00:00:04 -!- augur_ has quit (Remote host closed the connection).
00:30:14 -!- icarot has joined.
00:31:48 -!- jconn has quit (Ping timeout: 248 seconds).
00:34:07 <icarot> I know nothing of this channel, but after reading the logs with topic jumps to Russian, logic gates, opaque puns - and who even knows what else - I had to come here.
00:34:28 <HackEgo> icarot: 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.)
00:34:35 <elliott> usually I'd expect the logs to drive people away
00:35:22 <icarot> It must have been an unusual log, then. Or maybe it's just me.
00:35:46 <Bike> well, oerjan's been messing with intercal for the last twenty minutes
00:36:01 <elliott> I thought it was just Fueue
00:36:04 <Bike> fueue, intercal, same shit
00:36:52 <oerjan> i haven't messed with intercal for a decade
00:37:37 <Bike> COME FROM is an integral part of nondeterministic programming
00:38:01 <oerjan> http://oerjan.nvg.org/esoteric/intercal/
00:39:06 <Bike> "please read me a story"
00:40:06 <elliott> I love oerjan's comments in that program
00:40:16 <Bike> galois field mines
00:41:00 <elliott> monqy: have you read it? you should
00:41:25 <monqy> ive read parts of it
00:41:52 <oerjan> ...are you still talking about my interpreter
00:43:18 <oerjan> the galois field was nice
00:56:55 <HackEgo> ihcahroht: Wehlcohme to the ihntehrnahtiohnahl huhb fohr ehsohtehrihc prohgrahmmihng lahnguahge dehsihgn ahnd dehployhmehnt! Fohr mohre ihnfohrmahtiohn, chehck ouht ouhr wihki: http://ehsohlahngs.ohrg/wihki/Maihn_Pahge. (Fohr the ohthehr kihnd ohf ehsohtehrihca, try #ehsohtehrihc ohn ihrc.dahl.neht.)
01:22:33 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
01:23:23 -!- Bike_ has joined.
01:23:31 -!- Bike has quit (Ping timeout: 240 seconds).
01:37:28 -!- augur has joined.
01:37:32 -!- sirdancealot has quit (Ping timeout: 255 seconds).
01:40:02 -!- sirdancealot has joined.
01:42:38 -!- azaq23 has joined.
01:44:46 -!- Arc_Koen has quit (Quit: Arc_Koen).
02:25:26 -!- Bike_ has changed nick to Bike.
02:28:13 <kmc> http://upload.wikimedia.org/wikipedia/commons/6/65/Fucking_map.png
02:29:04 <Bike> They should upload a cleaner version...
02:29:19 <kmc> yeah look at those jpeg artifacts in a png :(
02:29:44 <Bike> it's simplistic enough to be a svg, really
02:29:45 <kmc> i wonder if there is research into fancy ways of de-jpegifying line art
02:30:44 <shachaf> http://rentheatmap.com/sf.html
02:45:50 -!- WeThePeople has joined.
02:48:50 <Lumpio-> One area looked really cheap at first but it turned out to be a lake
02:48:55 <Lumpio-> Wonder if you can rent an apartment on the lake
02:49:37 <shachaf> Nothing wrong with living in lakes.
02:50:29 <kmc> shachaf: http://i.imgur.com/uwtBNUf.png
02:51:24 <shachaf> Hmm, your nick looks weird in uppercase.
03:09:45 <kmc> Lumpio-: haha
03:09:49 <kmc> well http://www.lrb.co.uk/v35/n03/rebecca-solnit/diary does talk about "water plots"
03:12:21 <zzo38> What does the "S.A.C." Fieldata code mean?
03:12:51 <Lumpio-> You sure do ask the most random questions.
03:13:00 <shachaf> fungot's are even more random
03:13:00 <fungot> shachaf: more simply put: siod sucks as a general purpose ( similar, and i'd like to see that mystical forest powers, but this time on the impact of the introduction to theoretical computer, fnord of the fnord here, just above me, asked me to do that in the " better" language
03:13:09 <shachaf> fungto: that's not a question..................................................
03:13:13 <shachaf> fungot: that's not a question..................................................
03:13:14 <fungot> shachaf: agora alice c64 ct darwin discworld europarl ff7 fisher ic irc* jargon lovecraft nethack pa speeches ss wp youtube with this jaybad... he is like the punching him? does it involve? doing it the way you would calibrate a normal. pm showed the bq a parking citation. if confronted, you will say you are only here to deliver a message and then i'll put his shit.... w.t.f lol
03:13:43 <fungot> shachaf: agora alice c64 ct darwin discworld europarl ff7 fisher ic irc* jargon lovecraft nethack pa speeches ss wp youtube* 54, 55, 56, 57, as bryant still had his glory days ahead of them must have brought it with blood is, he's a total of two conversion rates, in that line, that i have gone totally wanting, and then you can choose fnord, but don't
03:13:49 <fungot> shachaf: i just wrote :p ( what was i thinking there.... :d
03:14:12 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot* homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
03:14:53 <fungot> Selected style: irc (IRC logs of freenode/#esoteric, freenode/#scheme and ircnet/#douglasadams)
03:15:34 <shachaf> fungot: riddle me a riddle
03:37:32 -!- jconn has joined.
03:43:29 <shachaf> kmc: did you read that book
03:43:58 <shachaf> either the one i mentioned the other day
03:44:13 <kmc> probably not
03:45:07 -!- WeThePeople has quit (Quit: Leaving).
03:53:26 -!- trout has quit (*.net *.split).
03:53:26 -!- Deewiant has quit (*.net *.split).
03:53:26 -!- Sanky has quit (*.net *.split).
03:57:01 -!- variable has joined.
03:59:59 -!- Deewiant has joined.
04:32:49 -!- TeruFSX has joined.
04:40:30 -!- RodgerTheGreat has quit (Quit: RodgerTheGreat).
05:02:03 -!- icarot has quit (Quit: leeave()).
05:02:17 -!- icarot has joined.
05:07:18 -!- icarot has quit (Ping timeout: 276 seconds).
05:13:57 -!- Bike has quit (Quit: Reconnecting).
05:14:00 -!- Bike_ has joined.
05:15:21 <oerjan> use cat for all your image compression needs
05:16:13 <shachaf> I thought png should be used to compress pictures of cats.
05:18:10 -!- oerjan has quit (Quit: Good night).
05:19:47 -!- TeruFSX has quit (Ping timeout: 260 seconds).
05:19:48 <quintopia> use cats to compress pictures of pngs
05:58:49 <monqy> shachaf was there any context for that
05:59:20 <shachaf> monqy dont cross-post................................................
05:59:35 -!- Frooxius has quit (Ping timeout: 256 seconds).
05:59:51 <monqy> i thought it'd be """""ironic""""""
05:59:54 <shachaf> you might say that im talking about
06:00:14 <zzo38> What are addictive monoids?
06:00:31 <shachaf> monoids that use a + and give you the first hit for free
06:00:33 <monqy> a bad pun, shachaf
06:01:10 <shachaf> monqy: the """""""ironic' part is that i crosspost sometimes??
06:01:17 <shachaf> monqy: also why are you reading #haskell but not talking
06:01:23 -!- Bike_ has changed nick to Bike.
06:01:26 <monqy> sometimes i say something!!!!
06:01:48 <monqy> i think i said something today? maybe??? or was that just yesterday
06:02:14 <shachaf> "we all appreciate your valuble contributions"
06:03:00 <monqy> ah yes today i gave a pointer on "map . map", yesterday I said something about pts, and a while ago it was.....catMaybes?
06:03:20 <lambdabot> (Applicative f, Traversable t1, Traversable t) => (a -> f b) -> t (t1 a) -> f (t (t1 b))
06:03:41 <lambdabot> Monad m => (a -> m b) -> m (m a) -> m b
06:03:46 <lambdabot> (Foldable t1, Foldable t, Monoid m) => (a -> m) -> t (t1 a) -> m
06:04:25 <shachaf> monqy: edwardk has been doing """awful things with these signatures
06:04:46 <monqy> that's pretty awful
06:04:54 <monqy> that's pretty awful too
06:05:05 <shachaf> and making performance changes with no benchmarks
06:05:38 -!- oonbotti has joined.
06:24:31 <shachaf> monqy: would it be better if i quit monoids
06:29:39 <zzo38> CPP is bad for Haskell due to the different comment syntax and the different use of apostrophes, and the different use of a backslash for line endings.
06:41:41 <ion> foldMap all the way down
06:42:26 <shachaf> I kind of wish Foldable required Functor. :-(
06:43:12 <ion> I heard you like maps, so i put a Functor in your Functor so you can fmap while you fmap.
06:46:12 <zzo38> But there might be some data types which can be foldable but not functor? (such as some GADTs)
06:46:42 <shachaf> Hmm, maybe GADTs are a good counterargument?
06:55:07 <zzo38> I think Foldable could be made to be defined in terms of toList; and I think there is a free theorem which makes it equivalent (specifically, that [] is a free monoid, which makes it a backward monoid transformer).
06:58:41 <zzo38> Do you think I am correct?
07:01:16 <elliott> it's not very good to define Foldable in terms of toList
07:01:23 <elliott> since you lose tree structure
07:04:09 <zzo38> But Foldable is forced to lose tree structure anyways due to associativity of monoids.
07:05:46 <elliott> yes, but part of the reason it's so nice is that you can rely on the balancedness of it when folding e.g. a tree
07:05:53 <elliott> if you have, say, a monoid you use for searching for something in a tree
07:06:20 <elliott> then a toList based version will have worse asymptotic performance
07:07:10 <zzo38> Yes, that can be a good reason not to define a Foldale instance in terms of toList. But that doesn't mean it is not mathematically equivalent.
08:02:24 <Sgeo> Maybe the math should be made to include performance complexity things
08:02:53 <Sgeo> 'a':"b" as not equivalent to "a"++"b" despite the same result
08:04:45 <Sgeo> Presumably there should be language support, time and space complexity in the types
08:04:45 <Sgeo> I have no idea how this would work though
08:07:38 <monqy> what are you talking about
08:08:46 <Sgeo> I want to be preventing from writing foo x = length x while accidentally assuming that foo x has O(1) time complexity
08:08:55 <Sgeo> I want the type checker to catch that
08:09:12 <Sgeo> And only successfully type check if I state that it has O(n) time complexity
08:09:23 <Bike> wow that sounds really weird
08:09:29 <Bike> like really incredibly what
08:11:12 <Sgeo> Prevent people from misunderstanding the time and space complexity of the algorithms they use
08:11:29 <Sgeo> So if something needs to run fast, they can be assured that it will run the way they expect
08:11:33 <Bike> haven't you heard cache locality makes O times obsolete
08:12:27 <elliott> Sgeo: you realise "a" ++ x and 'a' : x have the same time complexity right
08:12:33 <elliott> just really tiny different constant factors
08:12:42 <elliott> and optimised away completely by any half-decent compiler
08:12:50 <Bike> yeah what is that even, i'm confused
08:12:57 <Sgeo> elliott, hmm, ok
08:13:15 <Bike> it's frickin haskell, optimize that shit
08:48:57 -!- Bike has quit (Quit: warning).
08:52:34 -!- FreeFull has quit (Quit: Bye).
08:53:18 -!- azaq23 has quit (Quit: Leaving.).
09:34:00 -!- carado has joined.
09:55:04 -!- Sanky has joined.
10:05:23 -!- SimonRC has quit (Ping timeout: 260 seconds).
10:05:39 -!- SimonRC has joined.
10:06:15 -!- zzo38 has quit (Remote host closed the connection).
10:20:03 -!- zzo38 has joined.
11:00:54 <Jafet> size_t strlen(char const*) __attribute__((DUDE_MOVE_THIS_OUT_OF_YOUR_LOOP_HEADER))
11:13:52 <fizzie> __attribute__ ((optimize ("inline-small-functions", "inline-functions"), always_inline, inline_or_i_kill_you)).
11:18:41 -!- zzo38 has quit (Remote host closed the connection).
11:56:28 -!- asiekierka_ has quit (Quit: ZNC - http://znc.sourceforge.net).
11:59:54 -!- asiekierka has joined.
12:02:30 -!- oonbotti has quit (Ping timeout: 240 seconds).
12:13:24 -!- Lumpio- has quit (Ping timeout: 248 seconds).
12:21:29 -!- monqy has quit (Quit: hello).
12:40:10 -!- Lumpio- has joined.
13:13:41 -!- Frooxius has joined.
13:16:23 -!- sebbu has quit (Ping timeout: 245 seconds).
13:39:50 -!- sebbu has joined.
13:39:51 -!- sebbu has quit (Changing host).
13:39:51 -!- sebbu has joined.
13:42:17 -!- copumpkin has quit (Ping timeout: 252 seconds).
13:42:47 -!- copumpkin has joined.
13:47:39 -!- Arc_Koen has joined.
13:56:59 -!- Arc_Koen has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?").
13:57:24 -!- Arc_Koen has joined.
14:01:47 -!- boily has joined.
15:08:10 <Sgeo> So much for not allowing myself to use the computer until noon
15:09:20 <boily> it's not ok, it's 10:08.
15:09:57 <elliott> sorry about your incorrect timezone
15:10:35 <lambdabot> Local time for shachaf is Tue Feb 5 07:10:31 2013
15:10:40 -!- Phantom_Hoover has joined.
15:11:11 <lambdabot> Local time for boily is Tue, 05 Feb 2013 10:11:09 -0500
15:13:42 <boily> my Globolocaltime© is better than yours :p
15:14:00 <shachaf> I get three more hours every day...
15:21:45 <Sgeo> "and then, if you have a class which mixes in these traits, there are special rules which impl wins, depending on the order of the mixins"
15:21:50 <Sgeo> Give me Ada please
15:22:08 * Sgeo retreats back into the safety of a language that's supposed to make no assumptions
15:23:18 * boily pellets Sgeo with tidbits of PHP, just to keep him insane enough
15:23:27 <shachaf> So you like Ada after all?
15:23:45 <Sgeo> I still don't know Ada
15:24:07 <Sgeo> But it gave me a taste of disliking when languages make assumptions and the programmer has to guess at what it will assume
15:24:56 <shachaf> You should learn it a bit.
15:25:09 <shachaf> It's hardly fair to use it as an argument against other languages when you don't know it.
15:29:48 <Sgeo> I know 1<2 and 2>0 or 3<4 is invalid
15:30:25 -!- augur has quit (Remote host closed the connection).
15:30:29 <shachaf> You know how it's case-insensitive?
15:31:06 <Sgeo> Now I do. Makes sense though.
15:38:34 -!- carado has quit (Ping timeout: 246 seconds).
15:56:10 <kmc> "languages make assumptions and the programmer has to guess at what it will assume"
15:56:13 <kmc> isn't that
15:56:14 <kmc> every language
15:56:26 <kmc> except that you shouldn't "guess" you should understand
15:56:39 <kmc> every language but sgeolang?
15:57:14 <shachaf> There is no language but Ada and Sgeo is its prophet.
16:07:33 -!- hagb4rd has joined.
16:09:05 -!- epicmonkey has joined.
16:16:26 <hagb4rd> it even works. they finally managed to put me back online. i didn't even notice i haven't payed payed my electricity bills for months
16:17:50 <hagb4rd> maybe because i always pay once a year. but they seemed to change their policy.. and they have good argues
16:21:32 -!- Taneb has joined.
16:30:37 -!- hagb4rd2 has joined.
16:31:08 -!- hagb4rd has quit (Disconnected by services).
16:31:13 -!- hagb4rd2 has changed nick to hagb4rd.
16:41:43 <Sgeo> "So, unless first-class modules are something you get excited about, or you need Java interop for something, Scala isn't really worth the effort if you're already using Haskell, except as another excuse for broadening your experience of languages."
16:41:54 <Sgeo> First-class modules ARE something I get excited about!
16:42:37 <kmc> unless you need to work with other people or with existing code, Scala isn't worth the effort
16:42:56 -!- sirdancealot has quit (Ping timeout: 255 seconds).
16:43:13 <kmc> but a singular Haskell Genius can replace any code or person in O(1) time so it's fine
16:43:49 -!- sirdancealot has joined.
16:44:18 <boily> kmc: can a haskell genius replace himself?
16:44:54 <kmc> in O(0) time
16:45:14 <Sgeo> Scala's delimited continuation stuff in and of itself is interesting to me
16:45:19 <Sgeo> More-so than a Cont monad
16:46:32 <kmc> i didn't even know it has that
16:46:41 <kmc> it also has subtyping
16:46:54 <kmc> i find subtyping kind of gross, but i don't think it's wrong in principle for someone to find it interesting
16:47:04 <kmc> or "exciting" in words of the OP
16:47:24 <boily> I still haven't grokked the Cont monad. perhaps because I haven't seen it yet used in the wild.
16:48:36 <shachaf> I've been trying to figure out subtyping recently.
16:49:00 <shachaf> I bet it could be really good.
16:49:34 <Sgeo> kmc, I think this is utterly awesome https://github.com/urso/embeddedmonads
16:50:23 -!- augur has joined.
16:56:22 <kmc> shachaf: isn't it great that my org.cups.sid cookie is sent to every other localhost: app i visit?
16:57:03 <Sgeo> How many localhost apps is kmc using?
16:57:40 <shachaf> kmc: Cookies are shared between ports?
16:57:47 <kmc> apparently
16:58:06 <shachaf> I assumed they weren't, like JavaScript cross-domain things aren't.
16:58:18 <shachaf> Are you investigating cupsd?
16:58:18 <kmc> never assume two Web security rules are consistent
16:58:22 <kmc> or one rule between browsers
16:58:33 <kmc> i just noticed that the CUPS cookie keeps getting sent to the Django webapp I'm developing
16:59:03 <shachaf> kmc: On the other hand, most people never visit their localhost cups server.
16:59:32 <shachaf> Which means that when you want to try to CSRF it, you can just connect and have it choose the cookie based on time()!
17:00:06 <shachaf> And also the several other issues I've found.
17:00:34 <shachaf> Hmm, I still have a file with various notes in it.
17:01:14 <kmc> full-disclosure@
17:01:15 -!- AnotherTest has joined.
17:01:17 <kmc> the lazy option
17:11:38 <shachaf> Have you considered changing your name to olster?
17:14:11 <olsner> when I get old I might change it to oldster
17:15:53 <quintopia> you could move to northern ireland and change it to ulster
17:16:20 <olsner> isn't ireland full of IRA and terrorists?
17:17:44 <Taneb> Some of them are also "serious" "politicians"
17:18:04 <hagb4rd> they seem not have enough options for enemies on that island.. so they just get down to kill each other
17:18:50 <hagb4rd> i mean wan't that a conflict between catholics and protestants? (originally)
17:19:14 <hagb4rd> i mean, thank god they found a reason
17:20:13 <hagb4rd> but that's human nature. if there were no problems we would have to invent them
17:20:28 <kmc> i'm not sure to what degree it's actually a religious conflict and to what degree it's a conflict between two communities who happen to have different religions
17:20:49 <kmc> isn't it mostly about whether to be part of the UK or not
17:20:58 <hagb4rd> i guess it's not religious at all
17:21:02 <kmc> this is a common fallacy though
17:21:17 <kmc> americans tend to assume that the arab-israeli conflict is some millenia old religious conflict
17:21:28 <kmc> they don't understand how much of it dates back only to 1948
17:22:17 -!- DHeadshot has joined.
17:22:40 <hagb4rd> they fight for jerusalem since the city was founded
17:22:53 <TwilightSpockle> We decided it would be pretty cool to oust a bunch of people from their homes and move people of a conflicting religion in, then just sort of leave it and laugh from the sidelines.
17:23:04 <kmc> TwilightSpockle: i said "tend to" not "all americans believe this only"
17:23:23 <TwilightSpockle> (Although really we didn't "oust" them so much as they self-ousted when it was obvious that we fully intended to oust them)
17:26:47 <kmc> i disagree that it's a 'conflicting religion' though, except in specific cases like the temple mount
17:26:57 <kmc> jews and muslims live together peacefully in many parts of the world
17:27:02 -!- sebbu2 has joined.
17:27:11 <hagb4rd> yess.. back then, the reasons for the conflict might have been "constructed" and instrumentalized.. but after all that blood was shed, it is a real conflict.. you can't tell who started it, and maybe you're not even interested since your beloved once got killed by a bomb at breakfast on sunday morning
17:27:15 <kmc> places where the communities coexist organically rather than one being installed by force and then subjugating the other
17:27:17 -!- sebbu2 has quit (Changing host).
17:27:17 -!- sebbu2 has joined.
17:27:30 <kmc> hagb4rd: yeah
17:28:02 <olsner> ah, TwilightSpockle is indeed who I thought it was
17:28:08 <kmc> "Don't matter who did what to who at this point. Fact is, we went to war, and now there ain't no going back. I mean, shit, it's what war is, you know? Once you in it, you in it. If it's a lie, then we fight on that lie. But we gotta fight."
17:28:43 -!- sebbu has quit (Ping timeout: 272 seconds).
17:42:20 -!- hogeyui has quit (Ping timeout: 255 seconds).
17:48:19 -!- sebbu2 has changed nick to sebbu.
17:49:07 -!- azaq23 has joined.
17:49:17 -!- azaq23 has quit (Max SendQ exceeded).
17:51:35 -!- azaq23 has joined.
17:54:42 -!- AnotherTest has quit (Quit: Leaving.).
17:54:52 -!- AnotherTest has joined.
17:55:53 <kmc> http://t.co/adQRhvXD
18:06:27 -!- hogeyui has joined.
18:07:45 -!- hagb4rd has quit (Ping timeout: 256 seconds).
18:15:40 -!- Bike has joined.
18:23:21 -!- sirdancealot has quit (Ping timeout: 245 seconds).
18:24:08 -!- ais523 has joined.
18:31:32 -!- Nisstyre-laptop has joined.
18:36:48 -!- sirdancealot has joined.
18:44:53 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds).
18:46:41 <fizzie> kmc: According to W|A, "F/NaN C" is "69 755 K^2/nAn (kelvins squared per nanoan) (with temperatures converted to kelvins)".
18:50:32 <fizzie> (Apparently "an" is a unit of electric current, with 1 An = 0.1602177 A = 0.01602177 emus of current.)
18:51:11 <fizzie> TwilightSpockle: "unit officially deprecated", sadly.
18:51:34 <shachaf> Aren't miles officially deprecated too?
18:51:43 -!- FreeFull has joined.
18:51:57 <TwilightSpockle> They can take our furlongs and fortnights, but how DARE they take our emus.
18:52:44 <shachaf> i think we still have fortnights TwilightSpockle......................................
18:53:12 <kmc> is emu from cgs?
18:53:16 <TwilightSpockle> I think that ellipsis indicates that you paused for a fortnight...
18:53:28 <fizzie> "ElectroMagnetic Unit".
18:53:50 <fizzie> "The EMU unit of current, biot (Bi), also known as abampere or emu current, --"
18:53:52 <kmc> cgs is wacky
18:54:07 <kmc> cgs units have different dimension from SI units for ostensibly the "same quantity"
18:54:53 <kmc> something something permittivity of free space something intro physics pass/fail
18:55:02 -!- Phantom_Hoover has joined.
18:55:44 <fizzie> "Furthermore, within CGS, there are several plausible choices of electromagnetic units, leading to different unit "sub-systems", including Gaussian, "ESU", "EMU", and Heaviside–Lorentz. Among these choices, Gaussian units are the most common today, and in fact the phrase "CGS units" is often used to refer specifically to CGS-Gaussian units."
18:55:51 <fizzie> It sounds all very sensible and wise.
18:56:11 <kmc> yeah i think i mean gaussian cgs
18:56:25 <kmc> whatever Purcell, Electricity and Magnetism uses
18:56:27 <Sgeo> Why do I always use the royal we when describing how my code works, in, say, a comment?
18:56:33 <Phantom_Hoover> I like the fact that measuring the electric constant doesn't count as an experiment any more.
18:56:39 <Sgeo> I am turning into Oleg
18:56:47 <kmc> Sgeo: common practice in academic writing
18:57:14 <fizzie> "We" is somewhat unroyal in any paper with more than one author, though.
18:57:16 -!- stuntaneous has joined.
18:57:36 <kmc> i usually use either the academic 'we' or imperative statements with no subject ('// frob the woznitzes')
18:57:53 <kmc> i'll use 'I' when I'm talking specifically about me, rather than a notional reader or executor of the code
18:58:03 <Phantom_Hoover> what pronoun do a group of royals use to refer to themselves?
18:58:11 <kmc> so things like "// I couldn't make this work" or "// I have no idea why this is so fucked"
18:58:15 <fizzie> Even if there's just one guy doing all the work, and a couple of bureaucratic supervisors tagging their names on it, there's still nominally a group to talk about.
18:59:35 <fizzie> Phantom_Hoover: Maybe they use the "royal I", then.
19:01:44 <Sgeo> oh god why am I watching Star Trek bloopers bloopers ruin the magic
19:01:49 <shachaf> Academic "we" = "the author and the reader"
19:02:31 <shachaf> I guess "we show that ..." doesn't really work with my interpretation. :-(
19:02:34 <Bike> Sgeo: http://www.youtube.com/watch?v=KgspzYMdqdc
19:03:19 <fizzie> shachaf: Arguably the reader is such an integral part of the whole publication process, the use of "we show" is justified, in that the reader is also participating "in spirit". (At least if you really like to argue.)
19:04:05 <fizzie> I can't really find any non-W|A sources for the unit "an", the symbol of which is "An". It's not a terribly googleable word.
19:04:29 <Bike> war comes to long an
19:04:34 <Sgeo> Suddenly I'm nostalging for a location in Active Worlds that I have not managed to find in a long time
19:06:42 <Deewiant> fizzie: If you do find any, let this guy know: http://www.unc.edu/~rowlett/units/index.html
19:06:58 <ais523> hmm… an emu is a dekaämpere?
19:07:50 -!- sebbu has quit (Ping timeout: 255 seconds).
19:09:11 <boily> ais523: did you spontaneously and from your own free will use a diaeresis on an english word?
19:09:40 <TwilightSpockle> boily: You must be new here. (OK, you're not, but the point stands)
19:10:22 <Phantom_Hoover> then i changed computer and i couldn't work out how to set the compose key
19:10:23 <boily> I mean, the only place I see them are in the New Yorker.
19:11:08 <boily> Phantom_Hoover: don't listen to him. setxkbmap ca -variant multix
19:11:56 -!- monqy has joined.
19:12:01 <ais523> boily: it's a meme in this channel
19:12:09 <ais523> to use diareses in every context in which they even remotely fit
19:12:24 <ais523> although sometimes we forget
19:12:32 <boily> äïs523: now I see.
19:12:37 <TwilightSpockle> Y'know you're just using the diaeresis MARK, right? Diaereses are there whether you indicate them or not.
19:13:53 <shachaf> Anyway I use diæreses in English for words like coöperation even outside of this channel.
19:14:24 <TwilightSpockle> I have a sign on my cubicle that reads “Noöne appreciates diaeresis marks.”
19:14:44 <boily> TwilightSpockle: that's a good one.
19:14:51 <ais523> boily: diaeresis marks don't work like that
19:15:07 <ais523> although come to think of it, shouldn't it be diäeresis? or am I pronouncing it all wrong?
19:15:18 <fizzie> ais523: An abampere (i.e. the emu-cgs biot) of current going around in a circle with a radius of one centimetre produces a macnetic field of 2pi oerjans... I mean, oersteds at the center.
19:15:30 <fizzie> There seems to be no shortage of units, fortunately.
19:16:07 <ais523> fizzie: except it doesn't, because in order to get current to flow in a circle you need it to use a superconductor, and those act weirdly wrt magnetism, IIRC
19:16:27 <ais523> the standard experiment is that if you make a coil of superconductor, and put a magnet inside it
19:16:35 <ais523> because if it did, it would generate an infinite amount of heat
19:17:00 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
19:17:03 <TwilightSpockle> ais523: Well there go my plans on destroying the universe :(
19:17:47 <ais523> (in practice, typically it falls but very slowly, as the magnet itself interferes with the superconductivity)
19:18:18 <boily> strangely, I have a hunch that destroying the universe bmayo be easier than hacking egobot.
19:18:38 <boily> argh. stupid attributes. c07test
19:19:21 -!- sebbu has joined.
19:19:33 -!- sebbu has quit (Changing host).
19:19:34 -!- sebbu has joined.
19:21:43 <oklofok> do you know a tv show whose theme goes like 29219...........29219.6.........
19:21:45 <ais523> are you sending nonstandard color codes?
19:22:39 <boily> ais523: they were working an hour ago in another channel, following weechat's doc. no idea why they insert spurious chars here.
19:22:49 <oklofok> one symbol per beat and number n means C*440*2^(n/12) hz
19:22:53 <boily> another test that shouldn't work.
19:22:54 <ais523> was the other channel on another server?
19:23:04 <boily> yes, and it just worked here.
19:23:15 <boily> it is ctrl-c, then c, then a number.
19:24:04 <boily> so, as I was saying, destroying the universe bmayo be easier than hacking egobot.
19:24:29 <fizzie> Deewiant: W|A said 1 An = 0.1602177 A, and CRC Handbook says 1 eV = 0.1602177 aJ, it might be relevant. (Still, the handbook doesn't know anything specific about the an, while it does know about biots and emus of current.)
19:24:47 <boily> quintopia: thanks.
19:25:45 <Sgeo> Is this a bad thing to say to a recruiter
19:25:46 <Sgeo> "I don't entirely understand why the rsum must be a Word document. I wrote my rsum using LaTeX, and it thus only exists as a LaTeX document and as a .pdf. I can attempt to convert it into Word, but am not entirely sure of the purpose."
19:25:52 -!- Nisstyre-laptop has quit (Quit: Leaving).
19:26:22 <boily> Sgeo: some recruiters like word documents because they have automated software to extract practical infos from them.
19:26:48 <Sgeo> Hmm, might a converted document break those?
19:27:25 <fizzie> Deewiant: W|A also thinks you can say "1 au of electric current", to mean 6.62362 mA; the "hydrogen atom ground state current". Something that units page also doesn't list. (Although for that there are a few other hits here and there.)
19:27:32 <boily> probably. I'm an engineer, not a recruiter.
19:34:55 -!- Nisstyre-laptop has joined.
19:35:47 <Sgeo> meh, attached both slopily converted .doc and nicer .pdf
19:35:55 <Sgeo> I'm not especially interested in that job anyway
19:36:17 <Sgeo> Java, PHP, Perl, and Node.js
19:36:28 <Sgeo> Node.js is the only one of those I can see myself tolerating
19:37:05 <shachaf> I bet there are lots of good Ada jobs.
19:37:31 <fizzie> An "Ada job" must be some kind of a slang term for a sex act.
19:37:43 <Sgeo> Searching LinkedIn for Ada gets.... weird results
19:37:48 <Sgeo> Health & Wellness - Return to Work Specialist
19:37:52 <Sgeo> Building Manager at United Palace NY Cultural Arts Center / Church
19:38:08 <Sgeo> There seem to be more Scala jobs than Clojure jobs. I don't know why.
19:38:16 <Sgeo> I guess Scala has easier interop with Java?
19:39:07 <shachaf> It's probably a better language.
19:43:27 <boily> Java is very nice. I was afraid of its ecosystem, and had to learn the language very fast last year for oncoming projects.
19:43:50 <boily> I was reluctant at first, but now it is a very nice asset to know how to code in it.
19:47:18 <boily> hm. admitting you like java in this one channel may have been a major social faux pas.
19:48:54 <Bike> too mainstream?
19:49:05 <kmc> Bike: not pointlessly difficult enough
19:50:17 <Taneb> "Commons backs gay marriage bill" "RBS Investment bank head to quit" "Laundries 'product of harsh Ireland'"
19:50:23 <Taneb> Clearly these headlines are connected
19:52:10 <kmc> Java is okay, its conceptual basis is pretty clean and simple
19:52:59 <kmc> they kind of stopped short of building a full language around those ideas
19:53:16 <kmc> C# is better in this regard
19:55:19 <kmc> i wouldn't call the language "very nice" but i agree that it's an asset to know it
20:04:29 <boily> kmc: nice, in the sense that it is predictable, consistent, and utterly boring. which means that I can get my job done, and not think about it afterwards.
20:04:35 <kmc> yeah exactly
20:04:44 <kmc> this is also why businesses like it
20:05:01 <kmc> for building huge business software, using hundreds of programmers as replaceable parts
20:07:59 <boily> I guess my ideal language, the one that I dream of, would be a cross between java and python: statically typed like the first, and sane like the latter.
20:08:38 <monqy> being statically typed "like java" is pretty far from ideal
20:08:53 <monqy> covariant arrays, anyone
20:09:06 <shachaf> monqy: what bout "like eiffel"
20:09:17 <monqy> i've heard things about that
20:09:26 <monqy> function arguments being covariant right
20:09:45 <kmc> yeah java's type system is pretty weak sauce
20:10:13 <shachaf> 19:58 <edwardk> they treat function arguments as _co_variant
20:10:13 <shachaf> 19:58 <sclv> oh man, dolio's head just about exploded when he learned about that
20:10:16 <shachaf> 19:58 <edwardk> tey don't really grok the whole contravariance thing
20:10:18 <shachaf> 19:58 <shachaf> How does that work?
20:10:21 <shachaf> 19:58 <edwardk> it doesn't
20:10:24 -!- Nisstyre-laptop has quit (Quit: Leaving).
20:10:26 <monqy> yeah that's the quote i heard
20:10:33 <shachaf> oh did elliott paste it to you
20:10:50 <shachaf> you could just join #haskell-lens..........
20:11:00 <monqy> i've been considering it
20:11:42 <shachaf> just don't be too loud or edwardk might give you "commit access"
20:12:44 <shachaf> monqy: did you know #-lens once had "more users than #esoteric"
20:12:45 <boily> kmc: java's staticity is weak, but I prefer a little bit of challenge to plain perfection. besides, with something stronger like haskell's I'd be way too much distracted doing fun stuff instead of working.
20:13:09 <monqy> shachaf: i've heard
20:13:53 <shachaf> I cleared my /ignore list for #haskell and the channel has become markedly worse. :-(
20:16:59 -!- AnotherTest has quit (Quit: Leaving.).
20:19:41 <kmc> how big was it before?
20:20:24 <ais523> "function arguments are contravariant" is pretty much the whole thing that makes variance work
20:21:01 <ais523> actually, Java's problem is more being excessively sane, than insufficiently sane
20:21:09 <ais523> except in the libraries, those are really badly designed in parts
20:21:40 <shachaf> Eiffel allows covariant return and parameter types in overriding methods. This is possible because Eiffel does not require subclasses to be substitutable for superclasses — that is, subclasses are not necessarily subtypes.
20:22:44 <ais523> I can't think of any use for a covariant parameter to a function
20:22:49 <ais523> do you have any examples?
20:23:09 <shachaf> I don't know much about Eiffel.
20:23:13 <shachaf> (That was a Wikipedia quote.)
20:23:22 <shachaf> Apparently they do a lot of runtime checks?
20:26:19 -!- oerjan has joined.
20:43:43 <shachaf> @tell zzo38 CodensityAsk reminds me of type MendlerAlgebra f c = forall a. (a -> c) -> f a -> c (except that it's different)
20:46:17 <Sgeo> Oh god I'm drafting my blog post and it's starting to look like an essay
20:46:27 <Sgeo> It's taking paragraphs for me to get to my point :(
20:47:12 <Taneb> Hey, I wrote an essay about how Ook! is the only half-decent brainfuck derivative, and published it under a nom de somebody else
20:48:42 <kmc> photon hover
20:50:10 <shachaf> monqy: should i know things about galois connections
20:50:11 -!- Vorpal has quit (Quit: ZNC - http://znc.sourceforge.net).
20:50:49 -!- Vorpal has joined.
20:51:11 <monqy> shachaf: i know things about galois connections.........i have my reasons
20:51:19 <shachaf> monqy: what are the things you know
20:51:30 <shachaf> and do you know things about adjunctions too??????
20:51:39 <monqy> forget i said anything
20:51:42 <shachaf> monqy: no don't make it a secret
20:51:44 <monqy> what's a galois connection
20:59:43 -!- oerjan has quit (Quit: leaving).
21:01:29 <Vorpal> shachaf, you don't want to know
21:01:39 <Vorpal> there are some things man were not meant to know
21:02:17 <shachaf> yes i do..........................................................................
21:02:18 <Vorpal> one of those concerns galois connections.
21:02:28 -!- oerjan has joined.
21:02:43 <shachaf> oerjan: Do you logread when you disappear for 3 minutes?
21:03:22 <oerjan> not that i was finished logreading in the first place
21:04:08 <boily> <Vorpal> there are some things man were not meant to know <-- are you implying that shachaf is not a man?
21:04:59 <Vorpal> boily, he is something far more sinister. And man is pretty sinister to begin with.
21:06:00 <boily> I had this hunch that I should be careful around shachaf. he has this malevolent aura around him.
21:08:58 <kmc> sha256chaf
21:18:17 <boily> Ph'nglui mglw'nafh hmac shachaf fhtagn!
21:21:43 <Vorpal> boily, noo, you woke up shachaf!
21:26:02 <boily> no problem for me. he's far and away.
21:29:09 <Sgeo> shachaf will make you learn Ada if you're not careful
21:31:23 <boily> I'm armed with a squeaky rubber chicken. that should be enough.
21:40:25 -!- ogrom has joined.
21:41:13 <oerjan> that's not a chicken that's a shoggoth
21:41:53 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: rot256: not found
21:42:02 <Taneb> (rot256 should totally exist)
21:43:53 -!- WeThePeople has joined.
21:44:37 <shachaf> Sgeo: No, that's just you.
21:45:02 -!- Phantom_Hoover has joined.
21:45:25 <fizzie> I've recently-ish upgraded this desktubuntu from 12.04 to 12.10, and nowadays running pavucontrol seems to constantly eat a few % of CPU, except sometimes it eats a 100 % of CPU too.
21:45:28 -!- monqy has quit (Quit: hello).
21:46:13 <fizzie> It's already installed.
21:46:29 <olsner> if neither of those works, I'm out of tips
21:46:50 <fizzie> Maybe I should split the difference and "ninstall" PulseAudio.
21:47:04 -!- Phantom_Hoover has changed nick to Guest50856.
21:47:52 <olsner> make a script that repeatedly installs and uninstalls pulseaudio, adjust the duty cycle until it works
21:48:43 <oerjan> `ln -s /bin/cat bin/rot256
21:48:46 <HackEgo> ln: invalid option -- ' ' \ Try `ln --help' for more information.
21:48:47 <oerjan> `run ln -s /bin/cat bin/rot256
21:49:13 <oerjan> `run ln -s /bin/echo bin/rot256
21:49:15 <HackEgo> ln: creating symbolic link `bin/rot256': File exists
21:49:26 <boily> why do I hunger for shoggoth'n'chips now...
21:49:31 <oerjan> `run ln -s /bin/echo bin/rot256
21:49:53 -!- hagb4rd has joined.
21:56:43 <oerjan> <fizzie> ais523: An abampere (i.e. the emu-cgs biot) of current going around in a circle with a radius of one centimetre produces a macnetic field of 2pi oerjans... I mean, oersteds at the center. <-- WHEE!
22:00:37 -!- Taneb has quit (Quit: Leaving).
22:03:26 <hagb4rd> so you must have a spin of 1/2
22:04:18 -!- epicmonkey has quit (Ping timeout: 252 seconds).
22:05:13 -!- carado has joined.
22:06:11 * oerjan rotates 360 degrees and suddenly is upside down
22:06:39 <ais523> well, there are 4pi steradians in a sphere
22:06:44 <oerjan> hagb4rd: I CAN AFFERMI THAT
22:20:01 -!- WeThePeople has quit (Quit: Leaving).
22:20:42 -!- ogrom has quit (Ping timeout: 256 seconds).
22:24:14 <oerjan> god dammit my laptop has started making that occasional clicking sound again _and_ the housemate has started creaking with his chair again.
22:26:24 <olsner> maybe your laptop is a brick, you don't have a housemate, and you're imagining things
22:26:55 <ais523> oerjan: try banging on the case near the fan if it's clicking
22:27:00 <oerjan> last time it happened, it helped making sure jqs.exe wasn't running. but it isn't running now.
22:27:01 <ais523> in my case, a clicking sound means the fan has jammed
22:27:32 <ais523> and banging on the case has a 100% reliability on restarting it (typically with a delay of several seconds)
22:27:41 <oerjan> the fan was already running, alas
22:28:54 <ais523> in that case it's probably the hard drive, as the only other moving part
22:29:02 <ais523> and you should take backups ASAP because it may well fail soon
22:29:13 <oerjan> in fact the clicking only happens when the laptop is idle
22:29:15 -!- Zuu has joined.
22:29:23 <kmc> 'phpwm is an Xll window manager using embeded php to manage all events, creating a "scriptable" window manager.'
22:30:30 <kmc> that's x-ray lima lima, not X11
22:36:17 <kmc> Bike: are you just refusing to live in a world where this exists?
22:36:24 <kmc> can't say i blame you
22:36:26 <Bike> yeah, i do that a lot
22:36:38 <Bike> i like how "scriptable" is in scare quotes though
22:36:48 <Bike> "PHPWM User Group (phpwm) on Twitter" oh
22:37:18 <Bike> wait, this is west midlands, not window manager.
22:37:37 <ais523> oh dear, I'm in the west midlands
22:37:44 <Bike> huh, google has two different php-related things before it has the window manager
22:39:16 <olsner> if the window manager is only third worst, I don't want to know what the other two are
22:39:45 -!- sirdancealot has quit (Ping timeout: 256 seconds).
22:39:49 <Bike> a user group and a website manager
22:40:00 <olsner> php has users!? *gasp*
22:40:20 <Bike> there's actually a php user behind ais523 right now DON'T LOOK
22:40:24 <ais523> olsner: it's not surprising that it has users
22:40:28 <ais523> Bike: why can't I look?
22:40:35 <ais523> also there wouldn't be much room for a PHP user behind me
22:40:43 <ais523> my back isn't literally to the wall, but it's quite close to it
22:40:50 <Bike> he's actually outside
22:41:15 <Bike> you can't look because there's a wall there, you can't see through walls
22:56:28 * ais523 notices that Bike assumed it was an exterior wall, even though the sentences he stated don't technically assume that
22:58:20 <Bike> presumably if you go through enough walls you'll get to outside
22:58:36 <Bike> unless you're in some kind of toroidial installation
23:01:36 <Guest50856> Bike, ah, but in that case the floor eventually becomes a wall
23:01:37 <ais523> even if you were in a toroidal installation, you'd still get to outside if you went in a straight line backwards
23:01:47 <ais523> at least because of the curvature of the earth
23:02:38 <Guest50856> maybe if he's on the event horizon of a black hole
23:02:57 <ais523> Guest50856: backwards still exists then
23:02:57 <Bike> i think it's clear enough that i want to follow the floor when i'm talking about a building, and the PHP developers not therein
23:03:02 <ais523> it's just impossible for anything to move in that direction
23:03:19 -!- Guest50856 has quit (Changing host).
23:03:19 -!- Guest50856 has joined.
23:03:26 -!- Guest50856 has changed nick to Phantom_Hoover.
23:03:44 <olsner> hmm, your username is "phantomho"
23:04:08 <Phantom_Hoover> ais523, at the event horizon a straight line tangent to the surface is a circle around the event horizon.
23:04:40 <ais523> Phantom_Hoover: I don't think black holes actually change geometry like that
23:04:51 <ais523> at the event horizon, something going at the speed of light away from the black hole doesn't move at all
23:05:40 <Phantom_Hoover> at the event horizon, a lightlike curve tangent to the horizon is circular
23:05:56 -!- Frooxius has quit (Read error: Connection reset by peer).
23:06:42 <ais523> yeah, the position at which light would orbit the black hole is going to be some way away from the event horizon
23:06:47 <ais523> at the event horizon, it falls in
23:07:22 <Phantom_Hoover> anyway, point is, since something something geodesics the php advocate could well be in front of you
23:08:17 <fungot> Selected style: lovecraft (H. P. Lovecraft's writings)
23:08:32 <oerjan> fungot: what do you have to say about time traveling php advocates?
23:08:32 <fungot> oerjan: the next logical move was to get to the innermost recesses of the olive grove and the fnord of the
23:08:59 <oerjan> fungot: i was thinking maybe something more non-euclidean
23:09:00 <fungot> oerjan: " that refinery, though, it was to discover myself half sucked into a slimy expanse of hellish black fnord which extended about me in monotonous undulations as far as practicable. this, he guessed, was the
23:09:18 <olsner> fungot: what do you think about Xll?
23:09:20 <fungot> olsner: made of the original fragment during the work. the next day i carefully wrapped the green jade amulet and sailed for holland. what mercy i might gain by returning the thing to any ordinary or expected transportation route. his final idea was to have been the neighboring abyss which received the greatest fnord could have reproduced each subtle fnord and mental complexity they would have been the singular cold that aliena
23:09:27 -!- Frooxius has joined.
23:15:02 <shachaf> fungot: say something insightful
23:15:02 <fungot> shachaf: i did not open my eyes until the raucous clamour came loudly from a point as remote as new york, which joined on near 1830. there were hideous screams in the woods. he could turn and move and leap he could he would he would.
23:15:10 <shachaf> fungot: say something witty
23:15:10 <fungot> shachaf: i felt a chill which was not of the dead
23:15:15 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft* nethack pa qwantz sms speeches ss wp youtube
23:17:41 <shachaf> fungot: say something scary
23:17:42 <fungot> shachaf: the ground sloped upward to the east and see the arched flight of steps or its fnord.
23:18:13 <fungot> Selected style: alice (Books by Lewis Carroll)
23:18:24 <shachaf> fungot: say something blasphemous
23:18:24 <fungot> shachaf: " i should think it did go! why, i do believe,' said humpty dumpty, ' and i'm nearly sure they can't see me.
23:18:29 <Bike> fungot, bizarre math joke?
23:18:30 <fungot> Bike: one winter night, at half-past nine, cold, tired, and cross, and muddy, i had gone down upon one knee to help the fnord but much yet remains to be said.
23:18:47 <shachaf> fungot: stop saying fnord...............
23:18:47 <fungot> shachaf: they gazed in delight, while the great fnord of fnord and fnord his eyes. " isn't he kind, mister sir?"
23:19:01 <shachaf> fungot.................................................
23:19:01 <fungot> shachaf: but when she saw him fnord and fnord. and then there's the fnord, as well as fnord reading sauce with fish, or fnord, that alice quite fnord for it hadn't spoken before.
23:19:15 -!- DHeadshot has quit (Read error: Connection reset by peer).
23:19:21 <shachaf> fizzie: Could you disable the fnords?
23:20:07 <fizzie> Well, yes; in fact, I don't think the VariKN-trained models even have them.
23:20:10 -!- DHeadshot has joined.
23:20:37 <fungot> Selected style: iwcs (Irregular Webcomic scripts)
23:20:52 <olsner> pining for the fnords, as it were
23:20:57 <fizzie> fungot: Don't be a webcomic.
23:20:58 <fungot> fizzie: my name is dr. montana, sir.
23:21:06 <shachaf> can we have super mega comics
23:21:09 <fungot> shachaf: on the way, myth, god created the universe, a better universe! we choose only die fittest people of nigeria!
23:21:11 <olsner> fungot: pine for the fnords will you
23:21:11 <fizzie> fungot: /nick dr. montana, sir.
23:21:11 <fungot> olsner: no, the boy, do you?! i'm insulted!
23:21:11 <fungot> fizzie: a bit of a noble way of the most secure and well-guarded buildings in the world, we should at a time like a good spot of tea, then, why?
23:21:30 <fungot> olsner: ha! i fight the hippo, sea dog, and the bug to get the snakes, and 9... we only had ter put down 10%, not nothing. this isn't the time or place, but it has given me a splitting the profits for the nigerian government a charity? people! a wooden look, there's the mighty hunting creature! it's killed kyros and lambert, a travelling merchant, practised with bow, can i do to help, i meant! translation: the doors of the vati
23:22:04 <fizzie> The 'alice' style is perhaps a bit overly prone to fnords due to the large number of unique words.
23:22:49 <fungot> Selected style: speeches (misc. speeches from Project Gutenberg)
23:22:53 <olsner> what's the idea with the fnords?
23:23:13 <shachaf> fungot: can we have some churchill
23:23:16 <fungot> shachaf: that is what it has cost us fnord or worldly fnord is our reward that we can give nothing to the mass, that, if we do repel you by some wrong principle or practice. if we were to have indian budgets and indian estimates, if we pass this bill, and those not a few, or of the greatest names which ever distinguished the history of london, had not yet been taken down. we repaired and embellished the car under the wheels of
23:24:01 <fizzie> olsner: All words with very few occurrences (in this case, occurring only once, I think) have been replaced with a single token "UNK"; it's a reasonably standard thing to do in language models.
23:24:43 <olsner> to make the model smaller and easier to work with?
23:24:45 <fizzie> Of course it really only makes sense if the model is used to compute sentence likelihoods or what-have-you.
23:25:29 <fizzie> Yes, and anyway there wouldn't be enough ngrams where those words occur to get any particularly reasonable statistics for them either.
23:25:59 <fizzie> It's just that in the generative mode, it has to replace that token with *something*.
23:26:17 <Bike> you should use the phrase "hapax legomenon" because that's a good phrase
23:26:54 -!- boily has quit (Quit: Poulet!).
23:27:54 <olsner> `addquote <fungot> but when she saw him fnord and fnord. and then there's the fnord, as well as fnord reading sauce with fish, or fnord, that alice quite fnord for it hadn't spoken before.
23:27:56 <fungot> olsner: " therefore it is, that those motives to which mr mill pronounces to be a murderer merely for his pleasure. the passages in which the manufacturers of england are distinguishing. they hear these men speak broad. their tongue betrays them. their language had needed fnord for a debt, and of all christian sects the roman catholic clergy. had he permitted me to say what thought is meant to answer in favour of bringing about
23:27:58 <HackEgo> 953) <fungot> but when she saw him fnord and fnord. and then there's the fnord, as well as fnord reading sauce with fish, or fnord, that alice quite fnord for it hadn't spoken before.
23:29:26 <fizzie> Given that it's not possible for the model to see any unseen data, by definition, since it's not going to see any data at all, I could just drop everything that mentions the UNK. (Or for the small datasets, perhaps just build full vocabularies, though I suppose that might end up adding to the amount of verbatim quoting, since any occurs-only-once word would automatically force repetition of ...
23:29:32 <fizzie> ... whatever followed it until it drops out of the context.)
23:29:52 <shachaf> fungot: tell me about monoids
23:29:52 <fungot> shachaf: while our historians are practising all the arts of civil life. at the siege of jerusalem on a far shore i smoothed with tender hand, through months of pain, the sleepless bed of hyde:
23:29:59 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches* ss wp youtube
23:30:27 <shachaf> fungot: gimme some meaningless platitudes
23:30:28 <fungot> shachaf: mr. morley: then let me assure him, perhaps the most hardly treated. in november 1660, he was cheered and refreshed, and soothed, and asked him where he was treated excited the murmurs of the sea: who is now a large body of men who honor and appreciate your pilgrim stock as much as the classical feast with which the aristocrats were now treated, and to all the dignity and fnord with mystic awe: the proud sultana of the
23:31:48 * Sgeo wonders if CyberCoders is a bad idea or whether he should continue to try to find jobs through it
23:31:56 <shachaf> fungot: did i talk too much :'(
23:31:58 <fungot> shachaf: memoirs, popularity of, as a lexicographer, been under the necessity of education, not presuming to dictate any plan or system respecting it, i do verily believe i have been the true construction of his answer, if sincere, as frederick of prussia answered, pour fnord fnord de fnord occupy a large space in his works have an individuality of character which seems to have hated democracy only because he had not more time
23:32:09 <olsner> fungot: treat me the murmurs of the sea
23:32:10 <fungot> olsner: " thus much in general as such, shall not govern himself? when the white man: you are certain there is going to be in his place in this assembly. he said that his son couldn't tell a lie! but he never had any chance. it might be said that it was only because events had proved, in the discussion on affairs in fnord who wants to pay more regard to one poor man than to the true statesman. he thinks of the place was open, i
23:32:33 <fizzie> Anyway, I've farmed the model training to varigram_kn these days; it does automatic pruning (and has options for both "drop all words with less than ndrop occurances" as well as "remove n-grams containing OOV words"), so the fnords shouldn't be much of a problem.
23:33:01 <fizzie> I should really take the backoff weights into account some day, though.
23:38:53 <fizzie> Also, I wonder if there are any actually forced (as opposed to just highly probable) loops in any of the styles. From what I recall, e.g. the infamous sword that alone can't stop isn't really an unescapable one.
23:39:20 <fungot> Selected style: ct (Chrono Trigger game script)
23:39:27 <fizzie> fungot: So how about them swords?
23:39:27 <fungot> fizzie: as long as you keep crono in your heart, the day of lavos"... go to " leene square" 1000 a.d.? yes, i'd have done something very brave! he's probably up north, to guardia!!! let's toast our land! now we'll have some peace! magus is a tad on the spooky side. our only hope.
23:39:38 <olsner> fungot: swords swords swords
23:39:38 <fungot> olsner: yes, it's been awhile prometheus! by the way, the wings! now this is a treasure there... found the treasure!! probably filled with mystical forest powers are being drained. why not? then you should leave quickly!
23:39:50 <fizzie> Magus: a tad on the spooky side, yet our only hope.
23:41:10 <kmc> "The toothpaste provided to soldiers at most military bases in America contains extremely high levels of guns-- but little to no actual toothpaste."
23:43:04 -!- WeThePeople has joined.
23:44:48 <shachaf> i heard you didn't actually write phantom-hoover.tumblr.com.....................
23:45:33 <olsner> I think he used a ghostwriter
23:47:26 -!- sebbu has quit (Ping timeout: 272 seconds).