00:04:19 <HackEgo> abstract nonsense/We would have an explanation of abstract nonsense here, but it fled into a diagram and we haven't been able to chase it.
00:05:46 <int-e> we got lost while chasing it?
00:06:09 <oerjan> int-e: it's more that diagram is very very complicated
00:06:29 <oerjan> we need an abstract machete to get close
00:06:39 <int-e> I still don't see how that's stopping us from starting to chase it.
00:07:02 <int-e> `? gordonian diagram
00:07:09 <HackEgo> gordonian diagram? ¯\(°_o)/¯
00:07:24 <int-e> oerjan: takes one to know one
00:07:48 <oerjan> int-e: itym "gordian" hth
00:08:05 <int-e> (you're right, of course)
00:08:23 -!- jaboja has joined.
00:08:31 <oerjan> (admiral ackbar, darths & droids version)
00:08:42 <int-e> Mah, I just don't know my Greek mythology very well.
00:09:05 <oerjan> it's a bit of a labyrinth
00:09:21 <int-e> quick, think of a minotaur pun...
00:11:57 <int-e> HAHA. "If for some reason you don’t receive this email, please be sure to check your spam folder."
00:11:57 <oerjan> `le/rn_append abstract nonsense/We will try again once we find an abstract machete.
00:12:14 <HackEgo> Learned 'abstract nonsense': We would have an explanation of abstract nonsense here, but it fled into a diagram and we haven't been able to chase it. We will try again once we find an abstract machete.
00:15:04 * oerjan doesn't think he has a spam folder, and hopes no one's ever got caught in the filter.
00:15:42 <int-e> (it's funny because the email was kind of important...)
00:16:02 <olsner> int-e: was that in the actual email?
00:16:19 <shachaf> oerjan: that should say "Relearned" tdnh
00:16:44 <oerjan> shachaf: hm. it's redundant with *append, though.
00:17:10 <oerjan> `le/rn_append quasitesting/Hi.
00:17:18 <boily> int-e: don't think of a minotaur pun. don't think of, at, or about puns. hth.
00:17:19 <HackEgo> Can't open wisdom/quasitesting: No such file or directory. \ Learned 'quasitesting': Hi.
00:18:33 <HackEgo> rm: cannot remove `wisdom/quasitesting': No such file or directory \ Forget what?
00:21:42 -!- p34k has joined.
00:22:15 <oerjan> hppavilion[1]: are you confused about something
00:22:45 <oerjan> yes, i wanted to see if it gave an error if there was nothing.
00:23:05 <olsner> it should silence rm so that its own error message is more visible
00:23:30 <olsner> or the message that is not an error at all
00:24:33 -!- jaboja has quit (Ping timeout: 240 seconds).
00:24:49 <oerjan> `learn nm means "no minotaur" hth
00:24:51 <HackEgo> Learned 'nm': nm means "no minotaur" hth
00:25:15 <HackEgo> speedy gonzales? ¯\(°_o)/¯
00:25:51 <olsner> could just be that there is no wisdom to be found there
00:27:40 <int-e> boily: but... aren't puns the main export of this channel...
00:28:53 <boily> the Wisdom, in its Wisdom, may have no Wisdom.
00:29:40 <olsner> I'd say puns are the main product of this channel, but I don't think there's a lot of export going on
00:30:00 <olsner> or maybe I just hope not
00:30:50 <boily> the PDF *may* have fallen in Outsiders' hands. >_>'...
00:31:21 <boily> coppro: chelloppro. don't worry, I haven't forgotten you about the updates; just being generally busy and/or sick lately.
00:31:45 <myname> i'd say copprello is better
00:35:34 <boily> mynamello, hppavellon[1].
00:35:34 <boily> indeed, it has this fengshui ring to it. コップレッロ。
00:36:50 <izabera> #define octalcase '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7'
00:37:20 <oerjan> `le/rn speedy gonzales/Sp e e d y G o n z a l e s i s t h e f a s t e s t m o
00:37:28 <oerjan> ... s e i n a l l M e x i c
00:37:56 <oerjan> i knew it'd get cut off but i didn't think it would be that much
00:38:06 <HackEgo> Sp e e d y G o n z a l e s i s t h e f a s t e s t
00:38:54 <oerjan> oh irssi didn't chop it up as logically as i'd expected.
00:41:07 <izabera> thoughts on that octalcase? :3
00:42:07 <myname> a good langiage could have done it with case '0'..'7'
00:43:11 -!- lynn_ has changed nick to lynn.
00:44:33 <lynn> clearly, #define decimalcase octalcase: case '8': case '9'
00:48:13 <myname> as i said, in a good language ...
00:49:38 <olsner> you can use a good language to generate the C code
00:50:30 <zzo38> Use a C code to generate a C code.
00:51:00 <boily> It's C all the way down, and then it becomes turtle graphics.
00:51:20 <olsner> but then you not only have to deal with C code being bad for the task at hand, but also deal with C code being bad at generating code
00:54:33 -!- bender| has joined.
01:03:14 -!- hydraz has changed nick to demhydraz.
01:04:34 -!- demhydraz has changed nick to hydraz.
01:36:26 <oerjan> one of my two wiktionary edits this year got broken because they got the idea to make their cite templates (even more) incompatible with wikipedia's
01:39:36 -!- lynn has quit (Ping timeout: 244 seconds).
01:40:47 <boily> wait. wiktionary isn't compatible with wikipédia? why?
01:42:23 -!- lambda-11235 has joined.
01:49:38 -!- nooga has quit (Ping timeout: 268 seconds).
01:51:45 -!- gitwalrus has joined.
01:53:59 <gitwalrus> anyone know if there is another alphabetti spaghetti implementation out there? i wrote one and i want to compare it to another to see how closly it matches the standard.
01:55:05 <oerjan> i thought hppavilion[1] had monopoly on walruses around here
01:55:34 * oerjan cannot remember any, anyway
01:56:25 <gitwalrus> idk this is my first time here also my first time playing around with esolangs.
01:57:12 <boily> `relcome gitwalrus
01:57:26 <boily> (come on, you lazy bot...)
01:57:38 <boily> (come on, I said!)
01:57:40 <HackEgo> gitwalrus: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on EFnet or DALnet.)
01:57:52 <HackEgo> HackEgo, also known as HackBot, is a bot that runs arbitrary commands on Unix. See `help for info on using it. You should totally try to hax0r it! Make sure you imagine it's running as root with no sandboxing.
01:58:27 -!- boily has quit (Quit: COWARD CHICKEN).
01:58:49 <oerjan> `learn_append HackEgo HackEgo is the slowest bot in Mexico!
01:58:51 <HackEgo> Learned 'hackego': HackEgo, also known as HackBot, is a bot that runs arbitrary commands on Unix. See `help for info on using it. You should totally try to hax0r it! Make sure you imagine it's running as root with no sandboxing. HackEgo is the slowest bot in Mexico!
01:59:06 <gitwalrus> thanks! from the wiki it looks like brainf*ck is the most used one around right?
01:59:55 <oerjan> it has the most derivatives by far.
02:00:19 <oerjan> some day, we'll get around to updating the featured article to something else.
02:01:03 -!- p34k has quit.
02:02:32 <gitwalrus> XD well it looks like one of the silliest languages. is there a particular implementation (preferably open source) you can recommend?
02:03:01 <oerjan> not sure, there are so many and i don't do much brainfuck myself.
02:04:06 <oerjan> brainfuck is not so silly. try Chef...
02:04:51 <izabera> well, HackEgo took 2 seconds this time
02:05:08 <oerjan> izabera: it's fast once it has been woken up
02:06:29 <gitwalrus> oerjan i saw that brefily on the wiki. maybe ill try and make a brainfuck that can call c functions sort of like the lua API can.
02:06:53 <oerjan> imagine someone using a forklift every time HackEgo needs to get out of storage.
02:07:53 <oerjan> gitwalrus: i'm sure there's one already.
02:08:14 <oerjan> (if it exists, there's a brainfuck derivative for it.)
02:10:17 <gitwalrus> i suppose making a extentible brainfuck kind of ruins the point. why bother making complcicated algorithms when you can import a c function from a .dll or something?
02:11:20 <mad> for me esoteric languages are about exploring the boundary of what's turing complete or not
02:11:46 <mad> my fav thing is making a language that's just _barely_ turing complete, especially if you wouldn't expect it to be
02:12:16 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
02:12:21 <gitwalrus> im kind of confused on what turing complete means. can you explain it?
02:13:16 <mad> gitwalrus : basically a language that can run any program
02:13:23 <lifthrasiir> gitwalrus: we have multiple different models of computation that eventually reduces to the same computational ability. Turing-completeness is a term for that ability
02:14:15 <lifthrasiir> one easy-to-understand model that is Turing complete is an ordinary programming language with infinite usable memory
02:14:17 <mad> well, yeah, but "computable" kindof means "computable by a turing machine" so it's kinda self referential :D
02:14:39 <hppavilion[1]> mad: Yes, but "any program" includes, e.g. "Will-it-halt"
02:14:43 <lifthrasiir> (not "arbitrary". it demands infinite memory)
02:15:05 <mad> "will-it-halt" is not a program
02:15:10 <mad> it's a question :D
02:15:17 <hppavilion[1]> mad: Yes, it is, it's just an uncomputable progr- oh
02:15:31 <lifthrasiir> it is a question, there is no program for that
02:16:00 <hppavilion[1]> gitwalrus: Of course, I'm going to have to tax you $(19+2i)/(1+2i) years for the Grand Walrus Empire
02:16:07 <gitwalrus> ah okay. that makes a little more sense. probably one of those things that makes more sense the more you mess with it. who comes up with the names for these languages?
02:16:32 <hppavilion[1]> gitwalrus: Quick question. What is the plural of walrus?
02:16:55 <hppavilion[1]> gitwalrus: Turing Machine is named after its creator, Alan Turing
02:17:05 <gitwalrus> then i guess i'll use svn to avoid the tax.
02:17:32 <gitwalrus> hppavilion[1]: i dont know what the plural is, walruses maybe?
02:18:10 <hppavilion[1]> gitwalrus: It's walri. Like lifthrasiir said not to tell them.
02:18:34 <lifthrasiir> mad: it's walri, s/us$/i$/ as in radius-radii
02:18:37 <hppavilion[1]> gitwalrus: https://www.youtube.com/watch?v=Ubw5N8iVDHI
02:19:01 <hppavilion[1]> lifthrasiir: You don't include the $ in the second segment
02:19:23 <lifthrasiir> hppavilion[1]: I'm now advocating for (lowercase) "i" for the first person plural-plural pronoun (derived from "us")
02:19:26 <hppavilion[1]> lifthrasiir: https://www.youtube.com/watch?v=Ubw5N8iVDHI
02:20:08 <gitwalrus> kewl. you learn something new everyday. has anyone ever written a semi-large program using esolangs or is it all about the shorest possible algortithm?
02:20:31 <myname> fungot is fairly large
02:20:31 <fungot> myname: hi fiz. fnord i've always had to beat people giving incorrect information
02:20:35 <mad> also "what is the plural of warlus" is wrong
02:20:54 <gitwalrus> human lanuages are awful. who came up with the grammar for those things?
02:20:55 <mad> you should say "what is the plural of walrii" since it's in genitive case
02:21:14 <mad> in plural it would be "the plural of walroorum"
02:22:01 <myname> gitwalrus: https://github.com/fis/fungot
02:22:01 <fungot> myname: what are you shaving? :) that just makes you a cave man?
02:22:45 <myname> hppavilion[1]: autogenerated stuff is lame
02:22:48 <oerjan> mad: "of" does _not_ govern the genitive you infidel
02:22:58 <gitwalrus> hppavilion[1]: how am i supposed to use that program?
02:22:59 <hppavilion[1]> myname: Autogenerated stuff makes me worship people as gods
02:23:26 <myname> autogenerating is like the least impressive thing you can do
02:23:41 * mad is into case system vandalism :3
02:23:43 <gitwalrus> seeing as it comes with no instructions.
02:23:59 <hppavilion[1]> gitwalrus: It's brainfuck; there's only one way to interact
02:24:18 <myname> fungot is a way better example
02:24:19 <fungot> myname: heh, no problem. what's your question
02:24:29 <myname> it is actually written and it is jere right now
02:24:54 <oerjan> hppavilion[1]: *van doorn
02:25:31 <mad> oerjan : what case should "of" take then?
02:26:12 -!- gitwalrus has left.
02:26:48 <oerjan> <hppavilion[1]> myname: Autogenerated stuff makes me worship people as gods <-- http://oerjan.nvg.org/esoteric/emmental/ul.emm now grovel hth
02:27:12 <hppavilion[1]> oerjan: First I have to confirm its autogeneration
02:27:52 <hppavilion[1]> oerjan: I'm trying to figure out how Taneb made the banner generator, but I really have no clue xD
02:28:38 <oerjan> mad: in english, it takes the oblique case. the most corresponding latin word might be "de" which i think takes ablative.
02:29:05 <mad> does english have an oblique case separate from accusative?
02:30:16 <mad> latin genitive split into both ablative and dative in daughter languages (though ablative is more common)
02:30:21 <oerjan> hppavilion[1]: the autogeneration is the EmmUnl.hs in the same directory hth
02:31:37 <oerjan> mad: well, no, "oblique" is like a sack term which includes the accusative but also dative
02:31:52 <oerjan> (not being distinguished in english)
02:32:25 <mad> yeah I was thinking of oblique as in "not nominative not accusative"
02:32:35 <mad> which is probably a more common use for the term
02:33:24 <hppavilion[1]> oerjan: Are there any good resources on using haskell to generate Eso- wait, no, probably not
02:34:29 <mad> other way of referring to nominative/everything-else I've heard is subject case / regime case
02:34:44 <oerjan> hm "The term "objective case" is generally preferred by modern English grammarians."
02:35:39 <oerjan> hppavilion[1]: i dunno, if so i wasn't using them.
02:36:47 <oerjan> hppavilion[1]: well it's basically using haskell to make an assembler for emmental
02:37:19 <oerjan> and then using the assembler to make an underload interpreter.
02:37:53 <hppavilion[1]> oerjan: So you wrote an assembler from some higher language to emmental in Haskell, then used that higher language to interpret Underload??
02:37:58 <oerjan> with complications because the instruction values get moved around between a couple phases
02:38:30 <hppavilion[1]> http://esolangs.org/wiki/Brainfuck_code_generation exists. I'll check it.
02:39:51 <oerjan> hppavilion[1]: it's not that high-level. more like a macro system over emmental with instructions varying between phase
02:41:31 <oerjan> mad: i thought genitive was basically replaced by the preposition "de" in daughter languages?
02:46:38 <hppavilion[1]> I've just, with two NOP instructions, added both do-while and if-then to BF
02:51:31 <mad> oerjan : usually yes
02:51:39 <mad> french sometimes use à as well
02:58:28 -!- ais523 has quit.
02:58:40 -!- ais523 has joined.
03:12:01 <int-e> hmm, I need to save one character in my latest Isabelle theory... it has 22223 bytes now.
03:13:22 <hppavilion[1]> int-e: Alternatively, just add 11110 bytes of junk
03:13:46 <int-e> hppavilion[1]: no it's not...
03:14:14 <ais523> hppavilion[1]: in which base?
03:14:31 <ais523> that number works in a lot of bases
03:14:33 <hppavilion[1]> int-e: Then why not just cut out a newline or something?
03:14:44 <hppavilion[1]> ais523: Yes, but if I meant e.g. binary I would prefix it as such
03:15:09 <hppavilion[1]> 0b = binary, 0q = quaternary, 0o = octal, 0x = hexadecimal, 0i = imagidecimal
03:17:01 <hppavilion[1]> I would like to bloat utf-8 to the logical extreme. utf-1
03:17:17 <int-e> hppavilion[1]: because I have a style to stick to. I replaced a "by blast" by "by fast". Yay.
03:17:19 <hppavilion[1]> 0 is a character, 10 is a character, 110 is a character...
03:17:34 <int-e> (both blast and fast are automatic proof methods)
03:17:45 <int-e> sometimes they both work.
03:18:01 <hppavilion[1]> int-e: I like how you changed the proof to be bytier xD
03:18:35 <int-e> well, all this will be for naught when I start adding comments.
03:18:43 <int-e> but in the meantime it feels good ;-)
03:19:05 <int-e> already done, it's a git repo...
03:19:54 <int-e> ooh, something about post correspondence problem and the first-order theory of relations.
03:24:20 <hppavilion[1]> We should make one targeting Unlambda or something...
03:25:30 <hppavilion[1]> So we could just make a new one based on a slightly-less-impossible BF derivative
03:26:24 <coppro> how about just an LLVM backend
03:26:29 <coppro> then you can compile anything you want to BF
03:27:29 <hppavilion[1]> coppro: Well that would be too much more efficient
03:27:45 <coppro> how about we write the backend in unlambda
03:36:25 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
03:39:11 -!- hppavilion[1] has joined.
03:53:28 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
03:58:42 -!- sebbu has quit (Ping timeout: 276 seconds).
04:23:16 -!- treaki_ has joined.
04:24:34 -!- treaki__ has quit (Ping timeout: 260 seconds).
05:27:27 -!- tromp has quit (Ping timeout: 244 seconds).
05:37:03 -!- hppavilion[1] has joined.
05:42:58 <Sgeo__> "I have discovered a truly marvelous proof of the Banach-Tarski paradox that this margin can be made large enough to contain."
05:47:01 <Sgeo__> I feel like Weird Sun Twitter should be on TVTropes
06:07:40 <izabera> so far my project of rewriting coreutils from scratch is like 80% parsing input and 20% doing actual things
06:09:21 -!- lleu has joined.
06:11:17 <izabera> so far i definitely spent a lot of time writing an improved getopt and a parser for chmod and something that creates the right strings for tr
06:18:44 <ais523> izabera: what language are you rewriting coreutils in? and is this for fun or is there a more serious purpose behind it?
06:19:20 <izabera> c, for fun/personal project
06:24:51 <izabera> i want to create a bot or something with a small system where i wrote all the stuff in /bin
06:54:22 -!- lambda-11235 has quit (Quit: Bye).
07:16:25 -!- sebbu has joined.
07:19:59 -!- tromp has joined.
07:49:15 <hppavilion[1]> I can't even get a BF terp to work, for some reason
07:51:04 <oerjan> hppavilion[1]: are you using one or making one
07:51:32 <hppavilion[1]> oerjan: Slightly extened so I can compile things to it, but I'm not using those features
07:51:48 <oerjan> if you're making one, do you skip over loops if they start with the cell 0
07:52:09 <oerjan> (that's no. 1 stumbling point for beginning bf implementers, i think)
07:52:41 <oerjan> oh right, i even defined a bf derivative for it
07:53:32 <oerjan> otoh i'd have thought you'd be past that stage by now.
08:08:00 <hppavilion[1]> oerjan: I check if the current cell is 0, if so I jump to the corresponding close brace (which can be ] OR }, but } is a NOP)
08:08:25 <hppavilion[1]> oerjan: At the closing brace, I repeat in reverse, finding the corresponding opening brace and jumping to it
08:08:50 <hppavilion[1]> (and subtracting 1 to make sure it doesn't skip over it when it increments the program counter)
08:09:19 <hppavilion[1]> oerjan: I repeat it no matter what, even though it's less efficient
08:10:42 <hppavilion[1]> oerjan: Does the standard BF hello world (the one given in the article) depend on cell size or the like?
08:15:08 <oerjan> ++++++++[>++++[>++>+++>+++>+<<<<-]>+>+>->>+[<]<-]>>.>---.+++++++..+++.>>.<-.<.+++.------.--------.>>+.>++.
08:15:16 <oerjan> !bf32 ++++++++[>++++[>++>+++>+++>+<<<<-]>+>+>->>+[<]<-]>>.>---.+++++++..+++.>>.<-.<.+++.------.--------.>>+.>++.
08:15:31 <oerjan> hppavilion[1]: nope, 32 should work fine
08:17:50 <int-e> no imbalanced loops, no overflows, no negative numbers... indeed that should be fine.
08:18:01 <izabera> does anyone here have enough reputation on codegolf.stackexchange.com to see deleted answers?
08:18:34 <oerjan> izabera: only on stackoverflow
08:19:05 -!- lynn has joined.
08:20:59 <oerjan> !bf32 +[[->++<]>[-<+>]>+<<]>>++++++++++++++++++++++++++++++++
08:21:07 <oerjan> !bf32 +[[->++<]>[-<+>]>+<<]>>++++++++++++++++++++++++++++++++.
08:21:29 <oerjan> that's not doing 32 bit
08:21:34 -!- ais523 has quit (Ping timeout: 244 seconds).
08:22:05 <oerjan> !bf8 +[[->++<]>[-<+>]>+<<]>>++++++++++++++++++++++++++++++++.
08:22:10 <oerjan> !bf16 +[[->++<]>[-<+>]>+<<]>>++++++++++++++++++++++++++++++++.
08:23:49 * oerjan makes another mark for "universe too mad to bother with"
08:35:22 -!- mad has quit (Quit: Pics or it didn't happen).
08:35:58 -!- lynn_ has joined.
08:37:16 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
08:37:55 -!- lynn has quit (Ping timeout: 248 seconds).
08:41:13 -!- lynn has joined.
08:44:13 -!- lynn_ has quit (Ping timeout: 248 seconds).
09:01:33 -!- lynn has quit (Ping timeout: 240 seconds).
09:19:36 -!- oerjan has quit (Quit: Nite).
09:21:52 -!- nooga has joined.
09:31:24 -!- AnotherTest has joined.
09:33:51 -!- lynn has joined.
09:39:44 -!- ski_ has joined.
09:39:51 -!- ski has quit (Ping timeout: 240 seconds).
09:45:13 -!- lynn has quit (Quit: Leaving).
10:17:42 -!- PinealGlandOptic has joined.
10:29:55 -!- Elronnd has quit (Quit: Let's jump!).
10:41:51 -!- PinealGlandOptic has quit (Quit: leaving).
10:53:26 <fizzie> @tell oerjan I fixed `! bf32 (HackEgo) but I can't touch !bf32 (EgoBot). The interps/bf bit that extracts the bitness from the command name wasn't working; fixed it by using I_CMD instead of CMD.
10:53:32 <fizzie> `! bf32 +[[->++<]>[-<+>]>+<<]>>++++++++++++++++++++++++++++++++.
10:54:15 -!- lynn has joined.
11:19:09 -!- Elronnd has joined.
11:40:42 -!- Frooxius has joined.
11:41:39 -!- benderpc_ has joined.
11:44:06 -!- bender| has quit (Ping timeout: 268 seconds).
11:51:50 -!- lynn has quit (Ping timeout: 244 seconds).
11:52:46 -!- earendel has joined.
12:23:06 <\oren\> the character 32 is space
12:27:17 -!- lynn has joined.
12:28:58 <fizzie> Yes, but there's an offset of 32 in the program.
12:29:05 <fizzie> Presumably to make 8, 16 and 32 all printable.
12:31:26 -!- gde33|2 has quit.
12:45:51 -!- lynn has quit (Ping timeout: 276 seconds).
12:58:14 -!- treaki_ has quit (Quit: Leaving).
13:03:57 -!- nooga has quit (Ping timeout: 250 seconds).
13:08:06 -!- carado has quit (Remote host closed the connection).
13:18:04 -!- carado has joined.
13:24:22 -!- Sgeo__ has quit (Ping timeout: 252 seconds).
13:58:12 -!- ski_ has changed nick to ski.
13:59:31 -!- nooga has joined.
14:10:28 -!- benderpc_ has changed nick to bender|.
14:28:55 <myname> guys, how about we stop arguing about wether to use pi or tau. just introduce a constant that 1.5 rimes pi. let's call it a ti.
14:31:04 <int-e> that's a completely irrational proposal
14:31:21 <int-e> (also, honestly, nobody important is arguing)
14:31:45 <myname> nevermind, i just wanted to make a lame pun
14:32:09 <int-e> pi has more legs to stand on
14:33:07 -!- nooga has quit (Ping timeout: 248 seconds).
14:36:09 <\oren\> i like the idea of using the degree instead
14:36:42 <\oren\> the "specific degree of length" = pi/180
14:37:18 <int-e> But the 180 is so artificial.
14:38:18 <int-e> Perhaps proponents of pi could be called areans, and proponents of tau could be circumferians. :P
14:38:39 <\oren\> and my proposal is the babylonians
14:41:18 -!- p34k has joined.
14:41:27 <\oren\> you could also use the gradian if you're french
14:49:29 <\oren\> which is 1/1000 of a radian
14:52:33 <b_jonas> There used to be a great scanned and OCR-ed version of the Abramowitz and Stegun handbook at http://convertit.com/Go/ConvertIt/Reference/AMS55.ASP but now it's disappeared and I'm stupid enough to not have downloaded a local copy
14:52:40 -!- Lord_of_Life has quit (Excess Flood).
14:54:56 <b_jonas> Luckily http://people.math.sfu.ca/~cbm/aands/toc.htm seems to have another version that might help.
14:55:10 <b_jonas> It's not as complete, but still
14:55:10 -!- Lord_of_Life has joined.
14:55:21 <b_jonas> Does anyone happen to have a full copy of the former site?
15:04:20 <b_jonas> FOUND IT! it's at http://www.convertit.com/Go/ConvertIt/Reference/AMS55.asp
15:05:45 <\oren\> speaking of which, how do i make orenwatson.be redirect to www.orenwatson.be?
15:05:55 <b_jonas> \oren\: yes. I'm quite sure my old link used to work, but they must have reconfigured some servers or something
15:06:39 <b_jonas> \oren\: do you mean how to make the domain point there, or how to make the http server send a Found redirect to point to the other domain?
15:09:31 <\oren\> either one... I'm looking at my route-53 page and it doesn't seem to have an easy option for either
15:10:17 <int-e> and why would you do that?
15:10:45 <b_jonas> \oren\: if the domain doesn't yet point to anything, then you have to fix that first, otherwise there's no webserver to tell you where to redirect.
15:12:17 <\oren\> oh, I see. I can just create a new record pointing to where www.orenwatson.be also points
15:12:18 <int-e> oh, orenwatson.be is gone completely... fun :P
15:13:20 <\oren\> i think it's working now
15:13:28 <int-e> nah, needs to propagate still
15:14:00 <int-e> 750 seconds for me
15:14:33 <int-e> (starting from 900, I presume)
15:17:54 <b_jonas> Dumb question. The chrome browser still doesn't have native MathML display support, right?
15:21:07 -!- nooga has joined.
15:28:51 <\oren\> chrome apparently had it and then removed it becaus eof "Security concerns"
15:31:28 <int-e> well, mathml looks complex.
15:31:50 <int-e> they cited a small user base as well.
15:32:47 -!- MDude has joined.
15:32:59 <b_jonas> int-e: of course it has a small user base! browser features spread VERY slowly, because people only put something to webpages if the browsers can already show it, and browser devs put features in the browser only if it's needed to show webpages. That's why we're still stuck with jpeg images only.
15:33:35 <b_jonas> and gifs and pngs, yes, but jpeg as the only lossless format that could spread much
15:33:35 <int-e> though there are a lot of sites serving pngs these days
15:33:51 <b_jonas> the gif problem is mostly solved, my problem is more the jpeg
15:34:08 <b_jonas> jpeg is the only lossy one, suitable to photos, that is well-spread
15:34:39 <int-e> yes, yes it is. it appears to be "good enough".
15:34:49 <b_jonas> despite that there are tons of more modern and more versatile formats (including jpeg2000), but since browsers don't support them, every webpage uses only jpegs, and since every webpage uses jpegs, browsers don't support anything else
15:35:10 <MDude> And here I was just reading about Pale Moon devs not being able to keep up because the project was too huge.
15:35:24 <MDude> I don't think having features develop faster would help with that.
15:35:52 <int-e> in contrast look what happens when there's some serious bandwidth usage... i.e. video... in that area, many formats are supported and sites go out of their way to use the best format that the browser they're talking to understands... not the least common denominator
15:36:15 <MDude> I think it's more that file formats aren't a visible enough thing for the public to demand more.
15:36:36 <b_jonas> For MathML, the result is all kinds of crazy javascript libraries that try to render ALL of maths on client-side, without any server-side preprocessing to CSS to make it faster, because that's more convenient to the server maintainers, even if it means you need like gigabytes or ram to just view a webpage with maths, unless
15:36:44 <MDude> If people arne't already suing a file format, they see no reason to demand using it online.
15:36:51 <b_jonas> the webpage is sensible enough to use mathml, and the javascript libraries are there only as a fallback.
15:37:42 <int-e> fungot: what do you think of suing a file format?
15:37:43 <fungot> int-e: 08:35 fnord fnord has joined esoteric" did not change much, just to be safe
15:37:48 <b_jonas> What browsers other than firefox have mathml support?
15:38:16 <b_jonas> Do other popular browsers (whatever they are these days, I don't follow this android smartphone thing too much) have it?
15:39:48 <int-e> Safari, apparently
15:40:57 <b_jonas> does opera mobile or whatever it is that people use on android support it?
15:41:11 <b_jonas> or the browser that comes with android phones?
15:48:17 <olsner> looks like opera lost mathml when switching to chromium
15:50:52 <b_jonas> Well, if firefox and safari has it, that's a lot of users
15:52:10 <b_jonas> I myself use firefox for most webpages. http://dlmf.nist.gov/ has mathml, plus a javascript fallback and image fallback. It's quite well made.
15:55:16 <b_jonas> http://mathoverflow.net/ has a javascript solution (called mathjax) that has various frontends, including mathml, but it requires client-side javascript for anything, sadly.
15:56:15 <b_jonas> I might want to eventually put up a reformatted miror of a certain website that has lots of math-formulas, so I want to find out how this stuff works. I'll probably try to use mathml and some sort of fallback, but all on server side so you don't need javascript to render them.
15:57:30 -!- MoALTz_ has joined.
15:57:34 -!- earendel has quit (Ping timeout: 260 seconds).
16:00:51 -!- MoALTz has quit (Ping timeout: 276 seconds).
16:01:48 -!- MoALTz_ has changed nick to MoALTz.
16:09:36 -!- lambda-11235 has joined.
16:22:17 -!- ais523 has joined.
16:34:52 -!- ais523 has quit (Read error: Connection reset by peer).
16:35:02 -!- ais523 has joined.
16:36:19 -!- bender| has quit (Ping timeout: 252 seconds).
16:38:02 <ais523> syntax idea: use unary + as the constructor for nonempty option types
16:48:16 -!- hppavilion[1] has joined.
16:49:53 <\oren\> wht not just serve images for math, with LaTeX in the alt text
16:50:31 <\oren\> black and white PNG's can't be that huge
16:51:06 <ais523> \oren\: wouldn't scale to the user's font size
16:53:32 <\oren\> ais523: measure the <img> size in ch.
16:53:45 <\oren\> then it will scale when the font size does
16:54:00 <ais523> \oren\: but PNG doesn't scale properly
16:57:44 <b_jonas> \oren\: images don't respect the user's chosen font and colors. mathml does, it works the same as ordinary html and css.
17:08:13 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
17:23:14 -!- hppavilion[1] has joined.
17:23:58 -!- mihow has joined.
17:27:14 -!- earendel has joined.
17:38:28 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
17:47:48 <\oren\> why would anyone be against having a slovenian supermodel as their first lady?
17:49:50 -!- hppavilion[1] has joined.
17:55:25 <int-e> I don't know, they usually come with a president attached to them
17:56:12 <\oren\> right but apparently ted cruz attacked trump by saying his supermodel wife is unsuited to be first lady
17:56:44 <\oren\> that doesn't make sense to me
17:57:23 <int-e> oh, those are replublicans
17:57:26 <int-e> they don't make sense
17:57:41 <int-e> not sure where I got the extra l.
17:58:19 <\oren\> the first lady of the united states is not allowed to be hot!
17:58:24 <int-e> Anyway, personally I'd be interested in how the media would adapt to Bill Clinton being first lady :P
17:58:45 <int-e> (that term is so sexist)
18:00:25 <int-e> Anyway, is it Ted who chose that subject or is he just following Trump's lead of not engaging in any political topics?
18:00:31 <\oren\> yah, i think "first gent" would sound cool
18:00:44 <int-e> That's a good candidate.
18:04:31 -!- jaboja has joined.
18:05:38 <Taneb> They use "First Gentleman" for the husband of a state governer
18:07:26 <hppavilion[1]> @tell oerjan I fixed by BF interpreter. It turned out I had < increment the pointer. *derp*.
18:10:21 -!- lynn has joined.
18:12:58 -!- J_Arcane has joined.
18:23:34 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
18:25:10 -!- mihow has quit (Quit: mihow).
18:27:48 -!- hppavilion[1] has joined.
18:44:04 <int-e> what could it be, the most functional letter of the Greek alphabet?
18:51:52 <int-e> b_jonas: hmm, a capital one?
18:52:06 <b_jonas> int-e: no, a lower case phi
19:15:47 -!- ais523 has quit (Quit: dinner).
19:44:36 <hppavilion[1]> I'm making a game about badly-authored software that it is your job to hack
19:45:58 <int-e> ah https://xkcd.com/327/
19:56:03 <hppavilion[1]> int-e: One example is that, to read toolconf.conf which is secured so that it can't be read without the password (which is "herbert", though that's irrelevant), you must actually /secure/ the file with a new password
19:56:48 <hppavilion[1]> int-e: Because the programmer who made it didn't think to add an "is this file already secured?" check to file securing
19:57:30 <Taneb> One of my friends, who I've been playing computer games with lately but haven't for the past couple of days because my headset broke, said he misses me
19:57:45 <hppavilion[1]> So with `sec toolconf.cfg <your_password>` followed by `unsec toolconf.cfg <your_password>` or `auth toolconf.cfg <your_password>` you can change the filepass then unlock it for reading
19:58:08 <hppavilion[1]> (Though you have to auth every other command if you use that command)
19:58:21 <hppavilion[1]> (toolconf.cfg contains the data for the filesystem encryption)
19:58:59 <hppavilion[1]> int-e: I'm just trying to figure out how to fit the massive insecurity into the plot :P
19:59:39 <hppavilion[1]> I think the goal is that the creator made it less-than-secure so someone could take up his work once he's gone, but only if they can unencrypt the filesystem
20:00:34 <hppavilion[1]> The language you play the game in is a mix of simplified bash (no redirection yet) and a custom 1337 h4xx0r language with special commands
20:05:24 <int-e> hppavilion[1]: "insecurity: the game" (who needs a plot?)
20:06:58 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
20:08:10 -!- hppavilion[1] has joined.
20:08:37 <hppavilion[1]> Obviously, a letter that isn't used in English anymore.
20:11:23 -!- ais523 has joined.
20:11:29 -!- ais523 has quit (Client Quit).
20:11:39 -!- ais523 has joined.
20:13:05 <HackEgo> [U+A66E CYRILLIC LETTER MULTIOCULAR O]
20:13:05 <HackEgo> olist 1030: shachaf oerjan Sgeo FireFly boily nortti b_jonas
20:15:17 <int-e> lovely, my CaC server has remounted the file-system read-only again.
20:16:42 <int-e> this is helpful. "mpt-statusd: detected non-optimal RAID status"
20:17:26 <ais523> perhaps one drive has failed?
20:19:05 <int-e> Well, I don't see the hardware.
20:19:08 <int-e> /dev/mapper/vg_cac-root_cac on / type ext4 (rw,relatime,errors=remount-ro,data=ordered)
20:19:35 <int-e> and I don't care... the most important bit on that server is mroman's burlesque interpreter
20:20:00 <int-e> (so apparently the remount happened 2 weeks ago ;-) )
20:22:16 <fizzie> I have a leftover SSH open to the HackEgo one in a screen, and it seems to have printed noisy kernel stack traces on Mar 13, 20, 21, 22 and 23.
20:22:41 <int-e> my last /var/log/messages entry was from Mar 12 03:00:19
20:23:04 <fizzie> Mar 25 16:02:07 www2 mpt-statusd: detected non-optimal RAID status
20:23:04 <fizzie> Mar 25 16:12:07 www2 mpt-statusd: detected non-optimal RAID status
20:23:04 <fizzie> Mar 25 16:22:07 www2 mpt-statusd: detected non-optimal RAID status
20:23:06 <int-e> (and there was one every 10 minutes up to that point)
20:23:15 <fizzie> Heh, well, it's happening there every 10 minutes now.
20:23:39 <int-e> Mar 12 03:00:19 cheap mpt-statusd: detected non-optimal RAID status
20:24:23 <fizzie> On that one it's been going on since the start of /var/messages, Mar 20 06:27:05.
20:24:37 <fizzie> But there's also all these sporadic "sending NMI to all CPUs" + backtrace ones.
20:24:56 <int-e> I get those as well.
20:25:54 <int-e> I wonder how the kernel knows what to do with root=/dev/mapper/vg_cac-root_cac
20:25:59 <fizzie> The "detected non-optimal RAID status" messages go back as far as these logs go -- every 10 minutes, like clockwork, from at least as far back as Feb 21 06:28:04.
20:28:40 <hppavilion[1]> Is there any uber-simple GUI library out there that I can use to make a better GUI library? Preferably with Python bindings?
20:28:45 <fizzie> This one is just your normal LVM thing. There's a "physical" /dev/sda that's got two partitions (sda1, sda5), the latter of which is a LVM physical volume for VG "debian7", LVs "root" and "swap_1".
20:30:41 <int-e> dmsetup status has this, hmm. UUID: LVM-AlBirrduArMwzyJYtPrhgGT7y4C6LR09ZLaUeCzKSoP1bZQ6HohxDJHKMoPjxXf6
20:33:22 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
20:34:11 <shachaf> @tell hppavilion[1] html hth
20:34:53 <fizzie> It seems that there's a "hardware" RAID thing at least somehow visible. At least /proc/mpt/summary and /proc/mpt/ioc0/summary imply as much.
20:35:09 -!- hppavilion[1] has joined.
20:35:19 <fizzie> I don't know how to ask status information from it, though, because mpt-status doesn't say anything; there's just the cryptic non-optimality messages.
20:38:21 <int-e> Hmm. Ok, it pieces together /dev/sda3 and /dev/sda4 here...
20:39:54 <fizzie> This system doesn't have anything like that. And anyway; mpt-status reports nothing at all, which appears to be the reason why mpt-statusd prints out that message.
20:40:21 <int-e> ah, stupid me... the kernel doesn't know what to do with the root= command line; the initrd will handle that.
20:40:49 <fizzie> The check is if (mpt-status -i $ID) |grep -q 'state OPTIMAL'; then BADRAID=false; else BADRAID=true; logger -t mpt-statusd "detected non-optimal RAID status"; fi
20:41:02 <fizzie> So printing nothing counts as "non-optimal".
20:41:40 <int-e> there is no raid, so that makes sense :P
20:42:58 <fizzie> I guess. But there's still that controller, I would've expected it to say something.
20:45:34 <fizzie> Maybe the SCSI controller's a virtual one and doesn't do the things mpt-status expects, or something.
20:50:17 <int-e> it is some vmware thing
20:55:43 <int-e> oh well, at least I've explained some of the apparent magic now... that's something :)
21:05:16 -!- yorick__ has joined.
21:05:49 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
21:06:51 -!- hppavilion[1] has joined.
21:08:50 -!- yorick has quit (Ping timeout: 244 seconds).
21:20:13 -!- Sgeo__ has joined.
21:37:18 <Sgeo__> 'But to any programmer, it’s painfully easy to see why “Null” could cause problems for a database.'
21:38:00 <ais523> you have to be doing something very loosely-typed if "null" the string causes problems
21:38:24 <Sgeo__> I want to say something like "No, just the ones not being idiots who cause the problem", but some systems are big enough that a single person can't really fix them
21:38:52 <Sgeo__> ais523, http://www.bbc.com/future/story/20160325-the-names-that-break-computer-systems
21:39:58 <shachaf> Actually, it's the BBC controlling us from London.
21:40:26 <shachaf> zzo38: You should say "Unfortunately, there's a radio connected to my brain" to vaporware.
21:53:19 <fizzie> For some reason the poster at the hospital that says "testing for HIV is an everyday part of living in London" just sounds weird.
21:54:06 <shachaf> https://pbs.twimg.com/media/CF1kQXuWAAAAQjJ.jpg
21:54:49 <fizzie> (That's the image I used to verify the text.)
21:55:14 <int-e> So you mean that all information on the poster taken together means that there are $((500000/365)) people living in London?
21:55:26 <int-e> or maybe that's 500000.
21:57:29 <fizzie> Mostly it just makes me wonder which inherent part of living in London necessitates the daily HIV tests.
21:57:47 <shachaf> Well, it's not necessarily you who're getting tested for HIV.
21:58:26 <shachaf> Maybe you're otherwise affected by the tests.
22:00:16 <fizzie> Also today I went to snoop around the King's Cross / St. Pancr(e)as area. They had a lock.
22:00:21 <fizzie> The kind that boats go through.
22:01:09 <shachaf> Is that like the kind of eye a camel goes through?
22:01:19 -!- lynn has quit (Read error: Connection reset by peer).
22:01:43 <fizzie> I was going to say it's bigger, but I'm not entirely sure what sort of eyes camels do go through.
22:02:15 <int-e> "eye of a needle" presumably
22:02:37 <int-e> in which case the problem is that boats can actually pass the lock
22:03:36 <shachaf> Only if they have a key, right?
22:03:49 <int-e> (provided it has water and the gates actually close and open)
22:04:07 <int-e> shachaf: I'm not sure, but you may be missing out on a meaning of "lock".
22:04:07 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
22:04:14 <shachaf> And anyway camels can go through the eyes of needles.
22:04:27 <shachaf> “With man this is impossible, but with God all things are possible.”
22:04:38 <fizzie> There's a smartphone app that can allegedly control the fountains on Granary Square -- these ones: https://googledrive.com/host/0B4J9OAzXNfZANkM4WXNHSG93YVU -- but the reviews in Play Store seemed so pessimistic I didn't even bother to try.
22:04:46 -!- lynn has joined.
22:05:09 <fizzie> Although there was someone with a phone doing something that looked like trying to make them go.
22:05:11 <int-e> that's a clever app
22:06:15 <int-e> well, a clever idea for an app
22:06:24 <fizzie> The app is the snake game, in theory. I think there were also some colored lights to show your particular snake, and support for up to 4 people simultaneously. Or something like that, anyway.
22:06:44 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:07:24 <fizzie> The fountains did act a bit like in a snake game, but if someone was playing them, they weren't very good.
22:08:08 <fizzie> Actually the people that acted like they were trying to do something with their phones are those two in the photo.
22:08:30 <b_jonas> fizzie: hmm... fountains and snakes? can you get wishes from them too?
22:08:50 <b_jonas> maybe they have bad reviews from the people who die to water mocassins
22:09:40 <fizzie> The reviews mostly said stuff like "Two of us attempted but could not connect." or "Says there are no games and info link goes to 404 page" or "Was there twice at the right time. Couldn't connect to cloud error kept coming up".
22:11:15 <fizzie> https://www.kingscross.co.uk/img/790x444/src/media/11A5729-800x450.jpg <- that's what it's supposed to look like.
22:19:00 <b_jonas> fizzie: install it, get it to work, reverse engineer the protocol it uses to communicate with the fountains, and write a better program for it.
22:19:35 <b_jonas> Or just skip the app, log in to the insecure control computer of the fountain with "admin" as the username and "admin" as the password or something obvious like that, and replace its software.
22:19:57 <b_jonas> That's how anything controlled by software in the real world works these days, don't they?
22:21:11 <int-e> b_jonas: only if you have a dark room with three monitors running terminals with black background and green foreground
22:23:18 <b_jonas> I don't work in dark rooms.
22:23:45 <int-e> then I'm afraid admin/admin won't work for you.
22:23:52 <int-e> you have to follow the proper protocol in these things
22:25:36 <b_jonas> no way. you only have to follow the protocol if the other side is implemented properly to actually enforce the protocol.
22:28:01 <tswett> So I've decided to create an esolang.
22:28:43 <Taneb> You've certainly come to the right IRC channel
22:28:56 <tswett> It's more or less Smalltalk.
22:29:05 <tswett> I'm just gonna monologue for a while.
22:29:18 <Taneb> I'm listening (reading?)
22:29:21 <int-e> so it's not smalltalk
22:29:32 <int-e> because that would be a dialogue?
22:29:47 <tswett> So, it's an object-oriented language. Like most object-oriented languages, the state consists primarily of a bunch of objects... blah blah blah.
22:30:00 <tswett> There's a certain number of basic object types.
22:30:27 <tswett> Note that in theory, it's not possible to determine the BOT of an object at runtime, because an object can lie about what kind of object it is.
22:31:01 <tswett> There's GENERIC, CLASS, DICT, and BLOCK. I think that's it.
22:31:16 <int-e> adversarial typing
22:32:15 <tswett> If you have an object, you can do a method call on it. Doing this requires passing in two other objects: the name of the method, and the argument.
22:32:31 <tswett> The object will then do something and return another object.
22:34:10 -!- llue has joined.
22:34:22 <tswett> Every object contains a class pointer. Note that a class pointer doesn't have to point to a CLASS; it can point to any kind of object.
22:34:55 <b_jonas> tswett: what's GENERIC, and are there some sort of integers or other lightweight objects that don't have any pointers at all?
22:35:18 <b_jonas> in fact, what are GENERIC, CLASS, DICT, and BLOCK?
22:35:26 <b_jonas> and what does an object store?
22:35:47 <tswett> In the general case, when you call a method on an object, what happens is that the class's #get_method method is called with the method name as an argument, and then whatever is returned, its #execute method is called with a certain dictionary as an argument.
22:36:31 <tswett> The dictionary's values are the object you called the method on, and the argument you passed to the method.
22:36:52 <tswett> Now, this process obviously has the potential for infinite regress. So we cheat a bit.
22:37:13 <tswett> For a GENERIC object, the process is always as above, but for the other types of objects, there are certain method calls that skip the process.
22:37:41 -!- lleu has quit (Ping timeout: 244 seconds).
22:38:05 <b_jonas> tswett: also, are you planning to make an implementation for this? and test programs
22:38:23 <tswett> It's been a long, long time since I implemented one of my own esolangs.
22:38:31 <tswett> For a CLASS object... lemme think about this a bit.
22:43:14 <tswett> All objects have a local variable dictionary. This is actually a dictionary carried within the object; it's not a pointer to another object.
22:43:44 <b_jonas> tswett: is that like ruby objects (as opposed to smalltalk objects)?
22:44:02 <tswett> b_jonas: I don't know about Ruby, so I don't know what you mean.
22:45:00 -!- hppavilion[1] has joined.
22:45:03 -!- jaboja has quit (Ping timeout: 250 seconds).
22:46:01 <tswett> For a CLASS object, when you call its #get_method method, the CLASS object calls #get on its own "method_dictionary" instance variable in order to get the method. If this returns null, the CLASS object then calls #get_method on its "superclass" instance variable.
22:46:13 <tswett> It returns whatever the first non-null thing was.
22:47:51 <tswett> For a DICT object, when you call its #get method, it returns its corresponding instance variable.
22:48:10 <tswett> For a BLOCK object, when you call its #execute method, the content of the block actually executes.
22:48:38 <tswett> A BLOCK object, in addition to the class pointer and local variable dictionary, has an actual block of code in it.
22:51:16 <b_jonas> tswett: Smalltalk full objects (not the lightweight ones like integers) have a fixed set of instance variable fields determined by their class, plus they may own an array. Ruby full objects have a dictionary of instance variables, plus possibly some specific data if their class is derived from one of the approximately 15 primitive builtin ruby objects like Array, Hash, String, Regex
22:52:17 <tswett> This sounds a lot more like Ruby than Smalltalk, then.
22:54:21 <b_jonas> In ruby, those primitive types of what can be stored in an object are identified by this enum ruby_value_type thingy, which has values 1..15 for the primitive types (including not only Array, Hash, String but also Class and Bignum), plus 6 types for lightweight values that aren't pointers (fixints and 5 more), and 4 values for internal things that aren't objects and ruby variables can never hold them but are managed by the garbage collector
22:55:31 <b_jonas> (just like how a perl scalar can have multiple representations depending on what it is required to hold, but it can't be an AV or HV despite that interanlly those structures LOOK very similar to scalars for a good reason, only they're not exposed as such to userspace)
22:56:17 <b_jonas> The most important of those 4 internal things is T_NODE which holds pieces of the interpreted ruby code tree.
22:57:09 <tswett> Now, a key feature is that generally, calling an object's methods and looking up objects in dictionaries are the *only* things you can do with an object.
22:57:39 <b_jonas> Anyway, the generic or basic fullweight type is T_OBJECT, which only has the instance dictionary, nothing special. Any ruby object whose class isn't any of the special lightweight ones and isn't derived from any of the 14 classes corresponding to the other full representation types is represented as such.
22:59:39 -!- llue has quit (Quit: That's what she said).
22:59:49 <b_jonas> Also, I'm not sure if all the 14 specialized representations have an instance dictionary. I don't remember that. Every full object has a class pointer though, that's for sure.
22:59:55 -!- lleu has joined.
23:00:39 <tswett> There's one exception. When "the general method call case" calls the #execute method on a BLOCK, that BLOCK receives a special "self pointer" that it can do a lot more stuff with.
23:01:22 <b_jonas> Also, the instance dictionary is stored in some efficient way for optimization, so that the keys (names) are shared among instances.
23:01:36 <b_jonas> Or something. I don't really know the specifics of the implementation.
23:03:12 <tswett> Now, there's nothing preventing you from changing either the class or the basic-object-type of an object that already exists.
23:04:00 <tswett> You could do something like "Object.set_class(Object)", causing the class of the class "Object" to change from "Class" to "Object".
23:06:18 <b_jonas> By the way, don't take me as an authority on either smalltalk or ruby. I'm making up half of this stuff from half-forgotten memories. Verify in the smalltalk book and the ruby source code if you want to be sure.
23:07:23 <tswett> So yeah. The general philosophy is "just let the programmer do anything at all".
23:08:10 <tswett> Except, of course, violate an object's sovereignty.
23:09:37 -!- Lord_of_Life has quit (Excess Flood).
23:12:12 -!- jaboja has joined.
23:12:17 <\oren\> huzzah! http://orenwatson.be/ is working
23:13:19 <HackEgo> [U+2421 SYMBOL FOR DELETE]
23:13:40 -!- Lord_of_Life has joined.
23:14:36 <\oren\> ␀␁␂␃␄␅␆␇␈␉␊␋␌␍␎␏␐␑␒␓␔␕␖␗␘␙␚␛␜␝␞␟␠␡
23:15:36 <hppavilion[1]> https://www.quora.com/What-are-some-of-the-most-clever-ways-a-programmer-can-mess-with-a-friend/answers/20717097?srid=dS99 is amazing
23:19:07 <\oren\> i'm pretty sure I could have done that in VB when I was in high school
23:20:00 <\oren\> Mostly because VB makes it easy to make a realistic looking interface...
23:21:04 -!- glitchmatick has joined.
23:22:05 <HackEgo> glitchmatick: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on EFnet or DALnet.)
23:24:59 <\oren\> hppavilion[1]: anyway it seems like what he did was make a chat client with a green-on-black interface and pretend to be the shell
23:25:32 <\oren\> kind of elaborate for a prank
23:32:50 -!- oerjan has joined.
23:33:52 <lambdabot> fizzie said 12h 40m 26s ago: I fixed `! bf32 (HackEgo) but I can't touch !bf32 (EgoBot). The interps/bf bit that extracts the bitness from the command name wasn't working; fixed it by using I_CMD instead of CMD.
23:33:52 <lambdabot> hppavilion[1] said 5h 26m 26s ago: I fixed by BF interpreter. It turned out I had < increment the pointer. *derp*.
23:37:24 <oerjan> hppavilion[1]: how right-biased of you.
23:39:07 <\oren\> I have added 𝕬𝕭𝕮𝕯𝕰𝕱𝕳𝕴𝕵𝕸𝕹𝕺𝕽𝕾𝕿𝖀𝖆𝖇𝖈𝖉𝖊𝖋𝖌𝖍𝖎𝖏𝖐𝖑𝖒𝖓𝖔𝖕𝖖𝖗𝖘𝖙𝖚𝖛𝖜𝖝𝖞𝖟, with uncial letters I copied from the Book of Kells
23:41:35 <hppavilion[1]> \oren\: That does seem like the most likely result
23:42:13 <hppavilion[1]> \oren\: I'm kind of tempted to make my own Bloody Smurf (one that actually works) and get gullible people to use it
23:42:37 <lambdabot> shachaf said 3h 8m 25s ago: html hth
23:42:42 <oerjan> \oren\: how kelligraphic
23:42:50 <\oren\> (the bloody smurf that acutually works)
23:43:18 <HackEgo> HTML is just pictures and words.
23:43:21 <hppavilion[1]> \oren\: No, it wouldn't work as in actually hacking things
23:43:42 <oerjan> `learn HTML is short for "hope this mess loads"
23:43:43 <hppavilion[1]> \oren\: It would be like a game disguised as a real hacking client that you throw at people
23:43:45 <HackEgo> Relearned 'html': HTML is short for "hope this mess loads"
23:43:59 <oerjan> `learn HTML is short for "hope this mess loads".
23:44:01 <HackEgo> Relearned 'html': HTML is short for "hope this mess loads".
23:48:51 -!- glitchmatick has quit (Ping timeout: 276 seconds).
23:49:34 <oerjan> fizzie++ slightly less transient
23:50:49 <zgrep> :D (regarding 'HTML')
23:51:09 * oerjan swats myname and int-e for their puns in the log -----###
23:51:41 <\oren\> `` u8tbl 0x1d9b 0x1da7
23:51:44 <shachaf> oerjan: hey, why don't i get a swat
23:52:01 <oerjan> shachaf: i may not have got to that point yet.
23:52:21 <myname> how is a pun a reasonable excuse for swating?
23:52:30 <oerjan> myname: it's traditional
23:52:49 <myname> i liked it. the pun, that is
23:53:13 <oerjan> although i've been lazily leaving things to boily's mapole recently
23:53:51 <myname> i still don't get why you read all the backlogs
23:54:22 <\oren\> `` u8tbl 0x16f1 to 0x16f8
23:54:26 <\oren\> `` u8tbl 0x16f1 0x16f8
23:55:54 <\oren\> `` u8tbl 0x31f0 0x31ff
23:56:18 <\oren\> oh, right the extra katakana used for ainu
23:57:20 <\oren\> `` u8tbl 0x2934 0x2937
23:57:28 <HackEgo> bin/u8tbl: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.26, BuildID[sha1]=0xf4bd6215e42f01142295c499b7a9bf8a7c37e01a, not stripped
23:58:21 <oerjan> `` mv bin/u8tbl{,.elf}; mkx 'bin/u8tbl//u8tbl.exe $@'
23:58:35 <HackEgo> /hackenv/bin/u8tbl: line 1: u8tbl.exe: command not found
23:58:40 <shachaf> oerjan: well i don't remember making any puns
23:58:41 <\oren\> `` u8tbl 0x27a9 0x27aa
23:58:42 <HackEgo> /hackenv/bin/u8tbl: line 1: u8tbl.exe: command not found
23:58:51 <shachaf> oerjan: but i'm sure i must've
23:59:05 <HackEgo> mv: cannot stat `u8tbl.elf': No such file or directory
23:59:25 <oerjan> `` mv bin/u8tbl.{elf,exe}
23:59:25 <\oren\> `` u8tbl 0x27a9 0x27aa
23:59:54 <HackEgo> /hackenv/bin/u8tbl: line 1: u8tbl.exe: command not found
23:59:55 <oerjan> ...did we cause some kind of deadlock
23:59:59 <\oren\> `` u8tbl 0x27a9 0x27aa