quit (Read error: Connection reset by peer).
02:19:45 -!- TodPunk has joined.
02:26:41 -!- Jafet has joined.
02:27:45 <kmc> hey hey smoke weed everyday
02:27:57 <copumpkin> is that what you do these days, kmc?
02:28:25 <kmc> smoke weed a measure zero subset of days
02:28:34 <kmc> i mostly do boring things like work and eat pasta
02:28:41 <kmc> i am growing mushrooms in my house, but the non-psychoactive kind
02:28:58 <kmc> though the spores came from a sketchy website that included a baggie of opium poppy seeds as a "free gift"
02:29:43 <zzo38> Do you like music "Funiculi Funicula"?
02:31:25 <copumpkin> wouldn't say I particularly like it though
02:32:05 <copumpkin> where all techies get their drugs these days
02:32:07 <zzo38> (In case you forgot, I mean the music only; disregard the words)
02:32:45 <kmc> are you still mining bitcoins copumpkin?
02:33:24 <copumpkin> nah, stopped mining pretty quickly
02:33:31 <copumpkin> I still deal in them a fair amount though
02:34:56 -!- amca has joined.
02:43:02 <zzo38> I had idea of pieces in chess variant, such as the "opposer", which after moved, you can change any of adjacent pieces to opponent's pieces.
02:44:57 <zzo38> I use fancy quotes mostly with TeX only writing ``...'' it prints using starting/ending quotation mark
03:08:21 -!- ais523 has quit.
03:19:53 -!- ais523 has joined.
04:47:16 -!- zzo38 has quit (Remote host closed the connection).
05:15:40 -!- FreeFull has quit.
05:25:02 -!- ais523 has quit.
05:31:17 -!- ais523 has joined.
05:33:23 <ais523> oh wow, anagolf has malbolge now
05:44:29 -!- DHeadshot has quit (Read error: Connection reset by peer).
05:44:49 -!- DHeadshot has joined.
06:07:33 <ais523> coppro: golf.shinh.org
06:07:59 <ais523> the site's way more complex and better-run than it looks from the appearance
06:08:10 <ais523> basically, shinh is a really good programmer but not so great at graphics design skills
06:10:15 -!- ais523 has quit.
06:16:26 <fizzie> But I thought those skills *always* go hand-in-hand.
06:40:55 -!- copumpkin has quit (Ping timeout: 260 seconds).
06:42:47 -!- copumpkin has joined.
07:06:43 -!- DHeadshot has quit (Ping timeout: 246 seconds).
07:11:34 -!- epicmonkey has joined.
07:35:31 <mroman> But currently it's broken I think. (It worked earlier)
07:36:07 -!- epicmonkey has quit (Ping timeout: 246 seconds).
07:38:39 <mroman> It seems to work again.
08:20:12 -!- atriq has joined.
08:21:42 <atriq> Well, I've started using emacs
08:21:49 <atriq> It's going better than I expected
08:27:31 <fizzie> What did you use before you started using emacs?
08:27:34 -!- fungot has quit (Ping timeout: 246 seconds).
08:31:42 -!- nooga has joined.
08:39:48 -!- epicmonkey has joined.
08:46:20 -!- Zerker has joined.
09:28:20 -!- Zerker has quit (Remote host closed the connection).
09:39:42 -!- mindlessDrone has joined.
09:45:43 <atriq> `welcome mindlessDrone
09:45:54 <HackEgo> mindlessDrone: 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.)
09:46:36 <atriq> mindlessDrone, two questions
09:46:42 <atriq> First, are you in Finland?
09:47:20 <fizzie> atriq: I think we already did the welcome bit.
09:47:32 <atriq> fizzie, are you in Finland?
09:47:33 <fizzie> Maybe also the Hexland/Finham bit, I'm not sure.
09:47:44 <atriq> Are you in Hexham?
09:48:21 <mindlessDrone> ya you did the welcome thing ..im neither in finland nor in hexham
09:49:36 <atriq> Are you in Metzingen?
09:50:33 <atriq> Baden-Wrttemberg, apparently
09:50:36 <atriq> It's twinned with Hexham
09:52:16 <atriq> I don't think anyone here lives in Metzingen
09:52:31 <atriq> Even though it's statistically more likely than living in Hexham
09:53:32 <fizzie> Espoo is apparently twinned with Esztergom, Gatchina, Irving, Køge, Mumbai, Kongsberg, Kristianstad, Nõmme, Sauðárkrókur, Shanghai and Sochi.
09:53:39 <fizzie> I don't know if that actually means anything.
09:54:00 <fizzie> (But Sauðárkrókur is a fancy name.)
09:54:00 <atriq> Basically, they get to put on their signs "Twinned with Espoo!"
09:54:32 <fizzie> "Sauðárkrókur is a town in Skagafjörður."
09:55:40 <atriq> I was just reading it phonetically
09:56:15 <atriq> Although if I was translating the name I would call it Sodamouth
09:57:01 <atriq> Sometimes I can't follow me
09:59:23 <fizzie> atriq: "Directly translated to English the name would be 'Sheep-river-hook'."
09:59:47 <atriq> Yeah, but that's ridiculous
10:00:12 <fizzie> I don't know, krókur sounds very hooky to me.
10:00:35 <atriq> Sheepcrook, maybe?
10:18:27 -!- Arc_Koen has joined.
10:21:15 -!- monqy has quit (Quit: hello).
10:22:06 <Arc_Koen> well I had a driving lesson and for some reason I set my alarm clock but forgot to turn it on
10:22:19 <Arc_Koen> so I was waken up by someone ringing at the door
10:22:30 <Arc_Koen> and yesterday I went to bed at 4am
10:22:54 <Arc_Koen> the driving lesson was... different from usual
10:24:34 <Arc_Koen> at some point I stopped at a light
10:24:45 <ion> Cool. They have taken 11-word sequences from Wikipedia, taught an appropriate neural network using them and used the t-SNE algorithm to make a map of the most common words that tries to group words with similar feature vectors together. http://www.cs.toronto.edu/~hinton/turian.png
10:25:02 <Arc_Koen> after one minute he asked me why I was stopping... the light was GREEN
10:26:45 <ion> The neural net’s output in this case.
10:26:55 <Arc_Koen> is it like grouping chemical elements according to the molecules they can form?
10:27:38 <ion> If you can make a neural network figure that out by itself, then i guess so.
10:51:04 -!- mindlessDrone has left.
10:52:05 -!- mindlessDrone has joined.
10:55:59 <fizzie> Kinda reminds me of the good old WEBSOM.
10:57:34 <fizzie> Dimensionality reduction is a funny topic.
10:57:59 <fizzie> There were people fiddling with NeRV around here, I think.
10:58:21 <fizzie> http://research.ics.aalto.fi/mi/software/dredviz/ these guys.
10:58:43 <fizzie> I even wanted to NeRV something up, but by now I've completely forgotten what it was.
11:36:08 -!- mindlessDrone has left.
11:43:04 -!- nooga has quit (K-Lined).
11:54:17 -!- amca has quit (Quit: Farewell).
12:41:27 -!- ogrom has joined.
12:45:56 -!- Jafet has quit (Ping timeout: 255 seconds).
12:46:33 -!- Jafet has joined.
13:21:34 -!- elliott has joined.
13:22:46 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
13:30:13 <elliott> Sgeo: why do you keep objecting to your deregistration and then not doing anything
13:30:13 <lambdabot> elliott: You have 2 new messages. '/msg lambdabot @messages' to read them.
13:30:20 <elliott> I guess that contract thing but you did it before then too
13:31:26 -!- Taneb has joined.
13:32:13 -!- atriq has quit (Disconnected by services).
13:32:16 -!- Taneb has changed nick to atriq.
13:41:18 -!- elliott has quit (Remote host closed the connection).
13:42:10 -!- copumpkin has joined.
13:42:23 -!- copumpkin has quit (Client Quit).
13:43:37 -!- elliott has joined.
13:45:52 -!- copumpkin has joined.
13:46:25 -!- pumpkin has joined.
13:47:16 -!- copumpkin has quit (Disconnected by services).
13:47:17 -!- pumpkin has changed nick to copumpkin.
14:24:41 -!- boily has joined.
14:53:06 -!- AnotherTest has joined.
15:09:03 -!- epicmonkey has quit (Ping timeout: 244 seconds).
15:44:33 <Sgeo> Is codeeval.com a good way to look for a job?
15:46:16 <mroman> Browsing through Burlesque's Language Reference seems to take longer and longer :D
15:46:31 <mroman> Which means I'm on the right path :P
15:51:17 <Sgeo> Meh. I'm going to do it
15:51:22 <Sgeo> And have a honking good time
15:51:29 <Sgeo> ...that last line was weird
15:53:22 -!- mindlessDrone has joined.
15:55:17 <boily> Sgeo: the only proper way to honk is in by being zen doing python.
15:57:12 <mroman> AnotherTest: It has more features than I can remember having implemented ;)
15:57:54 <AnotherTest> mroman: remember not to forget referring to microsoft, as this was their amazing idea
15:58:01 <mroman> That's why it has a documentation.
15:58:33 <mroman> I'm going to assume that in every not small software project you have implemented more stuff than you could possibly remember without the documentation.
15:59:04 -!- Vorpal has joined.
15:59:41 <mroman> at least I don't know every sideeffect of any command in Burlesque
16:00:01 <mroman> So when I golf in it on anagol I have to consult the documentation often :)
16:00:44 <mroman> Of course, the documentation can't tell you which sideeffect you need
16:00:57 <AnotherTest> Do you already have support for complex numbers?
16:01:02 <mroman> so you just randomly read something and judge if it is useful
16:01:21 <mroman> but you are welcome to implement it ;)
16:02:01 <mroman> I'm no haskell pro either.
16:02:32 <mroman> I know the things one needs to know to get most things done in it.
16:02:37 <AnotherTest> Well, I have never written anything big in haskell. And I didn't learn how to use important features of the language such as types yet.
16:03:44 <mroman> It's surprisingly easy at the lower levels
16:03:56 <mroman> and surprisingly freaking complicated at the higher levels ;)
16:04:34 <mroman> Especially since haskell is nowhere near my field of study.
16:05:15 -!- boily has quit (Read error: Operation timed out).
16:05:48 <mroman> My field of study is... well...
16:06:57 <mroman> It's a mix of finance, physics, communication technology, mathematics, economy, law
16:07:41 <mroman> it's called information technology ;)
16:07:41 <atriq> The legal implications of communication technology in an ideal economic system
16:08:00 <mroman> I'm not privileged enough to study at a university
16:08:23 <AnotherTest> I'm not even old enough to study at a university!
16:08:44 <mroman> which means I don't have the right credentials to study at a university
16:08:49 <mroman> because I learned a job ;)
16:08:58 <AnotherTest> Well, maybe I were if I was invited or something - which will most likely not happen :(
16:09:12 <AnotherTest> I wrote a paper... maybe if they see that?
16:10:39 -!- augur has quit (Remote host closed the connection).
16:10:55 <AnotherTest> mroman: Can't everyone go to university if they want to?
16:12:38 -!- FreeFull has joined.
16:21:42 -!- boily has joined.
16:30:46 <Sgeo> Apparently Pythonistas hate reduce()
16:32:02 <Slereah_> C people do not like me either because sometimes I use goto
16:32:53 <elliott> python 3 lacks reduce iirc
16:34:46 <FreeFull> Slereah_: Sometimes goto is the cleanest way to do something
16:34:55 <FreeFull> That's why the linux kernel has goto in places
16:36:09 <Sgeo> This codeeval challenge says this company prefers solutions in Python
16:36:22 <Sgeo> But this makes me want to do it in Clojure
16:36:50 <FreeFull> Then they won't know what the hell
16:37:09 <Sgeo> I don't think codeeval even accepts Haskell
16:37:22 <Sgeo> I think this was weighing on my mind a bit when I started liking Clojure over Haskell
16:37:40 <elliott> good criteria to like a language by
16:46:56 <mroman> AnotherTest: No, you can't.
16:47:06 <mroman> Universities obviously don't accept everybody.
16:47:24 <mroman> So there is a very complex bureaucratic system which describes who they accpet.
16:47:50 <mroman> Obviously in the form of restrictions.
16:48:00 <mroman> There restrictions are different from where you come.
16:48:30 <mroman> If I were to take the entry exam they would require me to have knowledge in chemistry
16:48:41 <mroman> even if I'm going to study something which has nothing to do with chemistry.
16:48:47 <mroman> (obviously this is bullshit)
16:49:05 <mroman> however, if I waste three years at some other school (which does not have chemistry) I get a free pass
16:49:23 <mroman> so If I come from a different school they suddenly don't require me to have chemistry knowledge.
16:50:42 <mroman> also the way we study is completely flawed too.
16:51:07 -!- augur has joined.
16:51:20 <mroman> It's actually the university who tells you what you have to study
16:51:43 <mroman> Of course they hide that fact well behind bureaucracy.
16:52:07 <mroman> also students have a different set of pre-knowledge
16:52:36 <mroman> some study-courses require specific pre-knowledge while others don't.
16:53:01 <mroman> To study computer science at a university of applied sciences (so not a real university) you are not required to know ANYTHING.
16:53:16 <mroman> so obviously the first course in your studies is "how does a fucking browser work"
16:53:32 <mroman> which is completely bullshit for people who already know much.
16:53:49 <mroman> right now I'm in the third semester and we're being teached assembler.
16:54:03 <mroman> Obviously I already know how to do that.
16:54:33 <mroman> But I can't just say "I'm not going to learn anything for the exam"
16:54:44 <mroman> because they don't test if you know how to do stuff in assembler
16:54:55 <mroman> they test if you know how to do it in their specific setting
16:55:19 <mroman> so they use a custom assembler with a different syntax
16:55:52 <mroman> which means you have to learn the differences in the syntax from the assembler you know and the one they customly manipulated.
16:56:20 <mroman> also you have to write programs for a specific processor which means you have to study that processor.
16:57:00 <mroman> Not that it's hard to do that, but it costs time of course you could be spending learning completely new interesting things.
16:58:16 <mroman> Next semester we are going to have "operating systems"
16:58:19 -!- augur has quit (Remote host closed the connection).
16:58:33 <mroman> sounds interesting but it's at such a low level it's completely bullshit
16:58:54 <mroman> Oh. You don't say? There's multitasking? NO WAY! <- like that
17:00:58 <AnotherTest> mroman: there is no entry exam for computer science where I live
17:01:23 <AnotherTest> and I'm pretty sure that if you know the required basis of maths, that it's feasible to succeed
17:01:42 <mroman> In Switzerland you have to come from specific schools to be accepted or pass a standardised general entry exam
17:01:56 <mroman> (which tests also knowledge you will never use during your study)
17:02:20 -!- augur has joined.
17:02:40 <mroman> "So, you are good at math and wan't to study math? Let me just check how good you are in geographie, chemistry and french"
17:02:49 <mroman> (yes, in switzerland they test french)
17:02:56 <mroman> if you don't know french, you can't study anything.
17:03:09 <mroman> even if you don't actually need french for your field of study
17:03:20 <mroman> but that's how the system runs.
17:03:51 <mroman> and nobody needs french for their field of study unless they are studying languages or french itself.
17:04:41 <mroman> I spent 8 years of learning french just to be able to be at the point of study I currently am.
17:05:40 <mroman> Of course, I never had to actually use it so I actually I don't know any french any more ;)
17:05:42 <fizzie> There's a mandatory Swedish exam in Finland that's required to get a master's degree out of the "technical university" schools.
17:05:56 <mroman> And that just pisses me off.
17:06:12 <mroman> It's a waste of thousands of hours for nothing.
17:06:32 <kmc> fizzie: http://en.wikipedia.org/w/index.php?title=File:Pois_pakkoruotsi.svg
17:07:00 <mroman> I mean, yes you can argue that "having once learnd a foreign language is good if you want to learn some other language later in your life
17:07:10 <mroman> but that's so far fetched
17:07:30 <AnotherTest> Well we learn french and english at school here as a second language
17:08:10 <mroman> The only important foreign language today is english.
17:08:19 <mroman> And actually schools teach english at third grade
17:08:27 <mroman> which is ~at an age of 9.
17:09:18 <mroman> and you are literally going to have a hard time finding an adult who still knows french
17:09:23 <mroman> even if he had to learn it for 8 years.
17:09:52 <AnotherTest> well, we have to learn French and Dutch (since these are the most used languages here). I though French is one of the languages in Swiss?
17:09:57 <mroman> because there is just no use for french.
17:10:05 <fizzie> We learn Swedish in primary/secondary education schools (grades 1-12 counting numerically), and there's a lot of debate whether that's useful or not, but at least those are sort of more or less generic schooling; having to pass a Swedish exam as part of the tertiary education is IMO kinda stupef.
17:10:06 <mroman> AnotherTest: So are two other languages ;)
17:10:31 <fizzie> I was under the impression the language areas are rather strictly geographically split in Switzerland.
17:10:38 <mroman> There are 4 official languages in switzerland
17:10:48 <mroman> and more non official languages of course.
17:11:13 <Arc_Koen> AnotherTest: isn't the "call" in "stmnt := literal stmnt | call stmnt | ε" a little redundant?
17:11:28 <Arc_Koen> since then you define "literal := int | real | string | char | call"
17:11:32 <fizzie> At least everything went from French to German when we took a train from Geneve to Bern.
17:11:55 <mroman> On they way from Geneve to Bern you are crossing one of the language borders.
17:12:54 <mroman> Which means there are 3 parts of switzerland where I can't understand my fellow countrymen :)
17:13:06 <fizzie> Incidentally, we'll be passing through Switzerland (Geneva-Visp-St. Moritz-Tirano-Lugano-Milan; okay, so there are some dips into Italy) next summer.
17:13:15 <mroman> If I still knew french then it would only be two parts
17:13:22 <Arc_Koen> mroman: isn't it mandatory when you leave in a multiple-languages country to learn all of those languages?
17:13:36 <AnotherTest> Arc_Koen: You can remove the call stmnt production indeed
17:14:05 <mroman> That would require you to know 4 languages
17:14:10 <fizzie> Depends on the country in question.
17:14:20 <fizzie> Swedish is mandatory here, but then again we only have two.
17:14:34 <Arc_Koen> mroman: note that by "mandatory" I'm not talking about law or anything
17:15:00 <mroman> Learning 4 languages is hard
17:15:34 <mroman> So it's not really practically possible to understand everybody
17:15:36 <AnotherTest> Unless of course they happen to be programming languages :p
17:15:51 <Arc_Koen> well, one of them is already your mothertongue
17:16:12 <Arc_Koen> also you might want to include english as well
17:16:13 <mroman> Arc_Koen: We don't actually speak german here ;)
17:16:45 <fizzie> I think 4 is probably the most common number of languages "learned" (as in, at least a couple of years of it in school) in Finland; Finnish, Swedish and English to (practically) everyone, and then one other that's most commonly French or German.
17:17:04 <mroman> it's 3 completely foreign languages one would have to learn here.
17:17:12 <mroman> and one somewhat foreign language.
17:17:55 <Arc_Koen> can't you deduce roman from french and italian?
17:18:03 <mroman> and you have to learn different dialects
17:18:16 <mroman> Probably the same way you can deduce english from other languages
17:18:36 -!- nvt has quit (Ping timeout: 246 seconds).
17:18:55 <Arc_Koen> hmm the opposite would feel more natural for me
17:18:55 <mroman> it's helps you a little but it's still going to be hard.
17:19:43 <mroman> We deduce german from swiss german ;)
17:19:45 -!- ogrom has quit (Quit: Left).
17:19:50 <mroman> Which doesn't work all the time
17:20:05 <mroman> because the grammar and vocabulary is not exactly the same.
17:20:08 <AnotherTest> Arc_Koen: I have corrected the redundant production
17:20:43 <Arc_Koen> so basically stmnt is a string of literals
17:21:15 <mroman> Does anybode of you know somebody who works for youtube?
17:21:29 <mroman> If so, kick his ass ;)
17:21:31 <AnotherTest> Technically "call" isn't an actual literal but OK
17:21:46 <AnotherTest> and call is indeed a list of ids, separated by dots
17:22:05 <mroman> Everytime I jump back in time it reloads the whole video.
17:22:13 <mroman> which means I have to download it yet again
17:22:48 <mroman> and with my poor bandwidth this really sucks like hell
17:23:05 <mroman> Arc_Koen: Wait for it to buffer for 2 minutes
17:23:10 <mroman> then watch the first two minutes
17:23:24 <mroman> then in the timebar let's say you wan't to rewatch minute 01:00 - 02:00
17:23:38 <mroman> assuming that's it's going to play that from what you already downloaded
17:23:44 <mroman> it's going to restart the download
17:28:24 <mroman> now I don't have a linux ghc build environment anymore :(
17:33:18 <fizzie> You should just youtube-dl anything you're going to watch.
17:34:23 <kmc> my youtube-dl stopped working :/
17:36:27 -!- nvt has joined.
17:37:00 -!- atriq has quit (Quit: dinner).
17:40:14 <fizzie> One has to keep updating it every now and then. (Disclaimer: haven't youtube-dl'd anything in a while.)
17:48:28 <mroman> AnotherTest: How are things going with HELP?
17:48:35 <mroman> Does it have an include yet?
17:48:59 <AnotherTest> I seem to have lost interest in help a bit
17:49:15 <mroman> That's one of the most interesting features about a pre-processor I think.
17:49:20 <AnotherTest> But I have not a lot of time to work on esoteric things
17:50:02 <AnotherTest> If I have free time, I usually work on this networking applications I'm creating
17:50:28 <mroman> I'm planning to write something bigger in Burlesque
17:50:39 <mroman> and an include would have been nice :)
17:50:46 <mroman> I just use my own preprocessor
17:53:40 <mroman> It doesn't have really a lot of features, but it has include, ifdef, ifndef and define
17:53:52 <mroman> and was written while I was learning haskell.
17:55:21 <mroman> peer pressure has won.
17:55:27 <mroman> kmc: I rather shoot myself.
17:55:34 <kmc> i can't tell if m4 sucks or if it's just associated with things people hate (sendmail, autoconf)
17:55:37 <kmc> I think the answer is "both"
17:55:48 <AnotherTest> mroman: I just finished a part of the other thing
17:56:00 <olsner> kmc: I'd also guess "both"
17:56:12 <kmc> but my opinion of autoconf actually improved as a result of using it in earnest
17:56:16 <kmc> so i'm probably some kind of freak
17:56:25 <mroman> You are writing a paper about it?
17:56:33 <AnotherTest> The IEEE guy said it was good but I needed to give more quantitative stuff
17:56:46 <AnotherTest> mroman: about the networking thing, not about HELP (of course)
17:57:04 <shachaf> kmc: You missed edwardk's lens talk in SF!
17:57:49 <olsner> I wonder what m4 was supposed to be used for ... when do you want a macro language?
17:59:00 <AnotherTest> The paper might actually have been published directly if I had won a noble price I think
17:59:37 <boily> olsner: your hardware is just not powerful enough to do anything more demanding than macro substitution?
17:59:52 <fizzie> olsner: "Its primary use has been as a front end for Ratfor for those cases where parameterless macros are not adequately powerful."
18:00:26 <mroman> kmc: I actually can literally shoot myself with M4.
18:01:12 <fizzie> olsner: (Second sentence of the abstract of Brian W. Kernighan and Dennis M. Ritchie. The M4 macro processor. Technical report, Bell Laboratories, Murray Hill, New Jersey, USA, 1977.)
18:01:36 <kmc> "1972 - Dennis Ritchie invents a powerful gun that shoots both forward and backward simultaneously. Not satisfied with the number of deaths and permanent maimings from that invention he invents C and Unix."
18:01:48 <fizzie> "It has also been used for languages as disparate as C and Cobol -- is particularly suited for functional languages like Fortran, PL/I and C --"
18:02:14 <fizzie> Ah yes, the well-known functional languages Fortran, PL/I and C.
18:02:47 <fizzie> It continues "-- since macros are specified in a functional notation."
18:02:48 <kmc> they have functions!
18:03:00 <fizzie> It's a kind of different "functional language" there.
18:03:16 <fizzie> Either Ritchie or Kernighan. It doesn't exactly say which one.
18:03:27 <mroman> and lips looks functional.
18:03:43 <fizzie> It didn't say anything about how "functional" M4 is, though.
18:04:53 <fizzie> Ratfor seems to have been the primary use case, though.
18:06:36 <fizzie> There was a point of time where C's preprocessor didn't have macros with arguments either.
18:07:14 -!- atriq has joined.
18:07:28 <fizzie> ("Many other changes occurred around 1972-3, but the most important was the introduction of the preprocessor -- Its original version was exceedingly simple, and provided only included files and simple string replacements: #include and #define of parameterless macros. Soon thereafter, it was extended, mostly by Mike Lesk and then by John Reiser, to incorporate macros with arguments and ...
18:07:35 <fizzie> ... conditional compilation." (Dennis M. Ritchie. 1993. The development of the C language. SIGPLAN Not. 28, 3 (March 1993), 201-208. DOI=10.1145/155360.155580 http://doi.acm.org/10.1145/155360.155580)
18:07:47 <AnotherTest> mroman: just a warning, HELP is actually much worse :p
18:08:15 <mroman> I don't have a cpp build environment anyway
18:46:30 -!- elliott has quit (Read error: Connection reset by peer).
18:47:09 -!- elliott has joined.
18:47:10 -!- elliott has quit (Changing host).
18:47:10 -!- elliott has joined.
18:54:37 <FreeFull> fizzie: It's hard to do good macros with any language that actually has syntax =P
19:12:44 -!- AnotherTest has quit (Quit: Leaving.).
19:16:02 -!- ais523 has joined.
19:29:04 <ion> Some Finns received empty MMS messages timestamped at 1970 after a telecom operator did a software update. A newspaper interviewed the company’s Chief [roughly “mobile plan business”] Officer about it. He said “the testing of the messaging technology started already in the 1970s. Messages may have been left to the phone network back then.”
19:29:06 <ion> http://www.hs.fi/kotimaa/Aamulehti+Saunalahden+asiakkaat+saivat+viestej%C3%A4+70-luvulta+j%C3%A4rjestelm%C3%A4virheen+takia/a1305608399534
19:30:53 * kmc wonders whether to take FreeFull's bait
19:31:32 <kmc> you know, C has no syntax, a program is just a sequence of characters
19:31:35 <kmc> that is the only syntactic form
19:31:49 <kmc> what those characters *mean* when juxtaposed is of course a semantic question
19:32:16 <olsner> ion: good thing they upgraded their system to get rid of all those unsent messages then
19:32:32 <FreeFull> kmc: Then nothing has any syntax?
19:32:58 <kmc> apparently the reason the TTL field was added to IP is that in older internetworks, as time approached infinity, traffic consisted entirely of packets caught in routing loops
19:33:43 <kmc> FreeFull: it's reducto ad absurdum from the ridiculous "Lisp has no syntax" claim
19:33:43 <elliott> <ion> Some Finns received empty MMS messages timestamped at 1970 after a telecom operator did a software update. A newspaper interviewed the company’s Chief [roughly “mobile plan business”] Officer about it. He said “the testing of the messaging technology started already in the 1970s. Messages may have been left to the phone network back then.”
19:33:59 <kmc> <elliott> <ion> Some Finns received empty MMS messages timestamped at 1970 after a telecom operator did a software update. A newspaper interviewed the company’s Chief [roughly “mobile plan business”] Officer about it. He said “the testing of the messaging technology started already in the 1970s. Messages may have been left to the phone network back then.”
19:34:03 <elliott> <kmc> <elliott> <ion> Some Finns received empty MMS messages timestamped at 1970 after a telecom operator did a software update. A newspaper interviewed the company’s Chief [roughly “mobile plan business”] Officer about it. He said “the testing of the messaging technology started already in the 1970s. Messages may have been left to the phone network back then.”
19:34:25 <kmc> TTL=255 <elliott> <kmc> <elliott> <ion> Some Finns received empty MMS messages timestamped at 1970 after a telecom operator did a software update. A newspaper interviewed the company’s Chief [roughly “mobile plan business”] Officer about it. He said “the testing of the messaging technology started already in the 1970s. Messages may have been left to the phone network back then.”
19:36:04 <FreeFull> kmc: Lisp does have syntax though =P
19:36:19 <olsner> TTL=254 <kmc> <elliott> <kmc> <elliott> <ion> etc
19:36:21 <FreeFull> Not as much as many other languages, but it does
19:37:01 -!- standa_ has joined.
19:37:07 -!- standa_ has quit (Client Quit).
19:37:11 <kmc> yeah that's my point
19:37:48 <kmc> and my other point is that things like (let (x 2 y 3) ...) vs (let ((x 2) (y 3)) ...) are basically syntactic questions
19:37:57 <kmc> or at least, it is not unreasonable to call them syntactic questions
19:38:15 <kmc> ultimately words are arbitrary and the participants in a discussion should just agree on what they mean
19:38:36 <kmc> then i'm not sure if Kernel has syntax besides s-expressions :)
19:38:54 <kmc> (or besides s-expressions plus a trivial rule for the syntactic form of a combination)
19:39:08 -!- zzo38 has joined.
19:40:27 <kmc> o'shachaml
19:46:15 <ion> !rot13 ocaml
20:01:17 <Sgeo> In a dynamically-typed language, if I'm getting wrong but sensible looking results, is it at all plausible that it's a type error, or is logic error far more likely?
20:01:23 <Sgeo> Such as, getting 4 instead of 5
20:01:47 <Sgeo> Because the Python program I wrote is giving wrong answers, but I feel like if I screwed up at all, it's type-related
20:10:52 <Arc_Koen> I think one problem with Emmental is that its instruction-rewriting system cannot be rewrited
20:12:09 <Arc_Koen> well, you can overwrite it with something else to stop being able to rewrite, but you can't write a different system since the instructions you create can only be sequences of instructions that already exist
20:13:11 -!- epicmonkey has joined.
20:14:01 <zzo38> Then do you have to make up a new one which includes the instruction to rewrite the instruction-rewriting system?
20:14:27 <Arc_Koen> well yes, you can do that, but it won't introduce anything new
20:18:31 -!- ais523_ has joined.
20:19:08 <zzo38> That is what I thought
20:19:55 -!- ais523 has quit (Ping timeout: 260 seconds).
20:20:18 -!- ais523_ has changed nick to ais523.
20:22:46 <Sgeo> A bit of printing helps me find my error
20:22:51 <Sgeo> \n was getting into some of the strings
20:24:12 <Sgeo> Ok, so I'm still getting wrong results
20:30:11 <Sgeo> I'm actually getting _more_ wrong results
20:32:18 <olsner> hmm, I guess you've made the code more wrong then?
20:33:19 <Sgeo> Not necessarily.
20:38:51 <Arc_Koen> you probably corrected a mistake that was cancelling out another
20:40:40 <olsner> if two bugs are cancelling each other out that's less wrong than having either of those on its own uncancelled by the other
20:41:04 <Sgeo> olsner, not if they only happen to be cancelling each other out in one test case
20:51:29 <mroman> If two bugs are cancelling each other out is that even a bug?
20:54:06 <mroman> Maybe one can consider it a skill introducing two bugs which are cancelling each other out on purpose.
20:54:13 <olsner> no, it's not a bug - it's two!
20:54:25 <mroman> If there is no problem, is there a bug?
20:55:26 <mroman> Technically such a bug is not observable until you actually introduce a bug by removing the cancellation?
20:56:50 <olsner> there may be other ways to trigger either bug - let's say in 90% of cases both trigger and you're fine
20:57:13 <olsner> but some cases might only trigger one of the bugs, requiring it to be fixed
20:59:42 <atriq> If there are two bugs that cancel eachother out all the time, you've discovered a new algorithm
21:01:12 <mroman> I just wanted to troll about the definition of it being a bug if there is no observable problem ;)
21:01:37 <olsner> yeha, how can the code be wrong if it works?
21:02:11 <olsner> (was supposed to be "yeah", but I guess you can read it yeeehaw if you like)
21:02:15 <mroman> Technically if it works it can't be wrong :)
21:02:23 <Sgeo> I'm beginning to hate Python
21:02:29 <mroman> (If it really works in 100% of the time)
21:02:44 <mroman> If anything it's bad ugly messy code.
21:04:42 <fizzie> !c int a, b; a, b = 4, 5; /* look, ma, Python-style unpacking assignment */ printf("a = %d, b = %d", a, b); /* whoops */
21:05:00 <fizzie> I'd like to know how many Python programmers have tried that.
21:05:15 <mroman> I wish haskell had forward declarations
21:05:24 <Sgeo> It doesn't need it
21:05:48 <kmc> i,i .hs-boot
21:06:22 <mroman> It's one of these cases where A needs B and B needs A
21:06:36 <mroman> which in C(++) you can resolve with forward declaration
21:06:47 <mroman> the guys in #haskell told me
21:06:54 <mroman> "just merge the shit into a single file"
21:07:20 <fizzie> !c int a, b; a, b = (4, 5); /* maybe it works if I make it more like a tuple */ printf("a = %d, b = %d", a, b); /* whoops again */
21:07:20 <mroman> now it's a huge file with a lot of stuff in it :)
21:07:39 <mroman> https://github.com/FMNSSun/Burlesque/blob/master/Burlesque/Eval.hs
21:07:46 <mroman> ^- see. I would like to categorize builtins
21:07:55 <mroman> so I can put them into files by category
21:08:05 <mroman> but some builtins require functions from other categories
21:08:18 <mroman> which yields to this circular dependency (or whatever you guys call that)
21:08:49 <mroman> and apparently according to #haskell there is no way to do it without merging all into a single file :(
21:09:21 -!- atriq has quit (Quit: Leaving).
21:09:23 <mroman> also the case matching is kinda messy
21:09:30 <mroman> but I guess there is no better way to do it.
21:09:46 <mroman> and haskell tidy removes comments
21:09:54 <mroman> (which totally sucks btw.)
21:10:02 <Sgeo> Oh, I didn't notice that this thing counts y as a vowel
21:10:13 <mroman> it cleans up your indentation and stuff, but it throws away all the comments.
21:10:30 -!- Phantom_Hoover has joined.
21:10:40 <mroman> (Not that there actually are lots of comments in the code :) )
21:15:53 -!- mindlessDrone has left.
21:16:05 <mroman> It's funny how little the actual eval function is :)
21:16:54 <mroman> And I should probably switch to a Map of some sort instead of the linear tupel lookup
21:17:52 <FreeFull> I wonder what the smallest lisp written in haskell is
21:19:17 <Sgeo> Two test-cases working
21:19:21 <mroman> http://sources.defmacro.org/blaise/src/
21:19:25 <mroman> ^- probably not the smallest
21:21:55 <mroman> Let's write the smallest lisp interpreter in haskell then shall we
21:22:46 <mroman> One could actually do some regex hacks
21:22:54 -!- epicmonkey has quit (Ping timeout: 264 seconds).
21:22:57 <mroman> (:+ (:- 3 2) (:- 2 2))
21:23:15 <mroman> then one declares a data with a Data Constructor :+ and :-
21:23:33 <mroman> and just let deriving Read do the work
21:24:26 <mroman> (writing parsers is for fools if you can just let ghc generate one for you :D)
21:28:46 <mroman> that only works for infix
21:32:24 <FreeFull> You could use a regex hack to replace * - + with (*) (-) (+)
21:32:36 <FreeFull> Except you'll have to special-case negative numbers =
21:38:44 <mroman> http://codepad.org/EhQMqj7Z
21:38:50 <mroman> not quite how I imagined
21:39:06 <mroman> but it works so it can't be wrong :P
21:39:43 <mroman> also since this is PythonLisp whitespaces matters :D
21:39:56 <mroman> if you don't put enough whitespaces in there it's gonna crash.
21:40:33 -!- Arc_Koen has quit (Quit: that's dr. turing to you, punk).
21:45:12 -!- boily has quit (Quit: Poulet!).
21:49:01 <mroman> http://codepad.org/F3WR4M98
21:49:57 <mroman> not really lisp of course
21:50:12 <mroman> You'd have to treat defun and stuff.
21:51:01 <mroman> At least writing a simple lisp interpreter is actually very easy.
22:00:56 <kmc> where by "lisp" we mean "a vaguely lisp-like language intended to be easy to implement"
22:01:07 <kmc> actually implementing Common Lisp, Scheme, or Clojure is not easy :)
22:02:23 <kmc> (R5RS Scheme is not horrifically hard, either, but I wouldn't call it easy)
22:02:47 <FreeFull> Scheme and Clojure are closer to the "Lisp-spirit"
22:03:17 <kmc> i wonder what's the hardest part of R5RS to implement
22:03:20 <kmc> maybe dynamic-wind
22:04:02 <kmc> though maybe that's secretly easy in a CPS transform or something
22:04:04 <kmc> i haven't thought about it
22:06:56 <olsner> uh oh, the first hit for r5rs dynamic-wind is an oleg article :/
22:07:30 <elliott> kmc: call-with-current-continuation + dynamic-wind are A Fun Pair
22:07:43 -!- sivoais has quit (Ping timeout: 260 seconds).
22:07:57 <kmc> olsner: danger danger high voltage
22:08:06 <kmc> inherit the dynamic-wind
22:10:19 <olsner> "Fortunately, delimited control operators let application programmers write dynamic-wind themselves; that function is no longer a primitive, is no longer hard-to-explain, and no longer has to be provided by the implementation."
22:13:42 <Sgeo> Fuck yourself with a rusty rake
22:14:18 <Sgeo> One of the challenges is "Email Validation "
22:14:20 <olsner> (followup question: what are these delimited control thingies?)
22:14:31 <Sgeo> "You are given several strings that may/may not be valid emails. You should write a regular expression that determines if the email id is a valid email id or not. You may assume all characters are from the english language."
22:14:43 <kmc> lol "no longer hard-to-explain"
22:15:39 <olsner> if it wasn't hard, what's it doing on oleg's site in the first place?
22:15:42 <kmc> http://www.ultrapico.com/Expresso.htm
22:18:48 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
22:22:30 <olsner> Sgeo: i.e. this? http://ex-parrot.com/~pdw/Mail-RFC822-Address.html
22:23:38 <olsner> a correct answer might be that the language of valid email addresses is not regular and cannot be matched by a regexp
22:25:06 <kmc> that kind of fancy book-learnin' will get you nowhere
22:35:51 <elliott> <olsner> (followup question: what are these delimited control thingies?)
22:38:55 -!- Nisstyre-laptop has joined.
22:41:53 -!- copumpkin has joined.
22:42:25 -!- heroux has quit (Ping timeout: 252 seconds).
22:42:34 -!- heroux has joined.
22:44:02 -!- sivoais has joined.
22:51:37 <ion> http://soundcloud.com/g3ksan/troll
23:09:19 <ion> This recording is hilarious. :-D
23:09:49 <shachaf> ion: Are you using Control.Lens yet?
23:10:33 <ion> shachaf: I haven’t written any real code that uses it but i have it in my .ghci
23:10:56 <shachaf> Is "over both" the right way to do it?
23:11:34 <shachaf> I'll probably need to painstakingly learn the operators one by one.
23:11:53 <shachaf> toListOf (both.both) ((1,2),(3,4))
23:12:35 <ion> Well, its type is right and it does the right thing. :-P
23:12:50 <ion> shachaf: Btw, i added a bunch of ….Lens imports. https://gist.github.com/3909552
23:12:52 <shachaf> Sure, but maybe there's another way of doing it?
23:13:03 <shachaf> ion: I should probably do that...
23:13:24 <shachaf> Wasn't your .ghci about 10 lines before?
23:14:09 <ion> This is what i posted first. https://gist.github.com/3909552/fb2dd963e317571249d29f5c4d305dc3c233b707
23:15:06 <ion> diff -u0 <(curl -s https://raw.github.com/gist/3909552/a92d0b0506154300baaf2be86b011213bfc458cf/.ghci) <(curl -s https://raw.github.com/gist/3909552/170990e52c4499158f45d1738cdf10796f32513f/.ghci)
23:20:04 <Sgeo> http://blog.sigfpe.com/2006/12/evaluating-cellular-automata-is.html is this likely to be a good explanation of comonads?
23:20:37 <shachaf> ion: So have you figured out the zipper business?
23:20:48 <ion> Nope, haven’t even looked at it yet.
23:21:01 <elliott> Sgeo: everything sigfpe writes is good
23:21:47 <shachaf> zipper ("hello","world") % down _1 % fromWithin traverse % focus .~ 'J' % rightmost % focus .~ 'y' % rezip
23:22:08 <ion> But that’s the only thing i’ve looked at wrt. them.
23:24:45 <shachaf> kmc: Are you going to accept typos for the mosh bug bounty?
23:28:16 <kmc> maybe $0.01
23:28:23 <kmc> didn't you get some money mining typos in tarsnap?
23:28:28 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
23:28:53 <shachaf> kmc: $20 for running a spellchecker on the codebase.
23:28:54 <ion> I’m in tears laughing at this recording. :-D
23:29:05 <shachaf> I got $30 more for "minor bugs"!
23:29:27 <shachaf> kmc: Man, you guys are cheap. Tarsnap gives you a whole $1 per typo!
23:33:46 * ion switched from tarsnap to s3ql + S3
23:36:53 <shachaf> kmc: That's because C is as "close to the hardware" as it gets, right?
23:44:17 <ion> Thanks, Hanks. Thanks. http://youtu.be/-C2h6JSnM-E
23:49:59 -!- Phantom_Hoover has joined.
23:51:16 <kmc> shachaf: I learned how to type ő with Compose!
23:51:18 <kmc> Compose = o
23:51:33 <shachaf> I was just wondering the other day.
23:51:54 <shachaf> I settled for copying it from https://en.wikipedia.org/wiki/Paul_Erdős
23:52:04 <kmc> it works for u as well
23:52:24 <shachaf> kmc: Presumably it works for anyone who's using the default Compose layout.
23:52:28 <kmc> i wonder if other vowels with that mark exist in unicode (without combining characters i mean)
23:53:28 <shachaf> 04F3 CYRILLIC SMALL LETTER U WITH DOUBLE ACUTE [ӳ]
23:53:49 <shachaf> That's it, though, I think.
23:54:20 <shachaf> kmc: You should get a Unicode search thing! That way you can look things up easily.
23:54:31 <shachaf> I have a file with all of Unicode that I search with less +R
23:59:16 <kmc> ӳ is used in... Chuvash, a language I have never heard of