00:05:35 * Sgeo goes to read http://www.reddit.com/r/programming/comments/1ah6ie/algorithmic_trading_for_dummies/
00:08:11 <oerjan> what we need is dummy training for algorithms
00:09:19 <shachaf> kmc: There's going to be a mosh gsoc?
00:11:28 * Sgeo is slightly sad that he never did gsoc
00:11:37 -!- copumpkin has quit (Ping timeout: 252 seconds).
00:12:08 -!- copumpkin has joined.
00:17:42 <Jafet> > let i = 0 :+ 1 :: Complex CReal in (((i**i**i)**i**i)**i**i)**i
00:17:55 <Jafet> > let i = 0 :+ 1 :: Complex CReal in ((i**i**i)**i**i)**i
00:18:08 <Jafet> > let i = 0 :+ 1 :: Complex CReal in (i**i**i)**i
00:18:20 <Jafet> > let i = 0 :+ 1 :: Complex CReal in i**i
00:18:22 <lambdabot> 0.2078795763507619085469556198349787700339 :+ 0.0
00:18:39 <Jafet> (they're all real)
00:23:31 <Bike> hm, that's a bit neat, it's exp(i log i)
00:24:14 -!- sebbu3 has joined.
00:24:45 -!- sebbu3 has quit (Changing host).
00:24:45 -!- sebbu3 has joined.
00:26:36 -!- sebbu has quit (Ping timeout: 260 seconds).
00:33:25 -!- zzo38 has quit (Remote host closed the connection).
00:42:41 -!- monqy has joined.
00:44:24 <lambdabot> monqy: You have 1 new message. '/msg lambdabot @messages' to read it.
00:45:11 <monqy> it's alright? i read it while i'm waiting and i'm in the section about limits which is after the section about adjunctions and before the section about monads. that's where it is.
00:45:46 <shachaf> i sense a "monad tutorial in your future"
00:46:03 <shachaf> you should do exercise ddarius gave me about limits
00:48:00 <monqy> sure but maybe i should wait until im less busy??. ive only been sort of glancing over the exercises and working them a bit in my head since i usually read it when i dont really have time to fully work through them....in this "first pass" im focusing more on getting an intuition over an ability to actually work through things
01:11:31 -!- WeThePeople has joined.
01:14:03 -!- DHeadshot has quit (Read error: Connection reset by peer).
01:14:11 -!- DH____ has joined.
01:16:47 -!- monqy has quit (Quit: hello).
01:18:42 -!- DH____ has quit (Ping timeout: 264 seconds).
01:21:23 -!- DHeadshot has joined.
01:29:02 -!- Arc_Koen has quit (Quit: Arc_Koen).
01:48:03 <shachaf> Old School vs. Non-Coding Hackers
01:48:05 <shachaf> A hacker would not whine about missing code. Hackers see missing code as an opportunity to build. Hackers like to build.
01:48:24 <Bike> stop or i'll link you haskell porn
01:48:28 <shachaf> better call our hacker classifier to classify this one
01:50:09 <Sgeo> Maybe I should learn Frege
01:50:15 <Sgeo> Could introduce it at work, maybe
01:50:33 <Bike> they named a language after frege of all people?
01:51:28 <Sgeo> 'He is considered to be one of the founders of modern logic and made major contributions to the foundations of mathematics.'
01:51:30 <Sgeo> What's so bad about that?
01:51:56 <Bike> mostly he's just old
01:52:21 <Bike> where's my schönfinkel, man
01:52:48 <Bike> also he's kind of an anti-semite, though i guess that doesn't matter for much when you're a hundred years dead, eh
01:55:26 <Bike> would be kinda neat if there was a language named after some nyaya guy
01:56:21 <oerjan> Frege oder nich Frege, das ist die Frage
02:06:56 -!- oerjan has quit (Quit: leaving).
02:07:51 <Sgeo> Are there any good Linux distros for laptops from, say, the mid-to-late 90s?
02:10:22 <pikhq> Here's a nickle, kid. Buy yourself a real computer.
02:12:28 <Sgeo> Should I buy myself a better radio?
02:12:28 <Sgeo> https://lh5.googleusercontent.com/-F_thxN8B8q8/ThDsmMfdNrI/AAAAAAAAABs/oOMaW22O-CM/s537/11+-+1
02:13:20 <Sgeo> https://lh4.googleusercontent.com/-AM3tMmG_l6M/ThDsmRcTTHI/AAAAAAAAABw/X9kvx5G_PqA/s537/11+-+2
02:16:08 -!- carado_ has quit (Ping timeout: 256 seconds).
02:16:59 <Sgeo> http://www.fark.com/comments/7649896/Evil-defense-contractor-develops-graphene-membrane-that-makes-it-vastly-less-expensive-to-filter-water-for-drinking
02:19:16 <Phantom_Hoover> that radio is impressive in its ability to avoid any vestige of retro charm
02:19:42 <Sgeo> You mean it's not the sort of radio you think of when you think of 'old radio'?
02:25:08 <Bike> i have to ask why you'd ask for such a distro, sgeo
02:25:41 <pikhq> Sgeo: Your laptop there is undoubtedly worse than my last three cell phones.
02:26:41 <Bike> my netbook is worse than my phone. it's a bit sad.
02:27:22 <Sgeo> Bike, because it woild be fun to use old equipment like that?
02:28:34 <Sgeo> Why is Linux so much heavier than DOS?
02:29:14 <Lumpio-> I dunno perhaps it's because Linux actually does something
02:29:15 <pikhq> Because DOS proper is basically a BIOS abstraction layer?
02:30:06 <Bike> it's like, if you're going to retrocompute, at least do it with something old enough to be interesting
02:30:23 <pikhq> Or at least obscure enough.
02:30:31 <pikhq> At *least* do OS/2 on that sucker or something.
02:30:41 <Bike> like that PARC laptop i forgot the name of
02:30:41 <Sgeo> Windows 1.0 was supported until 2001 wtf
02:31:05 <pikhq> Running DOS on a mid-90s system is... basically boring.
02:31:06 <Sgeo> I think the laptop has a busted battery anyway :(
02:32:47 <Sgeo> Windows 1.0 used tiled windowing....
02:34:42 <Sgeo> Windows/386 is followed by Windows/286. Who was in charge of naming o.O
02:35:06 <Sgeo> o.O OS/2 is what became NT? I didn't really know what OS/2 was
02:35:43 <coppro> `addquote <Sgeo> o.O OS/2 is what became NT? I didn't really know what OS/2 was
02:35:49 <pikhq> OS/2 didn't really become NT.
02:35:53 <Bike> well, you could also say xenix became windows
02:36:00 <HackEgo> 985) <Sgeo> o.O OS/2 is what became NT? I didn't really know what OS/2 was
02:36:00 <Bike> you'd be mostly wrong but you still said it
02:36:07 <pikhq> OS/2 was supposed to be the successor to old Windows and DOS.
02:36:17 <pikhq> But NT replaced it instead.
02:36:18 <Sgeo> http://www.computerhope.com/history/windows.htm
02:36:26 <Sgeo> 'Following its decision not to develop operating systems cooperatively with IBM, Microsoft changes the name of OS/2 to Windows NT.'
02:36:41 <pikhq> That's very damned false.
02:37:54 -!- DHeadshot has quit (Ping timeout: 264 seconds).
02:48:11 -!- monqy has joined.
02:57:35 -!- hagb4rd has joined.
02:59:39 -!- zzo38 has joined.
03:00:17 <zzo38> I have read somewhere that the "par" operator in linear logic means that you have both, but you are not allowed to use them together. Is this correct?
03:01:06 <Sgeo> http://www.oldsoftware.com/
03:01:15 <Sgeo> This site looks like it's as old as the stuff it's selling
03:08:52 -!- monqy has quit (Quit: hello).
03:12:47 <Phantom_Hoover> i like how the beth numbers and the gimel function are both things related to cardinality despite the fact that beth and gimel look the same
03:13:39 <Sgeo> beth and gimel don't look the same
03:14:15 <shachaf> They don't look the same at all.
03:14:21 <pikhq> Yeah, and 入 and 人 are utterly identical.
03:14:40 <Phantom_Hoover> i was going off http://en.wikipedia.org/wiki/Gimel_function
03:14:52 <Sgeo> Hmm, I guess they could look similar in a fixed-with font, somewhat
03:15:01 <shachaf> Wow, that font is really awful.
03:15:06 <Sgeo> Although I could still easily tell the difference
03:15:48 <Sgeo> Does the standard latex font draw the dot in the beth?
03:17:31 <shachaf> Now, complaining about bet and kaf would be more justified.
03:18:05 <Sgeo> How about yud and English apostrophes
03:18:28 <shachaf> Those are different. י and '
03:18:41 <shachaf> You might also complain about ר ד
03:19:11 <Sgeo> Would not complain about those
03:19:42 <Sgeo> only occurs at the end of words, iircםAlso
03:19:47 -!- WeThePeople has quit (Quit: Leaving).
03:20:08 <shachaf> But ס can also occur at the end of words.
03:20:27 <Sgeo> One's boxy, the other's curvy
03:20:48 <pikhq> Sgeo: Now try distinguishing ユ and コ.
03:20:52 <pikhq> In sloppy handwriting.
03:21:12 <shachaf> pikhq: Easy: The first one looks like a ב and the second one looks like a כ.
03:21:16 <Bike> god why are foreign langauges all so messy *idly scribbles u's and v's all over the place*
03:21:26 <pikhq> shachaf: Smartass.
03:27:44 -!- WeThePeople has joined.
03:38:12 -!- Jafet has quit (Quit: Leaving.).
03:54:08 -!- aloril has quit (Ping timeout: 276 seconds).
03:59:22 <zzo38> I made a programming language where the return values of subroutines can be a number, or they can be one of the three special values: open bus, CIRAM bank 0, CIRAM bank 1.
03:59:29 -!- monqy has joined.
04:01:24 <zzo38> In the use it is for, it can be used, so it is not for general purpose.
04:03:25 -!- sebbu3 has changed nick to sebbu.
04:03:46 -!- sebbu has quit (Read error: Connection reset by peer).
04:04:06 -!- sebbu has joined.
04:04:33 -!- sebbu has quit (Changing host).
04:04:33 -!- sebbu has joined.
04:06:08 -!- aloril has joined.
04:06:21 <zzo38> Do you know of analog Verilog?
04:10:21 -!- hagb4rd has quit (Ping timeout: 256 seconds).
04:11:06 -!- hagb4rd has joined.
04:14:31 <zzo38> Can any C compilers make a optimization for inverted logic?
04:17:24 -!- Phantom_Hoover has quit (Remote host closed the connection).
04:20:04 <zzo38> Is there a fast way to erase a DRAM from software?
04:23:07 -!- edwardk has quit (Ping timeout: 252 seconds).
04:23:47 -!- edwardk has joined.
04:32:35 -!- sebbu has quit (Ping timeout: 255 seconds).
05:14:23 -!- hagb4rd has quit (Ping timeout: 256 seconds).
05:15:40 -!- azaq23 has quit (Quit: Leaving.).
05:39:00 -!- edwardk has quit (Ping timeout: 252 seconds).
05:39:34 -!- edwardk has joined.
06:17:29 <zzo38> Do you like this? http://wiki.nesdev.com/w/index.php/User:Zzo38/Metadata_INI
06:18:25 -!- kallisti has joined.
06:18:25 -!- kallisti has quit (Changing host).
06:18:25 -!- kallisti has joined.
06:18:43 -!- kallisti_ has joined.
06:22:07 -!- Nisstyre-laptop has quit (Ping timeout: 256 seconds).
06:22:41 <oklopol> "<zzo38> Can any C compilers make a optimization for inverted logic?" what do you mean?
06:23:34 -!- sebbu has joined.
06:24:02 -!- sebbu has quit (Changing host).
06:24:02 -!- sebbu has joined.
06:25:22 <zzo38> oklopol: In some cases the boolean logic will be inverted in some circumstances, and depending on the CPU architecture and other things, moving around the places where the logic is inverted might be more efficient.
06:25:50 <kmc> ah yeah, i learned that inverting gates (nand/nor) are faster that non-inverting gates (and/or)
06:26:00 <kmc> and that's probably a simplification
06:26:19 <kmc> zzo38: if you want to bulk-erase memory, then maybe non-temporal store instructions are useful, but I don't know for sure
06:28:42 <oklopol> (i thought this had to do with zzo's earlier bijective haskell thing.)
06:29:57 <zzo38> kmc: Yes, that is one example of what I mean, but that specific example would probably be more useful for hardware programming languages such as Verilog. However, that is not the only example. Even in software, there may be cases based on how the boolean values are used in various numbers and bitwise operations and branch instructions and so on.
06:30:42 <zzo38> oklopol: I did write about the bijective Haskell stuff in ##logic channel, and they say yes it can work, to make the Curry-Howard of the logic with numbers.
06:31:36 <zzo38> So, that is the Curry-Howard of the logic with numbers! Now you know.
06:32:08 <Bike> how is ##logic?
06:32:41 <zzo38> I did nothing other than asking and discussing that question, and only one other person replied, but it helped.
06:33:57 <shachaf> A person who's in this channel, too!
06:34:14 <zzo38> Yes, in that case it is.
06:34:31 <zzo38> They didn't answer on this channel, though.
07:18:02 <fungot> Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
07:24:11 <FreeFull> So a NAND/NOR is both more powerful and faster?
07:29:29 -!- FreeFull has quit (Quit: gotta go).
07:37:23 <kmc> FreeFull may never know if a NAND/NOR is both more powerful and faster
07:50:39 <fizzie> fungot: Is NAND/NOR more powerful and faster?
07:50:39 <fungot> fizzie: now, what if we entered it into sarahbot :) at least relative to his current level
07:51:05 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
07:58:03 -!- DHeadshot has joined.
07:59:43 -!- hagb4rd has joined.
08:36:48 -!- Bike has quit (Ping timeout: 264 seconds).
08:41:27 -!- epicmonkey has joined.
08:43:12 <Sgeo> I'm beginning to learn the hard way that just because someone sounds like they know what they're talking about, doesn't imply that they actually know what they're talking about
08:44:49 <zzo38> It can also be partially.
08:46:36 <Sgeo> A knowledgeable-seeming person on Reddit seems to be convinced that UTM == classical computers
08:46:44 -!- ThatOtherPerson has joined.
08:47:23 <zzo38> Someone made a user page in esolang wiki, which says "My web blog; visit the following website page" but actually it is linked to a Wikipedia article for an airline in Poland. The user page says they live in Austria, though.
08:47:28 <ThatOtherPerson> <Taneb> #esoteric is supposed to be about esoteric programming languages, but is really a couple of dozen people being weird
08:55:06 <Sgeo> "I may have been a bit lax with my terminology."
08:59:18 <HackEgo> bash: slist: command not found
09:00:54 <Sgeo> Someone baleeted it
09:02:28 <HackEgo> accesslog \ a.out \ bin \ brainfuck.fu \ canary \ Category:Self-modifying \ dbg.out \ dkVb20VL \ egobot.tar.xz \ etc \ factor \ foo \ foo.err \ foo.out \ ibin \ index.html \ interps \ karma \ lib \ link \ paste \ prefs \ prefs.bf \ quines \ quotes \ radio.php?out=inline&shuffle=1&limit=1&filter=*MitamineLab* \ sgfoobar \ share \ slist.orig \ slist.
09:02:49 <HackEgo> /bin/ls: cannot access slist: No such file or directory \ /bin/ls: cannot access slist: No such file or directory
09:03:01 <HackEgo> /bin/ls: cannot access slist*: No such file or directory \ /bin/ls: cannot access slist*: No such file or directory
09:05:28 -!- hagb4rd has quit (Quit: hagb4rd).
09:06:01 <fizzie> `run ls slist* # may be more what you want
09:06:16 <fizzie> Good old leftover revertation things.
09:10:11 -!- TodPunk has quit (Read error: Connection reset by peer).
09:10:25 -!- TodPunk has joined.
09:24:27 <zzo38> I have the book "Surely You're Joking Mr. Feynman" and I also have the entire text of the book in my computer. Therefore it is sometimes useful for searching text in it, even though I like to have the book, too.
09:26:34 -!- carado has joined.
09:26:52 <fizzie> A q-cumber is a quantum vegetable.
09:27:24 <zzo38> Other uses of such things is if one page got destroyed, to print out another copy. In this case it is difficult, but with the TeXbook, since the page correspond exactly, it can be done.
09:29:41 -!- edwardk has quit (Quit: Computer has gone to sleep.).
09:33:19 -!- DHeadshot has quit (Ping timeout: 258 seconds).
09:34:06 -!- DHeadshot has joined.
09:36:26 -!- DHeadshot has quit (Read error: Connection reset by peer).
09:37:26 <zzo38> I was playing Franchise Basketball, and on the free agent menu, it lists 1 to 14 on each page, but I pushed view 15, and it says some unusual things; their name is "f", they are 2079 feet and 6 inches tall, played 105 games without ever scoring any points, 15 years old, etc.
09:38:33 <zzo38> When trying 16, I get "vin ringdale", age -60, weight -15351 pounds, and an error message about "GETCOACH 0003".
09:39:20 <fizzie> 2079 feet 6 inches sounds quite high even for a basketballer.
09:39:37 <zzo38> fizzie: Yes, I think so too.
10:01:06 -!- ThatOtherPerson has quit (Quit: Page closed).
10:06:47 <zzo38> I have managed to win over one hundred games in a row.
10:09:21 <zzo38> Maybe the computer opponent is no good.
10:11:19 -!- sebbu has quit (Ping timeout: 256 seconds).
10:20:44 -!- sebbu has joined.
10:21:13 -!- sebbu has quit (Changing host).
10:21:13 -!- sebbu has joined.
10:36:46 -!- oerjan has joined.
10:37:23 -!- Jafet has joined.
11:00:50 <oerjan> <zzo38> I have read somewhere that the "par" operator in linear logic means that you have both, but you are not allowed to use them together. Is this correct? <-- that really sounds more like "with" to me
11:01:15 <oerjan> which is that you must use exactly one of them
11:04:33 -!- Lumpio- has quit (Ping timeout: 258 seconds).
11:30:32 -!- Lumpio- has joined.
11:31:09 -!- Nisstyre-laptop has joined.
11:31:47 -!- WeThePeople has quit (Quit: Leaving).
11:33:00 -!- Nisstyre-laptop has quit (Read error: Connection reset by peer).
11:36:39 -!- impomatic2 has joined.
11:36:53 -!- impomatic has quit (Ping timeout: 256 seconds).
11:43:31 -!- monqy has quit (Quit: hello).
12:09:50 -!- monqy has joined.
12:52:24 -!- jix has quit (Read error: Connection reset by peer).
12:52:44 -!- jix has joined.
13:17:34 -!- Phantom_Hoover has joined.
13:17:40 -!- boily has joined.
13:20:46 -!- ThatOtherPerson has joined.
13:33:48 -!- impomatic2 has left.
13:35:29 -!- metasepia has joined.
13:53:57 -!- ais523 has joined.
13:54:37 -!- sirdancealot has quit (Ping timeout: 245 seconds).
13:55:35 <oerjan> ais523: SPAM SPAM SPAM SPAM
13:56:17 <lambdabot> ais523: You have 1 new message. '/msg lambdabot @messages' to read it.
13:56:23 <lambdabot> elliott said 1d 13h 1m 26s ago: looks like we shouldn't link to sprunge from the wiki; your BF Joust interpreter has disappeared
13:58:19 <ais523> A Letter To Bunny Pattern is pretty awesome, incidentally, despite being spam
14:08:19 -!- ThatOtherPersonY has joined.
14:09:19 -!- ThatOtherPerson has quit (Ping timeout: 256 seconds).
14:09:56 <boily> ais523: when a spam page is deleted, is it deleted forever? I'm curous about this lagopedian pattern.
14:12:02 -!- oerjan has quit (Quit: leaving).
14:15:50 -!- ThatOtherPersonY has changed nick to ThatOtherPerson.
14:19:16 -!- ssue_ has quit (Ping timeout: 258 seconds).
14:20:27 -!- yiyus has quit (Ping timeout: 245 seconds).
14:23:37 -!- ssue_ has joined.
14:25:15 -!- impomatic2 has joined.
14:26:29 -!- sirdancealot has joined.
14:30:45 <ThatOtherPerson> `run echo -e "#/bin/sh\necho \"Cool stuff\"" > coolstuff.sh
14:30:54 <HackEgo> accesslog \ a.out \ bin \ brainfuck.fu \ canary \ Category:Self-modifying \ coolstuff.sh \ dbg.out \ dkVb20VL \ egobot.tar.xz \ etc \ factor \ foo \ foo.err \ foo.out \ ibin \ index.html \ interps \ karma \ lib \ link \ paste \ prefs \ prefs.bf \ quines \ quotes \ radio.php?out=inline&shuffle=1&limit=1&filter=*MitamineLab* \ sgfoobar \ share \ slis
14:31:13 <HackEgo> accesslog \ a.out \ bin \ brainfuck.fu \ canary \ Category:Self-modifying \ coolstuff.sh \ dbg.out \ dkVb20VL \ egobot.tar.xz \ etc \ factor \ foo \ foo.err \ foo.out \ ibin \ index.html \ interps \ karma \ lib \ link \ paste \ prefs \ prefs.bf \ quines \ quotes \ radio.php?out=inline&shuffle=1&limit=1&filter=*MitamineLab* \ sgfoobar \ share \ slis
14:31:34 <HackEgo> bash: coolstuff.sh: command not found
14:32:39 -!- yiyus has joined.
14:34:49 -!- edwardk has joined.
14:35:45 -!- Mathnerd314 has quit (Read error: Connection reset by peer).
14:38:04 <ais523> boily: it's not actually deleted, just hidden
14:38:07 <ais523> we can undo the deletion if necessary
14:38:25 -!- Mathnerd314 has joined.
14:43:38 -!- carado_ has joined.
14:44:17 -!- carado has quit (Quit: Leaving).
14:44:39 -!- edwardk has quit (Quit: Computer has gone to sleep.).
14:44:53 <Sgeo> Fun fact, deleting a specific revision entails deleting the page and undeleting all but one of the revisions
14:45:37 <ais523> we used to have to do that, but there's a better UI on it now
14:45:50 <ais523> and we can also keep the fact the revision existed public, but delete the contents
14:53:08 -!- sirdancealot has quit (Ping timeout: 255 seconds).
14:53:55 <Phantom_Hoover> it also used to be that if anyone made edits to another section of the page after the one you wanted deleted, you had to remove those too
14:54:15 <ais523> Phantom_Hoover: that's still the case, MediaWiki thinks in terms of revisions, not diffs
14:54:27 <ais523> so if you're trying to hide text, you have to hide every revision that includes that text
15:00:53 -!- sirdancealot has joined.
15:36:33 -!- Arc_Koen has joined.
15:41:23 -!- Bike has joined.
15:42:26 -!- FreeFull has joined.
15:56:45 -!- Nisstyre has joined.
15:59:30 -!- Taneb has joined.
16:04:25 <Taneb> We're the most welcoming channel on Freenode!?
16:04:57 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: wehlcome: not found
16:05:05 <HackEgo> Tahnehb: 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.)
16:09:37 <Taneb> Vorpal, I don't like Feedly
16:09:42 <HackEgo> 980) <elliott> also did you learn Haskell yet <Bike> i think i've got most of it now what are these "type" things
16:09:50 <HackEgo> 981) <boily> it's raining in newcastle, therefore the elliotts are distinct. <tswett> boily's Newcastle Theorem.
16:09:56 <HackEgo> 982) <ais523> there's more evidence that scammers exist, than that, say, the average Nigerian exists
16:10:02 <HackEgo> 983) <ais523> is it March 3? <olsner> on some dates, yes
16:10:10 <HackEgo> 984) <zzo38> Two eggs costs less than one egg, but if you buy two eggs, you must eat both. Does linear logic do this?
16:10:11 <Vorpal> Taneb, okay, haven't had much time to use it. Just got home today.
16:10:15 <Vorpal> Taneb, what is the issue with it?
16:10:26 <Taneb> I just find it awkward to use
16:10:44 <Vorpal> I can say I don't like the android client, though it is usable, but the desktop interface (in chrome at least) seems okay
16:10:53 <Vorpal> but yes, google reader is better
16:10:54 <HackEgo> 985) <Sgeo> o.O OS/2 is what became NT? I didn't really know what OS/2 was
16:13:22 -!- Nisstyre has quit (Read error: Connection reset by peer).
16:14:20 -!- nooodl has joined.
16:22:43 <ThatOtherPerson> This is quite a welcoming channel, I was welcomed to it before I even joined
16:23:19 <ais523> `welcome ThatOtherPerson
16:23:26 <HackEgo> ThatOtherPerson: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
16:23:41 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: names: not found
16:24:24 <fungot> Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
16:25:08 <ais523> what is ^list a list of?
16:25:13 <fungot> ThatOtherPerson: now i'll go make a feathejs.git it for
16:27:22 <Jafet> ais523: names listed by ^list
16:27:32 <ais523> I think some of the lists exist purely to confuse people about what they're lists of
16:27:42 <ais523> Jafet: I will laugh so much if that is the actual reason
16:28:04 <monqy> ais523: if only it were so simple. . .
16:28:10 <coppro> if memory serves, ^list is the homestuck list
16:28:26 <coppro> there's ^olist for oots
16:28:28 <coppro> not sure about the others
16:28:30 <ais523> why is one in fungot and the rest in HackEgo anyway?
16:28:31 <fungot> ais523: so is precise garbage collection means it knows exactly what you want to discuss that on ideologies, there are
16:28:48 <monqy> ais523: because there's only one "rest", presumably
16:28:51 <coppro> which I think that fungot has answered admirably
16:28:52 <fungot> coppro: how do you say that 5kz is not correct, but not copyright encumberance)) instead
16:28:57 <HackEgo> /bin/ls: cannot access bin/*list: No such file or directory \ /bin/ls: cannot access bin/*list: No such file or directory
16:29:08 <ais523> oh did someone delete `list sneakily?
16:29:08 <Jafet> `run echo bin/*list*
16:29:11 <HackEgo> bin/list bin/listen bin/lists
16:29:11 <elliott> hm how do globs work again
16:29:18 <HackEgo> tail -n +2 $0 | xargs echo; exit 0 \ list
16:29:24 <ais523> elliott: they work fine but you forgot the `run
16:29:26 <Sgeo> ais523, there was slist on HAckEgo which was the Homestuck list, but someone deleted it
16:29:32 <Sgeo> and ^list was a backup
16:29:43 <ais523> also what does `lists do?
16:29:57 <ais523> I'm trying to work it out
16:30:19 <ais523> oh, I see, it's a really convluted way to write "echo list"
16:30:48 <coppro> theorem: mathematicians need more pronounceable names
16:30:52 -!- Taneb has quit (Ping timeout: 245 seconds).
16:31:06 <ais523> designed so you can easily append to it
16:31:20 -!- Nisstyre has joined.
16:31:25 <monqy> coppro: ????????????????????????????????????????????????????????????
16:31:34 -!- nooodl has quit (Quit: Leaving).
16:32:15 <ais523> coppro: would it be OK to take one mathematician, and give him or her a lot of pronounceable names?
16:33:15 -!- nooodl has joined.
16:34:05 <coppro> monqy: proof: paul erwhat?
16:35:02 <ais523> bleh, now I've forgotten how to type a double acute
16:35:21 <ais523> it's perfectly pronounceable if you know how to pronounce Hungarian
16:35:30 <ais523> which is a very orthogonal and phonetic language
16:35:34 -!- Taneb has joined.
16:35:49 <coppro> yes, but I don't know how!
16:35:53 <monqy> oh i thought coppro was saying there was someone named paul erwhat and i was trying to figure out how thats hard to pronounce
16:35:57 <coppro> I'm a mathematician, not a linguist!
16:36:04 <monqy> but then i got lost in how good of a name erwhat is
16:36:34 <ais523> coppro: hungarian ő is basically like german ö but a bit longer
16:36:47 <ais523> whereas hungarian ö is like german ö but a bit shorter
16:37:02 -!- nooodl has quit (Client Quit).
16:37:06 <coppro> that would be helpful if I could pronounce german
16:37:54 <ais523> the closest sound in English is probably "er" or "ouer", but it's not exactly the same
16:38:25 -!- nooodl has joined.
16:39:25 <monqy> clearly the correct solution is to get on a first name basis
16:42:05 -!- lmt has joined.
16:42:16 <lmt> uhaeheuhuhaheheuaha
16:42:31 -!- lmt has set topic: The most welcoming channel on Freenode. You have been warned. | <Taneb> #esoteric is supposed to be about esoteric programming languages, but is really a couple of dozen people being weird | Newsflash: fungot has been writing spam for money. | Logs: http://codu.org/logs/_esoteric/ | uhueueheuehueahuahuehuaha.
16:42:37 <lmt> huehuehuehuehueha
16:42:41 <HackEgo> Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
16:43:01 <lmt> `welcome huehahauheueha
16:43:05 <HackEgo> huehahauheueha: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
16:43:22 <lmt> elliott: hue hue hue hue
16:43:40 <Bike> always good to hear.
16:43:44 <elliott> do you know words that have letters other than h, u, e or a in them
16:44:08 <Bike> well he said "elliott".
16:44:24 <Bike> wait elliott isn't a "real word" is it
16:44:48 <elliott> well they just copied those from other people
16:44:57 <HackEgo> 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.)
16:44:59 <elliott> you're not allowed to say hello to people
16:45:16 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/bin/hello: Permission denied \ /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: /hackenv/bin/hello: cannot execute: Permission denied
16:45:17 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/bin/hello: Permission denied \ /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: /hackenv/bin/hello: cannot execute: Permission denied
16:45:30 <Bike> `cat bin/hello
16:46:02 <Bike> `chmod +x bin/hello
16:46:07 <lmt> anyone here into minecraft?
16:46:11 <HackEgo> chmod: missing operand after `+x bin/hello' \ Try `chmod --help' for more information.
16:46:18 <ais523> lmt: that's not really what the channel's about
16:46:24 <monqy> lmt: Sgeo runs the minecraft joint around here
16:46:27 <ais523> redstone probably counts as an esolang
16:46:32 <Bike> `run chmod +x bin/hello
16:46:41 <lmt> ais523: have you *seen* the piston turing machine, it is amazing
16:46:47 <ais523> and there was the whole elliottcraft debacle
16:46:54 -!- ThatOtherPerson has set topic: The most welcoming channel on Freenode. You have been warned. | <Taneb> #esoteric is supposed to be about esoteric programming languages, but is really a couple of dozen people being weird | Newsflash: fungot has been writing spam for money. | Logs: http://codu.org/logs/_esoteric/.
16:47:02 <lmt> ais523: this turing machine https://www.youtube.com/watch?v=1X21HQphy6I
16:47:20 <ais523> lmt: it's a youtube link and I don't follow those at work (nor the rest of the time, very often, either)
16:47:28 <lmt> well it's a turing machine
16:47:30 <lmt> with MOVABLE TAPE
16:47:40 <ais523> effectively infinite, until it hits floating point issues?
16:47:42 <lmt> actual tape that actually moves
16:47:55 <lmt> no it's effectively finite because you need a piston once in a while to keep pushing it
16:47:55 <ais523> if it's finite I'm not interested ;)
16:48:02 <lmt> but still it MOVES
16:48:17 <lmt> i don't think i've ever seen that before
16:48:21 <lmt> a physically moving tape
16:48:52 <lmt> i suppose you could make it infinite by generating your world with a script
16:48:54 <ais523> lmt: perhaps you need to play more Rubicon :)
16:48:55 <elliott> lmt: there was a physical TM that did that too
16:48:59 <monqy> i have some tape over to my left but it's too sticky to move
16:49:02 <lmt> elliott: legos?
16:49:07 <elliott> lmt: http://aturingmachine.com/
16:49:19 <ais523> although that was finite due to the finite playfield, I made one with a moving cyclic tag queue
16:49:21 <lmt> that's awesome
16:50:19 <ais523> actually if you're making turing machines in physics simulators, moving the tape tends to be the easiest way
16:50:41 <lmt> as in minecraft
16:50:59 <lmt> btw have we ever resolved that issue
16:51:18 <lmt> of whether an infinite program that itself is generated by a finite algorithm qualifies for anything
16:51:36 <elliott> i think if the program is repeating it's considered ok
16:51:44 <elliott> since you have to do that for e.g. even game of life to qualify
16:51:50 <lmt> as in an infinite minecraft world
16:51:54 <lmt> with infinite tape in it
16:51:58 <lmt> and pistons to push it
16:52:09 <elliott> with minecraft the implementation has limitations that stop it being infinite, but I guess an "idealised Minecraft" with a predefined world algorithm is TC
16:52:13 <ais523> lmt: that's the issue I'm arguing with the wolfram people over
16:52:16 <elliott> also redstone doesn't work in unloaded chunks
16:52:24 <lmt> elliott: yeah
16:52:27 <elliott> so you have to specify that all chunks are simulated all of the time in idealised minecraft
16:52:28 <Bike> ais523: man have they still not published your thing
16:52:39 <lmt> ais523: they gave you the money right?
16:52:40 <Jafet> Conway's life works on a blank universe
16:52:41 <ais523> Bike: it's not that, they've been asking me for a revised version for years
16:52:54 <Jafet> It can (probably) make tape using a universal constructor
16:53:01 <ais523> Bike: and I don't want to just make a few minor corrections to keep their reviewers happy without actually resolve the underlying issue
16:53:08 <Bike> what's the underlying issue
16:53:10 <lmt> redstone breaks in horrible ways in unloaded chunks :(
16:53:32 <ais523> Bike: that I'm using an infinite nonrepeating initialization
16:53:39 <ais523> which is not in general considered OK
16:53:45 <ais523> so I need to explain why it's OK in this specific case
16:53:46 <lmt> i made an awesome music machine but once you walk away from it and come back, it's probably so broken that even a reset button won't fix it
16:53:57 <lmt> so you need to make sure you turn it off every time you walk away from it
16:54:03 <lmt> stupid redstone
16:54:07 <Bike> i thought your paper covered that already
16:54:24 <Bike> you can generate the pattern subTCly or whatever
16:54:29 <lmt> ais523: imo it should be okay if it's generated by a finite algorithm
16:54:30 <elliott> ais523: it isn't actually OK, is it?
16:54:39 <ais523> elliott: well I think it is or can be modified to be
16:54:43 <ais523> but defining that is hard
16:54:45 <lmt> oh and if it can be generated on demand
16:54:45 <elliott> Bike: well you can have two non-conventionally-TC machines that you can hook up to make a TC system, I believe
16:54:53 <lmt> instead of generating it all at once
16:54:55 <ais523> Bike: yeah but that's a really really informal argument
16:55:07 <ais523> I guess the really informal argument is "you can look at how the simulation works and it obviously isn't cheating"
16:55:12 <ais523> but mathematicians don't like that one for some reason
16:55:26 <lmt> as long as they give you money, who cares what they like
16:55:29 <nooodl> wow this http://aturingmachine.com/ thing is extremely cute
16:55:29 <ais523> elliott: I'm not sure; cpressey tried to do that and failed
16:55:32 <Bike> so what would a formal argument be because at some point it just seems like who cares
16:55:33 <lmt> also, wolfram sucks
16:55:45 <ais523> Bike: I'm not sure, why do you think I haven't submitted the revised paper?
16:56:39 <ais523> anyway wolfram pretty much discouraged me from working on it
16:56:42 <Bike> drawing strict boundaries between this-is-the-TC-system and this-is-the-background seems i dunno unworkable
16:56:56 <Bike> it's my understanding that wolfram is a megawanker
16:57:00 <ais523> when he asked me to go meet him in cambridge, and gave me a huge sheaf full of corrections they wanted, and he just basically assumed that there were no mathematical issues without reading it at all
16:57:08 <ais523> also left to eat lunch without offering me any
16:57:12 <ais523> luckily I'd thought to eat before the meeting
16:57:35 <lmt> at least you got to meet.. the great stephen wolfram
16:57:48 <ais523> there's at least one fallacious proof in ANKOS because it makes a similar incorrect assumption (that the details don't matter)
16:57:50 <ais523> but I managed to fix that one
16:58:29 <nooodl> his holiness stephen wolfram"
16:58:48 <monqy> a new kind of idk something
16:58:52 <elliott> pope for stephen wolfram (you guys know stephen wolfram is a position not a person right)
16:58:59 <elliott> monqy: a new kind of theology??
16:59:18 <Bike> oh right didn't we decide on a whatever-you-call-that-thing-you-do-with-Muhammad-peace-be-upon-him foir wolfram
16:59:24 <Bike> stephen wolfram fuck that guy or suchlike
16:59:32 <elliott> to date there have been 34 stephen wolframs
16:59:42 <Jafet> A new kind of stephen
16:59:44 <Bike> fuck that position?
16:59:46 <Bike> fuck that fuck
16:59:52 <ais523> but yeah, if people are interested in this, http://esolangs.org/wiki/1cnis is a start
16:59:53 <elliott> they all get expert makeovers to make them look like the previous one
17:00:05 <elliott> so far all of them have died in office
17:00:13 <Bike> elliott are you writing ankos fanfiction
17:00:21 <ais523> I'd love to prove that a legal initialization pattern for the 2,3 machine can be generated by 1cnis
17:00:25 <ais523> but so far I haven't even managed that
17:00:30 <monqy> i hear fanfiction has a nasty tendency to be erotic
17:00:47 <elliott> well ankos fanfiction can only have stephen wolfram in it i think
17:00:49 <ais523> I'd like to collaborate with #esoteric on this, at least; if you lot think it's important, it's important to me
17:00:52 <elliott> so it'd have to be necrophilia
17:00:58 <ais523> elliott: it can have other people whose job is to be irrelevant
17:01:02 <elliott> and i'm just not sure we can go down that route
17:01:31 <monqy> i'm sure someone would be into it
17:01:34 <lmt> "and now", stephen murmured into my ear as I unbuttoned his shirt, "we shall play... the game of life"
17:02:28 <Jafet> "Today's rule is rule 110"
17:02:37 <elliott> i think wolfram probably hates gol because it isn't one of his
17:02:42 <lmt> rule 34 automaton
17:02:46 <elliott> you could make some really stupid rule 34 joke here but
17:02:55 <lmt> ban elliott
17:02:56 <elliott> guys I hate myself and I hate everyone else
17:03:04 <Bike> i'd love to, ais523, but unfortunately i'm incompetent
17:03:10 <boily> what the wolfram is going on in this channel...
17:03:23 <ais523> boily: reading scrollback may or may not help
17:03:28 <nooodl> "ankos fanfiction" is literally just a setup for that joke
17:03:36 <elliott> boily: we're trying to recover lmt's vowels
17:03:42 <ais523> elliott: what you can do is link to the xkcd comic where he calls rule 34 on wolframs' rule 34
17:03:49 <ais523> and annoy everyone at once
17:03:55 <elliott> well ais523 I see a fundamental problem with that plan
17:03:59 -!- ChanServ has set channel mode: +o lmt.
17:04:00 <elliott> and it's that I'd never be able to sleep at night again
17:04:05 -!- lmt has set channel mode: -b dbelange*!*@*.
17:04:08 -!- lmt has set channel mode: -o lmt.
17:04:10 <monqy> is lmt banning elliott
17:04:19 -!- dbelange has joined.
17:04:19 <Bike> who the hell is dbelange
17:04:30 <Bike> `welcome dbelange
17:04:33 <HackEgo> dbelange: 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.)
17:04:42 <Bike> elliott: if my public service is any indication you can't do that anyway!
17:05:28 <boily> dbelange: do you speak french, by chance?
17:05:50 <dbelange> if that is your idea of esoteric
17:06:08 <boily> *pickled laughter*
17:06:17 <nooodl> oh this reminds me i have a French Question i can't find the answer to
17:06:38 <boily> hm? go ahead, maybe I'll answer. maybe I'll even answer correctly!
17:06:54 <monqy> maybe i'll answer but it sure as heck won't be correct
17:06:55 <nooodl> why's it "par consquent", but "en consquence"
17:07:17 <nooodl> "consquent" is an adjective, isn't it... how does "par" + adjective even work
17:07:20 <elliott> don't trust boily he's from qùébëč
17:07:25 <elliott> unfortunately i could not get compose to put an accent over that q
17:07:30 <tromp__> ais523: is the background pattern an automatic sequence? http://en.wikipedia.org/wiki/Automatic_sequence
17:08:07 <boily> elliott: I modified my layout the other day to be able to get about every possible accent. nothing works on q, to my knowledge :(
17:08:13 <ais523> tromp__: I don't know, but that looks like a pretty promising link
17:08:33 <tromp__> that should qualify it as sufficiently simple
17:08:43 <elliott> boily: do none of the combining characters render properly?
17:08:52 <monqy> http://wiki.answers.com/Q/How_do_you_type_an_accented_q
17:08:54 <Bike> is that a macron, are you a wizard
17:09:18 <boily> elliott: combining chars work properly, but there ain't no precomposed q with nothing that I know of.
17:09:21 <ais523> actually I've been using COMBINING MACRON a lot recently, in quickly typed up notes in gedit
17:09:50 -!- Broly has joined.
17:09:53 <Bike> how do you type a spanish Q?
17:10:18 <Bike> Q is Q in Spanish.
17:10:18 -!- md_5 has quit (Quit: ZNC - http://znc.in).
17:10:22 <ais523> Bike: combining character goes after the character it combines on
17:10:29 -!- nooodl has quit (Remote host closed the connection).
17:10:56 -!- nooodl has joined.
17:10:59 <Broly> i'm not esoteric but i am eccentric
17:11:07 <HackEgo> Broly: 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.)
17:11:19 <Broly> how dare you mention dal net in front of an efnetter
17:11:21 <boily> nooodl: I have no idea about the «conséquent/conséquence» thing.
17:11:26 -!- Broly has left.
17:11:33 <ais523> tromp__: oh, looking at the substitution point of view, I don't think that's enough power
17:11:48 <ais523> Bike: a major IRC server
17:11:48 <monqy> Bike: have you never heard of efnet
17:11:51 <monqy> Bike: have you never heard of dal
17:11:51 <ais523> presumably it's rivals with dalnet
17:11:59 <ais523> or Broly wouldn't have had that reaction
17:12:09 <Bike> rivals. irc rivals? is that a thing
17:12:14 <ais523> now I'm reminded of that bash.org oneliner, "what the fuck is wtf"
17:12:17 -!- Broly has joined.
17:12:17 <monqy> one time ive been on efnet and another time ive been on dal but i didnt actually DO anything there
17:12:22 <ais523> Bike: and it's probably a thing
17:12:32 <Broly> color script bitch ass nagas
17:12:37 <tromp__> even proving that it's not automatic would be a publishable result
17:12:43 <Bike> like the snake things?
17:12:47 <monqy> i hear dal is full of crepes
17:12:48 <Broly> yep the serpents from wow
17:12:54 <elliott> ais523: theyre snake people
17:12:57 <Bike> they're not from wow, they're from mythology :(
17:12:58 <nooodl> oops i stoly monqy's joke
17:12:59 <Broly> dalnet is probably pedo central and color scripts aplenty
17:13:01 <nooodl> do you want it back monqy
17:13:01 <ais523> elliott: yeah I know, from NetHack and the like
17:13:09 <monqy> nooodl: nah it's fine it's pretty bad
17:13:13 <elliott> monqy stole monqy's joke from xkcd :-/
17:13:16 <ais523> I don't think freenode has any major enemies
17:13:17 <Broly> i remember when i was 11 years old i pretended i was some guy from 5ive and would get nudies from random hoes on teenchat
17:13:28 <ais523> I don't really like that side of IRC
17:13:40 <ais523> sensible discussions about programming and the like are so much more interesting
17:13:57 <Broly> there are many more intelligible topics of conversation
17:14:06 <Broly> but they take effort
17:14:07 <Bike> not interested in sick porn, ais? prude
17:14:28 <monqy> porn of terminally ill people
17:14:30 <Broly> i am pretty sure every little 11 year old wanted nudies of hot chicks
17:14:34 <ais523> Bike: there are so many fetishes in existence, that I doubt there's anyone who's interested in /all/ porn
17:14:46 <Broly> not pedo porn, that's gross
17:14:54 <boily> Broly: who are you. you are not me.
17:15:04 <lmt> boily: yes he is
17:15:11 <Bike> boily: it's ok, he has a unique two letter prefix.
17:15:19 <ais523> what's that channel that kicks everyone from the channel simultaneously and sets it to invite only, then invites you?
17:15:20 <Broly> that sounds like a name for a big pet boil someone would keep
17:15:52 <Bike> illegal pornography
17:15:56 <monqy> some efnetter caught wind of something something dalnet
17:16:01 <ais523> ThatOtherPerson: I think we've been invaded
17:16:02 <elliott> ais523: /msg chanserv clear users #channel
17:16:03 <Bike> god knows, thatotherperson
17:16:06 <Broly> he suggested i check out dalnet
17:16:23 <ais523> oh, it's not as dramatic as it used to be
17:16:29 <Bike> robots aren't people broly
17:16:43 <Broly> it looked like an automated msg
17:16:45 -!- DHeadshot has joined.
17:16:49 <Broly> but it doesn't change the fact that it has dalnet in it
17:17:08 <monqy> the one time i connected to efnet and then quit i think the motd had a unicorn in it? that sounds about right.
17:17:10 <Broly> dalnet doesn't deserve to be in any pre-set message
17:17:14 <HackEgo> Broly: 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.)
17:17:16 <HackEgo> BROLY: 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.)
17:17:25 <ais523> hey, how did the color code get into `relcome?
17:17:31 -!- Broly has left.
17:17:39 <Bike> good job, elliott. gjelliott.
17:17:41 <ais523> I didn't even realise the channel was colors-allowed
17:17:58 <ion> `run cat bin/relcome
17:17:59 <Bike> you could +c if you dislike it i guess.
17:18:00 <HackEgo> #!/bin/sh \ welcome "$@" | colorize
17:18:01 <elliott> Bike: i actually apologised to him in /msg first time he quit because i had the feeling there were new levels of stupidity to reach that we deserved to see
17:18:14 <ion> `run welcome | hyphenate.fi | colorize
17:18:17 <HackEgo> Wel-co-me to the in-ter-na-ti-o-nal hub for e-so-te-ric prog-ram-ming lan-gu-a-ge de-sign and dep-lo-y-ment! For mo-re in-for-ma-ti-on, check out our wi-ki: http://e-so-langs.org/wi-ki/Main_Pa-ge. (For the ot-her kind of e-so-te-ri-ca, try #e-so-te-ric on irc.dal.net.)
17:18:22 <elliott> Bike: it was a long hard battle to get -c here don't undo our progress
17:18:29 <ais523> aha, the command I was thinking of was /cs recover
17:18:35 <nooodl> `run welcome | uuu | hyphenate.fi | colorize
17:18:37 <Bike> Oh, are freenode channels +c by default?
17:18:38 <HackEgo> Uuuuuuu uu uuu uuuuuuuuuuuuu uuu uuu uuuuuuuu uuuuuuuuuuu uuuuuuuu uuuuuu uuu uuuuuuuuuu! Uuu uuuu uuuuuuuuuuu, uuuuu uuu uuu uuuu: uuuu://uuuuuuuu.uuu/uuuu/Uuuu_Uuuu. (Uuu uuu uuuuu uuuu uu uuuuuuuuu, uuu #uuuuuuuu uu uuu.uuu.uuu.)
17:18:59 <ais523> "More precisely, everyone will be deopped, limit and clear will be cleared, all bans matching you are removed, a ban exception matching you is added (in case of bans Atheme can't see), the channel is set invite-only and moderated and you are invited."
17:19:03 <Bike> this place is no fun, let's move to dal, they know how to party
17:19:04 <ais523> that's pretty dramatic
17:19:16 <Bike> ais523: the final solution of irc moderation
17:19:17 <elliott> ais523: that sounds like fun, let's try it out
17:19:27 <Jafet> `run welcome | hyphenate | uuu | sed 's/dal/efnet/' | rainwords
17:19:30 <HackEgo> bash: hyphenate: command not found
17:19:35 <Jafet> `run welcome | hyphenate.fi | uuu | sed 's/dal/efnet/' | rainwords
17:19:38 <HackEgo> Uuu-uu-uu uu uuu uu-uuu-uu-uu-u-uuu uuu uuu u-uu-uu-uuu uuuu-uuu-uuuu uuu-uu-u-uu uu-uuuu uuu uuu-uu-u-uuuu! Uuu uu-uu uu-uuu-uu-uu-uu, uuuuu uuu uuu uu-uu: uuuu://u-uu-uuuuu.uuu/uu-uu/Uuuu_Uu-uu. (Uuu uuu uu-uuu uuuu uu u-uu-uu-uu-uu, uuu #u-uu-uu-uuu uu uuu.uuu.uuu.)
17:19:59 <elliott> `run cat `which rainwords`
17:20:00 <HackEgo> #!/usr/bin/python \ import random; w=[l.split() for l in open("/dev/stdin").read().split("\n")]; r=[4,7,8,9,2,13,6]; print "\n".join((lambda s: " ".join(chr(3) + str(r[(i+s)%len(r)]) + l[i] for i in range(len(l))))(random.randrange(0, len(r))) for l in w)
17:20:03 <ion> `run grep bin/*
17:20:06 <HackEgo> Binary file bin/fortune matches \ Binary file bin/tclkit matches
17:20:17 <elliott> what does rainwords do??????????????????????????????
17:20:22 <elliott> `run echo hello tset | rainwords
17:20:46 <fizzie> Also a predefined cycle of.
17:20:47 <monqy> i hear uk has a lot of rain dont you see those things
17:20:49 <elliott> oh i guess rainbows are technically made of rain
17:20:55 <Jafet> `run rainwords </usr/share/dict
17:20:56 <HackEgo> Python error: <stdin> is a directory, cannot continue
17:21:03 <Jafet> `run rainwords </usr/share/dict/words
17:21:05 <HackEgo> bash: /usr/share/dict/words: No such file or directory
17:21:20 <ais523> elliott: no they're made of sunlight
17:21:25 <ais523> the rain is responsible for making them
17:22:07 <Jafet> `fetch en.wikipedia.org/wiki/rainbow
17:22:11 <HackEgo> 2013-03-18 17:22:09 URL:http://en.wikipedia.org/wiki/Rainbow [138931/138931] -> "rainbow" [1]
17:22:17 <Jafet> `rainwords <rainbow
17:22:18 <elliott> ais523: technically they are made of rainbow
17:22:58 <Jafet> `run rainwords <rainbow
17:23:00 <HackEgo> <!DOCTYPE html> \ <html lang="en" dir="ltr" class="client-nojs"> \ <head> \ <title>Rainbow - Wikipedia, the free encyclopedia</title> \ <meta charset="UTF-8" /> \ <meta name="generator" content="MediaWiki .21wmf11" /> \ <link rel="apple-touch-icon" href="//en.wikipedia.org/apple-touch-icon.pn
17:23:12 <ion> `run welcome | hyphenate.fi | perl -CS -Mutf8 -pwe 'y/!-~/！-～/; y/ / /' | colorize
17:23:14 <lambdabot> Gregor: You have 2 new messages. '/msg lambdabot @messages' to read them.
17:23:16 <HackEgo> Ｗｅｌ－ｃｏ－ｍｅ ｔｏ ｔｈｅ ｉｎ－ｔｅｒ－ｎａ－ｔｉ－ｏ－ｎａｌ ｈｕｂ ｆｏｒ ｅ－ｓｏ－ｔｅ－ｒｉｃ ｐｒｏｇ－ｒａｍ－ｍｉｎｇ ｌａｎ－ｇｕ－ａ－ｇｅ ｄｅ－ｓｉｇｎ ａｎｄ ｄｅｐ－ｌｏ－ｙ－ｍｅｎｔ！ Ｆｏｒ ｍ
17:23:50 <ais523> why do we have so many welcome variants anyway?
17:23:53 <lmt> can i use rainwords as a syntax highlighting plugin
17:23:56 <ais523> I don't really understand why they're useful
17:24:00 <fizzie> It seems that colorize thinks of characters, not bytes.
17:24:40 <elliott> ion: we have a script for fulltext welcome
17:24:56 <fizzie> (How come there's no handy "widenize" keyword for that bit of Perl?)
17:25:14 <elliott> ais523: so you don't have to type that perl onelnier
17:25:21 <fizzie> ais523: It's all to make people feel extra-welcome, I think.
17:25:56 <ais523> I guess the topic is at fault
17:26:08 <ais523> elliott: do you still have optbot around?
17:26:12 -!- lmt has set topic: The least welcoming channel on Freenode. You have been warned. | <Taneb> #esoteric is supposed to be about esoteric programming languages, but is really a couple of dozen people being weird | Newsflash: fungot has been writing spam for money. | Logs: http://codu.org/logs/_esoteric/.
17:26:28 <ais523> (today is worse than average in terms of random line quality, so `pastlog will be marginally better than `log)
17:26:55 <HackEgo> 2003-09-10.txt:07:59:59: -!- clog has quit (ended).
17:27:19 <elliott> ais523: I don't have optbot, but it's easy to write it
17:27:21 <ais523> it'll act as an obituary for clog, except that it's still here
17:27:26 <elliott> that's what I usually do when I want optbot back
17:27:29 <HackEgo> 2008-03-06.txt:03:34:35: <oklofok> it's so awesome, you could say "awesome is cise", and not be wrong.
17:27:41 <elliott> last time I revived optbot a few people whineda bout it existing though
17:27:45 <ais523> and even vaguely ontopic
17:27:53 <ais523> elliott: insufficiently many for it to be an official bot?
17:27:59 <ais523> hey quick someone invent a new bf joust strategy
17:28:08 <monqy> istr knowing what optbot was but then i forgot
17:28:19 <elliott> ais523: well a certain few people were vocal about their dislike of its valuable topic-changing service
17:28:28 <elliott> ais523: whereby vocal means, one of them created a bot specifically to revert optbot's topic-changes
17:28:39 <ais523> antioptbot was hilarious
17:28:43 <elliott> n.b. this is really stupid
17:28:56 <monqy> antioptbot sounds cute
17:29:00 <ion> elliott: Yes, but it didn’t have hyphenate.fi in the middle.
17:29:09 <lmt> what's .fi
17:29:24 <ais523> or in this case, more likely language code for finnish
17:29:49 <elliott> http://hypenate.fi/ should be a web interface to hyphenate.fi
17:29:59 <ais523> what is it actually? NXDOMAIN?
17:30:04 <elliott> telnet hyphenate.fi a netcat interface, etc.
17:30:18 <elliott> hyphenate.fi appears to not exist
17:30:26 <Taneb> HackEgo runs on New Zealand English?
17:30:32 <ais523> why do we need a web service for everything anyway?
17:30:41 <ais523> I'd rather have a local executable for hyphenating Finnish, than a webservice
17:30:56 <lmt> i always knew finnish was an esoteric language
17:31:00 <monqy> but then how will you hyphenate finnish on your smartphone
17:31:01 <ais523> Taneb: my theory is that it runs on en_NZ specifically to cause people to ask why it runs on en_NZ
17:31:15 <ion> elliott: That would cost moneys. :-(
17:31:39 <ais523> because I don't think we have any new zealanders here
17:31:54 <ais523> whereas there is one former-regular who was south african
17:32:05 <fizzie> "Because people complained when I set it to zh_TW" is what Gregor last answered when it was asked.
17:32:06 -!- DHeadshot has quit (Ping timeout: 245 seconds).
17:32:30 <lmt> of course they would
17:32:34 <lmt> that's a political statement
17:32:37 <Jafet> `run sed -e '1,/mw-content-text/ d' -e 's/<[^>]*>//g' <rainbow | tr \\n \ | rainwords
17:32:39 <HackEgo> Double rainbow and supernumerary rainbows on the inside of the primary arc. The shadow of the photographer's head on the bottom marks the centre of the rainbow circle (antisolar point). A rainbow is an optical and meteorological phenomenon that is caused by
17:32:52 <Gregor> fizzie: And that is the reason.
17:33:08 <ion> It almost looks like a triple rainbow.
17:33:09 <elliott> hmm, those two magentas are very close
17:33:11 -!- DHeadshot has joined.
17:33:16 <fizzie> Gregor: Well, we'll all form our own opinions on that.
17:33:23 <monqy> the two magentas look pretty different to me
17:33:24 <ais523> oh wait, I know why I'm confused
17:33:26 -!- tos9 has joined.
17:33:29 <ais523> I set my client to filter out colors
17:33:32 <lmt> yes, very different magentas
17:33:37 <ais523> but some of them are showing anyway
17:33:56 <fizzie> ^rainbow2 has a different (longer) set of colors, but it's not any more rainbowy.
17:33:56 <fungot> ████████████████████████████████████████████████████████████ ...too much output!
17:34:23 <fizzie> It also has the greyscale gradient in it.
17:34:30 <lmt> that is not a rainbow.
17:34:59 <fizzie> ^rainbow Everything is a rainbow in your HEART.
17:35:00 <fungot> Everything is a rainbow in your HEART.
17:35:19 <elliott> monqy: they can't look different to you because we use the same irc client and terminal.......
17:35:37 <monqy> elliott: but do we use the same eyes alt. monitor configs alt. colorsets alt.
17:35:42 <ais523> you might have different color profiles on your screens
17:35:52 <elliott> monqy: i've used monqy's eyes since 2012-01-04
17:36:02 <elliott> they're a 2.17x improvement over my previous eyes
17:36:07 -!- edwardk has joined.
17:37:24 -!- md_5 has joined.
17:37:56 <Gregor> (He now has 4.34 eyes)
17:38:49 <ais523> `bfjoust does_this_work_yet (>)*8(>[-])*21
17:38:51 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: bfjoust: not found
17:38:54 <ais523> !bfjoust does_this_work_yet (>)*8(>[-])*21
17:39:08 <EgoBot> Score for ais523_does_this_work_yet: 15.9
17:39:29 <Gregor> Unfortunately, Codu is once again running near its load limit, though I don't precisely know why *sigh*
17:39:34 <ais523> !bfjoust does_this_work_yet (>)*8(>[>(>[-])*21])*21
17:39:37 <EgoBot> Score for ais523_does_this_work_yet: 9.9
17:39:41 <ais523> !bfjoust does_this_work_yet (>)*8(>[(>[-])*21])*21
17:39:45 <EgoBot> Score for ais523_does_this_work_yet: 7.1
17:39:51 <ais523> !bfjoust does_this_work_yet (>)*8(>[(>>>+++++[-])*21])*21
17:39:55 <EgoBot> Score for ais523_does_this_work_yet: 2.9
17:40:02 <ais523> I guess it doesn't work yet
17:40:57 <fizzie> fungot: How many eyes have you got?
17:40:57 <fungot> fizzie: ummm......what were we talking about again? :)
17:41:06 <fizzie> fungot: Eyes. Have you got them?
17:41:06 <fungot> fizzie: in implementations with fast call/ cc actually does have a point
17:41:50 * ais523 is reminded of Subtle Cough
17:42:02 <ais523> IIRC that only had three essentially different programs, didn't it?
17:43:22 -!- Taneb has quit (Ping timeout: 252 seconds).
17:43:44 <fizzie> I'm still leading the "ends a line with ? and a space" statistics: http://sprunge.us/Cfgg
17:44:32 <Gregor> Man, I'm at 3? When the hell did I ever boff up and do that...
17:44:42 <ais523> ooh, I'm not even on the list
17:44:51 <ais523> presumably you mean "? ", not " ?"?
17:45:10 <ais523> enigma has different floors !
17:45:40 <fizzie> ais523: You have a 1 in my logs, if one looks at the full list: http://sprunge.us/eGcV
17:45:43 <ais523> (the game never actually says that)
17:46:08 -!- Taneb has joined.
17:46:14 <monqy> wow i have one???how did that happen
17:46:22 <monqy> and not happen more
17:46:26 <HackEgo> 2012-01-12.txt:12:21:33: <ais523> <fungebob> cpressey: ever consider a timecube esolang?
17:46:39 <ais523> fizzie: I was quoting someone else, that doesn't count
17:47:01 <HackEgo> 2012-12-14.txt:23:05:57: <monqy> % of fungot quotes present in qdb actually verbatim???
17:47:07 -!- AnotherTest has joined.
17:47:40 <lmt> what's the syntax for log
17:48:05 <elliott> in fact, it doesn't actually exist. HackEgo is just a very helpful person
17:48:10 <lmt> ok then i'll just copy it without understanding anything
17:48:14 <lmt> `log <lament>.*\?\ $
17:48:37 <Bike> good implicit denial of chomskyism elliott
17:49:01 <lmt> Pitifully Crappy Regular Expressions
17:49:32 <ais523> it acts like a subset, at least
17:49:34 <Bike> perl incompatible irregular expressions
17:49:56 <ais523> OK, it isn't PCRE, but it's accepted every PCRE-compatible expression I've tried to feed to it
17:50:01 <ais523> although admittedly I haven't tried very many
17:50:09 <Bike> perl compatible regular expression compatible
17:50:38 -!- Mathnerd314 has quit (Read error: Connection reset by peer).
17:51:19 <Gregor> Bike: Perl Compatible Regular Expression Compatible Expression
17:51:35 <ais523> `log ^(.{2,}{2,}$(*COMMIT)(*FAIL)|.)
17:51:36 -!- Mathnerd314 has joined.
17:51:58 <ais523> that expression, if I've translated it correctly from Prolog, should match lines with a prime number of characters
17:51:59 <HackEgo> #!/bin/sh \ cd /var/irclogs/_esoteric \ if [ "$1" ]; then \ grep -P -i -- "$1" ????-??-??.txt | shuf -n 1 \ else \ file=$(shuf -en 1 ????-??-??.txt) \ echo "$file:$(shuf -n 1 $file)" \ fi
17:52:10 <Bike> library that determines if a given library has perl-compatible regexes
17:52:12 <lmt> might as well say perl compatible regular expression compatible regular expression
17:52:34 <ais523> Bike: drop-in replacement for that library
17:52:42 <Bike> what the heck is commit, oh man
17:53:20 <Bike> backtracking verbs. backtracking verbs
17:53:21 <ais523> Bike: it's like an indirect cut in Prolog, makes the entire expression fail if you backtrack into it
17:54:06 <Bike> man at least snobol didn't pretend its matching was regular
17:54:15 <ais523> for smaller cuts, you have (*SKIP) that causes the expression to only be able to match in a different place, and (*PRUNE) that works exactly like a prolog cut
17:54:42 <ais523> and really, the only reason Perl calls them regexes nowadays is that it has to call them /something/
17:54:49 <ais523> elliott: do Perl regexen count as an esolang?
17:54:52 -!- kallisti has quit (Quit: Lost terminal).
17:55:00 <elliott> ais523: if you call them "regexen" then yes.
17:55:10 <monqy> there's that `funny page, about perl isnt there. heh heh.
17:55:11 <ais523> the name determines whether they're an esolang?
17:55:16 <ais523> monqy: perl as a whole isn't
17:55:21 <ais523> but some embedded parts of it feel like it
17:55:24 -!- Taneb has quit (Ping timeout: 264 seconds).
17:55:29 <ais523> just like compile-time C++ is often considered an esolang
17:55:42 <Bike> ais523: you've seen that regex debugger that had a function that wasn't actually a function but instead looked through the file it was compiled in to find its call site and then analyzed the regex it was passed, right
17:55:43 <elliott> monqy: do you mean the esowiki page about perl. I like that one because of oerjan's interpretation
17:55:45 <monqy> http://esolangs.org/wiki/Perl um wiki begs to differ
17:55:46 <Bike> "pretty esoteric"
17:56:12 <Bike> elliott: that's beautiful
17:56:37 <AnotherTest> ais523: compile time C++ as in metatemplate programming and the likes?
17:56:45 <kmc> i guess drawing the line between 'esolang' and 'language well suited to esoteric programming' is like drawing the line between 'functional language' and whatever
17:56:49 <Bike> hm, what would happen if i made a page for C using that example from unix-hater
17:57:01 <kmc> functional langauge with chinese characteristics
17:57:03 <Bike> i'd start a horrible trend of mocking languages based on syntax probably
17:57:15 <AnotherTest> ais523: I guess, although it is used in real code (especially libraries) too
17:57:30 <ais523> AnotherTest: Boost doesn't count, it's probably the world's finest example of serious esoprogramming
17:57:36 <Bike> kmc: i think i need to adopt that idiom
17:57:58 <Bike> i think that boost parser thing, specifically, is the world's finest example of serious esoprogramming
17:57:58 -!- DHeadshot has quit (Read error: Connection reset by peer).
17:58:04 <AnotherTest> Also, there's probably a lot more libraries besides boost and loki
17:58:05 <ais523> Bike: all sorts of parts of Boost are wonderful
17:58:12 <Bike> yes but especially the parser
17:58:12 <ais523> I should probably read its source some day
17:58:22 <Bike> you know the one, the one that makes a parser out of templates
17:58:26 <ais523> comp.lang.c particularly likes the preprocessor library
17:58:26 <Bike> two kilobyte symbol names etc
17:58:32 -!- DHeadshot has joined.
17:58:32 <ais523> partly because that's the only part they're allowed to talk about
17:58:43 <kmc> people tend to judge Boost as a single entity when really it's a huge collection of unrelated libraries by different people
17:58:49 <kmc> many of which are quite simple and sane and essential
17:58:50 <AnotherTest> Great, unless you actually plan on using it (and if you don't have a few hours time to wait until compilation is done)
17:59:00 <Bike> kmc: which is why i'd rather pick out spirit.
17:59:06 <AnotherTest> some people actually wrote compile-time compilers
17:59:19 <ais523> kmc: yes but the insane parts are more fun to talk about
18:00:02 <ais523> I was going to say "why would you make a compile-time compiler anyway"
18:00:05 <ais523> then realised I could think of reasons
18:00:08 <ais523> not /good/ reasons, but…
18:00:28 <Bike> oh i look compile time compilers something something lisp something
18:00:43 <Bike> do compilers count as compile time compilers
18:00:44 <AnotherTest> ais523: one reason would be "because you can"
18:01:00 <AnotherTest> another one would be "because it looks cool"
18:01:09 <Bike> well, i guess JITs and other runtime-accessible compilers don't
18:02:00 <lmt> a compiler that compiles when it does
18:02:02 <AnotherTest> I'd say it is a compiler that compiles its input during compile-time
18:02:19 <ais523> Bike: ThatOtherPerson: my interpretation was "something written in language X, that contains code in language Y, which when compiled compiles language Y into language X and feeds it to the compiler it's being compiled with"
18:02:55 <Bike> ais523: seems pretty usual to me. constant regexes for example
18:03:53 <AnotherTest> Someone should write a compile-time compiler for C++ in C++.
18:04:05 <ais523> Bike: indeed, that's what I meant by saying I could think of reasons
18:04:11 <ais523> but compiling separately seems saner
18:04:19 <AnotherTest> Although, it could be said that that's very easy
18:04:23 <Bike> compiling separately meaning what
18:04:37 <Bike> having a seperate regex->target to go with language->target?
18:05:25 <ais523> and possibly adding the regex→target into the language→target compiler
18:05:32 <Bike> if you compile to language you could use it anywhere language works, though
18:06:19 -!- Taneb has joined.
18:06:54 <AnotherTest> actually who does anyone still /run/ programs written in C++ these days?
18:07:06 <Bike> was that supposed to be "why"?
18:07:14 <Sgeo> What's XChat written in? What are web browsers written in?
18:07:26 <Bike> because programs written in C++ are super common and work, i imagine
18:07:28 <Sgeo> What's KDE and various KDE programs written in?
18:07:47 <boily> fsvo work, but that goes for about everything.
18:07:56 <ais523> AnotherTest: regardless of what OS you're using, you're likely to be using a program written in C++ at the moment
18:07:56 <ThatOtherPerson> AnotherTest: it's going to be a whole lot faster than anything written in Python or Ruby
18:07:59 <Vorpal> AnotherTest, I very much doubt your computer is completely devoid of machine code that was compiled from C++
18:08:04 <ais523> (the chances are lowest on OS X, out of the widely used OSes)
18:08:07 <Bike> AnotherTest: most windows machines don't have C++ compilers, i don't think
18:08:21 <ais523> Bike: indeed, but most windows programs use precompiled executables
18:08:36 <Vorpal> ais523, on OS X iirc the driver framework in the kernel uses C++
18:08:43 -!- sebbu has quit (Read error: Connection reset by peer).
18:08:43 <Bike> i'm sorry, kmc. we have to establish our programming cred by disparaging languages we know not much about with no reference to specific implementations. it's a fact of life
18:08:50 <lmt> pretty much everything in OS X is written in "Objective C++"
18:08:55 <lmt> which is a superset of C++
18:08:59 <ais523> Bike: don't worry, C++ is a language you're allowed to dislike
18:09:05 <ais523> lmt: transparent trolling attempt?
18:09:11 <Vorpal> lmt, objective C afaik
18:09:11 -!- DHeadshot has quit (Ping timeout: 245 seconds).
18:09:18 <kmc> Objective C++ is a real thing
18:09:23 <kmc> kind of a monstrosity
18:09:23 <ais523> (actually objective C++ actually exists, but objective C is more common)
18:09:26 -!- sebbu has joined.
18:09:33 <ais523> kmc: I think gcc can compile it
18:09:37 <lmt> objective C++ is extremely common and commonly used by apple
18:09:38 <Vorpal> kmc, true, but it isn't used much on OS X afaik
18:09:46 <kmc> it has the ObjC object model and the C++ object model living together in sin
18:09:53 <Sgeo> How about Subjective C? Subjects instead of objects?
18:09:54 -!- sebbu has quit (Changing host).
18:09:55 -!- sebbu has joined.
18:10:07 -!- epicmonkey has quit (Ping timeout: 264 seconds).
18:10:08 <kmc> Sgeo: if you google you can find a semi-funny parody about Objectivist C
18:10:10 <Bike> puns about "subjective" and "objectivism" hurt man
18:10:19 <Gregor> If an object model lies with another object model as it does with a paradigm, it shall be stoned to death.
18:10:32 <lmt> objective c++ is pretty bad
18:10:38 <lmt> but it allows you to use c++ libraries
18:10:47 <Bike> kmc: i can't imagine C++ objects and objc objects being very uh, workable together, at all
18:10:50 <lmt> and everything is written in c++, so
18:10:56 <ais523> hmm… most C++ libraries are for games programming
18:11:00 <ThatOtherPerson> Oh, it already exists http://en.wikipedia.org/wiki/Subject-oriented_programming
18:11:01 <Gregor> Vorpal: What, you don't like Bible jokes? X-D
18:11:06 <monqy> ooh ooh i have another pun how about a langauge called.......c(incomprehensible squiggles)
18:11:14 <kmc> somebody pointed out that ObjC and C++ are a good demonstration of why syntax isn't everything... C++ is closer to C in syntax but is a beast of a language, 20 times more complicated than C
18:11:19 <ais523> interesting fact about Windows: the vast majority of native-code Windows programs are written in C++, but the system libraries they link against are entirely in C
18:11:28 <AnotherTest> ThatOtherPerson: wait, is this real or did someone just hack wikipedia
18:11:28 <kmc> ObjC has syntax that looks weird to a C programmer but is really a pretty modest extension to the language
18:11:36 <Vorpal> Gregor, I don't think that was the issue in this case. It was just cringe-worthy...
18:11:37 <lmt> Bike: they are just as workable as any kind of incompatible types
18:11:58 <ais523> AnotherTest: anyone (unless they don't have internet access or have been blocked) can create a page on Wikipedia, no hacking required
18:12:07 <ais523> (or do have internet access but can't use a computer for some reason)
18:12:21 <ais523> (or unless they have sufficiently short time to live that they die before they can finish creating the page)
18:12:25 <ais523> (or probably other things I haven't thought of)
18:12:27 <ThatOtherPerson> AnotherTest: if it's in Wikipedia it's real. You can use this fact to modify the universe around you
18:12:28 <Bike> « As illustrated in that paper, an analogy is made with the contrast between the philosophical views of Plato and Kant with respect to the characteristics of “real” objects, but applied to software ones. » programmers you need to stop doing this
18:12:46 <Bike> actually if everyone could just stop making kant agree with them that woud be great
18:12:53 <lmt> is object-oriented programming phallocentric
18:12:56 <Bike> this isn't the nineteenth century and you're not a philosopher
18:13:01 <ais523> ThatOtherPerson: that is actually sadly truer than it might look
18:13:02 <elliott> Bike: I'd like to stop doing that, but I Kant
18:13:10 <kmc> my CS degree automatically makes me an expert in philosophy, medicine, and law
18:13:16 <kmc> because those are for dumb people
18:13:19 <ais523> there are definitely nontrivial ways in which you can modify the universe via editing wikipedia
18:14:14 <Bike> i mean kay was definitely an avid plato reader when he came up with smalltalk but you can't just pretend simula and biology weren't super important
18:14:16 <Vorpal> ais523, hm. Would anyone sane use wikipedia for something important without doing further research?
18:14:27 <Bike> whoa kmc i thought you were serious for a second!!
18:14:31 <kmc> a lot of people do Vorpal
18:14:35 <Sgeo> Do sane people actually exist?
18:14:36 <Vorpal> kmc, I would argue that doing that makes you insane.
18:14:37 <AnotherTest> If there's zen programming, there should be like philosophic programming?
18:14:42 <kmc> Bike: i've realized not everyone knows when i'm kidding
18:14:43 <Vorpal> kmc, thus meaning that they are not sane
18:14:45 <Bike> zen programming, what the fuck is that
18:14:49 <ais523> Vorpal: perhaps; I needed an algorithm for topological sort
18:14:54 <Vorpal> Sgeo, interesting question
18:14:56 <Bike> also "sane" and "insane" are kind of dumb terms
18:15:04 <Bike> neither is even used medically any more!
18:15:05 <kmc> I think for a modest fee you could pay people to listen to everything you say (via hidden microphone / smartphone) and edit Wikipedia in real time so that you're always correct about stuff
18:15:10 <ais523> so I got one off Wikipedia (the algorithm, that is; I implemented the code myself), and it appears to work, and is currently in production code
18:15:13 <kmc> this would be a great trick
18:15:21 <AnotherTest> Bike: I have absolutely no idea what that is
18:15:22 <Vorpal> ais523, was it for use in something important? Or just a hobby project?
18:15:24 <ais523> kmc: you don't need money, just fame
18:15:32 -!- Taneb has quit (Ping timeout: 272 seconds).
18:15:34 <ais523> anything that stephen colbert says, for instance
18:15:45 <Bike> kmc: a recent boondocks episode had a judge check wikipedia to see if obama had legalized weed. good stuff
18:15:55 <ais523> although Wikipedia editors have taken to also listening to everything he says so that they can revert attempts by people to add it to articles
18:16:04 <Bike> "hm, doesn't look like he has. oh but several states..."
18:16:20 <ais523> Vorpal: it's in the Verity compiler
18:16:48 <ais523> although I may need to change it for computational order reasons, it turns out that linked lists of pairs are not the most efficient representations of directed possibly-acyclic graphs
18:17:01 <Vorpal> ais523, I wouldn't have done that myself. I would probably have started at wikipedia, but then gone on to check the sources it cited.
18:17:17 <kmc> Bike: that show still exists? woah
18:17:19 <Bike> you need cites for some topological sort pseudocode?
18:17:24 <Bike> kmc: i know, i was surprised too, but there it was
18:17:35 <Bike> episode was about Grandpa smoking weed. good episode
18:17:40 <ais523> Vorpal: yeah but checking to see if the algorithm looks reasonable, and the page hasn't been edited in ages, and there's no clear sign of vandalism
18:17:47 <Vorpal> ais523, that might work
18:17:53 <Vorpal> Bike, If I was going to use it for something in research or production yes
18:18:13 <Bike> i am sloppier than you then, i'd just implement it and check to see that it worked
18:18:33 <Bike> clearly this is why i'm unemployed
18:18:46 <Vorpal> Bike, well, I would be worried about there being some fault that only showed up in some edge cases, if the algorithm was non-trivial at least.
18:18:56 <ais523> Vorpal: it's also quite hard (although not impossible) to hide a logic bomb in a mathematical description of an algorithm
18:19:07 <Bike> i'd be worried about that from the cited pages too
18:19:16 <AnotherTest> Is Internet Relay Programming like human-objected programing?
18:19:18 <ais523> it's happened before now (most famously an accidental self-inflicted one on Knuth), but it's rare
18:19:37 <Bike> i mean we all know the story about that C pearls sort or w/e that had a wraparound error
18:19:53 <Vorpal> Bike, I don't know that story
18:20:08 <ais523> I also don't know that story
18:20:29 <ais523> the one I was thinking about was about concurrent garbage collection, which is a field inherently prone to bizarre edge cases and even corner cases
18:20:40 <kmc> many binary search implementations in C are broken because (left + right) / 2 can overflow
18:20:43 <Bike> basically a commonly repeated binary search algorithm had the usual (hi + lo) /2 to find the median
18:20:49 <Bike> http://googleresearch.blogspot.com/2006/06/extra-extra-read-all-about-it-nearly.html
18:21:11 <kmc> fucking C programmers
18:21:14 <Vorpal> yes, that would be one thing that would worry me
18:21:56 <ais523> Vorpal: that one's at least as likely to come up in a binary search written from scratch
18:22:02 * Bike realizes he made the same error in some quick code he spit out, though it would fall back to bignums so it's just a speed problem
18:22:09 <ais523> and even more likely to come up in academic paper implementations, which don't have to worry about overflow
18:22:13 <kmc> one of Ksplice's upstream Linux contributions was finding the dozens of binary search implementations in Linux, most of which were broken in this way, and replacing them with a library
18:22:32 <Bike> kmc: whoa whoa whoa, code reuse? in real life?
18:22:34 <Vorpal> ais523, which is why you should probably use your standard library if possible
18:22:36 <ais523> kmc: arguably if you have equal-sized ints and pointers, and you're sorting things larger than a single byte
18:22:43 <Phantom_Hoover> doesn't division work like normal in modular arithmetic
18:22:49 <Vorpal> ais523, that is probably the safest bet. Either that or a dedicated math library
18:22:51 <ais523> the overflow can't happen due to the numbers not working large enough
18:22:53 <Bike> what does "like normal" mean
18:22:57 <ais523> addition subtraction multiplication do
18:22:57 <kmc> ais523: maybe you binary search indices rather than pointers
18:23:02 <Bike> Phantom_Hoover: the problem is the intermediate hi+lo
18:23:03 -!- edwardk has quit (Ping timeout: 245 seconds).
18:23:20 <ais523> Bike: well this isn't even a pedantic "signed overflow is undefined" thing
18:23:24 <ais523> it breaks even on two's complement
18:23:42 <ais523> (btw, what's the most commonly used system that isn't two's complement?)
18:23:48 <olsner> everyone knows overflow doesn't actually ever happen
18:23:50 <Bike> sign and magnitude?
18:24:01 <ais523> no, I meant computer system
18:24:03 <Vorpal> ais523, sign bit, in floats
18:24:11 <Bike> sign and magnitude is a computer system...
18:24:15 <ais523> but two's complement doesn't work with floats
18:24:20 <Bike> i don't think any recent architectures use it of course
18:24:24 <ais523> Bike: well, processor + accompanying electronics
18:24:25 <Vorpal> ais523, you didn't specify integers only
18:24:42 <ais523> architecture is probably the word I was looking for
18:24:45 <Bike> ais523: i'm mostly sure that some old architectures used sign+magnitude for integers
18:25:21 <ais523> elliott: ooh, Verity reached a milestone recently: it got sufficiently good at interacting with other people's systems that it had its first endianness issue
18:25:22 <Bike> i want to say VAX but i say VAX for everything
18:25:35 <ais523> I feel that this is something of a milestone
18:25:36 <Vorpal> ais523, everything modern uses two complement for integer arithmetics afaik. And unsigned division/multiplication of course.
18:25:41 <fizzie> I don't know about relative magnitudes (pun not intended), but it's a well-known fact that sign-and-magnitude, one's-complement and two's-complement are the "big three".
18:25:51 -!- edwardk has joined.
18:25:56 <ais523> most research languages never get to have endianness issues because they handle all the representations-of-integers themselves
18:25:57 <Bike> older than i thought
18:26:20 <Bike> "This numeric representation system was common in older computers; the PDP-1, CDC 160A and UNIVAC 1100/2200 series, among many others, used ones'-complement arithmetic." okay real old here
18:26:37 <Vorpal> ais523, you might run into bit endianness in verity I presume?
18:26:47 <Bike> we should all use log log representation of everything
18:26:48 <Vorpal> unless I'm thinking of another language
18:26:53 <ais523> fun fact: one's complement is basically as easy to implement in hardware as two's complement, you feed the carry output from the last adder into the carry input of the first
18:26:58 <lmt> sometimes i interview co-op students and ask them what is the advantage of two's complement
18:27:14 <ais523> Vorpal: only possibly in drivers
18:27:18 <ais523> this was a byte endianness issue
18:27:20 <Bike> «Google's Protocol Buffers "zig-zag encoding" is a system similar to sign-and-magnitude, but uses the least significant bit to represent the sign and has a single representation of zero. This has the advantage to make variable-length quantity encoding efficient with signed integers.» ooh, neat
18:27:24 <ais523> lmt: "co-op students"?
18:27:32 <lmt> co-op students
18:27:34 <ais523> luckily I don't have to write the drivers any more
18:27:35 <fizzie> ais523: But you have a more difficult equality comparison, thanks to -0.
18:27:38 <ais523> lmt: as in, what are those?
18:27:38 <Vorpal> ais523, wasn't verity the hardware compiling language?
18:27:50 <lmt> they are students who get temporary internships at companies
18:28:00 <lmt> which somehow count towards their CS degree
18:28:04 <Vorpal> ais523, surely you can run into bit-endianness talking to native vhdl or verilog components then?
18:28:18 <ais523> Vorpal: no because VHDL and Verilog both abstract away bit endianness
18:28:30 <ais523> actually technically speaking VHDL doesn't, but indexing an array forwards will get you lynched
18:28:48 <Vorpal> ais523, I thought you could set whichever direction you wanted in VHDL at least
18:28:54 <ais523> writing std_logic_vector(3 downto 0) is fine
18:28:56 <fizzie> ais523: There's at least one balanced ternary computer too.
18:29:07 <ais523> writing std_logic_vector(0 to 3) will cause everyone to hate you and refuse to work with your code
18:29:21 <ais523> also I think you get at least a warning if you mix endiannesses
18:29:24 <Vorpal> ais523, does that change the endianness? I forgot heh
18:29:34 <ais523> or to put it another way, VHDL does have both endiannesses but one of them won and nobody uses the other one
18:29:57 <Vorpal> ais523, what if you want to change the endianness in order to talk to some external hardware though?
18:30:41 <ais523> Vorpal: you do that by redefining what internal signal is attached to what pin
18:30:58 <Vorpal> ais523, well, for a serial protocol maybe?
18:31:03 <ais523> which needn't necessarily be in any particular order, because the order your pins go in depends on what leads to the most convenient wiring
18:31:05 <Vorpal> I guess that is just over time though
18:31:13 <ais523> for a serial protocol it depends on how you write your modem
18:31:28 <Vorpal> Is modem really the right word here?
18:31:29 <zzo38> ais523: Hopefully, you know some things of Verilog like what I was asking of?
18:31:57 <ais523> zzo38: can you repeat the question? I might know the answer but didn't see the question
18:32:01 <fizzie> I don't think it counts as a modem if there's no MODulation going on.
18:32:07 <ais523> Vorpal: would "codec" be better?
18:32:20 <ais523> fizzie: there is modulation going on, though
18:33:34 <zzo38> ais523: I have various questions, actually. One is, can the I/O ports of a module be vectors that are split across different parts of the I/O list, rather than necessarily being all together?
18:33:35 <fizzie> ais523: Possibly it's an endec. (I think that's a wider concept?)
18:33:53 <Vorpal> ais523, By that definition almost all electronic is a modem
18:33:54 <ais523> encoder/decoder, rather than coder/decoder?
18:34:37 <fizzie> ais523: Yes, and to differentiate from the compressor/decompressor alt of codec.
18:35:38 <fizzie> Wikipedia page on endec suggests that endecs are hardware while codecs are software, but given the page, it's probably just some random jerk's opinion.
18:40:54 <kmc> never heard of endecs before today
18:40:56 <zzo38> ais523: I do have other questions too, such as do you know about Verilog with analog signals too?
18:41:23 <Vorpal> I wonder why my router things that one of my computer's names is in upper case
18:41:25 <ais523> zzo38: Verilog synthesizers don't handle analog signals, and the language does not have very good support for them
18:41:48 <ais523> also, the I/O ports of a module all have to be declared in the same place, but there doesn't have to be any other relation between them
18:41:55 <Vorpal> <kmc> never heard of endecs before today <-- same
18:42:40 <Vorpal> I don't think VHDL does analog signals either?
18:42:54 <Vorpal> Maybe simulation only thingy for logging or something?
18:43:39 <zzo38> ais523: Yes I know that they are declared in the same place, I mean can I declare PRG_A[11:0] and then some other I/O and then PRG_A[12:14] after that (the different direction is deliberate)?
18:43:51 -!- Taneb has joined.
18:44:13 <ais523> zzo38: no, you can't do that, you're declaring variables and types for them
18:44:31 <ais523> what you can do, and what I often do, is just assign the publicly visible ports to internal variables at the start of your code
18:44:35 <zzo38> Actually I mean inside of the I/O list for the module, rather than the declaration for the types.
18:44:56 <Vorpal> zzo38, try it and see what happens?
18:45:01 <zzo38> ais523: O, OK, if that is how it has to use?
18:45:11 <ais523> zzo38: that's the simplest way I know of
18:45:12 -!- ThatOtherPerson has quit (Quit: Leaving).
18:45:56 <zzo38> ais523: Why didn't they make it allowed?
18:46:22 <ais523> zzo38: because it's a programming language, you declare variables and types for them
18:46:35 <ais523> you don't declare half a variable, then the type of that half, and later on decide to add more on to the same variable
18:46:46 <zzo38> But there is a I/O list for the module, isn't it?
18:47:11 <ais523> yes, but that's still a list of parameters and types for them
18:47:18 <zzo38> I agree it doesn't make sense when declaring the type and variable generally, but in the I/O list it seem it should be allowed?
18:47:22 <ais523> in C, you don't do int foo(int a, short b, actually a is a long)
18:47:37 <ais523> you can argue that Verilog should be designed less like a programming language
18:47:48 <ais523> but it was designed as a programming language, and so follows similar conventions
18:49:15 <zzo38> ais523: Of course in C there is no such thing; it would mess up the calling conventions.
18:49:32 <ais523> zzo38: Verilog also has calling conventions, it couldn't be compiled or simulated otherwise
18:50:02 <zzo38> But aren't the Verilog calling conventions just the orders of the bits, or did I do something wrong?
18:50:15 <fizzie> Does MIT still have all their DNS records in UPPER CASE?
18:50:31 <fizzie> "169.22.9.18.in-addr.arpa domain name pointer WWW.MIT.EDU" APPARENTLY
18:50:59 <ais523> zzo38: well, they have inputs and outputs and inouts, that's also important
18:51:10 <ais523> and support for types other than bits at least in simulatoin
18:51:24 <ais523> in at least VHDL simulations, and probably Verilog too, you can even do string handling if you feel like it
18:52:35 <boily> fizzie: MIT has the whole 18.0.0.0/8 at their disposition.
18:52:55 <fizzie> boily: Well I have the whole 10/8 here SO THERE.
18:52:59 <fizzie> Okay, so it's not quite the same thing.
18:53:25 <Taneb> I congratulate you on your one and a quarter, fizzie
18:54:11 -!- azaq23 has joined.
18:54:37 <fizzie> Anyway, "Defense Information Systems Agency" has the whole 22.0.0.0/8, 26.0.0.0/8, 29.0.0.0/8 and 30.0.0.0/8. I'm sure they make good use of them.
18:55:12 <zzo38> ais523: Can module I/O be formats other than fixed number of bits, though? At least in some cases, a bit-count calling convention for module I/O might be useful, even though in some cases it might not be appropriate, possibly.
18:55:22 <boily> speaking of quarters, I have one from connecticut in my wallet.
18:55:42 <ais523> zzo38: Verilog tries to hide its type system by making everything bits
18:56:03 <ais523> but it has to interoperate with VHDL, which has quite an obvious type system
18:56:34 <ais523> fizzie: are those the people the Internet was created for in the first place?
18:57:04 <fizzie> ais523: They certainly have a big chunk of it, still.
18:57:49 <fizzie> I don't know how it differs from Army Information Systems Center (who have 6.0.0.0/8), but I suppose it's very different.
18:58:01 <fizzie> DISA has a Wikipedia page, the latter doesn't.
18:58:09 <ais523> fizzie: well that seems to refer to the army in particular
18:58:10 <dbelange> whoah guys hacking is OT on this channel
18:58:17 <ais523> who's a subset of the military
18:59:33 <zzo38> ais523: So would you know if such a thing as this would work with Verilog? http://wiki.nesdev.com/w/index.php/User:Zzo38/Mapper_768#Submapper_1
18:59:40 <fizzie> Also, I have here a dime and a cent, they've been rolling around my table since a visit to Portland, OR, in September, 2012. I have no idea what to do with them.
19:00:06 <ais523> zzo38: I'm a bit busy to look at that
19:00:27 <fizzie> It's not like currency exchange places would take them, and the value is so close to zero, it's hard to care terribly much, but on the other hand I don't want to just throw them away either.
19:00:34 <zzo38> The first sixty I/O ports of the main module of the Verilog code must correspond to the pins 01 to 60 of the 60-pin Famicom cartridge, in that order. This is followed by the pins for the PRG ROM, CHR ROM, non-battery PRG RAM, non-battery CHR RAM, battery PRG RAM, and battery CHR RAM.
19:00:47 <ais523> fizzie: do you have any children you could give them to?
19:00:51 <ais523> children often like foreign currency
19:00:55 <zzo38> The commands with $ at front might not be implemented, but should be safely ignored if not implemented. However, if there is a trainer ROM, there will be an additional command $trainer to access 8-bit numbers given the 9-bit address, and $battery_init which tell you if you need to initialize the battery RAM.
19:01:00 <zzo38> Analog commands may be used with the audio signals.
19:01:09 <zzo38> That is the relevant parts.
19:01:46 <boily> fizzie: I once found a French franc in my wallet. I have no idea how it got there.
19:02:02 <boily> ais523: I *could* count as a child, and I'm foreign.
19:02:21 <zzo38> I have some foreign money in my desk drawer too. I just keep it there for now.
19:02:37 <ais523> zzo38: you're not going to be able to synthesize analog circuitry with verilog, its execution model isn't suited to it
19:03:33 <zzo38> ais523: Would it be possible to omit those parts in a digital synthesis though, and then build the analog stuff separately?
19:04:00 <ais523> zzo38: it's possible to mark parts of VHDL to be ignored by synthesizers; I don't know if Verilog has that feature, but it wouldn't surprise me
19:04:12 <ais523> the execution model of the language is unsuited to expressing analog circuits, though
19:04:19 <Vorpal> ais523, is there any HDL for synthesizing analogue circuits?
19:04:20 -!- hagb4rd has joined.
19:04:24 <ais523> you could do it but only because Verilog is Turing-complete, it doesn't have any support for it
19:04:31 <ais523> Vorpal: not that I know of
19:04:42 <ais523> there are proprietary simulators that use netlists as input
19:04:48 <ais523> don't know if there's any standard for those yet
19:04:57 <ais523> I've never used one but I've read about them
19:05:09 <ais523> however, netlists are what typical synthesizers /output/, it's a very very low level format
19:05:12 <zzo38> I read that there is some extensions to Verilog for analog wires
19:05:33 <ais523> zzo38: the problem is that Verilog/VHDL's execution model is based around storing things in registers at clock edges
19:05:40 <Vorpal> sure simulators for analogue circuits exist. I even seen one at university, forgot the name of it.
19:05:41 <ais523> it's hard to create a reliable analog register
19:06:11 <ais523> Vorpal: I know, that's why I answered your question the way I answere dit
19:06:16 <Vorpal> How did the old analog computers work?
19:06:31 <ais523> Vorpal: they were basically a set of components that you joined together manually
19:06:36 <ais523> for electronic computers, with wires
19:06:41 <zzo38> Also, the $trainer and $battery_init commands are not supposed to be synthesizable either; they are meant only for emulation (and possibly could be ignored or switch off by conditional compiling if you are compiling to hardware)
19:06:44 <ais523> for the mechanical ones, by placing cogs at appropriate places
19:06:48 <ais523> to join shafts together
19:06:59 <Vorpal> ais523, but how did they store information I mean
19:07:03 <fizzie> boily: One used to occasionally-rarely get Italian 500 lire coins in place of the Finnish 10 mark coins, because they look -- http://digilander.libero.it/maggioref/500%20lire%20normali.JPG and http://i.ucoin.net/coin/178/1781/178195_2/finland_10_finnish_markkaa_1993.jpg -- slightly similar at a glance. (And the exchange rates as of the introduction of the euro were something like 500 ITL = ...
19:07:08 <Vorpal> what did they use instead of registers
19:07:09 <ais523> Vorpal: they didn't, that's what "analog" means
19:07:09 <fizzie> ... 1.54 FIM, so it was a good deal to the person who initially put the coin into circulation.)
19:07:21 <ais523> all data existed instantaneously
19:07:44 <Bike> Vorpal: are you familiar with data flow programs or synthesizers
19:07:46 <Vorpal> ais523, doesn't that analog just mean they are using a continuous signal, instead of two or more levels?
19:07:48 <Bike> analog computers are like that
19:07:49 <ais523> an analog computer calculates everything as a function of other things
19:08:01 <Bike> you could have an analog computer with registers
19:08:05 <ais523> common components would include add, multiply by constant, integrate, differentiate
19:08:12 <Bike> i don't think any built one really did, though
19:08:17 <Bike> since most of them were differential analyzers
19:08:23 <Vorpal> ais523, surely you could set up some sort of feedback loop though
19:08:31 <ais523> Vorpal: yes, they're not very useful without feedback loops
19:08:33 <Bike> yeah, a feedback loop could be used to store information
19:08:42 <Vorpal> <Vorpal> ais523, surely you could set up some sort of feedback loop though
19:08:42 <Vorpal> <ais523> Vorpal: yes, they're not very useful without feedback loops
19:08:44 <ais523> but they're time-continuous feedback loops
19:08:46 <Bike> that's how digital memory worked too :)
19:08:47 <boily> fizzie: USD and CAD coins are the same colours, same weight, same sizes, and are worth about the same. they get easily mixed at border areas.
19:08:48 <zzo38> ais523: Well, there is the esolang Proce, for analog stuff too, I have been told, and it include the operations you specified
19:08:54 <ais523> you get storage, but it's not particularly useful
19:09:01 <Vorpal> ais523, hey, delay lines
19:09:04 <ais523> or not particularly like registers
19:09:13 <boily> fizzie: still no explanation about that spontaneous franc, which is worth...
19:09:14 <ais523> Vorpal: delay lines don't work for analog storage
19:09:27 <ais523> digital delay lines include a "reshaping" circuit that's used to sharpen the corners on the 1s and 0s
19:09:28 <Vorpal> really? Don't they just store waves in mercury?
19:09:39 <Bike> Vorpal: basically a feedback loop is only like a register in that it can be held constant, to get information out you have it "send" it
19:09:41 <ais523> Vorpal: yeah, the problem is the data degrades a huge amount in transit
19:10:01 <ais523> anyway, you can also use an integrator to store data
19:10:06 <Bike> basically the analogy to registers is kind of useless
19:10:10 <ais523> leave input at 0, it retains its current value
19:10:16 <ais523> change input to 1 for a second, you're adding 1
19:10:19 <Bike> maybe you should just read some papers by Shannon or von Neumann and stop listening to me
19:10:23 <ais523> likewise you can subtract 1 by changing the input to -1 for a second
19:10:36 <ais523> it's just that these analog components don't behave a whole lot like digital components in general
19:10:40 -!- ais523 has quit.
19:10:40 <Bike> Shannon developed a model of analog computers, the GPAC, and there's been some not-ancient work on that
19:11:00 <Bike> somebody or another proved it "equivalent" in a weird way to turing machines
19:11:19 <boily> fizzie: ~20 ¢, so a gain of approximatively 5 cents.
19:11:22 <zzo38> The reason I want mix analog/digital is not only for Famicom mappers, but also, I have the idea I want to be able to use a Verilog code with Csound, somehow, in order to make a chip tune music with Csound.
19:11:42 <zzo38> Does someone other than ais523 have the suggestion of such thing (since ais523 has quit)?
19:12:04 <Vorpal> zzo38, write your own simulation in C? It is probably far easier
19:12:08 <Bike> von neumann wrote a few papers about analog, particularly i've read a paper of his that developed a model of making reliable neural networks (in a not-biological sense, but whatever) out of unreliable analog components
19:12:51 <Bike> hm or maybe they were just dataflow, but digital... it's been a while
19:13:01 <Vorpal> Bike, anyway we make reliable digital electronics out of analog components all the time. Transistors are not perfectly digital.
19:13:06 <zzo38> Vorpal: Yes, it could be written in C to make a Csound plugin, or in the Csound orchestra programming language, but I want to be able to use it with Verilog too
19:13:24 <Bike> von neumann was writing in like the 50s you realize
19:13:40 <Bike> transistors were barely extant
19:13:58 <Bike> anyway GPACs are more like what ais was talking about
19:14:04 <Vorpal> I'm not familiar with how vacuum tubes behave electrically
19:14:09 <fizzie> boily: Back when they introduced a new set of coins (in 1992) for the Finnish mark, many coin-operated vending machines (incl. the soda vending machine at my school... um, hypothetically) accepted the old 0.50 FIM coin as the 5 FIM coin, for a nice factor of 10. (They didn't look particularly similar, but the diameter and weight matched very well.)
19:14:20 <Bike> Vorpal: http://arxiv.org/abs/1203.4667
19:15:05 <Bike> i think synthesizers are a good intuitive model of this really
19:15:17 <Bike> you have something generating a wave, you pass it through a filter
19:15:32 <Bike> added bonus, you get to help zzo38 with whatever on earth he's doing
19:15:52 <Vorpal> is that really a bonus? XD
19:16:19 <fizzie> boily: Or come to think of it (and looking at those dates), maybe it was more like that some new vending machines introduced 5-10 years *after* the coinage change were the ones that confused them. But it was still a nice way to use a pile of leftover 0.50 FIM coins. Same machines also accepted old 0.20 FIM as 1 FIM, but that's just a factor of 5.
19:16:29 <olsner> Vorpal: Obviously it is.
19:16:41 <Taneb> Can anyone tell me the name of a play about the roman emperor Commodus (before he was emperor) that was written in rhyme and I think iambic pentameter
19:16:42 <Bike> zzo38 is the new craze that's sweeping the nation.
19:17:15 <Bike> also i don't understand half of what he's saying so i assume it must be interesting.
19:17:32 <boily> it is the '10s, and there is time to zzo38.
19:17:39 <zzo38> Bike: Then you must learn.
19:17:55 <Taneb> Bike, get a FAMICOM emulator and dev kit.
19:17:58 <Taneb> That is the first step
19:18:11 <zzo38> (assuming it must be interesting, is not a valid logic, but it is a valid opinion, I guess)
19:18:12 <fizzie> Is this year stamped in the cent the production year? Then I guess this is a reasonably old coin, since it says 1964 on it.
19:18:17 <Bike> is FAMICOM even an acronym, or are they just yelling
19:18:28 <Bike> i thought it was Family Computer
19:18:31 -!- epicmonkey has joined.
19:18:32 <zzo38> Bike: It is short for Family Computer
19:18:48 <Bike> FAMILY COMPUTER!!!
19:18:58 <Bike> also yes it's invalid zzo38, i'm no good with syllogisms
19:18:58 <Taneb> NINTENDO ENTERTAINMENT SYSTEM!
19:19:05 <Vorpal> how does a FAMICOM and a NES differ?
19:19:23 <Bike> they didn't have region coding in those days
19:19:25 <elliott> fizzie: I've never thought how machines recognise coins before now
19:19:37 <Taneb> Vorpal, different external design
19:19:40 <Bike> (not to say that NES and Famicoms aren't different, they are)
19:19:47 <Vorpal> Taneb, so internals didn't differ?
19:19:49 <Bike> elliott: didn't you ever read How Stuff Works as a kid
19:19:54 <olsner> Bike: ah, so by calling the systems different things people thought you couldn't import the games?
19:19:55 <Taneb> Not as far as I am aware
19:20:01 <Vorpal> Taneb, why different design though? Seems pointlessly expensive
19:20:03 <fizzie> Magic must be involved.
19:20:11 <Taneb> Vorpal, appeal to different markets
19:20:21 <Taneb> The US video game market had just crashed pretty badly
19:20:25 <Bike> olsner: back when the NES came out i don't think there was much appeal to importing games.
19:20:29 <olsner> "oh, only for Famicom? I guess I'll have to wait for the NES release..."
19:20:31 <fizzie> "In the United States, most vending machines have advanced currency detection techniques that can discern coins by reading the coins' "magnetic signature;" thus, many American vending machines will not take coins from other countries, even if their sizes are similar."
19:20:52 <fizzie> Advanced currency detection techniques sounds fancy.
19:21:09 <Bike> fizzie: originally adapted from cold war soviet sub detection
19:21:17 <kmc> didn't they use at least different video standards?
19:21:28 <Bike> oh, i meant with games specifically
19:21:31 <Vorpal> fizzie, iirc it boils down to applying a magnetic field and measuring the deceleration
19:21:36 <Bike> i think NTSC and PAL an stuff were around yes
19:22:00 <kmc> oh but japan uses NTSC
19:22:04 <elliott> Bike: that's way too many "t"s
19:22:11 <Vorpal> kmc, didn't Japan use NTSC?
19:22:18 <Vorpal> I thought only Europe used PAL
19:22:21 <zzo38> The NES/Famicom is mostly the same, although the cartridge is different. Also, the RF Famicom is different to AV Famicom.
19:22:24 <olsner> I thought japan used secam
19:22:31 <kmc> http://upload.wikimedia.org/wikipedia/commons/0/0d/PAL-NTSC-SECAM.svg
19:22:40 <Vorpal> olsner, wasn't that France?
19:22:41 <Bike> wow i've never even heard of secam
19:22:46 <kmc> Vorpal: most of the world uses PAL
19:23:02 <kmc> well maybe most?
19:23:05 <kmc> anyway there's the map
19:23:06 <Vorpal> kmc, true, but iirc SECAM in France and NTSC in US and JP?
19:23:06 <Bike> of course i don't actually know what NTSC and PAL are in the first place other than something ... something
19:23:09 <Vorpal> though I could be wrong
19:23:13 <kmc> SECAM is the french "different because we're french" standard
19:23:16 <fizzie> Hey, my cent -- http://cointrackers.com/coins/13649/1964-lincoln-penny/ -- is technically worth 20 cents!
19:23:21 <Bike> looks like russia uses it
19:23:21 <kmc> i didn't know russia uses it though!
19:23:29 <Bike> also the coveted mongolian market
19:23:37 <olsner> we once had a fancy japanese-brand VCR that did secam and other acronyms, I just assumed one of them would be "japanese tv"
19:23:38 <kmc> "Another explanation for the Eastern European adoption of SECAM, led by the Soviet Union, is that the Russians had extremely long distribution lines between broadcasting stations and transmitters... Long co-axial cables or microwave links can cause amplitude and phase variations, which do not affect SECAM signals."
19:23:55 <Bike> kmc: i'm thinking this map looks like the one from 1984.
19:24:02 <kmc> "According to this explanation, East German political authorities were well aware of West German television's popularity and adopted SECAM rather than the PAL encoding used in West Germany."
19:24:11 <kmc> Bike: haha
19:24:16 <zzo38> In most cases the same game can work in NES, Famicom, and some other clones.
19:24:19 <Bike> oh man french guiana
19:24:27 <Bike> all by its lonesome there, can't pick up any brazilian TV
19:24:33 <zzo38> Microphone is only in RF Famicom, though; AV Famicom and NES lacks this features.
19:24:44 <kmc> Western Sahara is the one island of NTSC in Africa
19:24:59 <Bike> kmc: that looks like no info
19:25:07 <Bike> which makes sense considering western sahara's population
19:25:08 <kmc> but it's greyed out because a) not a country, b) how many televisions are there anyway
19:25:23 <zzo38> Vorpal: Officially they just call it "Famicom" but I call it RF Famicom to distinguish from AV Famicom, since it is connected to the TV set by RF cables.
19:25:28 <Bike> bangladesh what are you doing
19:25:38 <Bike> that is bangladesh isn't it
19:25:44 <Vorpal> zzo38, it had a mic? What was that used for in practice?
19:26:10 <kmc> no that's myanmar
19:26:16 <Bike> Vorpal: http://en.wikipedia.org/wiki/Takeshi_no_Ch%C5%8Dsenj%C5%8D used it
19:26:22 <zzo38> Vorpal: Hardly anything, although there are some games (even ones written this year!) that use it, usually optionally.
19:26:27 <Bike> ugh i was confusing burma and thailand
19:26:41 <kmc> i forget are we calling it myanmar or burma now
19:26:50 <Bike> i don't remember either
19:26:51 <kmc> this is of course an intensely political question
19:26:53 <zzo38> (Famicom Hangman optionally uses the microphone to mix up the random number generator some more)
19:27:03 <Bike> That Fucked Up Country i guess
19:27:06 <olsner> fun fact: myanmar and burma are spelled the same way in myanmarian
19:27:21 <Bike> still can't believe they had a socialist government based on killing communists
19:27:46 <kmc> that is confusing
19:27:48 <Bike> ("bike the soviet union did that too" "shut up")
19:28:26 <kmc> the great thing about kinds of socialism is that there are so many to choose from
19:28:28 <olsner> Vorpal: one name is from how it was pronounced ages ago, the other is how they actually say it nowadays, but their writing system hasn't changes from "ages ago"
19:28:50 <Bike> what about spoken?
19:28:54 <Vorpal> olsner, which one is the modern pronunciation?
19:29:11 <olsner> Vorpal: fun facts do not come in such detail
19:29:20 <olsner> (that would make them not fun)
19:29:42 <Vorpal> olsner, you mean they would move on to being hilarious?
19:30:52 <olsner> I suspect the actual contemporary pronounciation is not remotely similar to either name, and in fact they all speak indonesian or somesuch anyway
19:31:38 <Vorpal> isn't there a military junta there?
19:31:42 <Bike> burma has a lot of languages
19:31:46 <Bike> yes, though they've been liberalizing
19:32:29 <Bike> burmese is the official language, though i don't know if that means it's actually popular (though i'd guess it is)
19:32:39 <Bike> "Burmese is spoken by 32 million as a first language and as a second language by 10 million" k then.
19:33:04 <Taneb> Burmese is one of the 5 languages with more than 5 million speakers that have a "th" sound like in English "with"
19:33:25 <nooodl> zzo38: what is my name today?
19:33:35 <olsner> Vorpal: fun fact #2: fun facts are usually at least partially wrong
19:33:49 <Vorpal> olsner, is that a fact?
19:34:04 <kmc> olsner: that's what makes them fun
19:34:19 <boily> Taneb: th and dh sounds are evil.
19:34:26 <kmc> "Note that Hong Kong and Macao, the two Chinese dependencies on the southern coast of China, are omitted on the map. Both city-states use the NTSC/NTSC J system."
19:34:29 <kmc> one country two systems
19:34:47 <Bike> PAL with chinese characteristics
19:34:58 <Taneb> I know someone who's lived in England all his life and has English as his first language and can't tell the difference between f and th
19:34:59 <shachaf> | fact n = n * fact (n - 1)
19:35:12 <zzo38> The only controls used by the current version of Famicom Hangman are the letter keys and the space-bar.
19:35:14 <Taneb> Other than that he's quite a good serious actor
19:35:27 <kmc> "Only Japan's variant "NTSC-J" is slightly different: in Japan, black level and blanking level of the signal are identical (at 0 IRE), as they are in PAL, while in American NTSC, black level is slightly higher (7.5 IRE) than blanking level. Since the difference is quite small, a slight turn of the brightness knob is all that is required to correctly show the "other" variant of NTSC on any set as it is supposed to be; most watchers mig
19:35:46 <Vorpal> Taneb, that doesn't make any sense
19:35:53 <Vorpal> they are rather different sounds
19:35:57 <Bike> by the way as long as we're talking about burma
19:35:57 <kmc> so technically NTSC and PAL are both color standards over a B&W standard which can vary
19:35:59 <Taneb> He associates them
19:36:05 <kmc> however NTSC is always used with "System M"
19:36:06 <Bike> i'd like to mention that the place is so fucked that wikipedia has this article
19:36:09 <Bike> http://en.wikipedia.org/wiki/Ceasefires_in_Burma
19:36:09 <Taneb> I think it's a psychological thing
19:36:11 <Phantom_Hoover> Taneb, i have a friend like that except he's scots-algerian
19:36:14 <Vorpal> Taneb, and I say that, not having the th sound in my native language
19:36:19 <nooodl> Taneb, albanian, arabic, danish, english, german, hebrew, portuguese, spanish, swahili, turkmen
19:36:31 <olsner> Bike: also http://en.wikipedia.org/wiki/Names_of_Burma since we were on that topic earlier
19:36:37 <kmc> http://en.wikipedia.org/wiki/PAL#PAL_broadcast_systems
19:36:53 <boily> Phantom_Hoover: I have a friend who's scots-algerian too. have we the same friend?
19:36:54 <olsner> (with an article as big as some countries)
19:37:06 <Vorpal> Taneb, the sound I do have an issue with is the "ch" sound. Like in chair and chest. Sometimes it seems to come in slightly different variants. Really hard to hear the difference there for me
19:37:15 <Phantom_Hoover> boily, i suspect there may be an age difference between them
19:37:28 <boily> Phantom_Hoover: maybe.
19:37:42 <Taneb> Vorpal, I think it's quite close to "dsh" or "tsh"?
19:37:49 <kmc> good thing we took digital TV as an opportunity to finally standardize world wide
19:37:50 <Taneb> But I also think th and f are quite close
19:38:19 <Vorpal> Taneb, Yeah, can't really hear the difference between those variants unless the person saying it over-exaggerates.
19:38:21 <shachaf> nooodl: Hebrew doesn't have the English "th" sound.
19:38:25 <Bike> hm... that got me thinking about standards... and I suppose IP is a standard that's been pretty universally adopted?
19:38:40 <Vorpal> Phantom_Hoover, Maybe. We don't really make that difference in Sweden, whatever it is
19:38:45 <nooodl> HebrewIraqiאדוני [ʔaðoˈnaj] (help·info)'my lord'Commonly pronounced [d]. See Modern Hebrew phonology
19:39:01 <Vorpal> Phantom_Hoover, I think it is the d/t thing Taneb is talking about
19:39:02 <nooodl> so it's probably only [] if you're some kind of -pronouncing asshole
19:39:21 <kmc> Bike: well it's adopted world wide, but lots of other standards coexist
19:39:26 <Taneb> nooodl, I'm referring to [θ]
19:39:28 <kmc> either independently or by tunneling one over the other
19:39:52 <Vorpal> Phantom_Hoover, also some voiced phonemes are kind of hard to pronounce for me, again we lack them in Swedish
19:40:09 <olsner> Taneb: isn't the -th in with a different sound?
19:40:26 <nooodl> that's pronounced /wɪð/...
19:40:26 <Vorpal> err, not voiced, wrong word
19:40:26 <olsner> voiced rather than voiceless, I think
19:40:32 <Vorpal> trying to find the right word
19:40:51 <kmc> we need OSI Model 2.0 which defines layers 8-14 which all run on top of HTTP
19:41:21 <Vorpal> Phantom_Hoover, the phonemes that when you pronounce them feels like your tongue is "buzzing", what are they called...
19:41:23 <boily> kmc: layers 5 and 6 are not even used. let's fill them before adding new layers.
19:41:42 <Vorpal> Phantom_Hoover, well, when you over-pronounce it
19:41:45 <Bike> kmc: xen over internet protocol
19:41:52 <olsner> boily: hmm, sounds like some kind of periodic table of models
19:41:57 <Taneb> Just looked it up, and with could be either th or th
19:42:05 <Taneb> I mean the one in "thin"
19:42:12 <Taneb> Not the one in "this"
19:42:13 <Bike> z vs s is the best example of voicedness imo
19:42:19 <Bike> just hold your adam's apple
19:42:31 <Vorpal> Phantom_Hoover, you have a lot more of then than Swedish iirc
19:42:44 <fizzie> Voicing doesn't have anything to do with the tongue, though.
19:43:18 <Vorpal> you mean the letters? No
19:43:24 <olsner> Phantom_Hoover: no, that's norwegian
19:44:21 <Vorpal> Phantom_Hoover, I think they share a common history though, after all in names "av" is sometimes written as "af" (that is like von in this case)
19:44:30 <boily> [ʁ] is the bestest phoneme of them all.
19:44:42 <Vorpal> boily, which one is that?
19:45:02 <boily> https://en.wikipedia.org/wiki/Voiced_uvular_fricative
19:45:09 <olsner> swedish v is like english v, I think ... but swedes tend to mix up w and v sounds in english, because the w sound is not common
19:45:43 <kmc> Bike: Syscalls As A Service
19:46:13 <Vorpal> boily, well okay, r as in Skånska I think?
19:46:37 <olsner> ah, "Southern dialects", yes
19:46:41 <Vorpal> yeah, sounds about right
19:47:02 <Vorpal> olsner, so a rolling r basically
19:47:36 <boily> it's not the rolling r, sadly.
19:47:39 <Vorpal> olsner, I like the sj-sound too
19:47:40 <fizzie> We don't have that at all. :/ (In fact, Finnish is reasonably modest when it comes to phonology.)
19:47:49 <olsner> the following example is funny, it says yangir, but it's the g that has the r sound
19:48:56 <Vorpal> boily, https://en.wikipedia.org/wiki/Voiceless_palatal-velar_fricative
19:49:11 <boily> Vorpal: that one is nasty.
19:49:18 <fizzie> What's "funny" (FSVO) about Finnish phonology is that for the most part, our graphemes map to the identical-looking IPA symbols. (Not quite all of them, but still.)
19:49:23 <Vorpal> boily, how so? I can pronounce it easily!
19:49:55 <boily> I think I can, but I don't know any Swedish people to check if I'm doing it correctly.
19:50:17 <olsner> "Though the acoustic properties of its [ɧ] allophones are fairly similar, the realizations can vary considerably according to geography, social status, age, gender as well as social context [...]"
19:50:21 <Vorpal> boily, I have a German relative who lived in Sweden since the 50s, and still has problems pronouncing it
19:50:38 <Vorpal> well, since the late 50s
19:50:52 <olsner> ... makes it sound a bit like swedes communicate only through ɧs
19:51:17 <Vorpal> olsner, well... Sju sjösjuka sjömän...
19:51:54 <olsner> you could just learn finland swedish instead and use ʃ
19:52:07 -!- edwardk has quit (Read error: Connection reset by peer).
19:52:13 <boily> I think the hardest Québec French sound isn't even a consonant. we extensively use /ɑ/.
19:52:56 <Gregor> TIL: Quebecois speak exclusively in vowels.
19:53:43 <fizzie> /ɑ/ is also the Finnish "a".
19:54:06 <Vorpal> English lacks the Swedish u too
19:54:07 <boily> /ɑ/ has disappeared from France French, mainly.
19:54:37 <fizzie> "But... it's just an a!"
19:54:50 <olsner> Vorpal: re "rolling r", I think you confused the names of the r's
19:55:11 <Vorpal> olsner, Which one is the skorrande r?
19:55:37 <Vorpal> which is what I was thinking of
19:55:45 <olsner> uvular or guttural, afaict
19:55:46 <Vorpal> English lacks ʉː right?
19:56:17 <boily> yes, it has nothing to that effect.
19:58:10 -!- edwardk has joined.
19:58:54 <fizzie> Possibly they should give each language some unique, non-shared phoneme, so that everyone'd have something to be proud of.
19:59:21 <Vorpal> fizzie, well Swedish has a couple of unique ones, with the ɧ
19:59:35 <fizzie> Phantom_Hoover: You can just add more subdivisions.
19:59:41 <Phantom_Hoover> there's this one fairly tame phoneme which isn't used by any languages
20:00:55 -!- ais523 has joined.
20:03:04 <Vorpal> Also fun in Swedish, anden and anden is pronounced differently depending on if it means "the duck" or "the spirit". Wikipedia lists the former as "[ˈa᷇ndɛ̀n] or [ˈan˥˧dɛn˩]" and the latter as "[ˈa᷆ndɛ̂n] or [ˈan˧˩dɛn˥˩]"
20:03:34 <shachaf> Half-voiced quidental-applasal darentive
20:03:52 <Vorpal> Phantom_Hoover, sure but is it the same type of difference
20:04:23 <fizzie> shachaf: Voiceless quidditch broomxirant.
20:05:01 <zzo38> I think the Famicom is possible to have keyboard and mouse connected at the same time, but I don't think any existing software uses it!
20:05:04 <ion> @karma ghoti
20:05:58 <Phantom_Hoover> there are a bunch of noun/verb stress differences there
20:06:08 <fizzie> You can tell our phonological poverty from the fact that most of the IPA symbols of Finnish phonemes are just ASCII. (Okay, so the mid-vowels technically have extra diacritics and there's a few exceptions, but still.)
20:06:37 <Taneb> You make up for your phonology with your conjugation, fizzie
20:06:47 <olsner> having any of your phonemes' IPA symbols in ascii is the sign of failure
20:07:38 <ais523> you know what more programming languages need? a "revert all state apart from the instruction pointer and call stack" command
20:07:44 <ais523> imperative languages, that is
20:07:57 -!- oerjan has joined.
20:08:04 <fizzie> Finnish has a lot of information-bearing phoneme lengths (tuli 'fire' vs. tuuli 'wind' vs. tulli 'customs/toll'), that I think is often mentioned as a curiosity.
20:08:22 <ais523> this would seem to be very useful for cleanup after an error
20:08:45 <ais523> you could do all the destructive updates you like but always be able to roll back
20:08:50 <Vorpal> ais523, hm, is there any imperative language that has that?
20:08:55 <ais523> hmm… that's more or less the exact opposite of a continuation
20:08:57 <ais523> Vorpal: not that I know of
20:09:03 <elliott> ais523: you need some scope for that
20:09:07 <shachaf> fizzie: English has these too. Compare: "elliott" and "eliot".
20:09:10 <Vorpal> ais523, it is kind of trucky if there is IO involved as well
20:09:22 <Vorpal> I suggest just using a reversible language?
20:09:26 <ais523> Vorpal: that doesn't have to be reversed
20:09:40 <ais523> this would be for algorithms anyway
20:09:46 <ais523> perhaps to respond to IO failures
20:09:48 <Vorpal> ais523, ah, so transactional memory then
20:09:59 <Vorpal> just roll back the current transaction
20:10:02 <ais523> yeah, transactional memory
20:11:23 <ais523> I've wanted it to be available repeatedly over the last week
20:12:00 <Vorpal> ais523, what about forking each time and then killing processes up till the point you want to revert to
20:12:06 <ais523> other people's programs, I think
20:12:09 <Vorpal> Since linux uses copy on write this should work
20:12:10 <ais523> Vorpal: that's how it's implemented in INTERCAL
20:12:16 <Vorpal> ais523, really? Niiice
20:12:35 <fizzie> Where "each time" presumably means "each time you do anything at all"?
20:12:37 <Vorpal> I wouldn't want to run intercal on cygwin then
20:12:44 <ais523> fizzie: each time you enter a stack
20:12:48 <Vorpal> fizzie, well, depend on the granularity you wan't rollback on
20:12:48 <ais523> Vorpal: interpreter threads, not OS threads
20:12:56 <ais523> you could do it with OS threads too but I don't want to try
20:12:58 <Vorpal> ais523, I mean OS ones
20:13:35 <ais523> seems a bit wasteful of pid namespace
20:13:48 <Vorpal> is it 32-bit on linux?
20:13:59 <ais523> it used to be 16-bit, and many programs still assume that so it's the default
20:14:04 <ais523> but the kernel supports 32-bit too
20:14:43 <fizzie> Seems a bit wasteful of things in general. (Sure, sure, copy-on-write, but there's still quite a lot of things in a process.)
20:15:08 <AnotherTest> Has anyone in here ever heard of someone who goes by the pseudonym "simpson"?
20:15:28 <ais523> fizzie: Linux lets you choose how much to copy when you create a process
20:15:33 <fizzie> ais523: It's not just the programs; seeing six-digits-or-more numbers in ps is confusing for a person, too.
20:15:39 <Taneb> I know someone called Simpson who gets called "simpy"
20:15:42 <pikhq> pid_t on my system is a signed 32-bit value.
20:15:47 <pikhq> ... why is it signed
20:15:59 <ais523> pikhq: for signalling process groups
20:16:00 <fizzie> pikhq: But what's your /proc/sys/kernel/pid_max?
20:16:00 <shachaf> pikhq: Because kill can take a negative argument?
20:16:00 <AnotherTest> Taneb: is this by any chance the simpson who is active at #python too?
20:16:06 <Taneb> AnotherTest, doubtful
20:16:19 <AnotherTest> (09:12:34 PM) simpson: AnotherTest: No, it's funny because I hack esoteric languages all the time and write compilers for fun. :3
20:16:20 <fizzie> (Mine is 2^15 exactly.)
20:16:30 <pikhq> fizzie: Blah, it is 2^15.
20:17:03 -!- lahwran has quit (Changing host).
20:17:03 -!- lahwran has joined.
20:17:22 <HackEgo> not lately; try `seen simpson ever
20:17:31 <Taneb> `seen simpson ever
20:17:58 <lmt> `seen lament
20:18:03 <HackEgo> not lately; try `seen lament ever
20:18:08 <lmt> `seen lament ever
20:18:14 <HackEgo> 2012-07-18 18:17:18: <lament> NOOOOOOOOOOOOOOOOOOOOOOO
20:18:33 <fizzie> ais523: PID_MAX_LIMIT in my random grep-kernel (3.something, anyway; haven't been building kernels lately) is a value that's not 2^31-1; in fact, it's a rather complicated value.
20:18:38 <olsner> lament laments, who'd've thought
20:18:42 <elliott> AnotherTest: he is in #haskell.
20:18:54 <fizzie> It's (CONFIG_BASE_SMALL ? PAGE_SIZE * 8 : (sizeof(long) > 4 ? 4 * 1024 * 1024 : PID_MAX_DEFAULT)), where PID_MAX_DEFAULT is (CONFIG_BASE_SMALL ? 0x1000 : 0x8000).
20:19:02 <HackEgo> 2013-03-18 20:18:58: <ais523> `seen ais523
20:19:14 <fizzie> Also, there's a comment saying "A maximum of 4 million PIDs should be enough for a while." on top.
20:19:15 <HackEgo> 2013-02-17 15:53:28: <ais523_> it's nice to have an OS such that your computer can overheat, crash, and restart before you ping out from IRC
20:19:23 <HackEgo> not lately; try `seen ais523__ ever
20:19:35 <HackEgo> 2011-03-31 15:54:19: <ais523__> indeed, it is
20:19:36 <HackEgo> not lately; try `seen ais521 ever
20:19:39 -!- zzo38 has quit (Remote host closed the connection).
20:19:51 <fizzie> So you can only raise /proc/sys/kernel/pid_max to 4194304.
20:20:03 <HackEgo> 2012-12-17 20:15:57: <ais521> Oops
20:20:26 <ais523> so was that me or someone else
20:20:30 <Vorpal> fizzie, that is such a random number
20:20:48 <fizzie> Vorpal: It "should be enough for a while", though.
20:21:15 <Vorpal> fizzie, true, mine is at 32768. Which is the default 16-bit thingy again I presume
20:21:26 <Vorpal> haven't had an issue so far
20:21:55 -!- AnotherTest has left.
20:22:15 <fizzie> Vorpal: There's a pidmap[] array, the size of which is related to pid_max, which might explain why it's not completely unlimited.
20:23:01 <boily> why is there a /proc/sys/kernel/hostname, and what is it doing in that directory?
20:23:09 <ais523> !c printf("%x", 4194304);
20:23:18 <ais523> fizzie: not that random
20:23:23 <ais523> boily: the kernel maintains its own idea of the hostname
20:23:28 <ais523> which is used at least by the uname system call
20:23:29 <fizzie> There's also some other default that relates pid_max to the number of processors, with PIDS_PER_CPU_DEFAULT being 1024 -- to match "original pid max of 32k for 32 cpus" -- and PIDS_PER_CPU_MIN of 8, presumably for sanity reasons.
20:23:47 <ais523> presumably that allows you to get at it, and maybe change it, without having to use a separate program like uname(1)
20:23:59 <boily> ais523: good argument.
20:24:17 <ais523> also I don't think uname can set the hostname, so you'd need /another/ separate program to set it
20:25:01 <ais523> which probably exists in section 8 somewhere but I don't know what it's called offhand
20:25:22 <oerjan> 16:36:34: <ais523> coppro: hungarian ő is basically like german ö but a bit longer
20:25:25 <oerjan> 16:36:47: <ais523> whereas hungarian ö is like german ö but a bit shorter
20:25:33 <oerjan> um german has length distinction in vowels too
20:25:33 <fizzie> pid_max is adjusted to be min(PID_MAX_LIMIT, max(pid_max, PIDS_PER_CPU_DEFAULT * num_possible_cpus())) right before the pid-bitmap is allocated.
20:26:57 <kmc> Linux actually supports running different processes with different ideas of what the hostname is
20:27:04 <kmc> clone(CLONE_NEWUTS)
20:27:23 -!- pikhq_ has joined.
20:27:27 <fizzie> There's all kinds of separatable namespaces these days, what with all the container-brouhaha.
20:27:29 -!- pikhq has quit (Ping timeout: 255 seconds).
20:27:44 <ais523> kmc: I used that in Web of Lies, then forgot about it
20:27:56 <ais523> and now I remembered again
20:28:11 <fizzie> And of course you can have multiple PID namespaces.
20:28:13 <Vorpal> ais523, I think hostname sets it?
20:28:53 <Vorpal> ais523, though my hostname is dragon according to the file, and hostname, but hostname -f lists dragon.lan
20:29:10 <kmc> the new hotness is that unprivileged users can now use these various namespaces
20:29:18 <Vorpal> ais523, and hostname -A hangs??
20:29:23 <kmc> which is currently a great source of privilege escalation holes
20:29:23 <ais523> how long before weboflies doesn't have to be root?
20:29:27 <kmc> i don't know
20:29:27 <Vorpal> oh no, it is just slow
20:29:33 <fizzie> "The FQDN is the name getaddrinfo(3) returns for the host name returned by gethostname(2)"; that's where hostname -f comes from.
20:29:40 <kmc> have only been following this via spender's rants on twitter
20:30:04 <ais523> it feels a bit weird for a program to be able to plausibly claim to be init without even being root
20:30:24 <Sgeo> I love how even the actor is half-convinced, even though he knows the truth
20:30:39 <Vorpal> <oerjan> um german has length distinction in vowels too <-- like Swedish you mean?
20:30:51 -!- sirdancealot has quit (Ping timeout: 240 seconds).
20:30:59 <kmc> "I once convinced a woman that I was Kevin Costner, and it worked because *I* believed it!"
20:31:03 <olsner> ais523: how so? sandboxing ought to be safe and freely available for non-privileged users
20:31:25 <ais523> olsner: it can feel weird without being a bad idea
20:31:38 <Vorpal> <kmc> which is currently a great source of privilege escalation holes <-- heh really?
20:31:48 <oerjan> Vorpal: oh. in context, that should be like hungarian actually.
20:32:24 <Vorpal> <Sgeo> I love how even the actor is half-convinced, even though he knows the truth <-- is there a context to this?
20:32:35 <Taneb> kmc, that reminds me, I need to watch Waterworld
20:32:36 <Sgeo> Vorpal, Space Cadets
20:32:41 <elliott> is there a context to anything sgeo says
20:33:09 <Taneb> elliott, sometimes there's a context when he says "^list"
20:33:12 <Vorpal> <ais523> it feels a bit weird for a program to be able to plausibly claim to be init without even being root <-- pure user space qemu can easily do this. An LD_PRELOAD hack could claim it to another program as well
20:33:22 <ais523> Vorpal: qemu's a VM, it doesn't count :)
20:33:32 <ais523> LD_PRELOAD, I'll give you that (although it's not 100% reliable)
20:33:42 <ais523> actually weboflies could claim to be init without actually being init
20:33:48 <Vorpal> ais523, whatever valgrind does would work too
20:33:52 <ais523> but that doesn't really fit in with its philosophy
20:33:56 <Vorpal> which is not exactly LD_PRELOAD only iirc
20:33:59 <Vorpal> though that is involved
20:34:20 <pikhq_> Valgrind doesn't even touch LD_PRELOAD.
20:34:23 <olsner> can LD_PRELOAD be used to insert a new and improved vsyscall page/dso/whatchamacallit?
20:34:37 <Vorpal> olsner, hrrm... probably not
20:34:44 <Vorpal> that is kind of special isn't it?
20:35:01 <pikhq_> It compiles the binary to a specialized IR, runs transformations on the resulting IR, and then JITs.
20:35:02 <Vorpal> ais523, how did you handle the vdso btw? I forgot
20:35:25 <ais523> note that I never claimed that web of lies actually /works/
20:35:33 -!- edwardk has quit (Quit: Computer has gone to sleep.).
20:35:51 <Vorpal> ais523, worked any further on it since it was revealed?
20:36:12 <kmc> with a modified ELF interpreter you could just link your own vDSO substitut
20:36:15 <Vorpal> fizzie, speaking of JITs. Jitfunge? What happened to it
20:36:30 <ais523> Vorpal: since it was /revealed/? yes; last several months? no
20:36:31 <Vorpal> kmc, it is inserted into static binaries too
20:36:38 <ais523> I was working on it as it was revealed
20:36:51 <kmc> Vorpal: but very few static binaries are going to parse the vDSO and use things from it, afaik
20:36:56 <kmc> maybe static glibc does, i dunno
20:37:01 <kmc> it's a different story on i386
20:37:09 <Vorpal> static glibc does weird stuff
20:37:18 <kmc> where the address of one of the vDSO entry points (for system calls) is passed directly in an ELF auxv
20:37:35 <kmc> on amd64 only the start of the vDSO ELF image is passed, for use by the dynamic loader
20:37:56 <kmc> there is an example program somewhere in the Linux source tree for parsing that ELF and using functions from a static binary
20:38:04 <kmc> it's not that hard, but i don't know if anyone does it
20:38:15 <Vorpal> kmc, well ld.so itself probably does
20:38:23 <Vorpal> ld.so is a static binary after all
20:38:26 <kmc> does ld.so need to make fast gettimeofday calls
20:38:53 <kmc> to make system calls in general? it wouldn't need to, it could just use that entry point that's passed
20:38:59 <Vorpal> I guess static glibc uses the gettimeofday call
20:39:21 <kmc> basically they decided it's OK to make it somewhat difficult for static binaries to use fast gettimeofday etc, but not OK to make it difficult for them to make syscalls on i386
20:39:51 <pikhq_> Breaking kernel ABI is generally considered a bad thing.
20:40:26 <pikhq_> You can still run Linux 1.0 binaries on a modern system.
20:40:44 <kmc> yeah Linux is fanatical about that
20:40:46 <kmc> to their detriment
20:41:09 <coppro> on average it's a good thing
20:41:11 <kmc> there's a ton of compatibility cruft in the kernel and it's a perpetual source of bugs and security holes
20:41:21 <coppro> though it would be nice to do a prune
20:41:28 <pikhq_> Userspace has a bit worse compatibility cruft.
20:41:35 <pikhq_> Seriously, 32-bit off_t should DIE in a FIRE.
20:41:36 <kmc> coppro: i don't know, there is the Windows alternative where the kernel ABI is kept clean and modern, and programs are supposed to dynamically link libc etc
20:41:50 <coppro> kmc: dynamically linking libc is not a kernel thing
20:42:29 <ais523> coppro: the Windows kernel team is also responsible for the libc
20:42:41 <ais523> because the libc is the only documented/approved way to communicate with the kernel
20:42:51 <ais523> and in general Windows considers it part of the kernel
20:42:52 <coppro> yeah, but it isn't on Windows
20:43:10 <coppro> (and it shouldn't be because fuck glibc ;) )
20:43:11 <Vorpal> <pikhq_> Seriously, 32-bit off_t should DIE in a FIRE. <-- That is gone on 64-bit systems afaik?
20:43:13 <kmc> i'm saying they could change that
20:43:21 <Vorpal> well apart from multi arch obviously
20:43:22 <olsner> kmc: I guess moving the ABI outside the real kernel is nice, but they still have an ABI in those dlls they can't change
20:43:22 <kmc> yeah, exactly, fuck glibc, Linux team can make their own libc
20:43:27 <kmc> olsner: yeah
20:43:32 <kmc> there's no free lunch if you want binary compat
20:43:35 <Bike> ooh what's wrong with glibc
20:43:39 <kmc> you could have multiple versions of that libc though
20:43:42 <coppro> Bike: the devs are morons
20:43:44 <kmc> http://blog.sc5.fi/2013/03/sc5-pays-salaries-in-bitcoin/
20:43:52 <kmc> drepper is not a moron but definitely an asshole
20:44:03 <Bike> drepper's the only one i know of and he seems smart if a jerk yes
20:44:10 <Bike> though he doesn't work on it any more, does he?
20:44:11 <kmc> he's rather bright, I enjoyed a lot of his whitepapers (published by Red Hat) regarding linkers, futex, etc
20:44:12 <elliott> drepper doesnt work on glibc any more does he
20:44:30 <pikhq_> Drepper totally left glibc.
20:44:37 -!- Taneb has quit (Ping timeout: 245 seconds).
20:44:38 <Bike> anyway i want specifics
20:44:45 <pikhq_> It's been having a major series of compliance bugfixes of late.
20:44:46 <Bike> "the devs are morons" could have so many exciting consequences
20:46:05 <pikhq_> He's quite bright, but not a good choice of maintainer.
20:46:13 <Vorpal> moving emulation and support of old APIs out of the kernel does have the advantage of reducing the amount of code running at elevated privileges though
20:46:26 <kmc> yeah, but Linux has never really been interested in that
20:46:28 <kmc> or in security in general
20:46:36 <kmc> the upstream kernel devs are actively hostile to security
20:46:47 <Vorpal> that makes little sense
20:46:59 <kmc> they intentionally obfuscate regarding which commits are security-relevant
20:47:23 <kmc> like if Linus fixes a trivially exploited root hole, the commit message is usually 4 paragraphs of irrelevant blather
20:47:26 <kmc> eg http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=799c10559d60f159ab2232203f222f18fa3c4a5f
20:47:28 <Bike> so how's SELinux or w/e work?
20:47:47 <kmc> i'm talking about their attitude toward bugfixes, not grand frameworks
20:47:51 <kmc> i don't know as much about the latter
20:47:56 <olsner> is that based on the idea that the fixes will expose security holes that linux has had or something?
20:48:10 <Bike> that commit doesn't mention security at all other than the site of the reporter. lol.
20:48:18 <olsner> or just that they don't care about figuring out whether their bugfix fixed something exploitable?
20:48:20 <coppro> security shouldn't be mentioned until a fixed release is pushed
20:48:50 <coppro> given the amount of time it takes for changes to filter from mainline to distros, it would be irresponsible to say "this fixes a root exploit" in commit logs
20:49:10 <pikhq_> Security fixes in particular filter from mainline to distros in the course of hours.
20:49:20 <kmc> coppro: so instead let's make the kernel team of each distro reverse-engineer the intent of upstream?
20:49:29 <Vorpal> kmc, only the last paragraph mentions security at all
20:49:40 <kmc> i don't think there is a *private* channel for these things, either
20:49:41 <Vorpal> and even then not directly
20:49:46 -!- sirdancealot has joined.
20:49:46 <kmc> they just dump fixes and you're supposed to figure out what's relevant
20:50:00 <kmc> some fixes do get tagged with CVE numbers, but not reliably
20:50:04 <elliott> ksplice wasn't invited to the secret privtae cahnnel
20:51:40 <kmc> yeah maybe :(
20:52:06 <boily> what's that mysterious secret privtae cahnnel?
20:53:26 <elliott> can't tell you it's secret
20:53:51 <fizzie> The guy running SC5 is technically an acquaintance (or at least once removed) from the university. And the bitcoin thing was mentioned by the news media.
20:53:59 <olsner> oracle could do lots of evil if they got told about security issues
20:55:14 <kmc> ah yes evil oracle
20:55:42 <ais523> fizzie: oh right, one of the most confusing/surprising things I've seen: two people from different schools in a chemistry competition started talking
20:55:47 <kmc> no matter what Oracle does, the Internet hates them for it
20:55:58 <ais523> and never having met before, nonetheless deduced that they had a friend of a friend in common
20:56:04 <ais523> i.e. four degrees of separation
20:56:05 <kmc> when Red Hat obfuscates their kernel patches to make it harder for Oracle to compete, it's somehow Oracle's fault
20:56:19 <ais523> the surprise is not in the degree of separation, but in the fact that they managed to work it out
20:57:43 <elliott> ais523: did they work it out independently? that would be more impressive
20:58:21 <ais523> elliott: via communicating with each other
20:58:30 <ais523> I'm also not sure what possessed them to start working it out
20:59:05 <Bike> sometimes i think i'm missing out on "hacker culture" when y'all talk about oracle being evil or whatever but then i notice it seems kind of terrible
20:59:40 <ais523> Bike: there are multiple such cultures, some of which are terrible
21:00:01 <ais523> there's reasonable evidence for oracle being dubious, at least, so it depends on how much you care
21:00:13 <ais523> there's actually a whole forum on tdwtf for oracle-bashing
21:00:17 <kmc> yeah I don't claim Oracle is not evil
21:00:24 <ais523> although it was added mostly as a joke
21:00:35 <kmc> I think Google / Apple / Microsoft / etc. are also pretty evil
21:00:38 <kmc> big companies tend to be evil
21:01:07 <ais523> Microsoft is at least reasonably consistent
21:01:09 <kmc> and yes "hacker culture" is a cesspool
21:01:27 <Vorpal> ais523, Apple is fairly consistent too, no?
21:01:29 <ais523> kmc: what culture are you talking about, specifically? some examples would help
21:01:35 <Vorpal> ais523, patents patents patents
21:01:38 <kmc> eh I rant about it enough as is
21:01:40 <ais523> Vorpal: that's more recent
21:04:00 <Sgeo> http://gkoberger.github.com/stacksort/
21:04:10 <ais523> Sgeo: already seen it from reddit
21:04:20 <ais523> incidentally, I was planning to implement an esolang like that
21:04:29 <ais523> except instead of doing sorting, you'd give it a number as the prorgam
21:04:30 <Bike> "If you like running arbitrary code in your browser," DO I
21:04:47 <ais523> and it'd take the majority opinion of submissions to the matching anarchy golf program
21:04:56 <Bike> holy damn, it actually works
21:05:09 <Bike> after 30 tries
21:05:18 <ais523> also, you run arbitrary third-party code in your browser every time you view an advert that isn't just a static image or text
21:05:33 <Bike> yes, that's why i didn't really care (though i have those blocked usually)
21:05:57 <Vorpal> ais523, I don't allow that. NoScript.
21:06:24 <Vorpal> This thing is probably safer than ads though
21:06:25 <ais523> Vorpal: indeed, many people don't allow that, but most people do
21:06:56 <Bike> i live dangerously
21:13:32 <oerjan> <ais523> whereas there is one former-regular who was south african
21:13:58 <oerjan> there has been at least one new zealander too
21:18:28 <oerjan> `pastelogs monkey.*joined
21:18:41 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.5435
21:20:37 <oerjan> `seen GreaseMonkey ever
21:20:44 <HackEgo> 2012-01-12 23:09:25: <GreaseMonkey> multitasking
21:26:15 <boily> was there ever a her here?
21:26:49 <boily> (I won't even ask for itses. every bot has its own bot here.)
21:27:57 <fizzie> fungot: Do you have your own bot?
21:27:58 <fungot> fizzie: so there is something called " 3m". this has nothing to to but i don't
21:28:02 <elliott> there is a her™ in the channel right now :P
21:29:58 <lmt> i miss vixey :(
21:35:11 <ais523> nickserv says that 3m isn't taken, so presumably we can use it as the name for fungot's bot
21:35:12 <fungot> ais523: what those numbers mean? riastradh ( i assume you're awake). and of course the function call, so you're probably right
21:35:34 <ais523> elliott: well alise uses female pronouns
21:36:07 <elliott> fungot: 3m isn't a number, it was a string
21:36:07 <fungot> elliott: but if stalin generates such labels, a more fnord friend of mine worked for a while while fully aware of the efficiency argument is small. if the child process
21:36:20 <elliott> fungot: i guess maybe you meant 3 million by it??
21:36:20 <fungot> elliott: not really related, but it's not really
21:36:29 <Bike> stalin is such a great name fora compiler
21:36:37 <Vorpal> <elliott> there is a her™ in the channel right now :P <-- who?
21:36:53 <fungot> elliott: we all hate windows, and all
21:36:57 <Vorpal> I know there has been a few before
21:37:07 <elliott> fungot has prescient accuracy
21:37:08 <fungot> elliott: personally i like writing programs in single expressions. :d) tai jotenki, mut ei ainakaa
21:37:18 <Bike> :d), what the fuck is that
21:37:22 <boily> fungot's speaking in tongues again...
21:37:23 <fungot> boily: yeah, that's a trivial prob.,. neither do activists. they've all got more pressing concerns. the gc needs to be a bunch of
21:37:24 <Phantom_Hoover> and of course most of the lurkers are of schrodinger's gender
21:37:43 <Bike> you lock the gender in a box to try to show how absurd copenhagen is?
21:37:44 <Vorpal> speaking of single expression, I might still have that irc bot in a single nested lambda expression in python you wrote around somewhere elliott
21:37:49 * boily lightly shakes fungot in a circular motion
21:37:50 <fungot> boily: now i'm telling miss piggy about money market funds!)
21:39:13 <Phantom_Hoover> it's the "trying to tell whether your nose is still there by feeling it with your tongue" smiley
21:39:27 <Bike> fuck you i can't do it :(
21:39:59 <fizzie> ais523: Given that "3m" starts with a number, I'd be kind of surprised if it were taken.
21:40:11 <fizzie> (Cf. ":leguin.freenode.net 432 * 3m :Erroneous Nickname")
21:40:33 <elliott> ais523: he was addressing me
21:40:34 <ais523> that's a really strong statement for something so minor
21:40:42 <ais523> elliott: yes, that's why I'm confused rather than angry
21:40:43 <elliott> omg i dont care that Bike said damn you to me
21:41:08 <Vorpal> <Phantom_Hoover> it's the "trying to tell whether your nose is still there by feeling it with your tongue" smiley
21:41:09 <Vorpal> <Bike> fuck you i can't do it :(
21:41:14 <Bike> oh we talked about this before and i forgot
21:41:21 <ais523> Fiora: we were discussing diversity in channel regulars
21:41:24 <fizzie> Ooh, vestigial #douglasdams Finnish bits again.
21:41:25 <Bike> Vorpal: there was someone on Ellen a few days ago who did it
21:41:29 <Fiora> elliott: and it's all bike's fault too. what a horrid person he is
21:41:54 <Bike> Vorpal: a tv show (yes, a chat show) hosted by Ellen deGeneres
21:41:54 <shachaf> i can confirm that Bike is horrid
21:41:59 <shachaf> Bicycle is not horrid, on the other hand
21:42:02 <ais523> they're like interview shows except with less content in the questions
21:42:05 <shachaf> imo Bicycle is a better name
21:42:09 <Bike> what about Bicyclidine
21:42:23 <lmt> neurodiversity in channel regulars
21:42:37 <Bike> i'm sure there are plenty of autists
21:42:52 <Fiora> there's probably at least one or two neurotypical people here?
21:43:02 <Bike> sometimes i wonder why "psychodiversity" isn't a thing but it turns out nobody even cares about neurodiversity so
21:43:03 <shachaf> Bike: (btw usernames gotta start witha lowercase letter but "we'll let it slide")
21:43:23 <lmt> Fiora: who
21:43:24 <shachaf> is a psychodiver like a psychonaut
21:43:29 <Vorpal> <ais523> they're like interview shows except with less content in the questions <-- so no hard hitting journalistic questions?
21:43:32 <Fiora> ummmm I don't know
21:43:34 <fizzie> Vorpal: Guinness says the current record-holder for "longest tongue" reaches 9.75 cm (3.8 in) from top lip to tip.
21:43:49 <ais523> Vorpal: I'm not entirely sure why they're popular
21:43:50 <fizzie> http://www.guinnessworldrecords.com/media/1698229/Chanel%20Tapper%20main.jpg looks like it'd have no problems reaching the nose.
21:43:50 <Vorpal> fizzie, is that in a human?
21:44:10 <Bike> yeah, Ellen is kinda boring, I just happened to see it on TV.< 1363644030 973217 :Frooxius!~Frooxius@cust-101.ktknet.cz QUIT :Read error: Connection reset by peer
22:00:33 <fizzie> fungot: Do you have anything to say for yourself?
22:00:34 <fungot> fizzie: i am an idiot! i am overwhelmed by the amount of
22:00:46 <fizzie> Vorpal: Well, it seems quite contrite, at least.
22:00:48 -!- Frooxius has joined.
22:01:03 <fizzie> fungot: It's all right, just don't let it happen again.
22:01:04 <fungot> fizzie: uh, that's not what i meant
22:01:13 <fizzie> fungot: What did you mean, then?
22:01:13 <fungot> fizzie: did that answer your question, and even i have that
22:01:54 <fungot> boily: ( i'm not really sure what you're doing? code?
22:02:11 -!- aloril has quit (Ping timeout: 256 seconds).
22:05:36 -!- edwardk has joined.
22:16:05 -!- aloril has joined.
22:19:32 -!- Nisstyre has quit (Quit: Leaving).
22:23:49 <boily> ah, cosmetics ingredients, you'll always surprise me. «Sodium Kokum Butterate». what the fungot is that.
22:23:50 <fungot> boily: beautiful women carrying pitchers of water will come up with something
22:42:47 <fizzie> (But the comment about separate checkouts sounds true too.)
22:43:29 <HackEgo> accesslog \ a.out \ bin \ brainfuck.fu \ canary \ Category:Self-modifying \ dbg.out \ dkVb20VL \ egobot.tar.xz \ etc \ factor \ foo \ foo.err \ foo.out \ ibin \ index.html \ interps \ karma \ lib \ link \ paste \ prefs \ prefs.bf \ quines \ quotes \ radio.php?out=inline&shuffle=1&limit=1&filter=*MitamineLab* \ rainbow \ sgfoobar \ share \ slist.ori
22:43:39 <boily> `learn comedogenic is something that causes comedy when applied to the skin, e.g. an accelerated cream pie in parabolic motion.
22:43:56 -!- Frooxius has joined.
22:44:03 <boily> I can now go eat. 'night all!
22:44:06 -!- boily has quit (Quit: Poulet!).
22:44:08 -!- metasepia has quit (Remote host closed the connection).
22:44:51 <HackEgo> boily is Canadian or something. We are not sure about Canada's existence.
22:45:09 <ais523> indeed; we are sure that Ottawa exists
22:45:12 <ais523> but not 100% sure it's in Canada
22:45:16 <olsner> I think boily is the man eating chicken
22:45:53 <olsner> shachaf: we can't put it in wisdom until we know it exists
22:46:41 <shachaf> `run mv wisdom/ngevd . && grep -i exist wisdom/* && mv ngevd wisdom/
22:46:51 <HackEgo> wisdom/boily:boily is Canadian or something. We are not sure about Canada's existence. \ wisdom/olsner:olsner seems to exist at least.
22:46:53 <oerjan> <olsner> Phantom_Hoover: no, that's norwegian <-- DON'T BE SILLY
22:47:43 <olsner> oerjan: what are you going to do about it? swat me?
22:47:56 <elliott> I'm not sure Ottawa exists.
22:48:06 <elliott> for a start, I'm not sure the plane ais523 was on actually flew to Ottawa and not some kind of fake Ottawa.
22:48:13 <elliott> also, I'm not sure ais523 exists, so his testimony is unreliable.
22:48:40 <ais523> elliott: well, either a fake ottawa is distinguishable from a real ottawa, or indistinguishable
22:48:50 <olsner> let's just define Ottawa as whatever he flew to
22:48:56 <ais523> if it's indistinguishable, then it's reasonably philosophically an ottawa
22:49:10 <elliott> ais523: it's distinguishable by not being the real ottawa, by definition
22:49:32 <ais523> otoh, there's no proof that the plane landed in canada
22:50:00 <ais523> although I have enough of a grasp of geography, combined with looking out of the windows a lot, to suspect that it's at least approximately geographically in the area where Canada is rumoured to be
22:50:31 <ais523> elliott: also, as for not being sure I exist: the alternative is that someone invented me, and that thought is horrifying
22:50:55 <olsner> err, but if someone invented you, you would exist
22:52:47 <olsner> in fact, with all this talking about whether canada exists or not, haven't we invented canada?
22:52:48 <ais523> well doesn't the fact that I accidentally assumed I exist
22:52:51 <ais523> mean that I'm not lying about it?
22:53:03 <ais523> olsner: do we know enough about Canada to be able to invent it via mere description?
22:53:31 <ais523> the salient facts about it are that it's rumoured to exist, there are esolangers who claim to be aware of it (even inside it), and that it's won Agora
22:54:27 <olsner> hmm, how much do you need to know about something in order to invent it? do we have to know that we invented it? do we have to know its name even?
22:54:50 <ais523> I guess you need to be able to come up with plausible replies to questions about it
22:54:55 <ais523> "I don't know" stops being plausible after a while
22:55:02 <lmt> i believe these concerns are adequately addressed in Tlön, Uqbar, Orbis Tertius
22:55:51 <elliott> ais523: we could apply an ontological argument type thing
22:56:05 <elliott> canada is the greatest thing, something is greater if it really exists as opposed to just being rumoured to exist...
22:56:19 <elliott> or perhaps all we need to do is agree that it is more canadian for something to exist than for it not to exist
22:56:24 <elliott> and obviously, canada is maximally canadian
22:56:59 <ais523> elliott: this reminds me of the argument that god must exist due to being defined as having all positive properties, including existence
22:57:58 <shachaf> I liked Raymond Smullyan's arguments about that!
22:58:11 <olsner> i somehow think that canada would become more canadian the less it exists
22:58:24 <elliott> ais523: yes, that's the ontological argument...
22:58:40 <ais523> elliott: wow, it actually has a /name/?
22:58:51 <ais523> shachaf: I probably learned about it from Smullyan
22:58:52 <elliott> ais523: of course, Gödel even made his own version
22:58:59 <ais523> elliott: it strikes me as not particularly convincing
22:59:15 <lmt> everyone made a version
22:59:21 <lmt> it's a popular argument
22:59:37 <shachaf> He had some fun variations.
22:59:39 <elliott> ais523: "the argument does not, to a modern mind, seem very convincing, but it is easier to feel convinced that it must be fallacious than it is to find out precisely where the fallacy lies" -- Bertrand Russell -- Wikipedia
23:00:05 <lmt> a bunch of famouse people after russell worked with it
23:00:19 <ais523> elliott: the problem is that it's a circular argument
23:00:35 <ais523> in order for it even to matter that something has particular properties, it first has to exist
23:00:45 <elliott> what do you mean by "matter"?
23:01:06 <lmt> i define unicorns as horses with horns that exist
23:01:15 <lmt> therefore, unicorns exist
23:02:00 <ais523> elliott: if something doesn't exist, then the universe is the same regardless of what properties it would have if it existed
23:02:20 <elliott> lmt: that doesn't really match the structure of the ontological argument at all
23:02:25 <shachaf> ais523: imo everything exists
23:02:42 <elliott> in particular, you don't even assert that the definition of unicorns implies that unicorns necessarily exist
23:03:13 <lmt> but it does
23:03:14 <kmc> i smell curry's paradox
23:03:17 <lmt> because it says "they exist"
23:03:20 <lmt> right in the definition
23:03:29 <ais523> kmc: which one is that?
23:03:37 <shachaf> ais523: "If this sentence is true then 1=0"
23:03:41 <elliott> kmc: I don't think the ontological argument is like Curry's paradox, really
23:03:58 <elliott> I'm not saying there aren't good arguments against it but I don't think these are it :P
23:03:58 <shachaf> Existence is not a property that some things have and some things don't.
23:04:24 <shachaf> It isn't the case that there are unicorns, but none of them have the property of existence.
23:04:27 <shachaf> There are just no unicorns.
23:04:31 <ais523> curry's paradox is logically equivalent to epimenides' paradox, isn't it?
23:04:48 <elliott> shachaf: I think it's more what you call "god" in the argument is equivalent to "idea_of_god"
23:04:58 <Arc_Koen> shachaf: who are you kidding? ALL unicorns have the property of existence
23:04:59 <elliott> and then the predicate of it existing means that "something which matches this idea exists".
23:05:31 <ais523> Arc_Koen: that statement is one I think everyone agrees with
23:05:50 <olsner> all unicorns have all properties?
23:06:39 <shachaf> Because P = P -> Q ---> P = !P || Q?
23:06:41 <ais523> about the closest I could get to what I was looking for is "http://en.wikipedia.org/wiki/List_of_fictional_ungulates"
23:07:02 <elliott> ais523: I don't think everyone accepts that every element of the empty set has all properties
23:07:23 <ais523> I was trying to find a list of fictional unicorns
23:07:24 <elliott> like I think paraconsistent type stuff might avoid that?
23:07:49 <lmt> oh Everyone knows that dragons don't exist. But while this simplistic
23:07:49 <lmt> formulation may satisfy the layman, it does not suffice for the scientific
23:07:49 <lmt> mind. The School of Higher Neantical Nillity is in fact wholly unconcerned
23:07:49 <lmt> with what ____does exist. Indeed, the banality of existence has been
23:07:49 <lmt> so amply demonstrated, there is no need for us to discuss it any further
23:07:51 <lmt> here. The brilliant Cerebron, attacking the problem analytically,
23:07:54 <lmt> discovered three distinct kinds of dragon: the mythical, the chimerical,
23:07:56 <lmt> and the purely hypothetical. They were all, one might say, nonexistent,
23:07:59 <lmt> but each nonexisted in an entirely different way ...
23:08:00 <ais523> but Wikipedia appears not to list them
23:08:01 <lmt> -- Stanislaw Lem, "Cyberiad"
23:08:21 <ais523> lmt: no copyvios, please
23:09:00 <lmt> http://en.wikipedia.org/wiki/List_of_legendary_creatures_by_type
23:09:08 <lmt> Unicorn is listed under "Fertility and human sexuality"
23:09:16 <lmt> omg my innocence :o
23:09:21 <ais523> lmt: the innuendos are far too easy
23:09:30 <olsner> Borges commented, "Not granting me the Nobel Prize has become a Scandinavian tradition; since I was born they have not been granting it to me."
23:10:25 -!- epicmonkey has quit (Ping timeout: 245 seconds).
23:10:27 <olsner> now that he's dead, are we still observing the tradition or did that become impossible?
23:10:58 <ais523> olsner: now I'm reminded of the headmaster at my secondary school attempting to create traditions via decree
23:11:10 <ais523> if the decree is effective enough, it might eventually become a tradition
23:11:17 <ais523> but you can't immediately cause something to become a tradition
23:12:53 <ais523> "how do you know that unicorns exist? well, have you ever seen a unicorn that /doesn't/ exist?"
23:13:28 <olsner> did you make that up now or is it a quote?
23:13:52 <ais523> it's not a direct quote of anything, but it may have been influenced by things I saw in the past
23:14:07 <shachaf> Everything you say has been influenced by things you saw in the past.
23:14:39 <olsner> now, if he doesn't exist, can he have seen things in the past?
23:15:09 <kmc> mouth full of bees :'(
23:15:56 <olsner> mouthless alien with two eyes and eight brains
23:16:17 <lmt> https://www.youtube.com/watch?v=6dQtFpuylpA
23:16:21 <shachaf> Bee bearding is the practice of wearing several hundred thousand honey bees on the face, usually as a sideshow-type demonstration at agricultural shows.
23:16:25 <shachaf> Several hundred thousand?!
23:16:42 <lmt> shachaf: that just means a hive
23:16:56 <lmt> hives are pretty big
23:17:04 <shachaf> alt. bees are pretty small
23:17:39 <lmt> they are very small if you squash them
23:18:49 <oerjan> if you squash some, the rest will go into a rage iiuc, which is a bad idea if you are using them for your beard.
23:19:38 <ais523> well wearing bees strikes me as a bad idea in the first place
23:19:39 <lmt> that's why you squash them before guing them to your chin
23:19:49 <lmt> using a commercial grade bee squasher
23:19:59 <ais523> I would recommend against it
23:20:10 <oerjan> ais523: so one should beware of wearing bees?
23:20:11 <lmt> it's better than wearing nothing at all
23:20:25 <shachaf> imo wearing nothing at all is preferable to wearing bees
23:22:20 <lmt> wearing bees is worth it just for the puns
23:23:11 <lmt> http://totallycoolpix.com/wp-content/uploads/2011/23072011_coolest_pix_week_29/week29_022.jpg
23:23:24 <lmt> i have no idea what that display says
23:23:35 <lmt> is it a bee counter
23:23:53 * shachaf is not a fan of bees on a personal level.
23:23:59 <elliott> presumably it's counting down time or something
23:24:05 <lmt> bees are your friends though
23:24:10 <shachaf> I'll let them bee if they let me be.
23:24:28 <oerjan> lmt: shachaf doesn't _want_ to befriend them
23:24:31 <kmc> presumably it counts down and then displays BEEEES
23:24:39 <lmt> oh it is a scale
23:24:49 <lmt> Wang won the competition after attracting 26 kg (57 lbs) of bees on his body in 60 minutes
23:25:13 <olsner> how does warring weary bees compare to wearing wary bees?
23:25:50 <lmt> if i had some bees, i would bee your friend
23:26:00 <kmc> ais523: it could be like that ant problem from ICFP
23:26:15 <ais523> oh yes, although that one seems kind-of limited
23:26:35 <kmc> we did that one in school except the prof came up with a strange variation
23:26:48 <kmc> if an ant is surrounded, it dies and turns into 6 food
23:26:54 <shachaf> ais523: would that have a busy bee function
23:26:59 <kmc> but additionally, you can drop 1 food on the middle of your ant hill and get 1 new ant
23:27:08 -!- Borgrel has joined.
23:27:18 <kmc> so the winning strategy is...
23:27:28 <Sgeo> ZipList fs <*> ZipList xs = ZipList (zipWith id fs xs)
23:27:41 <Sgeo> Any particular reason to use id instead of ($)?
23:27:50 <Sgeo> ($) seems like it describes the intent more
23:28:15 <ais523> ($) is the same as id, right?
23:28:20 <kmc> @let ego = id
23:28:23 <ais523> it's obvious I'm just not thinking atm
23:28:30 <shachaf> ais523: Yes, with a different type.
23:28:31 <ais523> so I'm not sure which way it's obvious
23:28:34 <olsner> shachaf: if you use spaces where you don't have to
23:28:35 <Sgeo> ($) is type restricted id
23:28:41 <ais523> oh, only works on functions
23:28:59 <shachaf> There's also this variation: ($) !f = \x -> f x
23:29:02 <ais523> this reminded me of someone talking about Verity's potential issues with accepting incorrect code
23:29:12 <ais523> someone accidentally wrote bitwise OR (|) rather than parallel composition (||)
23:29:14 <olsner> "($)" saves one character compared to " id "
23:29:22 <ais523> and it didn't reject it because it treats commands as 0-bit numbers
23:29:29 <ais523> and it knows how to bitwise OR those
23:29:40 <ais523> but the synthesizer complained that it didn't
23:29:44 <ais523> which is interesting, I guess
23:29:47 <kmc> is verity the categorical dual of coverity
23:29:50 <ais523> should probably add a specific error message for that
23:29:56 <ais523> kmc: I don't think so, although that would be interesting
23:29:59 <HackEgo> Borgrel: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
23:30:14 <ais523> it's more like algol disguised as not-algol, with a particularly hardware-focused set of primitives
23:30:21 <Sgeo> kmc, sorry from derailing you from your game
23:30:24 <elliott> shachaf: isn't ($) defined as f $ x = f x?
23:30:25 <olsner> kmc: would that be something that hides bugs in code instead of finding them?
23:30:27 <Sgeo> What's the winning strategy?
23:30:34 <kmc> "algol disguised as not-algol" <-- like every imperative language am i rite
23:30:36 <shachaf> elliott: Yes. That's why it's a variation.
23:30:40 <ais523> Sgeo: well in game semantics you're not trying to win
23:30:43 <ais523> just reach the end of the game
23:30:45 <elliott> <ais523> ($) is the same as id, right?
23:30:48 <elliott> <shachaf> ais523: Yes, with a different type.
23:31:01 <kmc> rude to show your ⊥ in public
23:31:02 <Sgeo> > (undefined id) `seq` ()
23:31:12 <elliott> kmc: well most imperative languages don't disguise themselves as ML
23:31:15 <elliott> which is what verity seems to do
23:31:19 <ais523> Sgeo: you have that backwards
23:31:23 <ais523> > (id undefined) `seq` ()
23:31:31 <ais523> not that it makes a difference, but…
23:31:39 <ais523> well it does, the test wouldn't work otherwise
23:31:44 <ais523> just the broken test produced the same result as the correct test
23:31:53 <ais523> > (undefined `id`) `seq` ()
23:32:02 <ais523> ooh, didn't know Haskell would let me do that
23:32:16 <ais523> this seems like a minor technique in writing obfuscated Haskell
23:32:36 <shachaf> What, sections of non-operators?
23:33:12 <ais523> it's like currying just backwards
23:33:28 <lambdabot> <hint>:1:6: parse error on input `)'
23:33:33 <ais523> OK, I can't section it both ways
23:33:38 <shachaf> ais523: I consider that a bug in Haskell. :-(
23:33:42 <lambdabot> <hint>:1:4: parse error on input `('
23:33:52 <elliott> I wish `expression` worked.
23:33:53 <shachaf> `` takes an identifier, not an expression.
23:33:55 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `: not found
23:34:10 <ais523> elliott: clearly you need directed quotes
23:34:25 <ais523> or alternatively, both ` and “
23:34:29 <elliott> IIRC Haskell papers usually do that
23:34:30 <oerjan> elliott: `(...)` wouldn't be ambiguous
23:34:31 <shachaf> elliott: Isn't it great when you don't have to write parentheses somewhere?
23:34:38 <elliott> oerjan: but it would be ugly.
23:34:40 <shachaf> Like matching on (a, x:xs)
23:34:42 <ais523> elliott: yeah, because LaTeX does that by default and it's awkward to tell it not to
23:34:53 <elliott> oerjan: I think I've had a good example of `f x` being useful before
23:35:00 <elliott> maybe involving a higher-order function f
23:35:29 <elliott> maybe just only allow nesting if you use parens.
23:35:33 <oerjan> ais523: i like to use (`f` x) rather than flip f x sometimes
23:35:45 <elliott> so `...`...`...` nests iff you have ( in the first ... and ) in the last.
23:36:18 <nooodl> x `compare (`on`) fst` y
23:36:36 <elliott> nooodl: that's compare on fst
23:37:09 <nooodl> oh the ()s were "nesting"
23:37:39 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: (compare: not found
23:37:41 <nooodl> i forgot (`f`) is a thing
23:40:39 <nooodl> wow, i scrolled up a bit and i'm tired and i read about everyone talking about how "<$> is the same as id, right?" and i thought i was going insane
23:41:12 <oerjan> > (1,2) & compare `on` fst $ (3,4)
23:41:14 <lambdabot> cannot mix `Data.Function.on' [infixl 0] and ...
23:41:36 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: on`: not found
23:41:37 <oerjan> > (1,2) & (compare `on` fst) $ (3,4)
23:42:21 <elliott> you made this bed shachaf. you have to lie in it. in fact I'm lying in it for you. it's comfortable
23:42:45 <oerjan> > (compare `on` fst) ?? (3,4) $ (1,2)
23:43:33 <lambdabot> Functor f => f (a -> b) -> a -> f b
23:44:38 <Bike> it's got too many parentheses
23:44:42 <nooodl> hmm... it's defined generally for functors?
23:45:06 <shachaf> Well, one generalisation of flip is that.
23:45:11 <shachaf> (r -> a -> b) -> a -> r -> b
23:45:28 <shachaf> There are other potential generalizations, of course.
23:45:40 <shachaf> Just like (.) can generalised to fmap or (Control.Category..)
23:45:43 <olsner> ais523: whales are ungulates and narwhals have one horn ... they must be unicorns
23:45:49 <Bike> generalerization
23:46:34 * elliott doesn't think anyone has yet used (??) for an f that isn't ((->) r)
23:47:41 <oerjan> elliott: i don't see why, it's a useful addition to applicative syntax
23:47:44 <nooodl> i guess you could generalize it for ((->) a) as well: (??') :: Functor f => (r -> f b) -> f (r -> b)
23:47:57 <elliott> oerjan: I'm not saying it's inherently unuseful
23:48:11 <nooodl> (iirc there's a word for "unuseful")
23:48:14 <elliott> I'm saying edwardk claimed it'd be useful because lens uses (a -> f b) things all over the place and as far as I know this hasn't actually turned out to be true :P
23:48:29 <elliott> nooodl: hmm, that looks like Distributive
23:48:55 <lambdabot> Perhaps you meant `distrib' (imported from Control.Lens)
23:49:22 <elliott> {--} distribute (++) "abc" "def"
23:49:25 <elliott> http://hackage.haskell.org/packages/archive/distributive/0.3.1/doc/html/Data-Distributive.html
23:49:29 <elliott> the one true generalisation of (??)????
23:49:32 <nooodl> i can't think of an implementation for (??') ugh
23:49:40 <oerjan> elliott: in fact i think ?? (maybe not with that syntax) should be a Functor method, like <$
23:49:40 <lambdabot> Functor f => f (a -> b) -> a -> f b
23:49:47 <elliott> nooodl: it's not implementable
23:50:15 <elliott> oerjan: ok but look at distribute. isn't that totally the best flip generalisation
23:50:34 <elliott> I wonder if (Functor f, Distributive g) => f (g a) -> g (f a) is doable
23:50:38 <elliott> which would be a strict generalisation of (??)
23:50:51 <elliott> that's exactly what I wrote.
23:50:57 <elliott> so distribute is a strict generalisation of (??)
23:51:07 <elliott> shachaf: you have to promise not to tell edwardk.
23:51:29 -!- Borgrel has quit (Quit: ChatZilla 0.9.90 [Pale Moon 19.0.2/20130308232348]).
23:51:55 <nooodl> "comapM [...] The dual of mapM: comapM f = fmap f . distributeM"
23:51:59 <nooodl> you people and your co's
23:52:09 <elliott> well Distributive is Cotraversable
23:53:09 <nooodl> would "comap f = fmap f . distribute = fmap f . flip", that sounds like a good function
23:53:45 <nooodl> wait. i think it's actually something extremely boring
23:55:14 <Bike> imo this channel should do numerical analysis instead of these oneliners
23:55:44 <Phantom_Hoover> why do you have to ruin something moderately cool like analysis with numbers
23:56:13 <lmt> fuck numbas
23:56:39 <lambdabot> (?f::f b1 -> b, Functor f) => f (a -> b1) -> a -> b
23:57:12 <lambdabot> Functor f => (f b1 -> b) -> f (a -> b1) -> a -> b
23:57:18 <elliott> Phantom_Hoover: flip flip is not flip
23:57:28 <lambdabot> (a -> ((a1 -> b -> c1) -> b -> a1 -> c1) -> c) -> a -> c
23:57:30 <lambdabot> Functor f => a -> f (a -> b) -> f b
23:57:32 <lambdabot> (a1 -> ((a -> b -> c) -> b -> a -> c) -> c1) -> a1 -> c1
23:57:34 <elliott> :t flip flip flip flip flip
23:57:35 <lambdabot> (a1 -> ((a -> b -> c) -> b -> a -> c) -> c1) -> a1 -> c1
23:57:40 <elliott> :t flip flip flip flip flip flip
23:57:42 <lambdabot> (a1 -> ((a -> b -> c) -> b -> a -> c) -> c1) -> a1 -> c1
23:58:01 <elliott> :t flip flip flip flip flip flip flip
23:58:02 <lambdabot> (a1 -> ((a -> b -> c) -> b -> a -> c) -> c1) -> a1 -> c1
23:58:06 <elliott> strange, I was sure that would break the cycle.
23:58:30 <lambdabot> (Functor f1, Functor f) => f ((f1 (a -> b1) -> a -> f1 b1) -> b) -> f b
23:58:36 <oerjan> :t (??) (??) (??) (??)
23:58:38 <lambdabot> (Functor f1, Functor f) => f1 ((f (a -> b) -> a -> f b) -> b1) -> f1 b1
23:58:44 <oerjan> :t (??) (??) (??) (??) (??)
23:58:46 <lambdabot> (Functor f1, Functor f) => f1 ((f (a -> b) -> a -> f b) -> b1) -> f1 b1
23:58:51 <nooodl> > let cool f = fmap f . flip in cool (\f -> [f 1, f (f 1)]) (+) 5