00:00:01 <Fiora> so like the LHC gets like 11 petakelvins
00:00:21 -!- n2liquid has joined.
00:00:21 <Fiora> elliott: I'm roughly 50.5 attoparsecs tall! seeeee
00:00:29 <elliott> Phantom_Hoover: http://video.google.com/videoplay?docid=-5054356894457127152
00:00:55 <zzo38> "Tonne" is confusing with "ton" so I think "tonne" should not be used.
00:00:58 <elliott> Fiora: i think this means you may actually be shorter than me
00:01:33 <pikhq_> Bike: Well, the derived Planck units are often quite reasonable.
00:01:50 <pikhq_> Bike: For instance, 1 Planck impedence is about 30 ohms.
00:02:21 <Fiora> elliott: wait how tall are you
00:02:45 <pikhq_> 4.63309e113 Pa? Jesus.
00:02:46 <oerjan> `frink 1 attoparsec -> m
00:02:51 <elliott> i am like 5 feet 1 inch or something
00:02:57 <elliott> maybe i grew since the last time i figured that out
00:03:28 <zzo38> Another thing you can do other than use metres and so on, is if the other units are wrong size, use other base units. So, such things as lightyear and so on, etc.
00:03:40 <pikhq_> 0.4 petayottayottayottayottapascal?
00:03:57 <pikhq_> 0.4 PYYYYPa is awesome.
00:03:58 <Fiora> an attoparsec is like 3cm
00:04:11 <Fiora> attoparsec per microfortnight is the best though :3
00:04:23 <pikhq_> Bit over an inch? Huh.
00:04:25 <Bike> about a foot per second or something, right?
00:05:31 -!- nooodl__ has joined.
00:05:46 <Fiora> it's like a cm/s I think
00:06:01 <olsner> iirc, a lightnanosecond is about a foot
00:08:16 -!- nooodl_ has quit (Ping timeout: 246 seconds).
00:11:48 <pikhq_> Huh. Planck charge is also fairly reasonable. About 12 times larger than the elementary charge.
00:12:34 -!- ifnspifn has quit (Read error: Connection reset by peer).
00:12:40 -!- ifnspifn_ has joined.
00:13:03 -!- lambdabot has joined.
00:13:24 <Bike> the charge of an electron
00:13:30 <pikhq_> Phantom_Hoover: No, the elementary charge is the magnitude of the charge of an electron.
00:13:32 <Bike> is called the "elementary charge"
00:13:50 <pikhq_> It's reasonable in many contexts.
00:15:08 <pikhq_> It's still more reasonable than most of the Planck units... I mean, shit, there's an SI prefix for that one. :P
00:16:57 <Fiora> planck impedance: 29.979 ohms?
00:17:31 <Fiora> there's a planck momentum too, it's not big
00:17:38 <Fiora> I wonder what the planck magnetic field is
00:18:17 <pikhq_> Magnetic flux density?
00:18:22 <Fiora> 2.15 * 10^53 teslas says a quick google? ohgod
00:18:32 <Fiora> "215000 yottayottateslas"
00:18:46 <fizzie> HTK's timestamps are in units of 0.1 microseconds. I've always presumed it's because it's accurate enough for each sample of 16 kHz audio to have an integer time in those units (unlike microseconds; one sample is 62.5 us), but nanoseconds would've resulted in too large numbers.
00:18:54 <Phantom_Hoover> i think that's above the point where maxwell's equations go to hell and the vacuum becomes polarised
00:19:08 <Fiora> I think that happens at like, about 10^40 times less? XD
00:19:12 <elliott> polarised, n. really cold (it's cold at the north and south pole)
00:19:20 <Bike> "megasecond" is my favorite usual unit, i think
00:19:20 <zzo38> Or just make up new prefixes.
00:19:47 <zzo38> Someone once told me, they read they wanted to add another prefix "heva"
00:19:59 <olsner> fizzie: might also be that they stole their time unit from windows
00:20:01 <Bike> wasn't there a petition to make "hella" official, or something
00:20:06 * oerjan swats elliott for confusing nouns and adjectives -----###
00:20:54 <olsner> n., n. separator between dictionary entries and their definitions
00:21:09 <zzo38> I don't want "zeppo" because it is similar to "zepto"
00:21:10 <oerjan> well if we have hella, then clearly the opposite prefix should be heava
00:21:11 <Phantom_Hoover> it may have been in either the feedback or letters sections of the new scientist
00:21:48 <zzo38> I think they wanted "heva" one above "yotta" although I am unsure.
00:21:51 <fizzie> olsner: I suppose it's possible, though I don't think it's a very Windows-oriented piece of software.
00:22:14 <fizzie> I've seen at least one "serious-looking" SI extension proposal.
00:22:34 <Fiora> they have to add bitsy too
00:22:38 <Fiora> so then you can have an itsybitsymeter
00:22:38 <olsner> itsy-, bitsy-, teeny- and weeny-
00:22:44 <pikhq_> http://sourceware.org/bugzilla/show_bug.cgi?id=11984 *grin*
00:22:47 <elliott> oerjan: fuck you i'm a disctionary
00:22:49 <Fiora> an itsybitsy spider would be very small indeed
00:22:59 -!- sebbu has joined.
00:23:02 <zzo38> But I am OK if you want to add other prefixes, whether it is "groucho" or whatever else
00:23:17 <oerjan> elliott: is that like a book using discworld spelling conventions?
00:23:44 <elliott> oerjan: who publsiehs the definitions around here
00:24:08 <zzo38> There is also "myria" not used much anymore, although it is a metric prefix for a myriad (ten thousand).
00:25:30 <pikhq_> Be handy for CJK folk.
00:25:50 <pikhq_> (CJK all group numbers in myriads.)
00:26:47 <zzo38> Yes, I know that too.
00:27:02 <Bike> oh, that's probably where knuth got the idea, isn't it.
00:27:34 -!- sebbu has quit (Client Quit).
00:27:36 <zzo38> At least I think so.
00:28:12 <fizzie> "Angstro" has been suggested for 10^(-10). Perhaps to get rid of ångström as a separate unit. (It even has the "m" there.)
00:28:59 <zzo38> I think it would be OK.
00:29:56 <fizzie> (Don't know whether the suggestion kept the diacritics. At least å would be free for abbreviations.)
00:30:27 <zzo38> Yes, and then "angstrom" becomes short for "angstrometer". But yes you should use the lowercase a with ring above for this prefix (uppercase A with ring above is angstrom)
00:31:14 <zzo38> Make a list of these things.
00:31:36 <fizzie> Then you also have 1 åÅ = 10 zm.
00:32:00 <zzo38> I suppose it does allow you to do that although I do not think it would be a good idea.
00:32:23 <fizzie> One ångströångström does sound a bit silly.
00:32:42 * pikhq_ approves of the ångströångström
00:33:35 <pikhq_> Though maybe that should be "ö¨å" (can't be bothered to find the combining char)
00:34:20 <elliott> > The excuse for not fixing this does not make sense.
00:34:21 <elliott> What doesn't make sense is you.
00:34:21 <lambdabot> <hint>:1:52: parse error (possibly incorrect indentation)
00:34:33 <oerjan> http://sv.wikipedia.org/wiki/Fat_(m%C3%A5tt)
00:34:34 <olsner> you mean an å with a diaeresis?
00:35:11 <olsner> oh, 1 fat is 480 osmunds ... pretty thin fella this Osmund
00:35:25 -!- sebbu has joined.
00:35:37 <zzo38> What is the equation for the Railsback curve?
00:35:56 <pikhq_> olsner: Needs both, clearly, as otherwise that'd be a diphthong.
00:36:32 <olsner> not really, since öå doesn't become a diphtong in the first place
00:37:26 <oerjan> olsner: unless you're very drunk
00:37:47 <pikhq_> I will diäresisize all vowels that are adjacent!
00:38:21 <oerjan> föå en öål til föår föan
00:39:45 <oerjan> olsner: or from skåne, probably
00:39:52 <pikhq_> Which is why it's "oërjan" not "Ørjan". >:D
00:41:05 <nooodl__> does this: "å̈" look fine to you guys
00:41:16 <oerjan> nooodl__: excellent plain a, there
00:41:18 <Bike> no, it looks like you have some kind of growth on your head there
00:41:31 <nooodl__> it's supposed to be... an a with an ö above it, essentially
00:42:36 <pikhq_> nooodl__: That composed poorly.
00:44:13 <oerjan> http://www.google.no/search?hl=no&safe=off&tbo=d&site=&source=hp&q=sk%C3%B6%C3%A5ne&oq=sk%C3%B6%C3%A5ne&gs_l=hp.3...122718.172765.0.173062.7.7.0.0.0.0.219.750.3j2j1.6.0...0.0...1c.1.T-0dj36Ttyc gives a bit of hits
00:49:13 <elliott> ion: I think he meant rwbarton?
00:50:17 <ion> <shachaf> Unless someone like ion wants to do it.
00:50:28 <ion> Dunno whom he meant, but i was asking about it in response to that. :-)
00:50:46 <olsner> "someone like ion" might refer to you
01:02:13 -!- nooodl__ has quit (Ping timeout: 245 seconds).
01:11:37 -!- nooodl__ has joined.
01:22:20 -!- Tod-Autojoined has changed nick to TodPunk.
01:38:53 -!- nooodl__ has quit (Ping timeout: 245 seconds).
01:40:08 <n2liquid> Do you guys sometimes discuss about practical (albeit experimental) programming languages?
01:40:39 -!- sebbu has quit (Read error: Connection reset by peer).
01:40:53 -!- augur has quit (Remote host closed the connection).
01:41:19 <olsner> there's some Finnish too, but opinions differ as to wether that counts as a language at all
01:41:23 -!- sebbu has joined.
01:41:23 -!- sebbu has quit (Changing host).
01:41:23 -!- sebbu has joined.
01:41:53 <olsner> ... shachaf started it
01:42:06 <oerjan> if haskell counts, then we do.
01:42:25 <n2liquid> Ok, Haskell; what about new programming languages?
01:42:26 <oerjan> also some have had less esoteric projects, i recall Gregor's plof
01:42:29 -!- Sgeo__ has changed nick to Sgeo.
01:42:32 <pikhq_> n2liquid: Well, it's not esoteric programming, so *yes*.
01:42:44 <n2liquid> Or languages that extrapolate concepts like object orientation
01:43:39 <n2liquid> Especially not stuff that's too focused on mathematical concepts
01:43:49 <lambdabot> oerjan says: i only do impractical things
01:44:39 <n2liquid> oerjan, is plof on the esolang wiki?
01:44:56 <oerjan> no, because it's not esoteric
01:45:11 <oerjan> well, _maybe_ it's linked from Gregor's user page.
01:45:27 <pikhq_> n2liquid: Plof is more a research language than anything else.
01:45:28 <Sgeo> Oh, practicaal?
01:45:31 <oerjan> glass is OO, esoteric, and also gregor's :)
01:46:56 <oerjan> http://plof.codu.org/wiki/
01:51:37 -!- atehwa has joined.
01:52:11 <zzo38> We do discuss experiemental programming and stuff too sometimes
01:53:11 <n2liquid> I've been munching some concepts for a programming language for some time now
01:53:28 <zzo38> Although I suppose we also sometimes discuss completely different things too
01:53:28 <oerjan> i recall edwardk had a language project when he came here, although i think he's gravitated towards "everything fits better into haskell"? also everything edwardk is by definition the opposite of "not stuff that's too focused on mathematical concepts".
01:54:17 <oerjan> in the same way, some haskellers seem to go on to agda.
01:54:22 <n2liquid> It's nothing bat-crazy like what's often explored in academic research projects or esoteric languages
01:55:41 <Sgeo> Has there been any work on Plof recently?
01:55:54 <Sgeo> Mostly, I just want to read awesome documentation, even if the language isn't well implemented
01:56:05 <Gregor> Why hallo thar, people talking about me >_>
01:56:17 <oerjan> well the fythe VM for it was worked on in february, at least, i see
01:56:33 <Gregor> Fythe was last poked at a few months ago.
01:56:45 <Gregor> It's possible that I'm currently contractually obligated not to touch it, so I'm not :)
01:57:22 <Gregor> Just for two more weeks now *shrugs*
01:57:47 -!- Phantom_Hoover has quit (Remote host closed the connection).
01:57:50 <Gregor> The Fythe engine works great, and I have some neato ideas for making it work better, I just haven't bothered to finish porting Plof to it yet.
01:58:13 <Gregor> What with my being distracted by silly irrelevant stuff like pursuing a PhD and making money and watching ponies.
01:58:14 <oerjan> any relation to your ioccc entry? :P
02:00:33 <Gregor> Lest it's not obvious, my IOCCC entry has no practical application whatsoever. Anything to make it not suck would also make it unportable, defeating the whole purpose.
02:02:30 <Bike> what entry is this?
02:02:50 <n2liquid> What do you guys think about LLVM?
02:03:18 <Gregor> Bike: http://www.ioccc.org/years.html#2011_richards
02:05:14 <Bike> well that's... impressive
02:06:20 <zzo38> I happen to like LLVM, although I also think some things missing, such as supporting bytes other than 8-bits, and supporting ARM2
02:07:50 <n2liquid> I know I wouldn't be into programming language design without LLVM
02:08:17 <Bike> is the point of all this preprocessor stuff to make the source resemble dc code
02:11:29 <ion> gregor: Is there a description of how the JIT is done somewhere?
02:14:04 <Gregor> ion: The description was in your heart all along.
02:18:51 <Fiora> that JIT is really amazing
02:19:08 <Fiora> it manages to hide the opcodes and stuff too, wow
02:25:33 <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:29:35 <oerjan> `addquote <Fiora> omg <Fiora> that JIT is really amazing [...] <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:29:38 <HackEgo> 859) <Fiora> omg <Fiora> that JIT is really amazing [...] <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:30:12 <shachaf> elliott: Two spaces are the devil. :-(
02:30:18 <oerjan> elliott: i checked with `quote, and the first example that came up had only one
02:30:32 <elliott> oerjan: two spaces between the first two messages, I mean
02:30:40 <shachaf> kmc: The GHC inliner is annoying me. :-(
02:30:55 <shachaf> I should probably read that one paper, _Secrets of the GHC Inliner_.
02:30:56 <HackEgo> 683) <Phantom_Hoover> Here in Scotland we have a rigorous and well-tested theory of brothels.
02:31:03 <HackEgo> 341) <GregorR> How to make a tasty deep-fried treat: 1) Buy ingredients: Large vat of boiling oil, dry ice and a small Filipino boy. 2) Place Filipino boy in dry ice until frozen solid. 3) Shatter now-frozen Filipino boy into boiling oil. 4) Wait fifteen minutes, drain and enjoy! <CakeProphet> I have the weirdest boner right now.
02:31:27 <HackEgo> 43) <Deewiant> Reality isn't a part of physics
02:31:45 <HackEgo> *poof* <Fiora> omg <Fiora> that JIT is really amazing [...] <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:31:52 <oerjan> `addquote <Fiora> omg <Fiora> that JIT is really amazing [...] <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:31:55 <HackEgo> 859) <Fiora> omg <Fiora> that JIT is really amazing [...] <Gregor> I hear if you listen carefully to the rustling wind on a warm night with a full moon, you can hear the sound of the JIT building ARM functions.
02:32:16 <Bike> wait, why did you do that
02:32:40 <oerjan> ion: the rules for [...] are even more complicated.
02:32:50 <ion> You should use “ ” and see if anyone notices.
02:32:51 -!- augur has joined.
02:33:33 -!- Vorpal has quit (Ping timeout: 276 seconds).
02:34:04 <oerjan> Bike: elliott is a space nazi
02:34:14 <oerjan> we don't want to annoy him
02:34:26 <elliott> Bike: the formatting rules shall not be broken
02:34:27 <ion> Like in Iron Sky?
02:35:03 <shachaf> `addquote <elliott>bike:the formatting rules shall not be broken
02:35:06 <HackEgo> 860) <elliott>bike:the formatting rules shall not be broken
02:35:09 <Bike> i was just about to do that
02:35:19 <Bike> i'm so uncreative.
02:35:34 <HackEgo> *poof* <elliott>bike:the formatting rules shall not be broken
02:35:37 <oerjan> what is an appropriate THAT'S THE JOKE link?
02:35:49 <Bike> the simpson's clip?
02:36:29 <oerjan> i guess that is the top google hit
02:36:36 <kmc> shachaf: inliner? i hardly know her!
02:36:37 <Bike> i'm going to pretend that apostrophe was supposed to be there. "The Simpson" is a nickname of a bigfoot-like internet legend, known for posting photographs of firearms with jokes engraved on them.
02:36:49 <oerjan> ion: http://www.youtube.com/watch?v=xECUrlnXCqk
02:37:37 <elliott> fizzie: what was that connect command again
02:39:26 -!- Lumpio_ has changed nick to Lumpio-.
02:44:22 -!- Nisstyre-laptop has quit (Ping timeout: 246 seconds).
02:45:15 -!- aloril has quit (Ping timeout: 276 seconds).
02:49:03 <c00kiemon5ter> oh, it's december o.O late happy new month people o/
02:49:27 -!- augur has quit (Remote host closed the connection).
02:50:36 <zzo38> Yes, is December now, and is Advent tomorrow, I think.
02:50:44 <shachaf> Finland isn't real, elliott.
02:50:52 <shachaf> Do you actually know anyone who's ever been there?
02:51:01 <pikhq_> Finland is drunk Japan. No more, no less.
02:51:25 <pikhq_> Well, when you're drunk... :P
02:51:31 <HackEgo> Finland is a European country. There are two people in Finland, and at least nine of them are in this channel. Corun drives the bus.
02:51:39 <zzo38> I added some commands in my Csound plugin, including "meanfilter", "multiguide", "pairecho", etc
02:52:06 <oerjan> it would appear the bus has got lost
02:52:31 <zzo38> The "pairecho" command makes two delay lines which are allowed to interfere with each other and the amount of interference and output levels can be adjusted at x-rate.
02:52:39 * oerjan tried to read pairecho as something latino
02:52:54 <zzo38> Why? Do you speak Latino?
02:53:05 <Bike> ladino's cool.
02:53:17 <oerjan> ...it looks vaguely spanish.
02:53:31 <Sgeo> monqy, elliott Fiora
02:53:33 <pikhq_> Nah, oerjan only speaks onital.
02:53:34 <zzo38> It is meant to be English.
02:57:07 -!- jfischoff has joined.
02:58:37 -!- elliott has quit (Quit: leaving).
03:00:19 -!- elliott has joined.
03:00:39 <elliott> Alright, I got insecure IPv6 working.
03:00:44 -!- elliott has changed nick to Guest55032.
03:01:04 <shachaf> imo Guest55034 > Guest55032
03:01:19 -!- Guest55032 has changed nick to elliott.
03:12:25 -!- Arc_Koen has quit (Quit: Comment on appelle un mec qui pilote un avion ?).
03:14:26 <ion> 68060 > 55032
03:15:39 <ion> λ> compare Guest55034 Guest55032
03:17:01 <Bike> how did that happen?
03:17:08 <pikhq_> let compare = flip compare
03:21:52 <ion> data Guest = Guest55034 | Guest55032 deriving (Eq, Ord)
03:23:00 <shachaf> data Comparing = Eq | Ord deriving (Eq, Ord)
03:23:45 <Bike> note to self do not ask questions
03:24:39 <oerjan> i'd be more impressed if he made it work in lambdabot
03:28:02 <shachaf> oerjan: You should set this channel -n
03:28:09 <shachaf> So elliott can spam it without joining.
03:29:18 <shachaf> oerjan: Also, you should kick me for spamming.
03:29:19 <zzo38> If there is Haskell library for defining moves of chess pieces, what should it be called?
03:30:30 <zzo38> I mean the module name, not the package name, though.
03:32:37 <oerjan> Control.Games.Board.Chess.Moves.Class
03:36:04 <Fiora> oh gosh I got quoted
03:36:21 <Bike> you've really hit the bigtime now.
03:36:23 <shachaf> @remember Fiora oh gosh I got quoted
03:36:36 <shachaf> No, lambdabot is for high-quality quotes.
03:36:38 <shachaf> @forget Fiora oh gosh I got quoted
03:36:51 <lambdabot> sorear says: [emacs haskell mode] not fucked up, just well documented
03:37:10 <lambdabot> copumpkin says: a monad is just a lax functor from a terminal bicategory, duh. fuck that monoid in category of endofunctors shit
03:37:18 <elliott> so Fiora has gone through two phases of initiation
03:37:30 <elliott> excessive welcoming and being quoted
03:37:34 <Bike> is she going to have to write an eodermdrome interpreter too
03:37:38 <elliott> next up is the goat sacrifice
03:37:49 <oerjan> Bike: no, no one actually managed that yet
03:37:51 <elliott> Bike: yes but that usually comes later
03:37:51 <Bike> oh i'm so up for that
03:38:04 <Bike> oerjan: isn't that the joke
03:38:24 <oerjan> she has to spend a few weeks agonizing over how to do it before she fails, though
03:38:36 <Bike> i got bored/frustrated, turns out np-complete problems are annoying, especially when you look at graph rewriting and then look at thue and think "why"
03:38:51 <oerjan> elliott: he just chose a different failure mode, known as "losing the source"
03:39:06 <Fiora> what's an eodermdrome interpreter @_@
03:39:23 <Bike> Fiora: http://esolangs.org/wiki/Eodermdrome it's an esoteric language, or "esolang" for short
03:39:37 <oerjan> anyone whose eyes look like @ isn't sane enough to be told!
03:41:08 <Fiora> yes I know what an eslolang is
03:41:20 <Fiora> oh no. it's /that/ one
03:41:27 <Bike> the one i was failing at, yes
03:41:42 <shachaf> Fiora: "what is an eslolang"
03:41:53 <Bike> i did write a mascarphone interpreter when arc_koen was talking about it the other day, though. completely untested, as the gods demanded
03:42:16 <Fiora> sorry I'm like randomly being pulled away and things because I'm at my parents' house this weekend
03:42:27 <oerjan> does this mascarphone interpreter have good call rates?
03:42:58 -!- Jafet has quit (Quit: Leaving.).
03:43:13 <Bike> what does that mean? if it's some subquestion of "is it efficient" the answer is still no
03:43:37 * oerjan notes strong winds above Bike's head
03:44:06 <Bike> i have congenital joke-blindness, you bigot!
03:44:49 <oerjan> oh. i'm afraid you'll probably not survive in this channel, then.
03:44:56 <Fiora> oh. subgraph isomorphism is the hard one
03:45:01 <Fiora> and graph isomorphism is the almost-hard one
03:45:18 <Bike> i actually got the subgraph isomorphism bit, but the replacement semantics are not what i assumed
03:47:00 <oerjan> there's the "some vertices need to have no extra edges" bit
03:47:47 <Bike> yeah, basically i didn't pay enough attention to the description and wrote it wrong, so, frustration when i found out
04:04:06 <shachaf> ion: So do you do the thing where you capitalize your sentences, but not the word "I"?
04:04:51 <olsner> if a sentence starts with I, do you capitalize the following word instead?
04:04:58 <shachaf> I was wondering whether you do that on purpose or what it is that you do.
04:05:14 <shachaf> olsner: I meant treating "I" as any other word, and capitalizing it like any other word.
04:05:19 <shachaf> Rather than giving it special treatment.
04:05:31 <ion> I’d like to learn the reasoning for the special treatment of “I”. You don’t capitalize “You”.
04:05:52 <Bike> probably something to do with the venerable medieval lack of punctuation or spaces
04:06:13 <shachaf> ion: So it is a thing you do on purpose?
04:06:32 <zzo38> Bike: Yes I am guessing something like that
04:07:17 -!- david_werecat has joined.
04:07:42 <ion> Well, i capitalize “I” in anything more formal, but i find the exception strange.
04:08:10 <shachaf> i ONCE CONSIDERED DOING IT THAT WAY
04:08:22 <n2liquid> I agree, but I just got over it
04:08:47 <n2liquid> There are enough oddities in my own language to be questioning english
04:08:53 <olsner> Write All Sentences in Some Kind of Title Case
04:09:08 <Gregor> I read an article that suggested that the only reason “I” is capitalized is because printers thought “i” looked ugly.
04:09:18 <pikhq_> Bah, we should just capitalize all Nouns, as is the historical Practice.
04:09:24 <shachaf> Nowadays printers have no free will.
04:09:32 <elliott> I have a 1-2 ms ping to the london freenode server from my server
04:09:44 <oerjan> shachaf: that's what they _want_ you to think
04:09:45 <Gregor> elliott: Not particularly wild?
04:09:53 <ion> The same guys who thought you should move a subset of punctuation following a right quotation mark inside the quotation? Screw those guys.
04:10:01 <pikhq_> Free Will is for Plebians, Peasants, and Fools. We educated Souls know that there is no such Thing as free Will.
04:10:10 <pikhq_> There is only the Dictates of Physics.
04:10:22 <elliott> 64 bytes from sturgeon.freenode.net (83.170.94.214): icmp_req=1 ttl=56 time=1.06 ms
04:10:26 <elliott> 64 bytes from sturgeon.freenode.net (83.170.94.214): icmp_req=3 ttl=56 time=1.16 ms
04:10:29 <elliott> 64 bytes from sturgeon.freenode.net (83.170.94.214): icmp_req=4 ttl=56 time=1.15 ms
04:10:31 <shachaf> Gregor: I looked for a taquería and didn't find one.
04:10:32 <elliott> 64 bytes from sturgeon.freenode.net (83.170.94.214): icmp_req=5 ttl=56 time=1.15 ms
04:10:34 <elliott> 64 bytes from sturgeon.freenode.net (83.170.94.214): icmp_req=6 ttl=56 time=1.13 ms
04:10:38 <elliott> don't tell me that's not wild
04:10:47 <Gregor> shachaf: Where the heck are you in SFO that you can't find a taqueria???
04:10:49 <ion> elliott: You seem to have some packet loss.
04:10:50 <pikhq_> elliott: ... How many lightmilliseconds are you from London?
04:10:59 <shachaf> Gregor: I didn't look very far, admittedly.
04:11:30 <pikhq_> Though "a while" is not a quantity of lightmilliseconds, vague or otherwise. ;)
04:11:53 <oerjan> `frink 1 while -> lightmillisecond
04:11:54 -!- david_werecat has quit (Ping timeout: 255 seconds).
04:12:01 <kmc> shachaf: are you in san francisco san francisco or are you in epa
04:12:04 <HackEgo> Syntax error: <String>, line 1, near column 2 \ 1 while -> lightmillisecond \ ^ \ 1 error(s) occurred during parsing.
04:12:09 <kmc> epa gangnam style
04:12:17 <shachaf> kmc: At the moment San Francisco San Francisco.
04:12:41 <shachaf> kmc: San Francisco is the best San Francisco, don't you think?
04:12:41 <kmc> where abouts
04:12:59 <elliott> pikhq_: anyway you should help me get ipv6 working
04:13:00 <kmc> http://en.wikipedia.org/wiki/San_Francisco_(disambiguation)
04:13:03 <elliott> i can connect to freenode via ipv6
04:13:04 <elliott> i can connect to freenode via ssl
04:13:06 <elliott> but not both at the same time
04:14:21 <shachaf> Gregor: Are you doing a PhD thing in San Francisco?
04:14:39 <Gregor> shachaf: I'm doing an internship thing in San Francisco.
04:15:30 <pikhq_> Is it true that a soulectomy is a job requirement?
04:15:48 <kmc> most of my coworkers used to work at oracle :(
04:15:54 <kmc> they must be zombies now
04:16:22 <shachaf> kmc: Are they using Python?
04:16:26 <shachaf> That's a sure sign of being a zombie.
04:16:31 <kmc> i don't understand
04:16:46 <shachaf> Do you have lots of coworkers now?
04:17:11 <kmc> of whom five used to work at oracle
04:17:18 <kmc> shachaf: i suppose
04:17:21 <olsner> six down from however many it was before the zombies started munching brains
04:17:33 <kmc> there are four founders; i am employee #1 and employee #2 started two days after me
04:18:28 <kmc> that's right
04:21:25 <shachaf> kmc: My sister wants to move to Cambridge.
04:21:30 <shachaf> Unfortunately it's the wrong one.
04:21:59 <shachaf> How do I persuade her that Mid-Cambridge, MA is the best city?
04:25:41 <kmc> it's not the best city
04:25:44 <kmc> but pretty good
04:26:07 <kmc> why is your sister moving to the other cambridge?
04:26:21 <shachaf> I think she wants to go to university there.
04:31:52 -!- n2liquid_ has joined.
04:31:52 -!- n2liquid has quit (Read error: Connection reset by peer).
04:39:34 -!- ogrom has joined.
04:42:07 -!- oerjan has quit (Quit: Gnith).
05:00:18 <zzo38> Once I read something by someone who capitalized every word
05:00:52 <monqy> the !!!batch specification? I guess that's not every word
05:01:12 <zzo38> No, I mean many sentences.
05:01:35 <zzo38> Not only that but the lines were also numbered even though the line numbers are not related to text items.
05:04:54 <shachaf> monqy: have you learned lenses yet
05:05:02 <shachaf> monqy: there's a new thing on the block to learn
05:05:16 <zzo38> What do co-lenses do?
05:05:26 <shachaf> zzo38: The dual of what lenses do.
05:07:18 <zzo38> It still doesn't help. Can you be more specific what its type and so on is? Does it make a category like lenses do?
05:08:58 <shachaf> Lenses aren't really a category, are they?
05:09:11 <shachaf> Lens s t a b = (s -> a, (s,b) -> t)
05:09:39 <shachaf> Colens s t a b = (b -> t, s -> Either t a)
05:11:49 <zzo38> If you have the one with only two parameters instead of four, then there is the lens category (including the Category instance).
05:12:25 <monqy> shachaf: where is colens
05:12:45 <shachaf> It's called Projection in lens.
05:14:48 -!- elliott has quit (Quit: leaving).
05:14:56 <shachaf> @tell elliott You know what you should do?
05:19:59 -!- elliott has joined.
05:20:19 <elliott> Well, I have SSL + IPv6 working.
05:20:20 <lambdabot> elliott: You have 2 new messages. '/msg lambdabot @messages' to read them.
05:20:32 <elliott> But it's broken SSL certificate verification. :(
05:20:53 <elliott> Seems to be a result of this bug: https://bugs.launchpad.net/ubuntu/+source/irssi/+bug/573256
05:21:05 <shachaf> elliott: If you wrote a lazy syntax highlighter for Haskell.
05:21:12 <elliott> I think I could work around it just by adding all the freenode servers I want to connect to to the irssi network rather than the round-robin?
05:21:51 <elliott> you sent me a lambdabot message
05:21:56 <shachaf> 21:21 <shachaf> elliott: If you wrote a lazy syntax highlighter for Haskell.
05:22:09 <shachaf> Hmm, I might've changed my tense-thing in the middle there.
05:22:48 <shachaf> elliott: but http://www.reddit.com/r/haskell/comments/144biy/pretty_output_in_ghci_howto_in_comments/ :'(
05:23:58 <Sgeo> Edit 2: You will need to either :seti -XNoMonomorphismRestriction or give myPrint an explicit type annotation. I happened to have the former already.
05:24:17 <Sgeo> Shouldn't rewriting myPrint to not be pointfree work?
05:24:21 <shachaf> Sgeo: I don't care about that.
05:24:27 <shachaf> elliott: Someone bountied my stackoverflow answer!
05:24:52 <shachaf> Hmm, or maybe they bountied one of the other answers.
05:25:26 <Sgeo> Or actually, nothing's wrong with a type annotation, right?
05:25:29 <elliott> the points come automatically
05:25:50 <shachaf> But the asker person didn't pick best answer!
05:26:05 <elliott> then it's assigned automatically
05:26:22 <elliott> shachaf: you should upvote http://stackoverflow.com/questions/9050725/call-cc-implementation/9050907#9050907 so my top-voted answer isn't to the question "How do you identify monadic design patterns?"
05:26:54 <shachaf> Hmm, you could get people to downvote that one instead.
05:26:57 <elliott> hmm i don't even like how the latter answer words it at the start
05:27:02 <elliott> shachaf: yeah but then i would lose internet points
05:27:34 <shachaf> elliott: but how did you get a billion internet points without every getting more than 50 votes on an answer
05:28:23 <elliott> writing one great answer that immediately gets a huge number of votes doesn't even help that much anyway due to the 200 rep cap
05:29:05 <elliott> fizzie: are you there i need help with this certificate thing!!
05:29:33 <elliott> feel free to peruse the long and tedious history of posts at http://meta.stackoverflow.com/ if you want an answer
05:40:16 <elliott> ion: are you the ion in http://bash.org/?152037 by the way
05:40:57 <pikhq_> Nicknames should be unique!
05:46:29 <Sgeo> I should go read the Facebook TOS
05:48:55 <Sgeo> "For content that is covered by intellectual property rights, like photos and videos (IP content), you specifically give us the following permission, subject to your privacy and application settings: you grant us a non-exclusive, transferable, sub-licensable, royalty-free, worldwide license to use any IP content that you post on or in connection with Facebook (IP License). "
05:49:05 <Sgeo> Why do they need a transferable and sub-licensable license?
05:50:09 <Bike> so that they can grant reproduction rights to advertisers and other licensees and stuff
05:51:09 <Sgeo> Honestly, I understand the rest of it, from a technical perspective
05:51:26 <Sgeo> Just in terms of showing photos to peoples friends when their privacy settings are set such, for example
05:51:57 <Sgeo> "You will not post unauthorized commercial communications (such as spam) on Facebook."
05:52:08 <Bike> the main thing to get about facebook is you're the product, not the customer. the customers are ad companies and zynga and companies like that. the TOS is oriented to help the cusotmers.
05:52:36 <Sgeo> How does one get authorization? Many companies have Facebook pages, which I can only assume are used for commercial communications
05:53:28 <Sgeo> "You will not use Facebook to do anything unlawful, misleading, malicious, or discriminatory.
05:53:36 <Sgeo> Uh. I'm not allowed to lie on Facebook?
05:54:15 <Bike> TOSs aren't exactly meant to be followed to the letter, it's just for legal cases and shit
05:54:33 <pikhq_> Sgeo: Yup. So, if you're gay you better be out.
05:55:26 <Bike> I wonder if "don't mislead" would cover being gay and not telling anyone. Since you're only "misleading" them in that they assume you're straight because that's the cultural norm.
05:55:49 <pikhq_> Curses be unto heteronormativity.
05:56:13 * Bike makes traditional hand gesture of agreement
05:57:05 <Sgeo> What about April Fools jokes?
05:57:13 <Sgeo> Such as "just got arrested for drunk driving"
05:57:14 <pikhq_> Not actually misleading.
05:58:00 <Sgeo> Had an April Fools Day where I was just posting statuses saying how I was going to a party, got drunk, had unprotected sex with some girl, and got arrested for drunk driving.
05:58:04 <Sgeo> I don't think anyone believed it
05:58:17 -!- elliott has left.
05:59:20 <Sgeo> "You will not transfer your account (including any Page or application you administer) to anyone without first getting our written permission.
05:59:40 <Sgeo> That... seems annoying for companies. Unless the company is considered the administer the page, and not individual employees?
06:01:01 <Sgeo> "You will not tag users or send email invitations to non-users without their consent. Facebook offers social reporting tools to enable users to provide feedback about tagging.
06:01:58 <Sgeo> ...how does that even work? The tagging mechanisms don't require consent, do they? And sending email invitations to non-users... um, it's an invitation, how do you get consent to send an invitation?
06:02:06 <Sgeo> Although I think that that section is geared more towards developers
06:03:29 <Sgeo> In a section for developers: "You will not directly or indirectly transfer any data you receive from us to (or use such data in connection with) any ad network, ad exchange, data broker, or other advertising related toolset, even if a user consents to that transfer or use."
06:03:50 <Sgeo> That's pretty blatant in terms of what Bike was saying before
06:03:54 -!- elliott has joined.
06:04:25 <Sgeo> hi elliott. helliott.
06:12:10 <Sgeo> CALIFORNIA CIVIL CODE 1542 is useless, isn't it?
06:13:54 <elliott> i hear monqy knows about california
06:14:33 <Sgeo> From what the Facebook TOS quoted, it's a thing about releases (of liability I guess) not applying in all circumstances. So, all an organization needs to do to get a fully general release is say that the other party waives it
06:14:56 <elliott> are you still talking about facebook's terms of service jesus
06:15:43 <Sgeo> I stopped. And didn't say everything I wanted to say, after you joined, because I figured that that's why you left
06:16:02 <elliott> if #esoteric said only things i wanted it would be a very different place
06:23:01 <n2liquid_> ... I see this channel is full of respectful people, eh
06:26:21 <shachaf> kmc: I missed the train south and now the next one won't be for another ~1.5 hours.
06:28:05 <elliott> n2liquid_: we must be in different #esoterics
06:28:23 <Bike> my esoteric is better than your esoteric.
06:29:46 <n2liquid_> Well, I've only been here for this night, but it does seem peaceful
06:36:00 -!- sebbu3 has joined.
06:36:17 -!- sebbu3 has quit (Changing host).
06:36:17 -!- sebbu3 has joined.
06:36:48 -!- sebbu has quit (Ping timeout: 245 seconds).
06:37:09 -!- sebbu3 has changed nick to sebbu.
06:48:03 -!- sebbu2 has joined.
06:48:03 -!- sebbu2 has quit (Changing host).
06:48:03 -!- sebbu2 has joined.
06:48:28 <elliott> @tell fizzie does the client certificate thing work as good as sasl now that you're using it? with the cloak-guaranteed-to-take-effect-before-joining-channels stuff
06:50:58 -!- sebbu has quit (Ping timeout: 245 seconds).
06:51:45 -!- Gregor has changed nick to BabsSeed.
06:52:20 -!- BabsSeed has changed nick to Gregor.
07:01:19 -!- augur has joined.
07:02:13 -!- Gregor has changed nick to TheSmooze.
07:57:51 -!- ogrom has quit (Quit: Left).
08:11:56 -!- aloril has joined.
08:16:07 -!- c00kiemon5ter has quit (Read error: Operation timed out).
08:20:11 -!- jfischoff has quit (Quit: jfischoff).
08:20:30 -!- sebbu2 has changed nick to sebbu.
08:23:52 -!- jfischoff has joined.
08:45:22 -!- epicmonkey has joined.
08:50:51 -!- n2liquid_ has quit (Remote host closed the connection).
08:52:23 -!- jfischoff has quit (Quit: jfischoff).
09:08:59 -!- Sgeo has quit (Ping timeout: 244 seconds).
09:15:05 -!- Bike has quit (Ping timeout: 260 seconds).
09:16:01 -!- zzo38 has quit (Remote host closed the connection).
09:22:33 <fizzie> @tell elliott I haven't actually tried reconnecting after setting it up. It worked for the first time, I think, but then again I still had the server pass thing set up, and that works too most of the time.
09:37:05 -!- punisher00 has joined.
09:41:26 -!- Sgeo has joined.
10:11:47 -!- ifnspifn_ has quit (Quit: ifnspifn_).
10:13:11 -!- punisher00 has quit (Read error: Connection reset by peer).
10:13:59 -!- ifnspifn has joined.
10:15:39 -!- ifnspifn has quit (Client Quit).
10:21:47 -!- ifnspifn has joined.
10:21:54 -!- ifnspifn has quit (Client Quit).
10:53:44 -!- carado has joined.
10:54:32 -!- Phantom_Hoover has joined.
11:28:30 -!- Vorpal has joined.
11:56:24 -!- nooodl__ has joined.
12:11:13 -!- monqy has quit (Quit: hello).
12:24:56 <Sgeo> "In fact, the odds of dying from a scorpion sting are one in 300 million. To put this in perspective: Your odds of dying by simply falling over in the shower are one in 65,000. In other words, if you find a scorpion in your shower tomorrow morning, the shower stall itself may still be the greater danger.
12:24:56 <Sgeo> Read more: http://www.cracked.com/article_19171_5-things-that-arent-nearly-as-dangerous-as-hollywood-thinks.html#ixzz2Dtf6HJYx"
12:25:05 <Sgeo> ...screw you thingy
12:25:23 <Sgeo> Am I allowed to facepalm at the misunderstanding of probability, or should I just accept it as a joke
12:26:00 <Phantom_Hoover> is that 1 in 300 million the population-wide probability of dying from a scorpion sting
12:26:24 <Sgeo> It links to http://www.bookofodds.com/Accidents-Death/Death-Rates/Odds/The-odds-a-person-will-die-from-being-stung-by-a-scorpion-in-a-year-are-1-in-299-400-000-US-2006
12:28:21 <Sgeo> It could still be true, I guess, but there's not enough information.
14:21:55 <FreeFull> Sgeo: The reason why the odds of dying from a scorpion sting are so low is that scorpions don't live in most places that people do
14:22:14 <FreeFull> If there is actually a scorpion there, your odds definitely go up
14:29:32 <Phantom_Hoover> your ability to state what everyone else had already implicitly figured out is truly a boon to the channel
14:31:33 <FreeFull> Well I am Captain Obvious, if that's not obvious
14:41:25 -!- david_werecat has joined.
14:51:22 -!- lambdabot has quit (Ping timeout: 244 seconds).
14:54:34 -!- lambdabot has joined.
15:15:58 -!- c00kiemon5ter has joined.
15:20:53 -!- c00kiemon5ter has quit (Remote host closed the connection).
15:21:28 -!- c00kiemon5ter has joined.
16:21:28 -!- Nisstyre-laptop has joined.
16:22:53 -!- yours_truly has joined.
16:33:31 -!- yours_truly has quit (Quit: Leaving).
16:42:03 <Vorpal> <Phantom_Hoover> please stop being captain obvious then <-- who should be captain obvious then?
16:42:45 <fizzie> But then the discussion can go astray.
16:42:46 <lambdabot> fizzie: You have 1 new message. '/msg lambdabot @messages' to read it.
16:43:02 <Vorpal> Phantom_Hoover, I always thought of Captain Obvious like a superhero, like Captain America or whatever
16:43:13 <Vorpal> now I'm curious as to what was originally intended in that phrase
16:44:05 <fizzie> "By your inanities combined, I am Captain Obvious!"
16:45:12 <Vorpal> then we obviously need a Captain Obvious
16:45:21 <kmc> fuck captain planet
16:45:34 <elliott> fizzie: imo you should do some testing & also help me out w/ my new ssl problem
16:45:34 <lambdabot> elliott: You have 1 new message. '/msg lambdabot @messages' to read it.
16:45:39 <lambdabot> fizzie said 7h 23m 7s ago: I haven't actually tried reconnecting after setting it up. It worked for the first time, I think, but then again I still had the server pass thing set up, and that works
16:45:54 <kmc> his agenda of taking pollution "down to zero" puts one in mind of the fanatical de-industrialization agenda of the khmer rouge
16:46:39 <elliott> #esoteric, number one channel for sociopolitical analysis of captain planet
16:46:56 <kmc> shachaf: did you make it to a train
16:46:58 <Phantom_Hoover> captain planet throws gi in a labour camp because she looks too intellectual
16:47:06 <kmc> or did you just have to run along the tracks at high speed yelling "CHOO CHOO MOTHERFUCKER"
16:48:02 <Vorpal> Phantom_Hoover, heh what?
16:48:44 <Phantom_Hoover> i have never seen captain planet or indeed heard of it beyond 80s pop culture references
16:49:15 <fizzie> I have watched it, it was broadcast in Finland.
16:49:28 <kmc> in college we had a house office named "captain planet"
16:49:39 <kmc> they were charged with taking care of all the living things in the house
16:49:41 <kmc> primarily the hot tub
16:50:07 <fizzie> Phantom_Hoover: It was also made in 1990-1996, so it's curious it appears in 80s pop culture.
16:50:28 <kmc> it's a complex ecosystem
16:50:46 <kmc> haven't seen
16:51:39 <fizzie> Helsinki University has a greenhouse warmed by computer exhaust heat on top of the Exactum building.
16:51:59 <kmc> fizzie: nice
16:52:02 <fizzie> http://www.youtube.com/watch?v=WmmfmYID1Yw -- you can see it there.
16:52:26 <fizzie> Or you can see the rack, but the vaguely transparent plastic thing behind it is a greenhouse.
16:53:20 <fizzie> (I saw the link somewhere a while ago.)
16:54:04 <kmc> heh he has a shirt with the pac man kill screen
16:55:07 <Phantom_Hoover> yeah, the it crowd is good with having actual nerdy references in the background
16:56:25 <Phantom_Hoover> there are some eff posters around the place too, and a picture of bob dobbs
16:57:22 <kmc> 'This show's like "The Big Bang Theory" except it doesn't suck a Big Franch Dick.'
16:57:27 <kmc> well put, random youtube idiot
16:58:06 <kmc> (franch, as everyone knows, is a condiment composed of a mixture of french and ranch dressing)
16:58:17 <Phantom_Hoover> even assuming that's a baffling typo for french, since when were french dicks known for their size?
16:58:51 <kmc> well if they were, the adjective wouldn't be necessary
16:59:04 <kmc> insert coq joke here
16:59:51 <kmc> 'Twenty years ago, I wrote a comedy in which a scientist accidentally kills God and feels really terrible about it. Meanwhile, his former lab assistant goes on to fame and fortune by inventing something called "Franch" -- a salad dressing that's half-French, half-ranch.'
17:00:14 <Vorpal> <fizzie> http://www.youtube.com/watch?v=WmmfmYID1Yw -- you can see it there. <--- lol what?
17:00:18 <Vorpal> servers out in the open?
17:00:26 <fizzie> Vorpal: It's not out in the open.
17:00:35 <fizzie> Vorpal: There's a transparent plastic in front of the rack.
17:00:59 <fizzie> Vorpal: But yes, it's a "Experimental Free Air Cooling setup".
17:01:32 <Vorpal> fizzie, looks like he is removing snow directly from the servers but oh well
17:02:14 <Vorpal> fizzie, what are they talking about btw?
17:02:31 <fizzie> There's a translation in the comments, though it's a bit off.
17:02:35 <Vorpal> "Person 1: this is a really stupid idea" "Person 2: No no, it will just work"
17:03:03 -!- Bike has joined.
17:03:19 <fizzie> It's something like "Here is Mikko doing server administration." "Works well!" "And that's how we do a little clean<cuts off>"
17:03:39 <kmc> reminds me of http://www.afrotechmods.com/
17:04:17 <kmc> in particular http://www.afrotechmods.com/papercooling.htm
17:04:35 <fizzie> https://lh5.googleusercontent.com/-6F8oAMFyZtk/T2iGlBvcFaI/AAAAAAAACVs/GJwiMc8Oo94/s1152/IMG_20120320_145825.jpg shows the other side of the rack.
17:04:55 <Vorpal> fizzie, it might become too cold for the servers as well hm
17:05:21 <fizzie> Vorpal: The blog mentions they haven't had any problems at -30 degrees Celsius or so.
17:05:23 <elliott> <META NAME="GENERATOR" CONTENT="Mozilla/3.04Gold (WinNT; I) [Netscape]">
17:05:26 <elliott> this page is as old as kmc
17:06:25 <fizzie> Vorpal: Most of their computer centers are more traditional, I think this is more of a hobby experiment.
17:06:39 <kmc> http://afrotechmods.com/stupid/memory/memory.htm
17:08:48 <Phantom_Hoover> (i forget, is that one of the things everyone's seen but some twat always posts it)
17:09:29 <olsner> why doesn't the heat of the servers melt the snow and create horrible water problems?
17:10:36 <fizzie> olsner: That's exactly what one of the comments asked, I think.
17:10:49 <fizzie> I don't have any sort of answer.
17:11:10 <fizzie> Except that water flows down, I suppose you can sort of guide it elsewhere.
17:11:31 <olsner> I suppose it could be cold enough that nothing in the server is above freezing
17:11:36 <fizzie> The linked blog has a lot of content about the design of the box around it.
17:12:19 <fizzie> The snow is going to melt before summer, anyway, so I'm sure they've considered water.
17:12:30 <fizzie> Also, it sometimes rains.
17:19:22 <kmc> what if a moose tries to eat the servers
17:20:07 <arcatan> the servers are on the roof of the university building, so no mooses there (hopefully)
17:26:50 -!- quintopia has quit (Ping timeout: 265 seconds).
17:27:40 -!- quintopia has joined.
17:31:01 -!- jfischoff has joined.
17:48:43 -!- Arc_Koen has joined.
18:03:06 -!- jfischoff has quit (Quit: jfischoff).
18:03:25 -!- zzo38 has joined.
18:12:49 -!- jfischoff has joined.
18:17:50 -!- ogrom has joined.
18:19:05 -!- ogrom has left.
18:42:54 -!- sebbu2 has joined.
18:42:54 -!- sebbu2 has quit (Changing host).
18:42:54 -!- sebbu2 has joined.
18:46:23 -!- sebbu has quit (Ping timeout: 245 seconds).
18:47:52 -!- jfischoff has quit (Quit: jfischoff).
18:50:26 -!- jfischoff has joined.
18:51:48 -!- AnotherTest has joined.
18:52:44 -!- sebbu2 has changed nick to sebbu.
19:03:51 <Phantom_Hoover> have you by any chance done that thing you've claimed you'll do for like a week
19:08:59 -!- ifnspifn has joined.
19:14:47 <ion> elliott: I’m not.
19:15:02 <ion> elliott: It’s a fake ion.
19:17:31 -!- ifnspifn has quit (Quit: ifnspifn).
19:17:45 <elliott> ion: did you see that fancy new edwardk exported unsafeCoerce
19:18:50 <ion> I saw the paste, didn’t look at “vacuous” yet.
19:19:05 <elliott> ion: the implementation is vacuous = unsafeCoerce
19:19:16 <elliott> because if you have a legit functor f, you can implement f Void -> f a that way
19:19:30 <elliott> unfortunately, people can lie!
19:19:41 <elliott> so (Functor f) => f Void -> f a actually means forall f. f Void -> f a
19:19:48 <elliott> which means Iso Void Void -> Iso Void a
19:19:55 <elliott> which means (a -> Void, Void -> a)
19:20:04 <elliott> er, with a forall on each, of course
19:20:07 <elliott> which means compose them and you get a -> b
19:33:56 -!- david_werecat has quit (Ping timeout: 255 seconds).
19:36:16 <zzo38> You could write it with empty case blocks if it were allowed, such as (fmap $ \x -> case x of {})
19:37:03 <zzo38> But you could use undefined instead.
19:38:59 -!- ifnspifn has joined.
19:42:24 -!- ogrom has joined.
19:44:43 -!- epicmonkey has quit (Ping timeout: 245 seconds).
19:52:27 -!- atriq has joined.
20:00:59 -!- FreeFull has quit (Quit: Rebooting into new kernel).
20:09:42 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
20:10:44 -!- carado has quit (Ping timeout: 246 seconds).
20:10:45 -!- FreeFull has joined.
20:10:46 -!- ais523 has joined.
20:12:10 -!- oerjan has joined.
20:13:04 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
20:13:13 <elliott> oerjan: I found an exported unsafeCoerce bug in an edwardk package!
20:13:37 <fizzie> !help im trapped in a #esoteric factory
20:13:38 <EgoBot> Sorry, I have no help for im_trapped_in_a__esoteric_factory!
20:13:42 -!- david_werecat has joined.
20:13:43 <elliott> oerjan: You know the void package?
20:13:55 <elliott> And its unsafeCoerce trick for implementing vacuous :: (Functor f) => f Void -> f a?
20:14:04 <elliott> Because that works for all functors.
20:14:07 <elliott> WELL, http://hpaste.org/78675
20:14:10 <kmc> elliott: pfft found that bug like a year ago
20:14:16 <kmc> srsly though good work
20:14:30 <oerjan> !malbolge ('&%:9]!~}|z2Vxwv-,POqponl$Hjig%eB@@>}=<M:9wv6WsU2T|nm-,jcL(I&%$#"`CB]V?Tx<uVtT`Rpo3NlF.Jh++FdbCBA@?]!~|4XzyTT43Qsqq(Lnmkj"Fhg${z@>
20:14:43 <elliott> it is funny because this is the second time one of edwardk's "safe" unsafeCoerces has turned out to result in an external unsafeCoerce
20:15:00 <elliott> admittedly the first one was only in the git version of lens and never released
20:15:01 -!- carado has joined.
20:15:07 <kmc> http://hpaste.org/52660
20:15:27 <kmc> different bug i guess
20:15:38 * oerjan is testing the Malbolge hello world to see what it actually prints, because of a strange edit war on wikipedia.
20:16:31 <fizzie> oerjan: But wouldn't that be ORIGINAL RESEARCH?
20:16:35 <elliott> kmc: did you tell edwardk :P
20:18:31 <EgoBot> invalid character in source file
20:18:39 <EgoBot> invalid character in source file
20:18:49 <EgoBot> invalid character in source file
20:18:50 <shachaf> kmc: YOur bug is the same one, I think, or pretty close.
20:18:52 <fizzie> oerjan: You must write a published book about Malbolge Hello Worlds, and then wait until someone else writes a book that refers to your book, and then you can rely on that.
20:19:00 <kmc> elliott: i did
20:19:31 -!- AnotherTest has quit (Quit: Leaving.).
20:19:31 <oerjan> elliott: ok so it's an unsafeCoerce that is only safe if the Functor instance actually satisfies the laws, presumably?
20:19:33 <elliott> he said he'd fix it when i showed him this one!
20:19:47 <elliott> oerjan: any data type that is an actual functor will admit that unsafeCoerce implementation, I think
20:19:48 <FreeFull> Or is asm some other esoteric language
20:19:55 <elliott> or at least, I can't think of a counterexample
20:20:00 <elliott> maybe there is one with fancy GADT type family stuff
20:20:16 <fizzie> FreeFull: There's "asm" in both the esoteric and other lists.
20:20:58 <FreeFull> !c int main() { printf("%d\n", 1); return 0;}
20:21:01 <elliott> for some reason I thought of this unsafeCoerce bug while trying to get to sleep
20:21:08 <elliott> after not having looked at or used the void package in ages
20:21:46 <FreeFull> !c int main() { int arr[1]; int i; for(i=0;;i++) { arr[i] = i; } return 0; }
20:21:48 <EgoBot> ./interps/gcccomp/gcccomp: line 52: 17341 Segmentation fault /tmp/compiled.$$
20:22:01 <fizzie> FreeFull: The !asm runs /interps/gcccomp/gcccomp assembler which will try to gcc yourcode.s.
20:22:15 <fizzie> FreeFull: So you need to write it in x86-64 AT&T assenbler.
20:22:35 <fizzie> That's not what it looks like.
20:22:43 <EgoBot> ./interps/gcccomp/gcccomp: line 52: 17509 Segmentation fault /tmp/compiled.$$
20:22:57 <fizzie> mov 1, %eax would be intel mov eax, [1].
20:23:04 <FreeFull> Ah, without $ it's a memory address
20:23:12 <fizzie> There's also .globl main; main: pushq %rbp; movq %rsp, %rbp; prependend automatically, and movl $0, %eax; leave; ret; suffixed.
20:24:09 <nooodl__> can !asm do anything reasonably nifty
20:24:18 <elliott> fizzie: wanna hear a joke?
20:24:37 <fizzie> !asm .extern puts; .data; hello: .asciz "hello there"; .text; mov $hello, %rdi; call puts
20:27:02 <oerjan> FreeFull: btw the error you got on the Malbolge is because every position in the program allows a varying set of 8 bytes, namely the ones which decrypt to actual commands at that spot.
20:27:18 <oerjan> and no other byte is allowed at that spot
20:28:05 -!- ogrom has quit (Quit: Left).
20:30:05 -!- ogrom has joined.
20:31:28 -!- ogrom has left.
20:35:29 -!- Phantom_Hoover has joined.
20:35:55 -!- ashet4 has joined.
20:39:41 <Phantom_Hoover> why would i look at the topic, it's not like it ever contains any useful information
20:40:45 <oerjan> <Phantom_Hoover> your ability to state what everyone else had already implicitly figured out is truly a boon to the channel <-- hey don't be too hard on him, i was tempted myself. even while logreading.
20:41:37 -!- oerjan has set topic: Babies are usually not eaten owing to their repugnant looks, as well as their viscosity and unpleasant habits. | http://codu.org/logs/_esoteric/.
20:41:54 -!- elliott has set topic: http://codu.org/logs/_esoteric/.
20:42:02 * oerjan swats elliott -----###
20:42:24 -!- elliott has set topic: .
20:42:42 -!- Phantom_Hoover has set topic: ^H^H^H^H^H^H^H^H^H.
20:43:00 * oerjan wonders if that actually breaks any clients
20:43:24 <atriq> My client doesn't support topics shorter than 28 characters
20:43:44 <oerjan> Phantom_Hoover: that might explain it.
20:43:54 -!- elliott has set topic: caret h.
20:43:59 -!- nooodl__ has set topic.
20:44:00 -!- oerjan has set topic.
20:44:10 -!- atriq has set topic: carrot ache.
20:44:17 <elliott> oerjan: how do you insert the actual Ctrl+H with irssi
20:44:35 <elliott> oerjan: doesn't work for me :(
20:44:37 -!- nooodl__ has set topic: /topic: the new way to chat.
20:45:12 -!- atriq has set topic: /topic: the new way to chat | How arre you? | *-r.
20:45:14 <oerjan> hm i may have set a binding myself
20:45:15 <nooodl__> on x-chat ctrl+shift+u-8 works but that's probably not the case in irssi
20:46:06 <oerjan> elliott: ^ the above binding is what i have, and allows me to do that with any control char
20:46:45 <elliott> FreeFull: that's not what i want to do
20:46:53 <oerjan> i probably set it to resemble vim in that respect
20:48:40 <elliott> http://en.wiktionary.org/wiki/forget,_when_up_to_one%27s_neck_in_alligators,_that_the_mission_is_to_drain_the_swamp
20:49:37 <Bike> is that an actual term anyone has ever used
20:50:04 <elliott> Bike: let's not forget, when up to one's neck in alligators, that the mission is to drain the swamp. it doesn't matter whether anyone has *used* the term.
20:50:37 <oerjan> elliott: i've also bound home to scroll_start and end to scroll_end
20:50:46 <atriq> Why is a lot of things making me feel guilty
20:51:04 <atriq> It's not as if those nuns didn't deserve to be guillotined...
20:52:23 <nooodl__> elliott: thanks for that idiom
20:52:25 <oerjan> because i kept wanting to do that, while they by default just duplicated ^A and ^E. iirc.
20:52:32 <elliott> nooodl__: please dont use that idiom
20:55:00 <fizzie> oerjan: They are beginning_of_line and end_of_line by default, yes.
20:55:26 <elliott> does ecape_char even have a default binding
20:55:41 <fizzie> It does not. At least it wasn't bound to anything for me.
20:55:49 <fizzie> Now it is bound to ^V IT'S SPREADING
20:56:50 <elliott> maybe i can bind something to type "/win " for me
21:03:03 -!- hagb4rd2 has joined.
21:03:57 -!- hagb4rd has quit (Disconnected by services).
21:04:04 -!- hagb4rd2 has changed nick to hagb4rd.
21:13:45 -!- epicmonkey has joined.
21:15:29 <FreeFull> oerjan: But how am I meant to get to the start or end of what I typed then
21:16:43 <oerjan> FreeFull: ^A and ^E, i said
21:17:25 <oerjan> in my initial setup, those had the same bindings as home and end. so i changed one pair to something else useful.
21:18:11 <elliott> 21:16:34 -!- TuxBlackEdo [~TuxBlackE@unaffiliated/tuxblackedo] has joined ##crypto
21:18:14 <elliott> 21:17:16 <TuxBlackEdo> hey i just met you, and this is crazy, but here is a cyphertext, crack it for me maybe?
21:18:20 <elliott> regretting joining this channel so i could let lambdabot in it
21:18:49 <oerjan> elliott: well if you can now find a one-line lambdabot command to crack it, all will be well.
21:19:46 <FreeFull> oerjan: I have ^A as my tmux key
21:19:58 <FreeFull> I have that make a literal <CTCP>
21:20:53 <oerjan> FreeFull: fine, we can assume there's a reason irssi allows personal settings.
21:21:25 <FreeFull> Because otherwise nobody would use it?
21:21:49 <zzo38> This IRC client makes CTRL+A send a literal CTRL+A for such purpose; to send other controls literal requires CTRL+P at first.
21:26:27 <fizzie> I keep using ^A and ^E for start_of_line/end_of_line even though it's under screen; I just ^Aa all the time. It's very stupid.
21:26:37 <elliott> have you considered changing your screen key
21:27:03 <fizzie> I think that would just make me type a lot of ^As everywhere.
21:27:04 <FreeFull> ^A is the most convienient for me
21:27:37 <FreeFull> The default key for tmux is ^B or some shizz
21:28:31 <fizzie> It's not as if I could think and operate a computer at the same time, so I have to keep doing the stupid thing.
21:29:52 <elliott> fizzie: Is there a way to do a "/window list" that lists in the current buffer?
21:30:18 * oerjan suddenly imagines an intelligent race that cannot think and act at the same time, but must do all actions according to (short) preplanned algorithms
21:30:30 <fizzie> elliott: You can not have a status window at all, I think that would do it.
21:30:43 <hagb4rd> fizzie: you could try to pawlow self-conditioning
21:30:46 <elliott> fizzie: that is not terribly satisfying
21:31:01 <elliott> fizzie: i would be ok with a way to write a key combo that switches to the status window and does /window list and then types "/window " for me i guess
21:31:02 <hagb4rd> fizzie: everytime you use type ^A bite yourself
21:31:06 <elliott> so i could do like ^W <number><enter>
21:31:52 <shachaf> elliott: What about Alt number?
21:32:00 <oerjan> and once an algorithm has started, it cannot be halted, except according to its own rules, and no thinking can happen simultaneously.
21:32:29 <elliott> shachaf: that doesn't show me a list. this is for when i forget which number is which
21:32:39 <fizzie> elliott: You can probably do that switch-to-status thing.
21:32:51 <shachaf> elliott: Just type the name in?
21:33:15 <oerjan> and the algorithms are too short to simulate any significant intelligence.
21:33:22 <elliott> "/win goto the-full-name" is a bit long.
21:33:28 <elliott> I would accept "/w substring"
21:34:25 <oerjan> hm it occurs to me that such an intelligence species would get around this by cooperation.
21:34:47 <FreeFull> I should bind scroll_start and scroll_end to something
21:35:14 <fizzie> elliott: "/bind X multi change_window 1;command window list;insert_text /win " (with the trailing space) where X is your key, I think.
21:35:16 <atriq> elliott, are you getting addicted to lens?
21:35:17 <kmc> switch (core.opcode_index+Fetchb()) {
21:35:18 <kmc> #include "core_normal/prefix_none.h"
21:35:18 <kmc> #include "core_normal/prefix_0f.h"
21:35:18 <kmc> #include "core_normal/prefix_66.h"
21:35:36 <hagb4rd> intelligence seems not be the best method for multitasking
21:35:39 <elliott> atriq: more like addicted to HAPPINESS
21:35:54 <elliott> fizzie: Okay, let's try this. (Is ^W bound to anything by default?)
21:36:05 <fizzie> elliott: It's bound to the usual delete-word thing.
21:36:38 <Phantom_Hoover> what are lenses and can they be given a succinct, incomprehensible definition in terms of category theory
21:37:39 <atriq> There to Functors as something else is to Applicatives
21:38:26 * oerjan vaguely recalls something about that
21:38:31 <atriq> No, something more category theoryish
21:39:21 <zzo38> A lens is not a functor, although there is functor from category of isomorphisms of (->) to Lens, I think.
21:39:22 <oerjan> i thought lens had a spot where you put in either Functor or Applicative and the former gave you an ordinary lens
21:39:27 <atriq> Phantom_Hoover, basically, they use functors to generalize functions and setters to the SAME THING
21:40:05 <zzo38> But perhaps they could be made on other categories too, I don't know
21:40:40 <elliott> Phantom_Hoover: store comonad coalgebra or something
21:41:05 <elliott> ah: "Lenses are the coalgebras for the costate comonad"
21:41:41 * oerjan might perhaps look at lens properly some day.
21:41:56 <oerjan> although maybe not until after upgrading the platform.
21:42:08 <FreeFull> fizzie: I just made lots of alt+shift and then ^Z bindings
21:42:14 <oerjan> i think i'm at least two versions behind.
21:42:22 <FreeFull> I'll do ^Z-shift or ^Z-alt when I'll need it
21:42:23 <atriq> What would an algebra in the state monad even look like
21:42:24 <fizzie> Conans are to good cooperation what nans are to good operation.
21:42:30 -!- sebbu2 has joined.
21:42:30 -!- sebbu2 has quit (Changing host).
21:42:30 -!- sebbu2 has joined.
21:42:54 <oerjan> fizzie: i thought Conans were barbarians.
21:43:13 <atriq> oerjan, it's cool, nans are grandmothers
21:43:15 <zzo38> I guess algebra of state monad would be like ((s -> (a, s)) -> a)
21:44:49 <FreeFull> A grandmother certainly isn't a number
21:44:54 <oerjan> except that Conan that is a comedian. i guess his nan is rather median, then.
21:45:02 <hagb4rd> and conan is an instance of barbarian
21:45:05 -!- sebbu has quit (Ping timeout: 265 seconds).
21:45:33 -!- epicmonkey has quit (Ping timeout: 245 seconds).
21:47:48 -!- atriq has quit (Ping timeout: 244 seconds).
21:49:47 -!- david_werecat has quit (Ping timeout: 260 seconds).
21:52:14 -!- carado has quit (Ping timeout: 246 seconds).
21:52:27 -!- Frooxius has quit (Ping timeout: 244 seconds).
21:54:52 <oerjan> Phantom_Hoover: do the elves walk around saying "siege heil"?
21:59:02 <Vorpal> <fizzie> Nan the Cobarbarian. <-- I read that as "Nam"
21:59:14 <Vorpal> and was wondering why you were joking about Indian food
22:03:43 <Vorpal> hm, is that how it is written in Swedish though? *checks*
22:03:45 <kmc> woah that's a palindrome
22:04:06 <Vorpal> Phantom_Hoover, yeah you are right
22:04:51 <fizzie> Phantom_Hoover: Sometimes also plain 'nan'; I refer to it as "not-a-number bread" personally.
22:05:02 <FreeFull> palindrome is disappointing because it's not autological
22:05:35 <FreeFull> fizzie: I haven't seen bread that was a number
22:05:37 -!- david_werecat has joined.
22:06:07 <fizzie> "Nam" is also a Finnish interjection much the same as the English "yum".
22:06:16 <fizzie> http://en.wiktionary.org/wiki/nam#Interjection agrees.
22:06:18 <Phantom_Hoover> fizzie, i bet you're really popular with all the local takeaways
22:07:27 <fizzie> Phantom_Hoover: I don't think I've ever had to mention it to one of those; the not-a-number designation has pretty much been confined to discussions at home about what to buy from the store.
22:08:21 <kmc> shachaf: do you know how hard it is to segfault DOSBox with bad guest code? if you answered "not at all hard" then you are correct
22:08:43 <kmc> printf '\x62\xe5' > foo.com && dosbox foo.com
22:09:00 <fizzie> Hey, Volapük also has the word "nam".
22:09:07 <FreeFull> kmc: Hell, I have it happen even with valid DOS programs
22:09:12 <fizzie> "Etymology: Reversal of “man”, from Latin manus (“hand”)."
22:09:26 <shachaf> kmc: That's not very hard.
22:10:28 <fizzie> DOSBox doesn't check segment limits either. :/
22:11:53 <FreeFull> Its OPL emulation seems better than worse though
22:12:21 <fizzie> You can poke the VGA memory with a xor ax, ax; mov ds, ax; mov eax, 0xa0000; mov byte [eax], 42 in it.
22:12:59 <FreeFull> What would happen on a 386 machine in 16 bit mode with that code?
22:14:47 <fizzie> I am not entirely sure, but I'd guess an exception in at least virtual-8086 mode; maybe not in real-address mode.
22:15:27 <fizzie> Perhaps in real-address mode too.
22:15:30 <kmc> sadly that segfault is only a call to a NULL function pointer
22:16:04 -!- Mathnerd314 has quit (Remote host closed the connection).
22:16:55 <fizzie> "In the real-address mode, vector 13 is the segment overrun exception" just going by the name, that could happen.
22:19:34 -!- nys has joined.
22:21:17 -!- Vorpal has quit (Read error: Operation timed out).
22:22:17 <nys> so i just got this idea
22:22:20 <nys> http://pastebin.com/UnfL0Z7T
22:22:30 <shachaf> kmc: Looks like segfaults aren't very hard to come by.
22:22:41 <shachaf> Hmm, or maybe I was just wrong.
22:26:09 <kmc> what did you observe?
22:26:39 -!- aloril has quit (Ping timeout: 260 seconds).
22:28:08 <shachaf> Hm, printf '\x44\x51' > foo.com && dosbox foo.com
22:28:25 <shachaf> Unlike 62 e5, this isn't an invalid instruction.
22:28:58 <FreeFull> So overwrite whatever was on top of stack
22:29:20 <shachaf> I don't *think* that's supposed to segfault it...
22:29:24 <zzo38> But will the program terminate, or probably it will result something invalid and break it?
22:29:29 <kmc> segfaults my Debian DOSBox 0.74 but not the SVN trunk version
22:29:34 -!- aloril has joined.
22:30:32 <FreeFull> shachaf: I think the top of the stack would contain the return to dos
22:31:58 <shachaf> kmc: Anyway there are a bunch of other 2-byte .com files that will crash it.
22:32:09 <kmc> did you try them all?
22:32:33 <zzo38> To return to DOS using the stack I think you will need a RET instruction, though.
22:32:46 <shachaf> I generated them all but trying them is a hassle.
22:32:49 <FreeFull> Otherwise the CPU would just go past your program and then execute whatever
22:35:02 <kmc> seems the guest can install a callback for certain events (not the ISA-provided interrupt vector mechanism but some DOSBox thing) and the default callback is "jump to address 0 in the host"
22:35:32 -!- david_werecat has quit (Ping timeout: 248 seconds).
22:36:30 <zzo38> Is that implementing the "mount" command?
22:37:08 -!- nys has quit (Read error: Connection reset by peer).
22:41:15 -!- david_werecat has joined.
22:42:19 -!- monqy has joined.
22:44:31 <shachaf> kmc: How did you come across it?
22:46:06 -!- nys has joined.
22:47:39 <monqy> have you really not
22:47:42 <monqy> started the fortess yetr
22:47:48 <monqy> are you ever going to..........................
22:48:47 <Phantom_Hoover> first, get df working (come on you must have done this)
22:48:59 <elliott> i have to come up with a way to make sure Phantom_Hoover won't cheat!!
22:49:06 <elliott> something about cats and explosoins
22:49:12 <shachaf> elliott: imo don't play df btw
22:49:26 <monqy> playing df means not fixing lens : )
22:49:55 <shachaf> monqy: What happened to your nose?
22:50:59 <monqy> alt. can't you see it????
22:53:17 -!- Nisstyre-laptop has quit (Quit: Leaving).
22:54:05 -!- Nisstyre-laptop has joined.
22:57:38 <oerjan> @tell ais523 http://esolangs.org/wiki/Talk:Carriage#Representation_erasure
23:21:34 <Arc_Koen> my god it's full of (5-legged) stars
23:25:45 -!- jfischoff has quit (Quit: jfischoff).
23:45:41 <oerjan> i think five-pointed is more common, although my maniackal laugh was not about that.
23:47:02 <hagb4rd> where do i know that from?
23:47:25 <oerjan> there's even a wp article http://en.wikipedia.org/wiki/Five-pointed_star
23:50:33 <hagb4rd> though i find legs more accurate than points since it 10 of them
23:51:37 <oerjan> Arc_Koen: um isn't the mutual recursion in ocaml solved the same way with types as with functions, by using "and"?
23:53:39 <Arc_Koen> and I had always thought "it's stupid that you can do mutual recursion for variables but not for types"
23:54:05 <Arc_Koen> that's kind of a problem I have, though
23:54:40 <Arc_Koen> for instance when playing a board game and my opponent does something unexpected I will usually think "uh, that's a weird move" rather than "ok, now, why did he do that?"
23:55:16 -!- nys has quit (Quit: quit).
23:56:10 <oerjan> obviously this is simpler in haskell >:)
23:56:11 <hagb4rd> on the other hand this would pretty much depend on the move
23:56:55 <hagb4rd> is there a native operator in haskell to get the cartesian product?
23:57:12 <Bike> how can it be simpler than "use and"
23:57:15 <oerjan> > liftM2 (,) "may" "be"
23:57:17 <lambdabot> [('m','b'),('m','e'),('a','b'),('a','e'),('y','b'),('y','e')]
23:58:51 <hagb4rd> couldn't it be more like may X be?
23:58:57 <oerjan> Bike: admittedly for values/functions you also need to use "rec", i think
23:59:33 <oerjan> hagb4rd: it's not _that_ often you need it i think...