00:00:41 <rdococ> should I unlearn that?
00:01:21 -!- Lilly_Goodman has joined.
00:02:04 <HackEgo> rm: cannot remove `/home/hackbot/hackbot.hg/multibot_cmds/env/.hg/store/data/canary.orig': Is a directory \ Done.
00:02:48 <HackEgo> ` is the prefix to greatness.
00:03:53 <hppavilion[1]> rdococ: There are two players running identical programs psedOSes
00:04:27 <hppavilion[1]> rdococ: The goal is to hack your enemy to the point that they're unable to play any further, and to continuously patch your computer to block incoming attacks
00:04:38 <hppavilion[1]> The code is badly-written and uncommented, and the docs are pretty poor
00:05:11 <rdococ> so yo have to edit their program to do stuff?
00:06:05 <hppavilion[1]> rdococ: It's constrained by simulated compile time
00:06:17 <rdococ> but does it have to be simulated?
00:06:32 <hppavilion[1]> (Every line you add is an extra 5 seconds before the changes take hold)
00:06:36 <rdococ> do they have access to multiple programs?
00:06:40 <hppavilion[1]> rdococ: It does, if we want the computers to not die
00:06:50 -!- Phantom_Hoover has quit (Remote host closed the connection).
00:07:14 <rdococ> you could edit one program of the other's computer, then edit another program while they're trying to patch the one you edited first
00:07:39 -!- Lilly_Goodman has quit (Ping timeout: 264 seconds).
00:08:12 <rdococ> what about negative data
00:08:16 <hppavilion[1]> rdococ: There'd be "log::pt", which is a file documenting changes. You have to refresh it every so often.
00:08:43 <hppavilion[1]> It wouldn't be line-by-line though, it'd be a sort of "AST measurer"
00:08:54 -!- Lilly_Goodman has joined.
00:10:25 <rdococ> why not actually simulate the program code in a sandbox environment?
00:10:45 <rdococ> it'd be fun if each program had different languages
00:11:22 <rdococ> we could even have some esoteric ones
00:12:01 <rdococ> also, hacking programming languages
00:12:13 <rdococ> what are lose conditions?
00:12:33 <hppavilion[1]> rdococ: Probably when all user input is disabled for the client
00:13:47 <hppavilion[1]> rdococ: I mean, I can think of one way (recursive AST scanning), but it'd be buggy
00:23:35 <\oren\> suppose a program was defined solely by stating what it cannot do
00:25:26 <rdococ> then an empty program can solve the halting problem
00:25:40 <\oren\> or might do something elsr
00:26:01 <\oren\> an empty program is allowed to compile into any program whatsoever
00:26:51 -!- oerjan has joined.
00:28:37 <\oren\> then you have some formal language for specifying program behaviour in terms of logical statements, each of which is always required to be false.
00:30:24 <rdococ> a normal language would be one specifying true statements
00:37:03 -!- Lilly_Goodman has quit (Ping timeout: 264 seconds).
00:37:37 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
00:42:01 <boily> /ħ/ isn't a fun sound hth
00:44:37 -!- Sgeo_ has joined.
00:46:03 -!- Sgeo has quit (Ping timeout: 240 seconds).
00:47:29 <rdococ> talking about logical problems
00:48:45 -!- Sgeo__ has joined.
00:49:29 <oerjan> talking about the logical problems that don't talk about themselves
00:50:03 -!- Sgeo_ has quit (Ping timeout: 240 seconds).
00:58:16 -!- tromp_ has joined.
01:02:42 -!- tromp_ has quit (Ping timeout: 260 seconds).
01:09:10 -!- jaboja has joined.
01:13:12 -!- Lilly_Goodman has joined.
01:15:37 <rdococ> is this problem false?
01:15:47 <rdococ> normal computer: CRASH
01:16:35 <rdococ> quantum computer: There is a 60% chance that the problem is true, and a 39% chance it is false.
01:19:06 <rdococ> and a 1% chance that you're trying to crash me
01:20:52 <lifthrasiir> the NLP module has learned to parse in the other way when the paradox arises
01:22:06 <rdococ> will you say no to this question?
01:22:30 <rdococ> normal computer: Fal--I mean tr--fa--tr
01:24:59 <boily> fungot: do you even paradox bro?
01:24:59 <fungot> boily: you want a high-quality fnord program in general. the general idea
01:25:16 <boily> fungot: that is enlightening. tdh.
01:25:17 <fungot> boily: this is an example from the wiki? looks suddenly sparse.) thanks for your answer. now you might have more luck rebinding space to enter when using the variable name
01:25:46 * boily rebinds the Sacrificial Space. “Fnord! Fnord! Fnord!”
01:30:40 <boily> huh. I thought we had a fnordwisdom in there.
01:31:31 -!- mihow has quit (Quit: mihow).
01:31:37 <oerjan> THIS WENT BETTER THAN EXPECTED
01:32:51 -!- mihow has joined.
01:40:13 * oerjan sits down to watch the temporal antipode effect
01:40:16 <boily> indeed. it's getting quite late, I say.
01:40:36 <quintopia> i can always know when he's going to bed
01:41:03 <HackEgo> quintopia is our resident tl;dr generator. He is flooded by thundercats and thunderdogs. He is the temporal antipode of boily despite living on the same continent.
01:41:29 <boily> speak to me of summer ♪
01:41:32 <quintopia> i'll go back in time 2 hours and overlap him a bit
01:41:38 -!- boily has quit (Quit: COMMITMENT CHICKEN).
01:42:31 <rdococ> even though it's 2:42 am where I am
01:42:51 <rdococ> which could be in africa
01:43:03 <oerjan> but is probably in britain
01:43:04 -!- XorSwap has quit (Quit: Leaving).
01:43:14 <rdococ> it could be in britain too
01:43:33 <rdococ> good idea for a name to an esolang
01:44:06 <quintopia> yes but what real country would it be in
01:45:37 <oerjan> fungot: is rdococ an AI or is there some other reason e's evading the question
01:45:37 <fungot> oerjan: anmaster must see.
01:45:39 <rdococ> I know it sounds duckoo but it's troo
01:45:55 <oerjan> Vorpal: fungot says you must see this hth
01:45:55 <fungot> oerjan: and no, that would explain why the video output isn't working properly
01:46:16 <fungot> rdococ: sorry, didn't read properly. a question of whether the loop was loopzored", i wouldn't expect car and cdr returned copies, hmph.
01:46:41 <rdococ> fungot: I said, am I an AI emulating a human brain?
01:46:41 <fungot> rdococ: i like to use _, then __, then ___, et cetera; and no record type descriptor operations such as vector-ref vector-set!.
01:46:48 <quintopia> what country do you currently reside, person currently to emself as rdococ?
01:47:23 <rdococ> I told you, I reside in an esolang
01:47:30 <quintopia> i'll just pretend you live in the same country as the server you are connected to for ease
01:47:43 <rdococ> what server am I connected to
01:48:08 <oerjan> rdococ: shouldn't that be an esoland, really
01:48:36 <rdococ> see? I'm not from britain at all. I'm from rdococland.
01:49:17 <rdococ> I thought it was on mars
01:49:40 <rdococ> sorry, there are still a few bugs in me
01:49:48 <oerjan> no, on mars you find london and o'wobble. that's what i recall from Blackadder, anyway.
01:50:15 <oerjan> it may have been the very last downer episode
01:50:30 <oerjan> he tries to get himself certified insane
01:50:47 <oerjan> by claiming he's from mars
01:51:12 <oerjan> alas, his superior is too smart/stupid
01:51:42 -!- Lilly_Goodman has quit (Ping timeout: 244 seconds).
01:52:53 -!- oerjan has set topic: Note: people with cloaks will be treated as if they're from Budapest | The international hub of esoteric programming language and font design | http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | The kitten typesetting channel.
01:52:59 <quintopia> oerjan: is your palmaris longus present/discernible?
01:53:27 <oerjan> ironically, b_jonas doesn't have a cloak
01:54:43 -!- Lilly_Goodman has joined.
01:55:03 <oerjan> quintopia: is it that big vertical one on the wikipedia picture? if so, yes
01:55:35 <lifthrasiir> https://twitter.com/johnregehr/status/715089198819241984
01:55:42 <oerjan> if i flex my hand just right, anyway
02:15:17 <quintopia> Lilly_Goodman: nosotros no ayudamos en espanol
02:16:03 <Lilly_Goodman> quintopia es que quiero saber como decir la hora en ingles
02:17:54 <quintopia> tu eres en el canal incorrecto, sabes?
02:19:45 <oerjan> quintopia: we've tried long ago
02:20:04 <Lilly_Goodman> quintopia quiero decir en ingles... son las nueve y media
02:21:35 <quintopia> Lilly_Goodman: ir a un canal de habla espanola por favor
02:25:00 -!- Lilly_Goodman has quit (Ping timeout: 250 seconds).
02:26:14 -!- Kaynato has joined.
02:29:07 -!- Lilly_Goodman has joined.
02:31:01 <Lilly_Goodman> hola alguien me ayuda a decir que son las nueve y media... en ingles???????????
02:33:38 -!- ChanServ has set channel mode: +o oerjan.
02:34:44 -!- oerjan has set channel mode: +b *!*canaima@*.dyn.movilnet.com.ve.
02:34:44 -!- oerjan has kicked Lilly_Goodman Tomar una pista.
02:35:06 -!- oerjan has set channel mode: -o oerjan.
02:35:44 <oerjan> izabera: she(?)'s been refusing to take a hint for over a week
02:36:49 <izabera> that doesn't make it less rude
02:37:27 <rdococ> why so many non english stuff
02:37:33 <oerjan> unless you're referring to "Tomar una pista" which i have no idea what means except it's what G.T. gives for "take a hint"
02:38:04 <rdococ> my brssim of dddipr rexhsaiuxzsted
02:38:40 <oerjan> izabera: i've been tempted for a week. she _refuses_ to speak in english, or to go anywhere people can understand her.
02:39:07 <oerjan> (well, she apparently doesn't know english well enough.)
02:39:48 <rdococ> ?- pun(programmingLanguage, linguisticLanguage)
02:39:58 <oerjan> also, those who _have_ tried to speak in spanish (mostly with G.T. which is horrible) have not been able to get any indication she's interested in our actual channel _topics_.
02:40:19 <rdococ> does she even know what this channel is about?
02:40:56 <oerjan> rdococ: well she's seen our `bienvenido message in spanish. i'm not sure she's at the point where she understands what it means.
02:41:52 <rdococ> haha "people with cloaks will be treated as if they're from Budapest"
02:42:13 <rdococ> oerjan: sounds like she uses google translate
02:43:20 <oerjan> izabera: she's either a strange troll or too maladapted for us to handle. i'm not sure.
02:43:37 <oerjan> rdococ: are you saying her spanish isn't right?
02:43:53 <rdococ> I was thinking she's faking it to be honest
02:44:08 <rdococ> since she didn't get the bienvenido message
02:44:13 <oerjan> rdococ: she hails from a venezuelan ip so it seemed logical it is her native language.
02:44:50 <oerjan> rdococ: could be though. however i meant "didn't get" as in pays no attention to it, as if she doesn't know what programming means.
02:45:19 <HackEgo> RDOCOCLIKESTOMAKELANGUAGESLIKETHIS
02:45:32 <rdococ> `learn rdococ Apparently from Budapest, but probably not.
02:45:38 <HackEgo> Relearned 'rdococ': rdococ Apparently from Budapest, but probably not.
02:45:57 <rdococ> `learn rdococ is apparently from Budapest, but probably not.
02:45:59 <HackEgo> Relearned 'rdococ': rdococ is apparently from Budapest, but probably not.
02:46:45 <rdococ> is there a programming paradigm that has not been done yet
02:47:57 <oerjan> paradigm-oriented, clearly.
02:48:10 <rdococ> how would that even work
02:49:29 <oerjan> izabera: also, she said she had been banned from #canaima-social(?) so she's presumably grating even to those who _do_ understand her.
02:49:50 <izabera> i've been banned from a dozen channels
02:50:02 <rdococ> I think I have been banned from a few
02:51:28 <oerjan> ok i get it, getting banned is no proof of anythin (except maybe that you're weird but we all are here.)
02:52:26 <oerjan> or that they're shit channels.
02:53:51 <rdococ> defocus is a shit channel?
02:54:03 <oerjan> THOSE WERE ALTERNATIVES
02:54:31 * oerjan perhaps shouldn't do joke shouting at this point
02:54:41 <oerjan> rdococ: like a platypus hth
02:54:53 <rdococ> what? I knew you weren't really angry
02:56:51 <HackEgo> [wiki] [[User:Rdococ]] M https://esolangs.org/w/index.php?diff=46701&oldid=46670 * Rdococ * (+14) yay added BASICER to my list
02:57:46 <rdococ> what do you think of BASICER?
02:58:10 -!- tromp_ has joined.
03:00:03 -!- lambda-11235 has joined.
03:03:15 -!- tromp_ has quit (Ping timeout: 276 seconds).
03:03:31 -!- heroux has quit (Ping timeout: 248 seconds).
03:03:41 <prooftechnique> I like the idea of a bas icer. Is that like a cake decorator who only does single layer cakes?
03:04:21 <Kaynato> rdococ: I am curious, what paradigm is Daoyu
03:04:24 -!- heroux has joined.
03:05:59 <rdococ> prooftechnique: my esolang
03:07:04 <prooftechnique> I was thinking like bas relief, but yeah, the principle is the same
03:07:52 <Kaynato> Have I gotten a chance to show you the language I have made?
03:08:15 <oerjan> BASRELIEF, the language you go to when you're just too fed up with BASIC
03:09:02 <Kaynato> Daoyu: https://esolangs.org/wiki/Daoyu
03:09:12 <prooftechnique> I'm trying to think of a way to implement conditionals with only GOTO just to make BASICER complete :D
03:09:28 <oerjan> prooftechnique: computed goto?
03:10:50 -!- tromp_ has joined.
03:11:38 <prooftechnique> I guess they kind of already exist, if you take the user as an oracle
03:12:01 <rdococ> there are no variables, how do you expect conditionals
03:12:08 <oerjan> arthur-merlin model, with the user as merlin
03:14:23 <Kaynato> @rdococ: is there a possibility you can help me with some C memory problems?
03:15:09 <Kaynato> You seem to be an experienced authority figure
03:16:14 <rdococ> well, I'm about to fall asleep, and I don't know all that much, ask someone else
03:16:53 <rdococ> but thanks for the compliment
03:17:31 <rdococ> channel.Sleep(awkward silence)
03:18:58 <oerjan> i'm sure there's some authority on C here.
03:19:08 <oerjan> i mean, there are IOCCC winners here.
03:19:26 <oerjan> this may or may not help in this case.
03:19:38 * oerjan pokes tromp_ and Gregor
03:19:55 <oerjan> (the latter only out of principle)
03:20:28 <oerjan> (and i'm not sure the former appreciates it either)
03:24:56 -!- shikhin has quit (Read error: Connection reset by peer).
03:25:24 -!- fungot has quit (Ping timeout: 246 seconds).
03:25:45 -!- sewilton has quit (Ping timeout: 246 seconds).
03:27:43 -!- Lymia has quit (Ping timeout: 246 seconds).
03:27:43 -!- fractal has quit (Ping timeout: 246 seconds).
03:27:43 -!- dingbat has quit (Ping timeout: 246 seconds).
03:28:54 -!- incomprehensibly has quit (Ping timeout: 246 seconds).
03:32:03 -!- jameseb has quit (Ping timeout: 246 seconds).
03:32:03 -!- nitrix has quit (Ping timeout: 246 seconds).
03:32:03 -!- glogbot has quit (Ping timeout: 246 seconds).
03:32:09 -!- esowiki has joined.
03:32:13 -!- esowiki has joined.
03:32:14 -!- esowiki has joined.
03:33:09 -!- esowiki has joined.
03:33:13 -!- esowiki has joined.
03:33:14 -!- esowiki has joined.
03:34:09 -!- esowiki has joined.
03:34:11 -!- glogbot has joined.
03:34:13 -!- esowiki has joined.
03:34:14 -!- esowiki has joined.
03:34:56 -!- sewilton_ has changed nick to sewilton.
03:35:27 -!- dingbat_ has changed nick to dingbat.
03:35:34 -!- tromp_ has quit.
03:36:16 -!- nitrix- has quit (Changing host).
03:36:16 -!- nitrix- has joined.
03:36:20 -!- nitrix- has changed nick to nitrix.
03:36:31 -!- incomprehensibly has joined.
03:37:52 -!- tromp has joined.
03:39:05 -!- tromp has quit (Remote host closed the connection).
03:39:53 -!- tromp has joined.
03:47:35 <HackEgo> [wiki] [[Special:Log/newusers]] create * Kc kennylau * New user account
03:47:46 <HackEgo> [wiki] [[///]] https://esolangs.org/w/index.php?diff=46702&oldid=40418 * Kc kennylau * (+4) /* Binary to unary conversion */
04:01:19 <HackEgo> [wiki] [[Binary to unary conversion]] N https://esolangs.org/w/index.php?oldid=46703 * Kc kennylau * (+1729) Created page with "A '''binary to unary conversion''' program is a program that can convert a given number in binary form to a unary form. It is mainly used to showcase [[Markov algorithm]] ([ht..."
04:01:21 <HackEgo> [wiki] [[Retina]] https://esolangs.org/w/index.php?diff=46704&oldid=45838 * Kc kennylau * (+110)
04:01:59 <HackEgo> [wiki] [[Binary to unary conversion]] https://esolangs.org/w/index.php?diff=46705&oldid=46703 * Kc kennylau * (+28)
04:02:57 <HackEgo> [wiki] [[Binary to unary conversion]] https://esolangs.org/w/index.php?diff=46706&oldid=46705 * Kc kennylau * (+92) /* Retina */
04:03:17 <HackEgo> [wiki] [[Retina]] https://esolangs.org/w/index.php?diff=46707&oldid=46704 * Kc kennylau * (+103)
04:07:03 -!- tromp has quit (Ping timeout: 244 seconds).
04:28:33 -!- Kaynato has quit (Ping timeout: 240 seconds).
04:32:35 -!- mihow has quit (Quit: mihow).
04:47:00 <oerjan> @tell hppavilion[1] <hppavilion[1]> http://dl.acm.org/citation.cfm?id=96559.96570 has &lgr; for some reason. I have no clue what that entity means <-- it's a lambda hth
04:51:06 <oerjan> @tell rdococ <rdococ> so there is no turing complete machine with only one register? <-- fractran.
05:01:18 -!- ais523 has joined.
05:03:48 -!- fractal has joined.
05:04:32 -!- tromp has joined.
05:09:22 -!- lambda-11235 has quit (Quit: Bye).
05:23:14 -!- tromp has quit (Ping timeout: 260 seconds).
05:23:49 <ais523> https://api.left-pad.io/?str=test&len=12&ch=%22
05:32:29 <Sgeo__> ais523, that's clearly a semver violation, they should put a version number somewhere. I suggest in the domain name just after .io
05:32:58 <ais523> is it a semver violation if you don't /have/ a version number?
05:37:07 <coppro> is it a semver violation if you use one of the noncompliant examples from the spec?
06:20:01 -!- tromp has joined.
06:22:28 -!- hppavilion[1] has joined.
06:26:27 -!- tromp has quit (Ping timeout: 260 seconds).
06:27:25 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
06:40:02 -!- tromp has joined.
06:45:45 -!- tromp has quit (Ping timeout: 268 seconds).
06:51:57 -!- lleu has joined.
08:21:49 -!- J_Arcane has quit (Ping timeout: 252 seconds).
08:42:08 -!- tromp has joined.
08:47:33 -!- tromp has quit (Ping timeout: 240 seconds).
08:49:42 -!- Effilry has changed nick to FireFly.
09:00:05 -!- ais523 has quit.
09:15:11 <b_jonas> brilliant! the ziped multiple html download of the POSIX spec from opengroup.org is so posixy it has two filenames that are equal case insensitively (_exit.html and _Exit.html) so you can't extract all of it on non-unixy file systems.
09:27:06 <oerjan> b_jonas: they may very well have the same contents, too
09:27:24 <oerjan> http://pubs.opengroup.org/onlinepubs/009695399/functions/exit.html describes those and exit
09:29:14 <oerjan> `fetch http://pubs.opengroup.org/onlinepubs/009695399/functions/_exit.html
09:30:01 <HackEgo> 2016-03-31 09:29:34 URL:http://pubs.opengroup.org/onlinepubs/009695399/functions/_exit.html [25309/25309] -> "_exit.html" [1]
09:30:05 <oerjan> `fetch http://pubs.opengroup.org/onlinepubs/009695399/functions/_Exit.html
09:30:08 <HackEgo> 2016-03-31 09:29:49 URL:http://pubs.opengroup.org/onlinepubs/009695399/functions/_Exit.html [25309/25309] -> "_Exit.html" [1]
09:31:03 <HackEgo> removed `_exit.html' \ removed `_Exit.html'
09:36:00 -!- jaboja has joined.
09:41:53 -!- jaboja has quit (Ping timeout: 244 seconds).
10:00:34 -!- jaboja has joined.
10:06:27 * oerjan thinks today's mezzacotta comic is strangely coherent.
10:09:14 <oerjan> i would recommend it to fungot, but e's not heeeeeeeeeeeeeAAAAA
10:28:51 <oerjan> http://www.mezzacotta.net/
10:38:32 -!- oerjan has quit (Quit: Later).
10:44:07 -!- tromp has joined.
10:46:15 -!- boily has joined.
10:49:46 -!- tromp has quit (Ping timeout: 252 seconds).
10:51:11 <myname> doesn't look that impressive
11:04:22 -!- jaboja has quit (Remote host closed the connection).
11:12:16 -!- dingbat has quit (Quit: Connection closed for inactivity).
11:20:16 -!- ^v has quit (Ping timeout: 264 seconds).
11:22:00 -!- boily has quit (Quit: ALMOST CHICKEN).
11:22:50 -!- ^v has joined.
11:22:51 -!- Deewiant has quit (Ping timeout: 244 seconds).
11:23:28 -!- Deewiant has joined.
11:46:32 <HackEgo> [wiki] [[Special:Log/newusers]] create * Midnightas * New user account
11:52:08 -!- Reece` has joined.
11:56:11 <HackEgo> [wiki] [[Pylongolf]] N https://esolangs.org/w/index.php?oldid=46708 * Midnightas * (+451) Created page with "= Pylongolf = Pylongolf is a simple stack-based language created by [http://midnightasgames.ga MidnightasGames]. == Adding to stack == Adding to the stack works differently t..."
11:56:25 -!- jameseb- has changed nick to jameseb.
12:05:22 -!- augur_ has quit (Read error: Connection reset by peer).
12:05:42 -!- augur has joined.
12:23:07 -!- ak48 has joined.
12:24:50 -!- ak48 has quit (Quit: node-irc says goodbye).
12:26:01 -!- Sgeo__ has quit (Ping timeout: 252 seconds).
12:29:25 -!- AnotherTest has joined.
12:40:03 -!- ^v has quit (Ping timeout: 248 seconds).
12:40:17 -!- augur has quit (Read error: Connection reset by peer).
12:40:55 -!- augur has joined.
12:41:03 -!- augur has quit (Read error: Connection reset by peer).
12:41:46 -!- augur has joined.
12:46:13 -!- tromp has joined.
12:46:57 -!- ^v has joined.
12:51:37 -!- tromp has quit (Ping timeout: 250 seconds).
13:09:34 <rdococ> am I the only one who didn't make a brain**** derivative when I first joined?
13:35:29 -!- Guest6809 has joined.
13:41:19 -!- p34k has joined.
13:50:48 -!- lambda-11235 has joined.
13:51:32 <b_jonas> rdococ: I still have to resist it VERY hard, because there's a good one I know but I want to make sure it's never made. I'll have to figure out how to do it elegantly without brainfuck, but I haven't yet.
13:53:03 <b_jonas> I'll probably have to adapt it to underload if possible.
14:06:34 <myname> did you made... any prooving technique
14:19:17 <b_jonas> prooftechnique: but what do you do with those users who are immune to mind tricks?
14:19:48 <b_jonas> I bet we have some of those on this channel
14:20:42 <prooftechnique> Then you just invoke descent theory and everyone gives up
14:29:15 <HackEgo> [wiki] [[Evil]] https://esolangs.org/w/index.php?diff=46709&oldid=30837 * Kc kennylau * (+165)
14:35:24 <b_jonas> Question. On windows 10, how do I configure the system to not turn off the monitor so quickly when on the lock screen?
14:35:35 <HackEgo> [wiki] [[Evil]] https://esolangs.org/w/index.php?diff=46710&oldid=46709 * Kc kennylau * (+105) /* 0 to 255 using only a, e, u, z (To be completed) */
14:36:15 -!- J_Arcane has joined.
14:36:46 <b_jonas> I tried to change the "Power Options" in control panel, but that only lets me change how quickly the monitor turns off when the login isn't locked.
14:39:26 <b_jonas> I tried a google search, and I found some webpages mentioning strange registry entries.
14:39:48 <b_jonas> Although most of them are about Windows 8 and they don't seem directly applicable.
14:42:21 -!- spiette has joined.
14:42:46 <b_jonas> also found another webpage with a different solution that might work: https://support.microsoft.com/en-us/kb/2835052
14:45:15 <rdococ> I have trouble trying to turn up the brightness on my monitor
14:45:26 <rdococ> no matter what setting I set, it always looks the same
14:48:16 -!- tromp has joined.
14:50:01 -!- lambda-11235 has quit (Quit: Bye).
14:54:34 -!- tromp has quit (Ping timeout: 240 seconds).
14:57:02 -!- bender| has joined.
15:04:40 -!- Kaynato has joined.
15:06:53 <rdococ> is there a number system where 1/0 = infinity?
15:07:41 -!- AnotherTest has quit (Ping timeout: 250 seconds).
15:09:56 <jameseb> rdococ: not if you want the field axioms to be satisfied, I think
15:14:24 <rdococ> that means x/0 = x * (1/0)
15:14:42 <rdococ> well, there will have to be multiple infinities
15:17:23 -!- Reece` has quit (Read error: Connection reset by peer).
15:22:32 -!- ^v has quit (Ping timeout: 260 seconds).
15:23:16 <rdococ> I always wondered why people keep talking about pi
15:23:22 <rdococ> I prefer the square root of 2
15:23:37 <rdococ> it is the everything number
15:24:41 <rdococ> because it's the number of my dreams
15:25:41 -!- ^v has joined.
15:32:04 <rdococ> I want to make a mathematical system which doesn't calculate things normally so I can do square root of -1 without adding a special case
15:33:34 <rdococ> in string substitution
15:34:17 <rdococ> sqrt($x) * sqrt($x) = $x
15:35:54 <rdococ> how would I make a super-turing complete programming language
15:39:39 <int-e> adding an oracle is the standard way of doing that
15:40:15 <int-e> if f(x) does not terminate then ...
15:40:51 <rdococ> but can we break that does not terminate super-complete instruction into smaller ones?
15:41:26 <rdococ> for example, are there other problems that turing complete machines can't solve?
15:42:44 <int-e> Well, most subsets of the natural numbers are uncomputable... so yes.
15:42:51 <int-e> The tricky bit is to define one.
15:43:50 <rdococ> in any system that sounds logical
15:45:24 <int-e> Would an oracle for the n-th digit of the halting probability (for, say, binary lambda calculus) solve the halting problem... I'm afraid it does?
15:49:31 <rdococ> how do we divide the halting problem into smaller instructions? basically, a super turing complete language where you need more than one instruction to solve the problem
15:51:50 <rdococ> infact, why not just specify that it computes everything in a finite time
15:51:55 <rdococ> then it's super-turing complete
15:56:31 <rdococ> programs that don't halt halt anyway
15:56:38 * int-e is actually hoping for a concrete definition of an oracle that's not computable but cannot solve the halting problem for TMs.
15:56:53 * rdococ wonders what an oracle is
15:57:23 <int-e> an external entity that answers interesting questions.
15:58:08 <rdococ> are you sure that the halting problem is even well defined?
15:59:34 <int-e> They're used in computability theory and also in complexity theory. For example a SAT oracle would take a boolean formula and immediately say whether it's satisfiable or not. So people talk about complexities relative to an oracle... P^{NP} would contain problems that can be solved in polynomial time and a polynomial number of queries to an NP oracle (which you can think of as a SAT oracle...
15:59:40 <int-e> ...because SAT is NP-complete).
16:00:16 <int-e> rdococ: yes, it is well-defined for Turing machines.
16:00:55 <rdococ> hey, turing complete machines can solve it for finite state machines
16:03:11 <int-e> the "if f(x) halts" construct is not well-defined if you allow recursion, due to its circularity... f := loop while f halts.
16:03:39 <b_jonas> I don't understand this. In the rust standard library, why is std::process::Command.spawn not an unsafe function, when it starts a process with an arbitrary executable and so can indirectly cause arbitrary memory access on your process too.
16:04:24 <int-e> now who would do such things... :P
16:05:01 <int-e> also, what distinguishes the spawned processes from any other processes that are already running on the system?
16:05:15 <int-e> by that reasoning, all code is unsafe.
16:05:37 <int-e> (which is quite close to the truth, but not a useful distinction to make... lacking the distinction)
16:05:56 <b_jonas> int-e: sure, if you run unsafe code _once_, and it's not actually safe, then you can get undefined behavior _later_.
16:06:12 <b_jonas> but the point is that running only safe code shouldn't be able to do thtat
16:06:24 <int-e> b_jonas: note that I don't know what kind of safety Rust is trying to model
16:06:34 <b_jonas> so opening files or starting processes should count as unsafe
16:06:42 <b_jonas> int-e: that is described in the rust book and manuals
16:06:47 * int-e doesn't even know this with complete certainty for Haskell... where he's a bit of an expert.
16:07:13 <b_jonas> well, haskell is sort of different
16:08:15 <rdococ> function f() { while f halts { } }
16:08:20 <rdococ> does the function halt?
16:08:26 <rdococ> the function halts if it does not halt
16:08:34 <int-e> b_jonas: I was merely trying to clarify that I was voicing an opinion not based on fact but at best on common sense.
16:09:09 <int-e> rdococ: I think I just wrote that, with slightly different syntaxc
16:09:26 <rdococ> make evaluation not go through the whole thing, but just one step - for example
16:09:41 <rdococ> evaluate((3 + 2) + 1) = 5 + 1
16:10:52 <rdococ> pairs of functions that cancel each other
16:11:06 <rdococ> successor(predecessor(x)) = predecessor(successor(x)) = x
16:11:22 <rdococ> sqrt(x^2) = sqrt(x)^2 = x
16:13:46 <rdococ> x * sqrt(y) = sqrt((x^2) * y)
16:14:49 -!- J_Arcane has quit (Ping timeout: 252 seconds).
16:14:58 <rdococ> 1 * sqrt(-1) = sqrt((1^2) * -1)
16:15:10 <int-e> it gets messy if you allow complex numbers
16:15:58 <rdococ> we could make a computer system that treats impossible numbers as if they were normal - basically a way to check if expressions evaluate to the same value without evaluating them
16:18:09 <int-e> Because of branches... and arbkitrary choices for "principal values". (The same is true for real numbers, but there the mess can be reduced to just discussing signs.)
16:18:10 -!- bender| has quit (Read error: Connection reset by peer).
16:23:25 <b_jonas> fungot, are you arbkitary?
16:27:08 <rdococ> i^2 = j^2 = k^2 means i = j???
16:27:59 <int-e> rdococ: are you looking at quaternions?
16:28:48 <rdococ> but it breaks rules I find comfortable
16:29:08 <coppro> rdococ: it's no different than 1^2 = (-1)^2
16:29:37 <int-e> rdococ: if so, note that multiplication there isn't even commutative, so (i-j)(i+j) = i^2 + ij - ji - j^2 != i^2 - j^2... so i^2 = j^2 does not even imply i = j or i = -j.
16:30:20 <int-e> (but the distributive laws continue to hold)
16:32:48 <int-e> (in fact, i^2 + ij - ji - j^2 = 2k whereas i^2 - j^2 = 0)
16:42:48 <\oren\> c++ seems to compile mmuch slower than an equivalent loc of c
16:43:25 -!- ais523 has joined.
16:43:36 <\oren\> i wonder how D and rust compare
16:43:49 <coppro> C++ needs to do semantic analysis during the parisng
16:44:04 <int-e> and expand templates during compilation
16:44:08 <rdococ> there's C, D, is there E?
16:44:16 <b_jonas> \oren\: you know there are compiler options to modify how quickly you compile, and in particular with gcc, -O (same as -O1) is great for improving compilation speed compared to -O2 (which does more optimizations)
16:44:21 <Kaynato> Could I ask for some c help?
16:44:25 <coppro> Kaynato: you can try ;)
16:44:32 <rdococ> I know there's F, perhaps G, idk about H or I, there is J, idk about K...well yet anyway
16:44:44 <b_jonas> also, don't use ancient versions of compilers, because some of those really are slow
16:45:02 <Kaynato> Ok, my program is crashing unexpectedly on typical input only of one sort
16:45:14 <b_jonas> rdococ: there are languages for every ascii letter, probably more than one language for half of them
16:45:18 <Kaynato> If I alter the input with inconsequential anything, it doesn't crash
16:45:28 <rdococ> what about Unicode letters
16:45:29 <Kaynato> If I put it in gdb, it doesn't crash
16:45:36 <b_jonas> rdococ: and there is a quite famous apl-like called K, and there was some crazy language called E somewhere
16:45:39 <coppro> Kaynato: you have undefined behaviour somewhere, probably
16:45:54 <coppro> Kaynato: start by turning off optimizations
16:45:56 <Kaynato> Other times calloc doesn't want to allocate 4 bytes
16:46:05 <coppro> possibly run with ubsan
16:46:07 <\oren\> uninitialized varriables?
16:46:20 <Kaynato> Optimizations are off, I am compiling for gdb
16:46:55 <Kaynato> All variables areinitialized
16:47:50 <\oren\> also check for division by zero
16:48:39 <coppro> rdococ: it's the name of a programming language specialized for use in the entertainment industry
16:49:04 <Kaynato> Can I run valgrind on windows?
16:49:18 <ais523> even if you could, it only runs Linux binaries
16:49:19 <coppro> ubsan should work though
16:49:29 <ais523> msan is Valgrind-like and probably works on Windows
16:49:43 <ais523> (and msan and ubsan complement each other, they catch different classes of bugs)
16:49:45 <\oren\> which compiler are you using?
16:51:06 <int-e> at some point you should start ruling out hardware problems, not sure whether you've reached it yet
16:51:35 -!- tromp has joined.
16:51:58 -!- dingbat has joined.
16:53:02 <b_jonas> (but chrome crashing doesn't mean that much)
16:53:24 <b_jonas> Kaynato: check if maybe your system's memory use is so high that programs run out of memory.
16:53:44 <int-e> it's not the most likely of causes... and it tends to produce intermittend rather than reproducible errors anyway
16:53:49 <b_jonas> or if the program you're running is allocating too much memory which is how it runs out.
16:53:52 <Kaynato> I mean, I can run this and have it allocate even a few mb
16:54:00 <Kaynato> it's just on this specific program that calloc fails for some reason
16:54:39 <b_jonas> Kaynato: usually that's just a memory corruption bug elsewhere in the program, eg. indexing past an array or using a stale or uninitialized pointer
16:55:04 <b_jonas> like a pointer to a freed object
16:55:53 -!- tromp has quit (Ping timeout: 244 seconds).
16:56:00 <Kaynato> it's really strange, I'm using really simple code and it causes this, while complex code doesn't
16:56:12 <Kaynato> I'll make a simpler example to see if the error is triggered
16:56:41 <Kaynato> Ok, so I can run the output file just fine
16:56:52 <Kaynato> But on the default of compile-interpret, it breaks
16:58:58 -!- hppavilion[1] has joined.
16:59:15 -!- mihow has joined.
17:01:06 <ais523> sounds like you have a memory corruption bug somewhere
17:07:09 <ais523> Kaynato: try -fsanitize=address -fsanitize=undefined when compiling with gcc (you'll need a recent version of gcc)
17:07:23 <ais523> that'll turn on asan (which I called msan earlier due to misremembering the name) and ubsan
17:08:07 <ais523> then run your proram as normal
17:09:26 <ais523> I'm not sure if this works on Windows yet; hopefully it does though
17:10:19 -!- Kaynato has quit (Ping timeout: 244 seconds).
17:10:32 <ais523> oh, Kaynato probably didn't see my reply :-(
17:10:53 <rdococ> if I were to make an infinity, it would be a neutral number like 0
17:11:01 <rdococ> okay, if you call it a number
17:11:34 -!- Kaynato has joined.
17:12:35 <ais523> <ais523> Kaynato: try -fsanitize=address -fsanitize=undefined when compiling with<ais523> then run your proram as normal gcc (you'll need a recent version of gcc) <ais523> that'll turn on asan (which I called msan earlier due to misremembering the name) and ubsan <ais523> I'm not sure if this works on Windows yet; hopefully it does though
17:13:19 -!- lambda-11235 has joined.
17:13:27 <rdococ> infinity is basically the 0 that's 1/0
17:13:40 <rdococ> infinity is like a second 0
17:14:30 <rdococ> I heard of this projective circle thing that goes from 0 -> 1 -> 2 -> 3 -> infinity -> -3 -> -2 -> -1 -> 0
17:15:24 <ais523> rdococ: right, that's basically projective geometry
17:15:28 <ais523> it works in more dimensions than just 1D
17:15:47 <ais523> and has some other fun properties, e.g. going past infinity and back from the other side makes an object into a mirror image of itself
17:16:21 <rdococ> so you can have complex projective numbers and cool extra transformations?
17:18:01 <ais523> I'm not sure if it works with complex numbers rather than just x/y coordinates, but it probably does
17:18:13 <ais523> because infinity is a point, not a line
17:18:48 <rdococ> x/y is the same thing as the complex plane
17:18:53 <rdococ> at least I think it is
17:18:55 <ais523> in 2D projective geometry, there is one "point at infinity" for each angle a line can have
17:19:09 <ais523> two parallel lines intersect at the point at infinity for their angle
17:19:16 <ais523> so there is a line at infinity that contains all the points at infinity
17:19:20 <rdococ> can a triangle have one angle in infinity?
17:20:16 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
17:20:38 <Kaynato> An angle of infinity is not particularly meaningful
17:21:39 <Kaynato> ais523: Riemann sphere is the likely one you are thinking about for complex numbers
17:22:07 <ais523> I'm not an expert on projective geometry, I've hardly even seen references to it
17:22:35 <ais523> rdococ: I don't see why you couldn't have a triangle formed out of two crossing lines, with the third at infinity
17:32:05 <int-e> ais523: because there's a whole line at infinity, and the two intersecting lines will intersect that one in different points.
17:32:46 <ais523> int-e: yes, thus forming a triangle
17:33:02 <ais523> one of the sides runs from one of the relevant points-at-infinity to the other
17:33:15 <int-e> ais523: oh, it wasn't clear that the line at infinity would be one of the sides
17:33:16 <ais523> the other two run from the points-at-infinity to a single point in "finite space"
17:33:20 <int-e> ais523: that's fine then
17:35:26 <int-e> oh, another parsing failure...
17:35:53 -!- oerjan has joined.
17:36:19 -!- lleu has quit (Quit: That's what she said).
17:39:33 <oerjan> @tell myname <myname> doesn't look that impressive <-- what do you mean, not impressive? all the panels share a theme despite not being just repetition of each other! (ok so the point is that it's usually far worse hth)
17:41:14 <ais523> oerjan: this is a mezzacotta comic?
17:41:56 <ais523> that's actually… mildly funny, at least
17:42:04 <ais523> you are right, it is unusually good for mezzacotta
17:45:18 -!- Frooxius has joined.
17:45:54 -!- spiette has quit (Quit: :qa!).
17:46:58 <lambda-11235> An old stack based language I wrote a long time ago, https://github.com/lambda-11235/tstk.
17:47:20 <oerjan> <rdococ> am I the only one who didn't make a brain**** derivative when I first joined? <-- technically, have i ever? the one i added recently already existed with a different name.
17:48:46 <oerjan> i've certainly implemented a few, though. although i don't think i implemented brainfuck _itself_ until i did it in Fueue.
17:50:02 <ais523> oerjan: did you invent small-finite-tape BF?
17:50:10 <ais523> like, with a single-digit length?
17:51:10 <oerjan> lambda-11235: why does github show README.html unrendered...
17:51:30 <ais523> oerjan: because it's not a Markdown file and github doesn't render anything but markdown
17:51:33 -!- zadock has joined.
17:51:57 <oerjan> ais523: cannot imagine so, there was already a proof by someone else that 5 were enough after all.
17:52:13 <lambda-11235> oerjan: I wrote it a long time ago, and for some reason I decided to use html for the README.
17:52:46 -!- tromp has joined.
17:53:21 <ais523> backspace is next to return, as usual
17:53:43 <ais523> hmm, why am I apologising for my typos, I used to just send them to #esoteric without caring
17:53:49 <ais523> presumably I'm on better behaviour than normal
17:53:53 <lambda-11235> oerjan: Write a program to convert html to markdown, and call it markup.
17:53:58 <oerjan> ais523: and you invented reversible brainfuck before i needed it for jolverine
17:54:08 <ais523> oerjan: I didn't realise you needed it
17:54:27 <oerjan> lambda-11235: doesn't pandoc do that? maybe it only goes the other way.
17:54:52 * oerjan doesn't really know pandoc
17:55:06 <ais523> now you're making me want to work on rtfm again
17:55:16 <oerjan> ais523: jolverine is reversible and has a tape, so it seemed like the obvious thing to try...
17:55:32 <ais523> (the basic idea is that it statistically analyzes text documents to work out what notation they use for headings, titles, etc., then converts into other formats)
17:55:42 <oerjan> and then i had to prove rev. bf TC first
17:55:43 <ais523> (I haven't even really started it yet, though)
17:57:06 <ais523> you know, looking at that page
17:57:15 <ais523> in retrospect I should probably define , to add the input to the current cell
17:57:22 <ais523> rather than crash the program if the current cell is nonzero
17:57:45 <oerjan> <ais523> presumably I'm on better behaviour than normal <-- hasn't that day passed
17:57:52 <lambda-11235> oerjan: Hold on, converting and tidying up the markdown.
17:57:55 <ais523> oerjan: it was earlier this month
17:57:57 -!- tromp has quit (Ping timeout: 246 seconds).
17:58:04 <ais523> however, I am sometimes well-behaved on other days too
17:58:04 <HackEgo> Agent “Iä” Smith is an alien with a strange allergy to avian body covering, which he is trying to retroactively prevent from ever evolving. On the 3rd of March, he's lawful good.
17:59:33 <oerjan> ais523: next you need to merge rtfm with aimake hth
17:59:50 <ais523> oerjan: it'd be possible
18:00:01 <ais523> aimake uses POD for documentation atm (both its own and that of programs it's installing)
18:00:32 <ais523> but that's partly because it ships with Perl, and partly because it's one of the few formats that has all the features that documentation really needs (the other is info and everyone hates info)
18:01:59 -!- idris-bot has joined.
18:02:11 <oerjan> i recall they changed the doc format for ghc last year.
18:02:35 <oerjan> although the new one _did_ lose a few features (deep nesting, i think)
18:02:51 <oerjan> but it was still considered a win.
18:03:02 <ais523> oh, hmm, asan and msan both exist and do different things, and valgrind does both those things
18:03:15 <ais523> (asan = check for addressability, msan = check for definedness)
18:05:23 <oerjan> from DocBook to ReStructuredText https://ghc.haskell.org/trac/ghc/wiki/UsersGuide/MoveFromDocBook
18:08:01 <oerjan> hm, email from norwegian tax authorities
18:09:23 <ais523> oerjan: well at least you're actually norwegian, so it makes more sense that you receive the email than that, say, I do
18:09:49 <oerjan> the spam filter marked it as possibly spam :P
18:09:56 <oerjan> but there's no link in it so...
18:11:29 -!- shikhin_ has changed nick to shikhin.
18:11:32 <oerjan> ais523: it's an email everyone gets, that the tax forms are ready
18:13:09 <oerjan> (everyone they've got email for, that is. i _almost_ managed to not use the official login long enough that they stopped considering my email valid)
18:13:58 -!- spiette has joined.
18:16:51 <ais523> there are three jobs you really want for a format like that: a) capable of representing arbitrary formatting; b) looks like a regular unformatted text file; c) easy for someone unfamiliar with the format to edit
18:16:57 <ais523> and it pretty much fails at all three
18:18:27 <oerjan> that's what they call an excellent compromise hth
18:20:08 <ais523> I've been considering making my own, going all-in on b) and only aiming for the other two to the extend they don't interfere
18:20:24 <ais523> was considering going so far as to allow people to write directives as sequences of spaces and tabs at the end of lines, Whitespace-style
18:25:36 <int-e> schlock in a holding cell, what did I miss?
18:25:37 <lambda-11235> ais523: I would want something that goes for a. A kinda of markup language that's turing complete with arbitrary drawing capabilities.
18:25:42 <b_jonas> ais523: I don't think perldoc has all the features that documentation needs. there's one feature I'm really missing
18:26:05 <ais523> lambda-11235: like HTML+JS, but less insane?
18:26:22 <int-e> (probably have to wait for the next strip for an explanation... sigh)
18:27:01 <b_jonas> well, at least with the current perldoc interpreters. it's possible that it could be fixed in the future by changing those.
18:27:11 <lambda-11235> ais523: I considered that, but yes, it needs to be more sane, and prettier to look at.
18:27:15 <ais523> b_jonas: you still haven't told me what the feature /is/
18:27:54 <lambda-11235> Like LaTeX, but easier to program and do graphics in.
18:28:04 <b_jonas> you know how perl has =begin and =for directives to put renderer-specific parts in a perldoc, so you can say something like eg. =for *HTML \n <img href="someimage.png"> to embed an image in HTML output
18:28:20 <ais523> =for aimake manualsection 6
18:28:37 <b_jonas> The feature I'm missing is the same thing in negative, so that you can easily put fallbacks for other renderers
18:29:04 <b_jonas> it's possible to do this for some renderers by putting the fallback text unconditionally and using some renderer-specific method to hide that text
18:29:04 <ais523> a "=for everythingelse"?
18:29:11 <b_jonas> ais523: no, more like a =for !HTML
18:29:26 <ais523> that wouldn't be useful unless you could hide from two or more renderers at once
18:29:32 <b_jonas> ais523: you can with =begin
18:29:40 <b_jonas> ais523: non-starred =being blocks can be nested
18:29:56 <b_jonas> obviously =for !HTML is the wrong syntax for it, because it's not backwards compatible
18:30:35 <int-e> `le/rn markdown/What will your markdown flavor be today?
18:30:48 -!- zadock has quit (Ping timeout: 276 seconds).
18:30:55 <b_jonas> we could use something like =for begin !\nHTML\n\n ... \n\n=for end !
18:31:08 <b_jonas> which would be ignored by existing renderers
18:31:19 <b_jonas> so the text inside would be kept
18:31:58 <b_jonas> you can still do specific hacks like =for HTML\n<!-- comment the fallback text (let's hope the HTML renderer doesn't emit a comment itself)\n\n
18:32:42 -!- lambda-11235 has quit (Ping timeout: 260 seconds).
18:33:07 <int-e> `le/rn markdown/The markdown flavor of the day is raspberry.
18:33:13 -!- hppavilion[1] has joined.
18:34:02 <oerjan> `` ls -l wisdom/password
18:34:03 <HackEgo> -rw-r--r-- 1 5000 0 39 Mar 5 17:33 wisdom/password
18:34:15 <HackEgo> The password of the month is qjkxbmwvz
18:34:19 <b_jonas> although it'd probably rather have to be \n\n=for *begin !\nHTML\n\n ... \n\n=for *end !\n\n with a star because it's a paragraph of non-pod content
18:34:24 <oerjan> a bit early to change.
18:34:39 <ais523> oerjan: is this related to our spam filter?
18:34:54 <oerjan> ais523: almost certainly not.
18:35:07 <ais523> on another site I use which has a wiki
18:35:18 <ais523> they put a password on their main site, behind a login
18:35:25 <ais523> as in, the site stated the password
18:35:36 <ais523> and the wiki CAPTCHA asked for it
18:36:00 <ais523> I was thinking that if we had real spambot trouble on the wiki we could make it ask for a wisdom entry
18:36:06 <ais523> so that people would have to come onto IRC to sign up
18:36:07 <b_jonas> Can I ask about web development here?
18:36:10 <ais523> (the CAPTCHA seems to be holding atm)
18:36:18 <ais523> b_jonas: it really depends on what the question is, I imagine
18:36:36 <oerjan> ais523: the CAPTCHA wasn't holding a while ago, but the filters did.
18:36:36 <b_jonas> nah, I should try to search first because they're FAQs
18:37:51 <oerjan> ais523: captcha still not holding
18:38:24 <oerjan> although i suspect fizzie hasn't tried changing it since he realized they were just solving them fast
18:38:55 <ais523> huh, one of those is a new spambot from the edits, just seems to be caught into an old filter
18:39:08 <ais523> in particular, it is trying to vandalize legitimate users' userpages
18:39:26 <ais523> but still doesn't know how to newline
18:40:18 <ais523> I'm glad we have AbuseFilter, that's a huge amount of spam we'd have to clear up manually otherwise
18:41:26 <ais523> you know, it's almost worth blocking 91.200.12.0/24 at this point
18:41:32 <ais523> lots of spam is coming from that range
18:42:26 <ais523> remarks: The pool is used other Department! remarks: In case of questions related to SPAM, HACKING, SECURITY remarks: Please contact directly abuse@vhoster.net
18:42:51 -!- zadock has joined.
18:42:56 <b_jonas> ais523: if you block it, the spam will come from other ranges you don't know about, so it's not worth blocking until they actually make edits
18:43:18 <ais523> b_jonas: they're making a ton, they're just all getting caught in the filter
18:43:19 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
18:43:23 <ais523> the block would help prevent them spamming up the spam filter :-P
18:43:42 -!- zadock has quit (Remote host closed the connection).
18:44:11 -!- MoALTz has joined.
18:44:55 <oerjan> ais523: btw i lost patience and banned Lilly_Goodman earlier, izabera thought it was rude though.
18:44:57 <int-e> the first page holds 2 weeks of edits...
18:45:08 <APic> ais523: /me would put „CRACKING“ or „BLACK-HAT-HACKING“ there instead of „HACKING“ ;)
18:45:15 <ais523> oerjan: on the wiki or IRC? I don't recognise the name
18:45:21 <ais523> APic: I didn't write it
18:45:36 <ais523> that also isn't the only error
18:45:39 <ais523> the grammar is not very good
18:46:46 -!- hppavilion[1] has joined.
18:47:18 <oerjan> ais523: you accidentally avoided seeing em active, i guess
18:47:40 <oerjan> it was one of the canaima users, just one who refused to accept that we didn't speak spanish
18:48:12 <oerjan> and who showed no indication of grasping what the channel was about
18:48:28 <oerjan> ais523: the venezuelan OS that sends us all the spanish-speakers
18:48:42 <ais523> oerjan: why is a venezuelan OS pointing spanish speakers at this channel?
18:48:46 <hppavilion[1]> oerjan: I've tried to get her to leave, but she doesn't seem to understand
18:48:56 <hppavilion[1]> ais523: I assume "esoteric" is a word in other languages
18:48:59 <lambdabot> oerjan said 14h 1m 58s ago: <hppavilion[1]> http://dl.acm.org/citation.cfm?id=96559.96570 has &lgr; for some reason. I have no clue what that entity means <-- it's a lambda hth
18:49:25 <oerjan> ais523: my latest theory is that it's just because #esoteric comes just before #espana alphabetically
18:50:04 <oerjan> ais523: @messages alone is quiet
18:50:15 <ais523> oerjan: I was testing the parser, not the functionality
18:50:26 <ais523> I don't actually have any messages right now
18:50:33 <oerjan> so perhaps people search for channels starting with #esp and it just shows up in the same window
18:50:36 <b_jonas> oerjan: oh, we could solve that by setting up a channel between the two
18:51:13 <oerjan> (this theory improves on the old one in that it doesn't require the canaima OS to include us erroneously)
18:51:13 <int-e> to be fair, extra-sensual phenomena are kinda esoteric
18:51:24 <int-e> or is that sensory
18:51:34 <oerjan> b_jonas: well, they might still have a channel list, so it might not be picked up
18:52:03 -!- AnotherTest has joined.
18:52:15 <b_jonas> oerjan: sadly it's not so simple. there are channels between, including #esperanto and #esprima and others, but they're not very popular
18:52:23 <b_jonas> so we have to set up a POPULAR channel in between
18:52:31 <b_jonas> one that search bots rate high
18:53:11 <b_jonas> oerjan: do you know what client they use? webchat or not, and which webchat?
18:53:32 <oerjan> btw s/OS/linux distribution/
18:53:54 <ais523> b_jonas: those aren't in between
18:54:06 <ais523> "esperanto" comes after "espana" and afet "esoteric"
18:54:30 <ais523> however, #esoteric-offtopic is between #esoteric and #espana
18:54:42 <ais523> wait, it doesn't exist
18:54:47 <hppavilion[1]> https://en.wikipedia.org/wiki/Canaima_%28operating_system%29
18:55:22 <oerjan> btw the automatic ban irssi chose is wide enough that it _might_ block a large number of canaima users. i'm not sure whether that is a bad thing at this point.
18:55:25 <ais523> I was last there in 2014
18:55:32 <b_jonas> although that could depend on the locale
18:56:18 <hppavilion[1]> The Venezuelan Government prioritizes FOSS in their administration
18:57:39 <b_jonas> hppavilion[1]: that's usually just a lie they tell in the news meaning that they don't want to pay money for the MS Office licenses this year but also don't help the administration people or bosses in any way for what they should do if they don't get licenses
18:57:51 <b_jonas> (or at least that's how it works here)
18:58:19 <hppavilion[1]> b_jonas: Yes, but the official story is comforting enough
18:58:26 <oerjan> <hppavilion[1]> ais523: I assume "esoteric" is a word in other languages <-- it's greek, so has been borrowed a lot. spanish seems to have it as esotéric{o,a}s?.
18:58:59 <ais523> hppavilion[1]: the Venezuelan economy is in shambles
18:59:04 <ais523> they basically based the whole thing around oil
18:59:18 <ais523> then when the oil price collapsed they became a lot poorer
18:59:21 <hppavilion[1]> I'm making a browser called WaterOwl or something.
18:59:45 <hppavilion[1]> I mean, we have shitty Ubuntu laptops at my local High School
19:00:30 <oerjan> <ais523> oerjan: I was testing the parser, not the functionality <-- afaik it will recognize either (1) a prefix of a unique command, or otherwise (2) edit distance <= 2 with unique best hit.
19:00:57 <b_jonas> oerjan: but wait, if they don't have canaima installed, then what sets their irc username to canaima?
19:01:47 <b_jonas> oerjan: although the two explanations aren't exclusive, they can use canaima but also look in some channel list
19:01:48 -!- Phantom_Hoover has joined.
19:02:16 -!- dingbat has quit (Quit: Connection closed for inactivity).
19:02:23 <oerjan> <b_jonas> oerjan: do you know what client they use? webchat or not, and which webchat? <-- gah i checked but i've forgot
19:03:07 <b_jonas> oerjan: wait... what if their software lists channels from some centralized canaima irc client list of most popular channels, so it's just a random channel that is popular among canaima irc users because the previous users clicked on that channel too, and that irc client is so bad that no sane person uses it
19:03:49 <b_jonas> oerjan: also worth checking when they join is what other channels they're joined to
19:04:36 <b_jonas> shouldn't we have the wiki url in the channel topic?
19:06:56 <rdococ> we should take a vote on what everyone thinks is the "circle constant"
19:07:44 <rdococ> hppa: well, you can pick any number
19:07:55 <rdococ> I choose i, or square root of 2
19:08:09 <oerjan> i'm with hppavilion[1] on i.
19:08:32 <rdococ> the square root of i? I think it's, um,
19:08:39 <oerjan> because e^(ix) gives a circle q.e.d.
19:08:56 <rdococ> it's crazy how e is related to circles...what is e even defined as?!
19:09:11 -!- Kaynato has quit (Ping timeout: 244 seconds).
19:09:23 <ais523> oerjan: was checking logs, she was here yesterday but before that wasn't here for over a week
19:09:28 <oerjan> b_jonas: someone snipped all the urls except the log ones a while ago
19:09:39 <ais523> so it's not a case of being persistent, but rather of checking back every now and then to see if the channel is Spanish again yet
19:09:45 <rdococ> pi is the area of a unit circle, tau is the --wait, do we even need a circle constant?
19:09:50 <b_jonas> oerjan: well the log url definitely has to be there
19:10:01 <hppavilion[1]> rdococ: You can use e^(ix) to prove that i^i is real
19:10:03 <oerjan> ais523: right, she went away for easter.
19:10:18 <b_jonas> (plus we should put both urls in a CHANSERV #esoteric PROPERTY SET too)
19:11:05 <hppavilion[1]> rdococ: The general form is e^(ix) = cos(x)+isin(x)
19:11:25 -!- Kaynato has joined.
19:11:27 <rdococ> how on earth do you get cosine and sine from e?
19:11:43 -!- b_jonas has set topic: The international hub of esoteric programming language and kitten typesetting | logs: http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | http://esolangs.org/ | Note: people with cloaks will be treated as if they're from Budapest.
19:11:58 <rdococ> e^(ix) = e^(sqrt(-1)x)
19:12:10 <ais523> rdococ: multiplying by e^i is basically a rotation of 1 radian
19:12:21 <ais523> thus e^it describes a circle
19:12:21 <hppavilion[1]> rdococ: So e^(ix) where x=π/2 is cos(π/2)+sin(π/2)
19:12:31 <ais523> if you take the x and y coordinates of a circle you get sin and cos
19:12:43 <rdococ> but where does e come into it?
19:12:46 <lambdabot> 0.7071067811865476 :+ 0.7071067811865475
19:12:56 <rdococ> put it this way, why is x^i so special when it comes to e?
19:12:56 <ais523> rdococ: it's more in reverse, e is defined as the number that makes this work
19:13:26 <rdococ> how would exponentation do that
19:13:35 <rdococ> actually, who needs angles
19:13:50 <ais523> rdococ: well you know how the derivative of e^x is e^x?
19:13:56 <rdococ> I want 1/sqrt(2) + i/sqrt(2) of that pie
19:14:11 <ais523> without any scaling, like you get if you use a number other than e?
19:14:12 <rdococ> the derivative of e^x is e^x? wow
19:14:18 <ais523> the derivative of sin x is cos x
19:14:26 <ais523> so the fourth derivative of sin x is sin x
19:14:27 <int-e> rdococ: only with respect to x
19:14:40 <ais523> this means that repeatedly differentiating a sin or cos doesn't change the magnitude of the answer
19:14:41 <rdococ> the fourth derivative of sin x is sin x, right...?
19:14:55 <ais523> same with e^x, and with no other base of an exponential
19:15:10 <b_jonas> ais523 or oerjan: can you put the three urls into a chanserv property, eg. /quote CHANSERV :#esoteric SET PROPERTY LOGS http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D
19:15:12 <ais523> so really you have to end up with e in the base if you implement sin and cos in terms of exponentials, otherwise they'd be changing the magnitude
19:15:20 <ais523> b_jonas: that's two URLs
19:15:34 <ais523> also nobody bothers to check chanserv properties
19:15:38 <rdococ> so why does e^i do what it does?
19:15:39 <ais523> it needs to be in the topic or the join message
19:15:49 <rdococ> I'm more confused about exponentiation by i
19:15:51 <b_jonas> ais523: the third is to a separate property: /quote CHANSERV :#esoteric SET PROPERTY WEB http://esolangs.org/
19:16:07 <b_jonas> ais523: "nobody bothers to check chanserv properties" => that's exactly why they're esoteric
19:16:11 <int-e> the wisdom link is gone too
19:16:48 <HackEgo> I think you might mean !logs
19:17:24 <ais523> how do you even read channel properties anway?
19:17:27 -!- lambda-11235 has joined.
19:17:34 <b_jonas> ais523: CHANSERV INFO #esoteric
19:17:57 <ais523> anyway there doesn't seem to be a standard for channel properties
19:19:44 <oerjan> <ais523> same with e^x, and with no other base of an exponential <-- (1/e)^x hth
19:19:54 <b_jonas> ais523: yes, they're free-form
19:20:01 <hppavilion[1]> e^(-π/2) is real -- Some math I don't know how to explain
19:20:14 <ais523> hppavilion[1]: or used #esoteric-blah
19:20:18 <ais523> that's what it was for, originally
19:20:26 <ais523> things that produce a lot of text/spam
19:20:31 <ais523> either pastes or commands
19:20:36 <b_jonas> hmm, maybe I should set chanserv property to the channels I lead before asking for it. although on those channels I just keep the info in the TOPIC, unlike here
19:21:27 <ais523> oerjan: doesn't that differentiate to -(1/e)^x ?
19:21:35 <ais523> not 100% sure, I'm not that good at doing calculus in my head
19:21:48 <ais523> oh I see, I said "magnitude"
19:21:54 <ais523> and you're nitpicking on that :-D
19:21:59 <rdococ> you're just doing math
19:22:12 -!- evalj has joined.
19:23:09 <evalj> ais523: |value error: hepl
19:23:11 <evalj> ais523: |value error: help
19:23:34 <b_jonas> ] '] is it''s short prefix I think'
19:23:34 <evalj> b_jonas: ] is it's short prefix I think
19:23:50 <b_jonas> evalj, help: would be how you invoke such a command
19:23:55 <b_jonas> but there's no help command I think
19:23:57 <evalj> b_jonas, jevalbot source is http://www.math.bme.hu/~ambrus/pu/jevalbot.tgz
19:24:29 <b_jonas> because "evalj: help" would be interpreted as the default command (evaluate J in session) with argument "help" which is a J statement
19:25:00 <ais523> ugh, now I remember a language or language parody in which writing "help" gave the response "there is no help for you"
19:25:08 <ais523> but can't remember the name of the language or alleged language
19:25:52 <oerjan> <ais523> and you're nitpicking on that :-D <-- actually i was just nitpicking on the f'''' = f case. if you consider magnitude in general you might also take e^(x*e^(ia)), a real.
19:26:30 <ais523> oerjan: there is definitely more than one f for which f''''=f, though
19:26:35 <b_jonas> ais523: I think that's a text adventure (colossal-cave-like)
19:26:55 <ais523> it's an obvious enough joke that maybe it's happened in multiple places
19:27:00 <b_jonas> although that would have to say something like "nobody can hear as you scream for help"
19:27:17 <b_jonas> "You call for help but nobody comes."
19:27:21 <ais523> hmm, now I have the idea for a dudley's dungeon comic
19:27:51 <oerjan> ais523: well yes, f''''=f is a linear combination of those with a = n*pi/2
19:28:39 <ais523> well, f(x) = sin x gives you f''''=f
19:28:43 <ais523> how does that fit into this framework?
19:29:06 <oerjan> sin x = (e^(ix) - e^(-ix))/(2i)
19:29:43 <oerjan> and a similar expression for cos
19:31:03 <evalj> b_jonas: 0.540302j0.841471
19:31:03 <ais523> oerjan: oh, I missed the linear combination thing
19:31:37 <oerjan> aww, i was hoping i'd blown your mind :P
19:33:27 * oerjan aims it at rdococ instead.
19:34:20 <oerjan> rdococ: btw ais523 alluded to a more general derivative. (a^x)' = ln a * (a^x), where ln a is the natural logarithm. so putting ln a = 1 gives you e there.
19:35:28 <oerjan> also ln x itself has a nice such property: (ln x)' = 1/x
19:36:07 <oerjan> (you could use that property to unravel all the other definitions, if you don't like to define e^x with an infinite series as is the usual alternative.)
19:36:33 <b_jonas> oerjan: so basically you define ln(x) as a simple integral?
19:36:52 <b_jonas> unlike e^x which you try to define as a differential equation or something
19:37:55 <oerjan> i am a bit partial to starting with (ln x)' = 1/x since that was the first method i learned.
19:42:23 <b_jonas> oerjan: I think the problem is that if you try to define e^x from a differntial equation, then you have to prove somehow that the differential equation has a solution, and probably also that it's unique, and that requires higher calculus, and might even end up being circular if you're not careful.
19:43:33 <b_jonas> oerjan: so I think it's better to take the power series e^x = sum_(0<=n) x^n/n!, prove that that converges everywhere, then prove that e^(xy) = e^x*e^y
19:44:45 <oerjan> b_jonas: you can also prove that ln has an inverse.
19:45:00 <b_jonas> and then from that prove that e^x is the limit of e^q where q is rational and goes to x and e^q is defined as rational power with repeated multiplications and square roots which you first have to prove as well-defined and monotonous
19:45:10 <oerjan> (monotone continuous function)
19:46:21 <b_jonas> That is, you define power in three ways, one for integer exponents, one for rational exponents and positive base with roots, and one for reals with power series, prove their nice properties (well-definedness and monotonity and identities) separately, then prove they're the same.
19:46:32 <oerjan> um your last line seems rather redundant with "e^x is continuous"
19:46:37 <b_jonas> But I think the linear differential equation has to enter it until much later.
19:46:45 <oerjan> (it was your last line when i started typing)
19:47:10 <b_jonas> oerjan: sure, you have to prove e^x is continuous, but you also have to prove that it equals to the other definition of e^q for rational q
19:47:15 <b_jonas> those two together mean it's the limit
19:48:09 <b_jonas> You can actually do that even before you define function limits, as long as you've defined series limits, because monotonous function with no jumps is an easier case
19:48:26 <b_jonas> you can just prove that e^x the unique _monotonous_ function that extends e^q
19:48:53 <oerjan> i'm really saying that you can shortcut much of the "prove their nice properties separately" by just proving e^x is nice enough and use that to investigate the others.
19:49:39 <b_jonas> oerjan: yes, of course, but you need to prove at least something like e^(x+y)=(e^x)(e^y) specifically to be able to prove it's a power function
19:54:52 -!- tromp has joined.
19:57:18 -!- lambda-11235 has quit (Read error: Connection reset by peer).
19:59:51 -!- tromp has quit (Ping timeout: 276 seconds).
20:05:03 -!- Kaynato has quit (Ping timeout: 240 seconds).
20:05:23 -!- fucrktheyhw has joined.
20:07:41 -!- oerjan has quit (Quit: Nite).
20:15:00 -!- fucrktheyhw has left.
20:27:43 -!- m0li has joined.
20:35:39 <HackEgo> U+1F4AF HUNDRED POINTS SYMBOL \ UTF-8: f0 9f 92 af UTF-16BE: d83ddcaf Decimal: 💯 \ 💯 \ Category: So (Symbol, Other) \ Bidi: ON (Other Neutrals)
20:36:19 <rdococ> the symbols don't show up to me
20:36:28 <lynn> They are emoji!
20:36:47 <rdococ> so what does this look like? :)
20:37:16 <lynn> 100 points, flower, thought bubble, money bag
20:39:10 <ais523> I should probably give up on assuming any sort of logical pattern to the way emoji are organized
20:40:28 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
20:55:41 <rdococ> what if you have 100 flowers in a thought bubble while you're holding a money bag in real life?
20:58:05 <ais523> I was hoping there'd be points symbols for numbers other than 100
21:16:48 -!- AnotherTest has quit (Ping timeout: 244 seconds).
21:23:26 -!- Kaynato has joined.
21:42:26 -!- Guest6809 has quit (Quit: This computer has gone to sleep).
21:55:54 -!- tromp has joined.
22:01:05 -!- fungot has joined.
22:01:13 -!- tromp has quit (Ping timeout: 244 seconds).
22:01:47 <fungot> shachaf: but it's ok... i have no idea how to find online help) and m-x ( execute command). you would see them as a countable set of countable sets? couldn't you just execute a move?
22:01:51 -!- hppavilion[1] has joined.
22:01:54 <fizzie> Also, exciting: they want me to decide who the next Mayor of London will be.
22:01:57 <fizzie> (Well, along with some other folks, I guess.)
22:02:17 <shachaf> fizzie: You're deciding about the Mayor of London and also about some other folks?
22:03:58 <fizzie> Yes, something about a London Assembly of London.
22:04:37 <HackEgo> fizzie is not fnord with a monad but the king of #esoteric, see http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/src/fizziecoin.jpg
22:06:43 <hppavilion[1]> It's probably done in every CAS, now that I think about it, but there's no standard
22:08:25 <hppavilion[1]> It's like regex, but it's for matching categories (the informal kind) of mathematical formula, rather than strings fitting a particular pattern
22:09:29 -!- spiette has quit (Ping timeout: 260 seconds).
22:17:38 -!- boily has joined.
22:20:06 <HackEgo> tomfoolery/tomfoolery is always factually inaccurate. always.
22:20:33 <ais523> You have chosen to open: fizziecoin.jpg which is: 23600 File (225 kB) from: http://codu.org What should Firefox do with this file? Open with: gedit
22:21:15 <shachaf> ais523: I don't understand why browsers don't have an option to override content-type when it's application/octet-stream.
22:21:30 <ais523> that's not application/octet stream, though
22:21:32 <fizzie> It's actually Content-Type: application/binary.
22:21:33 <ais523> assuming it's an actual jpeg
22:21:34 <shachaf> Most often I want it to be text/plain but sometimes other things.
22:21:38 <shachaf> Well, application/binary too.
22:21:42 <ais523> fizzie: is that a real content-type?
22:21:49 <shachaf> In fact I always want this option. But I especially want it in the download dialog.
22:21:56 <boily> his523. it's a real JPG! it's the fizziecoin!
22:22:12 <fizzie> ais523: I don't see it in http://www.iana.org/assignments/media-types/media-types.xhtml
22:22:19 <ais523> should be an "open in browser" option too
22:23:35 <shachaf> I guess allowing people to specify text/html can be problematic for XSSish reasons, maybe.
22:24:22 <ais523> Wikipedia refuses to serve anything as text/plain because IE is stupid
22:24:44 <ais523> it uses made-up content types instead like text/text or text/x-wiki if you try to force it to
22:24:58 <fizzie> hgweb used to guess content types from extension.
22:25:02 <ais523> the normal recommendation is to use text/css instead, which most browsers render like they /should/ render text/plain
22:25:04 <fizzie> They changed it for XSSish reasons.
22:25:23 <fizzie> (I mean, it served .html files as text/html and all that.)
22:26:04 <shachaf> boily: q: what content type does tra1n use?
22:26:05 <fizzie> Chromium just downloads the coin.
22:26:31 <boily> hellochaf. AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAmapoleAAAAAAAAAAAAAAAAAARGH!
22:26:33 <fizzie> "Content-Type: application/binary" + "Content-Disposition: inline" is kind of a strange combination, though.
22:26:45 <fizzie> "Inline this unspecified binary blob."
22:28:08 -!- jaboja has joined.
22:28:22 <ais523> it's more like saying "inline this unspecified bnry blob", it's a word that's sort-of like the correct one but isn't actually a real word
22:33:02 -!- jaboja has quit (Ping timeout: 260 seconds).
22:39:49 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
22:42:15 -!- ais523 has quit.
23:09:39 -!- FreeFull has quit (Ping timeout: 276 seconds).
23:09:57 -!- FreeFull has joined.
23:17:19 -!- hppavilion[1] has joined.
23:18:11 <HackEgo> glogbot/glogbot is a snitch, don't trust it.
23:21:21 -!- evalj has quit (Remote host closed the connection).
23:25:38 <HackEgo> persistence/Taneb invented persistence long ago, and it's been around ever since.
23:27:06 <int-e> . o O ( persistence is a property of properties of term rewriting systems )
23:28:27 <hppavilion[1]> https://www.quora.com/Which-is-the-most-unpleasant-programming-language-that-you-have-written-in is a godsend to #esoteric
23:31:56 <boily> aubergine is quite nice this time of the year. you should program in it.
23:32:37 -!- Lyka has joined.
23:33:09 <int-e> "Sure, there is a way to write $LANGUAGE the right way, but nobody does it." -- this seems applicable to many languages...
23:34:15 <int-e> (my ideas include mundane languages like PHP and JavaScript)
23:34:29 <Lyka> http://pastebin.com/CXrgJHdf
23:34:42 <Lyka> http://pastebin.com/S76vsBj9
23:36:05 <Lyka> examples of my current useless language
23:37:15 <hppavilion[1]> I'm going to answer it (I'm choosing Befunge) and point people to this channel with the promise of fungot
23:37:15 <fungot> hppavilion[1]: this is general discussion
23:37:29 <fungot> hppavilion[1]: but that's just me. dunno about converting linefeeds on windows.
23:37:53 <boily> Lyka: CISC architectures are nice. were you inspired by PDP assembly?
23:37:56 <fungot> hppavilion[1]: you can also use recursions. the only shortcomings that i'm aware of
23:38:05 <Lyka> what is fungot?
23:38:05 <fungot> Lyka: i think i have read the pdf, or 138 in the pdf
23:38:25 <boily> there's one fungot, and your fungot is fungot.
23:38:26 <fungot> boily: fnord iood helo' as the first example on the scheme-style page. i fnord 3 hours.
23:38:40 <boily> sometimes, he gets a little bit loopy...
23:39:02 <hppavilion[1]> fizzie: Awesome; that means I can mention that in my answer
23:39:15 <boily> fungot also tries to become sentient, but we're keeping him in check.
23:39:15 <fungot> boily: you have to resort to calling him. brr. that draft is chilly.
23:39:25 <fizzie> Well, I mean. The code is. Those babble models are "generated".
23:39:27 <boily> fungot: not chilly, only humid. it's getting warm here!
23:39:27 <fungot> boily: gah! what's with all the stuff
23:39:43 <hppavilion[1]> fizzie: What's the algorithm it uses to figure out how to talk?
23:39:46 <boily> fungot: it's spring. the seasun of puddles and mud and squirrels.
23:39:46 <fungot> boily: what is the value of x
23:40:04 <boily> fungot: 0 when not on the stack, otherwise whatever you paid for it when casting it.
23:40:06 <Lyka> boily: i once saw a refeerence sheet for 8051 asm.
23:41:20 <Lyka> the goal is that every command+argument structure is 16 bytes
23:42:35 <fungot> hppavilion[1]: it's probably actually worth 50e still since it looks, not to mention the pain that writing anything nontrivial in
23:43:46 <boily> Complex Instruction Set Computer. it was the fashion for mainframes to have as many different instructions they could manage to cram in.
23:44:04 <boily> then RISC happened, and now we're somewhere in the middle with current architectures.
23:45:35 <coppro> basically the way that a CISC worked is that they started to combine multiple instructions into one for efficiency reasons
23:45:43 * zgrep waits for EISC to happen, Esoteric Instruction Set Computer
23:46:01 * boily proudly touts the virtues of his aubergine ^^
23:46:02 <coppro> like having a variation of the add instruction that loads an operand directly from memory to avoid you having to load it yourself
23:47:51 <coppro> that's a super simple example, but if you look at a modern one like x86, it's extremely complex
23:47:55 <coppro> lots of ways to do the same thing
23:49:28 <coppro> AIUI, as the processors got better at optimizing and eventually started moving to things like microcode, they realized that the pipelining characteristics and relative code simplicity of splitting things back out seemed like a better approach
23:49:43 <coppro> the idea being that the processor could do the optimization that the compiler does in a CISC
23:49:57 <coppro> so they gave you lots of registers to work with, but comparatively few instructions
23:49:59 <zgrep> hppavilion[1]: I could, but [insert bad excuse here].
23:50:06 <coppro> the modern stuff is somewhere in the middle
23:51:21 <boily> RISC was the élément déclencheur for rigorous pipelining and orthogonality. but we still like some microcode happening in there, so middle ground seems to be the best course.
23:51:52 <boily> also, I hope to see the day were x86 disappears in purifying fire >:D
23:52:02 <zgrep> coppro: The modern stuff is also (il)logically backwards compatible, I think.
23:52:27 <boily> the Magic Blue Smoke Shall be Released, and Congeal Back into the Most Perfect Design Mankind can Dream of!
23:53:07 <zgrep> boily: That seems like an advertisement for recreational drug use.
23:53:27 <int-e> The idea of letting the compiler do all the optimizations just hasn't worked out, I think... dynamically you get more freedom because you know actual addresses, which allows moving, say, reads before writes even when a compiler couldn't.
23:54:14 <hppavilion[1]> zgrep: Yes, it's much better to force everyone to compile their applications 50 times for all the currently-in-use computers, /then/ to force users to learn enough to know what their computer's IS is so they can find the right executable in a massive list of executables with names that are probably just hexadecimal serial numbers :P
23:54:16 -!- Sgeo__ has joined.
23:54:47 <int-e> @google EPIC computing
23:54:48 <lambdabot> https://en.wikipedia.org/wiki/Explicitly_parallel_instruction_computing
23:57:57 -!- tromp has joined.
23:58:03 <hppavilion[1]> rdococ: Say, what programming languages do you use?
23:58:27 <rdococ> Lua, but I use it in the context of my favourite games website, Roblox
23:58:31 -!- tromp has quit (Remote host closed the connection).
23:58:44 <zgrep> hppavilion[1]: Technically, that's more so having a general agreed standard than backwards compatibility, but I see your point.
23:58:48 -!- tromp has joined.
23:58:51 <int-e> hppavilion[1]: some (CISC) instruction to heat the processor to a specified temperature would be useful for cooking
23:58:55 <rdococ> I've tried a browser implementation of BASIC.
23:59:13 <rdococ> (no, not that I've tried to implement it, but that I've used an implementation)
23:59:47 <rdococ> no, the other way round