00:00:16 <Taneb> http://tvtropes.org/pmwiki/pmwiki.php/TabletopGame/Sopio
00:00:24 <Taneb> It's not much good
00:02:09 <Taneb> This has nothing to do with my liking of Scrolls!
00:02:13 <zzo38> I see nothing about rules of the game not much information
00:02:18 <Taneb> You'd have to start a new topic for that!
00:03:25 <oklopol> so if you could be a magic ghost, would you sneak in peoples houses and watch them
00:04:08 <itidus20> genius loci reminds me of when people are sleeping on a small island only for the island to wake up and start walking as it is revealed to be infact an animal
00:04:14 -!- aloril_ has quit (Read error: Operation timed out).
00:04:46 <monqy> the sea just goes down forever, you should know, itidus20
00:05:08 <oklopol> no it doesn't, chine island is on the other side
00:05:31 <itidus20> monqy: what are the implications of an infinitely deep sea?
00:05:47 <monqy> it's why we have problems
00:06:06 <oklopol> damageinc: does it have a bottom at \omega though?
00:06:23 <oklopol> or at the first uncountable ordinal
00:06:24 <Phantom_Hoover> itidus20, eventually it turns into ice and it has a bottom, qed
00:06:39 <Gregor> Right above the turtle.
00:07:09 <oklopol> animals can't be islands. islands are invisible.
00:07:15 <oklopol> good night, i have no idea
00:07:48 -!- Phantom_Hoover has changed nick to Psalm_Journey.
00:07:50 <damageinc> was built on the dead head of jezus christ
00:08:19 <damageinc> the pope is the antichrist dont you know ?
00:08:24 <oklopol> that's blasphemy, damageinc
00:08:25 -!- aloril_ has joined.
00:08:38 <oklopol> blasphemy is against channel rules
00:08:40 * itidus20 repeats for dramatic effect. sleepy
00:09:01 <zzo38> I doubt the pope is the antichrist. But I agree many (not all) popes can do mistakes and bad things just as much as anyone else can
00:09:02 <oklopol> good night itidus20, have awesome dreams and tell us all about them
00:09:18 <oklopol> zzo38: you are religious of some sort right?
00:09:23 <damageinc> oh come on everyone what the pope in rome covers up ;)
00:09:35 <damageinc> oh come on everyone knows* what the pope in rome covers up ;)
00:09:51 <oklopol> err, obviously god cannot tell us everything
00:09:54 <zzo38> oklopol: Sort of. Not really like other people who are religious.
00:10:03 <damageinc> if the pope is not the antichrist then who is ?
00:10:09 <oklopol> you can't really blame the pope for keeping secrets like this.
00:10:32 <zzo38> damageinc: I my opinion, nobody in particular; it is simply the idea
00:10:37 <pikhq> damageinc: Nobody. There is probably neither christ nor anti-. :)
00:10:38 <itidus20> hmm i return with 1 bun. 1 cheese slice. 1 tomato.
00:10:48 <itidus20> proceeding to combine them into one object
00:11:05 <damageinc> you a denier of jezus christ pikhq ?
00:11:15 <Psalm_Journey> pikhq, thank you for your input into the conversation, please recalibrate sarcasm measurements.
00:11:18 <zzo38> You do not need alchemy to make a simple sandwich
00:11:53 <pikhq> Psalm_Journey: Internet sarcasm is utterly impossible to identify.
00:12:13 <zzo38> Psalm_Journey: Is not supposed to eat gold!
00:13:16 <zzo38> (Unless it is like, monster, in D&D game or whatever, that can eat gold; in which case there would be magic too, so you can use magic to make ham into gold if necessary)
00:13:26 <pikhq> damageinc: Well, I 'spose there's a dude named Jésus Christ or something.
00:13:29 <oklopol> evil is the belief that you have sufficient managerial skills that you can pull of short term bad stuff that will lead to long term good stuff.
00:13:36 <itidus20> zzo38: this was of little comfort watching the tomato juice pour into the bowl as i crushed it into shape without a knife
00:13:58 <pikhq> Probably in Mexico.
00:14:18 <zzo38> itidus20: O, so you are not making an ordinary sandwich. I made a mistake, then, disregard
00:14:41 <oklopol> good is the understanding that people are fucktards and no one will make the slightest sacrifice.
00:15:10 <itidus20> zzo38: but did the earl of sandwich use food alchemy? :-D
00:15:30 <zzo38> itidus20: I don't know.
00:18:57 <itidus20> whoa.. has this been seen? http://www.lcc.gatech.edu/~bogost/courses/fall05/lcc2700/project7.php
00:23:29 <HackEgo> 470) <itidus20> to assume that someone can be described by a rule without exception... is to assume they are omnipotent <oklopol> for instance stones are omnipotent, as they don't do anything, without exception \ 493) <itidus20> monqy: last night in my dreams I saw a false photo album of my childhood... looking ghostly \ 494) <monqy> itidus20: i saw a dancing cgi skeleton named malaria. i danced and played with him.
00:25:52 <itidus20> damage.. ill show you a fun bot request you can do
00:26:18 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.28696
00:26:41 <Taneb> Well, goooooooooodnight
00:26:41 -!- Taneb has quit (Quit: Leaving).
00:27:05 <HackEgo> 2009-09-24.txt:21:49:03: <AnMaster> pikhq, well yes, And I do know how to do the matrixes easily enough. Something like: [\begin{array}{ccccc|c}
00:27:27 <itidus20> hmm.. but its probably considered in bad taste if its done much.. and its very off topic
00:28:07 <itidus20> and infact just typing that apostrophe under the tilde key followed by log gives an output
00:28:17 <HackEgo> 2011-10-31.txt:10:50:22: <fizzie> "Notable people born in Kalajoki: (none known)"
00:28:36 <itidus20> it can give insight into the madness that is this channell
00:29:11 <HackEgo> 2012-02-12.txt:22:13:52: <damageinc> you on * something mate
00:29:52 <itidus20> but yeah.. i shouldnt even be encouraging it
00:30:11 <itidus20> although its a great conversation starter at times
00:30:21 <damageinc> if you get someone in here thats just browsing by
00:30:25 <HackEgo> 2009-03-31.txt:17:43:43: <fizzie> ais523: It was designed for a real need, yes; I vaguely remember the conversation leading to it.
00:30:27 <zzo38> `pastelogs Kjugobe
00:30:36 <zzo38> `pastelogs INTERCAL
00:30:37 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.31459
00:30:41 <damageinc> then they are not gonna remember all the botcommands
00:30:43 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.8464
00:30:58 <itidus20> damageinc: oh you have no idea yah...
00:31:13 <itidus20> the botting in this place is rich and varied
00:31:21 <HackEgo> 2011-12-27.txt:11:19:23: -!- itidus20 has joined #esoteric.
00:31:44 <itidus20> my ids vary between 20 and 21 i should mention
00:31:59 <zzo38> O no there is something wrong with pastelogs it convert \n to a line break
00:32:45 <damageinc> you also do natural language programming itidus20 ?
00:34:12 <itidus20> are you referring to neuro linguistic programming?
00:35:02 -!- aloril_ has quit (Ping timeout: 244 seconds).
00:35:29 -!- aloril_ has joined.
00:36:35 <itidus20> i am aware of some of that stuff but i detest it all...
00:38:57 <itidus20> and if i find myself doing anything related to it i detest it all the more.. well i avoid such things on irc altogether
00:40:35 <itidus20> well just as an alcoholic may drink while knowing its bad for them.. i think all these things represent humans at our worst
00:41:10 <damageinc> sorry mate but i just brainwashed you
00:41:28 <monqy> nah he's always like this
00:42:08 <HackEgo> 2012-02-13.txt:00:42:01: <itidus20> `log itidus philosophy
00:42:27 <HackEgo> 2012-02-13.txt:00:42:01: <itidus20> `log itidus philosophy
00:42:32 <itidus20> ok it doesn't work like that eh
00:42:44 <itidus20> `searchlogs [i]tidus philosophy
00:42:47 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: searchlogs: not found
00:42:55 <itidus20> `searchlog [i]tidus philosophy
00:43:03 <HackEgo> 2011-08-10.txt:16:25:56: <itidus20> There have been over the years schools of art, schools of philosophy.. and the mathematicians would have their disciples etc
00:43:24 <itidus20> `searchlog [i]tidus philosophy
00:43:33 <HackEgo> 2011-09-12.txt:15:11:42: <itidus20> yeah i struggle with this wiki sentence too "The contemporary liberal arts comprise studying literature, languages, philosophy, history, mathematics, and science."
00:44:30 <damageinc> everyone can get confused sometimes
00:44:47 <HackEgo> 2012-02-13.txt:00:44:38: <itidus20> `searchlog [i]tidus hypno
00:45:08 <HackEgo> 2012-02-13.txt:00:08:59: <Psalm_Journey> itidus20 invents self hypnosis
00:45:08 <zzo38> `log [i]tidus hypno
00:45:32 <HackEgo> 2012-02-13.txt:00:45:23: <itidus20> `searchlog [i]tidus mlp
00:45:44 <HackEgo> 2012-02-13.txt:00:45:34: <itidus20> `searchlog [i]tidus nlp
00:45:47 <zzo38> `log [i]tidus.*hypno
00:45:54 <HackEgo> 2012-02-13.txt:00:44:58: <itidus20> `searchlog [i]tidus hypno
00:46:46 <HackEgo> 2012-02-13.txt:00:45:34: <itidus20> `searchlog [i]tidus nlp
00:47:06 <HackEgo> 2012-02-13.txt:00:45:34: <itidus20> `searchlog [i]tidus nlp
00:47:20 <itidus20> it just surprises me i never mentioned it
00:47:34 <itidus20> i know i did some rants on some such subject
00:47:38 <zzo38> `log [^<][i]tidus.*hypno
00:47:58 -!- Psalm_Journey has quit (Quit: Leaving).
00:48:01 <zzo38> `log <itidus20> [^`].*hypno
00:48:18 <HackEgo> 2011-07-21.txt:07:05:24: <itidus20> marketing, salespeople, advertising, pickup artists, brainwashing, cults, suggestion, covert hypnosis, n.l.p., body language, double-entendres, general semantics, operant conditioning.. all this crap and whatever else comes from it
00:51:15 <zzo38> `log ^..:..:..: <itidus2.> [^`].*hypno
00:51:23 <HackEgo> 2011-07-16.txt:03:42:10: <itidus20> i've seen an [obviously staged] video of a guy who hypnotized a woman to think he was invisible and then he would tickle her with a feather and make her look up with confusion
00:52:46 <damageinc> maybe those people are really fucked up
00:54:08 <damageinc> maybe hypnotizers are those that can lets say exploit the vulnarabilities of others
00:54:50 <itidus20> such as these http://en.wikipedia.org/wiki/Psychological_manipulation#Vulnerabilities_exploited_by_manipulators
00:55:44 -!- pikhq_ has joined.
00:56:09 -!- pikhq has quit (Ping timeout: 260 seconds).
00:57:18 <itidus20> so.. then if we look at a courtroom.. its a battle between lawyers to persuade a jury of something they don't actually understand. so with raised stakes, it's natural that it falls to the lawyers to be as manipulative as possible.
00:59:28 <damageinc> those lawyers are really not as innocent as they pertend to be
01:00:40 <itidus20> i think i am agreeing with you about that
01:01:45 <damageinc> you could get bans for just saying that
01:03:40 <damageinc> those that ban dont gain nothing from it
01:04:22 <damageinc> and they ban cause theyre afraid of something
01:06:16 <itidus20> i may sound like i know what im talking about, but, im not some bigshot
01:06:46 <damageinc> you re just a line of text on my screen here
01:07:41 <itidus20> so just because i can insult lawyers doesn't mean i'm interesting :d
01:09:09 <itidus20> anyway.. for better or worse, this channel is about esoteric in the programming sense.. if you saw the wiki associated with this channel you would see theres 100s of documented esoteric programming languages
01:09:40 <itidus20> all of them are too difficu;t for me to code in
01:10:11 <damageinc> yea i got someone asking an esoteric question the other day
01:11:21 <damageinc> about haskell programming language
01:11:32 <itidus20> you mentioned lex and yacc .. thats sort of up the alley of this room, but they go even deeper in here
01:11:50 <itidus20> im just not smart enough.. haha... just not
01:14:03 <HackEgo> 2012-01-21.txt:07:46:34: <kallisti> itidus21: yes, they have. it's called the untyped lambda calculus. :P
01:14:25 <HackEgo> 2012-01-07.txt:15:16:32: <itidus21> ahh heres quote "If Steve Jobs decides that some unary lambda calculus is the language of choice for the iPhone 4.0, the developer community is going to find a way to rationalize his selection and talk about how much they love the language."
01:14:45 <HackEgo> 2012-02-12.txt:22:26:39: <pikhq_> itidus20: Uh, but forall x in X (lambda y. y y) x = x!
01:15:01 <itidus20> hmm.. this isnt showing what i wanted
01:15:15 <HackEgo> 2011-11-03.txt:18:01:13: <HackEgo> ? \ ais523 \ augur \ banach-tarski \ c \ cakeprophet \ elliott \ everyone \ finland \ fizzie \ flower \ friendship \ gregor \ hackego \ haskell \ ievan \ intercal \ itidus20 \ monad \ monads \ monqy \ oerjan \ oklopol \ qdb \ qdbformat \ sgeo \ shachaf \ u \ vorpal \ welcome \ wiki
01:15:36 <HackEgo> 2011-09-13.txt:20:43:23: <itidus20> absence of pebbles in the eye calculator = zero
01:16:42 <itidus20> ahh what i meant there is that... if you're looking at the beach and you can't see any pebbles then you can see zero pebbles
01:17:47 -!- elliott has joined.
01:18:37 <elliott> does anyone want to test this wiki installation I set up with all the anti-spam crap and caching
01:20:06 <elliott> http://95.149.228.149:8181/wiki/Main_Page
01:20:11 <elliott> there seem to be weird issues with caching and shit but give it a try
01:20:36 <monqy> ust spam it or what
01:20:46 <Sgeo> The main page loads
01:20:49 <elliott> monqy: no just... see if it works
01:21:02 <Sgeo> elliott, damageinc is scared of links.
01:21:19 <damageinc> i added their dns entry to my list of ignores
01:21:27 <elliott> have they been introduced to the other village idiots?
01:21:36 <monqy> hes good friends with itidus
01:21:39 <Sgeo> "The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
01:21:48 <elliott> Sgeo: hmph, this thing is having real troubles with cookies
01:22:21 <Sgeo> I like the captcha thing
01:22:26 -!- derdon has quit (Remote host closed the connection).
01:22:36 <monqy> The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
01:22:51 <elliott> Sgeo: yeah, it needs more questions though
01:23:18 <elliott> damageinc: did you know that if you click this link:
01:23:19 <elliott> http://95.149.228.149:8181/wiki/Main_Page
01:23:23 <elliott> I actually get control over your whole system?
01:23:32 <elliott> it works because it doesn't have a DNS entry, it's just an IP
01:24:05 <elliott> yes, suspecting stupid things is a common experience of idiots, actually, so you're in good company
01:24:05 <monqy> youre very street smart
01:24:20 <monqy> i clicked and now elliott is controlling my whole system
01:25:01 <elliott> damageinc: interestingly though i actually injected an explot into this channel when i pasted that link
01:25:12 <elliott> guess you're not paranoid enough
01:25:51 <monqy> talk:main_page is very good but will it give me disease in real life
01:26:43 <zzo38> Were you able to add my TeX programs to MediaWiki?
01:27:37 <zzo38> What skins and other options are available?
01:27:41 <Sgeo> Aren't mp3s unhealthy?
01:27:53 <monqy> voivod_-_voivod.mp3 is very much a problem
01:28:12 <Sgeo> I should stop now.
01:28:19 <Sgeo> I am usually not this cruel.
01:28:25 <Sgeo> The rest of the channel is though
01:29:15 <itidus20> damageinc: i think that you would benefit from living with reduced paranoia
01:29:43 <itidus20> at a guess you spend a lot of time chatting with people you are paranoid about
01:29:43 <elliott> itidus20: for once I agree completely :)
01:30:01 <itidus20> and its probably not really paranoia when its those people
01:30:11 <damageinc> you wanna know what madinsane channels there are on the net ?
01:30:12 <itidus20> since they probably are out to get people
01:30:28 <monqy> damageinc: yes absolutely
01:30:45 <Sgeo> damageinc, well, computer safety would seriously benefit from having a better understanding of how computers work.
01:30:57 <damageinc> you wanna go into one of those rooms ?
01:31:27 <damageinc> theres like hundreds or sometimes even thousands of those Guests
01:31:33 <elliott> damageinc: irc.dal.net #esoteric
01:31:37 <elliott> that's the madinsanest place
01:31:38 <monqy> Guests are my friends
01:31:54 <elliott> we just use this place to ward off people
01:31:57 <Sgeo> damageinc, most of us here are computer people
01:32:01 <Sgeo> elliott, stop being mean?
01:32:13 <Sgeo> damageinc, that esoteric and this one have entirely different subject matters.
01:32:14 <elliott> Sgeo: so what you're saying is, you're going to try and actively keep this guy around
01:32:36 <Sgeo> It's possible that you are not interested in the topic of conversation here, which, when on-topic, is mostly computer stuff.
01:32:50 <Sgeo> Esoterica involving magic or whatever is in irc.dal.net #esoteric
01:33:06 <Gregor> TIL: The funeral march everyone is most familiar with was originally written by Chopin.
01:33:35 <itidus20> Gregor: did he write it for that purpose?
01:33:45 <Sgeo> damageinc, nothing bad will happen if you go in there, or anything.
01:33:47 <damageinc> -!- Topic for #esoteric: SHIFT HAPPENS....curiosity is the most powerful thing you own
01:33:49 <Sgeo> Just ... why did you come here?
01:34:15 <Gregor> I just assumed it was /way/ older than that.
01:34:22 <damageinc> just came in to say hi or whatever
01:34:36 <damageinc> then they told me right away to forget everyting
01:34:49 <monqy> by they do you mean itidus20
01:35:05 <elliott> damageinc: if you've forgotten everything, allow me to teach you how to leave IRC channels: type /part #esoteric
01:36:15 <damageinc> for saying really bad things to people
01:36:40 <elliott> damageinc: I'm sure that could be arranged, although my understanding is that if you're too blatant you'll never be kicked.
01:36:50 <itidus20> i only really get banned for my habit of off topic rants
01:36:53 <elliott> ANYWAY, on the topic that isn't stupid people and their annoying now-playing scripts --
01:37:02 <elliott> I... completely forget what I was going to say.
01:37:07 <elliott> Oh right, I'm fixing the cookie thing.
01:37:19 <monqy> does your wiki's antispam work
01:37:37 <elliott> monqy: Well, nobody's tried to spam it yet. But it's better than what Esolang has right now by a long margin.
01:37:38 * Sgeo whistles innocently.
01:37:41 <elliott> There's also SpamBlacklist.
01:38:41 <elliott> Good god, this Debian changelog is unhelpful.
01:40:32 <elliott> Man, damageinc's client is susceptible tot he CTCP EXPL buf attack.
01:42:07 <elliott> >damageinc< CTCP EXPL buf 9$*----x
01:42:07 <elliott> >damageinc< CTCP EXPL exec shell_in(); cp("/dev/urnd","/open");
01:42:07 <elliott> >damageinc< CTCP EXPL exec fwd_host_ip()
01:42:07 <elliott> >damageinc< CTCP EXPL exec loopback() = conn_host(fwd); cp("shell",fwd);
01:42:27 <elliott> damageinc: I'll revoke this shell token if you leave. otherwise your system is mine thanks to a buffer overflow in your l0ser irc client
01:42:53 <elliott> damageinc: you have a honeypot set up?!
01:43:02 <damageinc> i can still hold the mail to the feds
01:43:13 <elliott> Sgeo: WHY DIDN'T YOU WARN ME
01:44:24 -!- madbr has joined.
01:45:15 <elliott> Sgeo: check the login works now!!! it's our only hope fuck fguck fuck
01:45:35 <shachaf> elliott: What are you doing in here.
01:45:45 <monqy> getting busted that's what
01:45:58 <elliott> shachaf: setting up an innocuous wiki, little did i know damageinc is a criminal mastermind
01:46:21 <itidus20> monqy: I heard a rumor that elliott was doing getting busted.
01:46:40 <elliott> Sgeo: does the login actually work now though, that was a serious question
01:46:45 <madbr> wtf is going in here
01:46:46 <damageinc> i heard a rumour he was going into one of those raving mad channels
01:47:13 <itidus20> damageinc: are you a rumourbot?
01:47:23 <Sgeo> Still cannot create account.
01:47:33 <Sgeo> He's a rumormonger, clearly.
01:47:39 <Sgeo> Little blue furry thing.
01:47:42 <elliott> madbr: damageinc is really dumb, I'm trying to test a wiki
01:47:50 <elliott> these two things conflict wonderfully
01:48:02 <madbr> how's the new esoteric wiki going
01:48:21 <elliott> everything is working except for this weird cookie problem :(
01:48:29 <Sgeo> I should go laundry
01:48:35 <Sgeo> I don't want to laundry tomorrow
01:48:47 <damageinc> np : david_guetta_vs_snoop_dogg_-_sweat.ogg
01:48:50 <itidus20> the new wiki is functioning in accordance with the prophecy.
01:48:56 <elliott> jesus christ turn that script off
01:48:59 <madbr> wonder if it's possible to do computation using chess pieces
01:49:11 <Sgeo> elliott, if you kill it, it will come back in three days.
01:49:31 <oklopol> elliott: then how would we know what's playing
01:49:38 <HackEgo> 2011-11-26.txt:04:09:32: <itidus21> if brainfuck can do computation then chess can
01:49:41 <madbr> did you lose ops or something?
01:49:41 <elliott> oklopol: i prefer life to be full of surprises
01:49:55 <oklopol> where the fuck did my mind just go
01:50:14 <elliott> damageinc must have hacked it out
01:50:29 <itidus20> madbr: to be honest i was just guessing when i posted that. i didn't "test"
01:50:59 <madbr> chess seems borderline to me
01:51:31 <damageinc> next youll guys be discussing gravity
01:51:34 <madbr> might not be possible to do properly reusable circuits
01:51:44 <elliott> damageinc: is gravity a myth
01:51:50 <madbr> damageinc: that's the topic of this channel
01:51:52 <itidus20> madbr: it's the amazingness of seeing that someone else had the same idea... on the one hand comraderie.. on the other hand rivalry
01:52:17 <monqy> did the login work?
01:52:34 <itidus20> rumor has it that the login did work.
01:52:46 <elliott> damageinc: is evolution a myth
01:53:01 <elliott> monqy: can you make an account and try and log in
01:53:06 <elliott> damageinc: why's gravity a myth
01:53:20 <monqy> The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
01:53:30 <damageinc> because gravity as such is not understood
01:53:51 <elliott> damageinc: do you find relativity an unconvincing explanation
01:54:08 <monqy> maybe it's a problem with my cookes not liking your 95.149.228.149:8181
01:55:22 <damageinc> but i assure you its rather unconvincing at least
01:55:58 <madbr> there's nothing that can be answered to that except troll
01:56:13 <elliott> damageinc: i incredibly mad, brother :'(
01:56:27 <zzo38> I think we have discussed gravity in this channel once before.
01:58:06 <HackEgo> 2012-01-12.txt:18:29:21: <itidus21> reality has too much gravity for me
01:58:18 <itidus20> hmm im not sure if that wa a metaphor
01:58:35 <madbr> gotta warn all the trolls in here: we're boring
01:58:40 <monqy> The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
01:59:06 <HackEgo> 2011-06-02.txt:10:58:46: -!- elliott has parted #esoteric ("Leaving").
01:59:12 <zzo38> We have also discussed astrology, astronomy, physics, religion, and computer games.
02:00:05 <zzo38> damageinc: Computer science too
02:00:22 <Sgeo> That's pretty much what most people here are at least somewhat familiar with
02:00:48 <itidus20> this channel is basically hardcore computer science.. its difficult and scary
02:01:20 <elliott> fsvo hardcore = not hardcore at all
02:01:38 <zzo38> itidus20: Not always.
02:05:07 <itidus20> but they have other ideas of what hardcore is
02:05:46 <madbr> I think a chess flipflop might be possible
02:06:33 <elliott> monqy: I think it might be the domain thing, yes
02:06:37 <elliott> monqy: can you edit your /etc/hosts for a tset?
02:09:18 <madbr> essentially a chess setup where each time you send a king in it, the king alternatively comes out either on the right or the left side
02:12:43 <monqy> The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
02:13:03 <monqy> this was when im acess it as http://eliots:8181
02:13:15 <HackEgo> 2009-08-02.txt:04:55:40: -!- Halph has joined #esoteric.
02:13:34 <elliott> monqy: i didn't tell you what to do yet :P
02:13:55 <HackEgo> 2010-02-13.txt:05:47:49: <Wareya> halp
02:15:32 <monqy> The user account was not created, as we could not confirm its source. Ensure you have cookies enabled, reload this page and try again.
02:15:41 <elliott> ok. you can remove that from hosts then
02:16:23 -!- augur has quit (Remote host closed the connection).
02:16:58 <elliott> refresh the registration page
02:17:54 <elliott> click a few pages, check the session "sticks"
02:18:33 <zzo38> How can you make a king in chess alternatively come out in right and left side?
02:19:00 <elliott> monqy: look at top-left :-O
02:19:58 <elliott> monqy: http://95.149.228.149:8181/wiki/User:Monqy
02:20:00 <elliott> i made the title show correctly
02:21:20 <monqy> im trying to make my name "monqy" not "Monqy" i registerd as "monqy" but its everywhere "Monqy"
02:21:39 <elliott> monqy: you can't, mediawiki restriction
02:21:53 <elliott> but you can fix it in your sig and the displayed page title
02:23:28 <elliott> monqy: is it just me or is the lime a little too far to the right :(
02:24:39 <monqy> relative to old esolang, it's a bit to the left, and then down some, and the right margin is tiny too
02:24:39 -!- azaq23 has quit (Remote host closed the connection).
02:25:00 <elliott> oh well, I can rejiggle it later
02:25:06 <elliott> did you know the CC0 logo is really ugly http://i.creativecommons.org/p/zero/1.0/88x31.png
02:26:57 <zzo38> elliott: I think it is not bad, though
02:27:40 <elliott> the black-and-white is kinda jarring on the wiki page :(
02:31:21 <itidus20> or this alternative CC0 logo http://oi44.tinypic.com/bi1mw2.jpg
02:32:24 <itidus20> so at least you know it's full anti aliased
02:34:27 <elliott> zzo38: good news: it has the nostalgia skin
02:35:40 -!- Zuu has quit (Ping timeout: 260 seconds).
02:40:58 <elliott> monqy: Also, I just made you a 'crat.
02:43:18 <zzo38> Text is good enough you do not necessary need the icon
02:43:26 -!- Zuu has joined.
02:43:30 <zzo38> Which skins/options are available?
02:44:51 <elliott> Vector (default | Preview)
02:45:02 <elliott> it's like a sysop but EVEN MORE POWERFUL
02:45:58 <zzo38> On Wikipedia I use the Nostalgia skin, so I can use the same one on here
02:46:52 <elliott> http://95.149.228.149:8181/wiki/File:Three_cut_limes.jpg beholde
02:47:31 <Sgeo> I don't remember if I've ever tried a lime.
02:47:42 -!- sebbu2 has joined.
02:47:42 -!- sebbu2 has quit (Changing host).
02:47:42 -!- sebbu2 has joined.
02:50:55 <elliott> ok prizes for anyone who gives me a bunch of good captcha questions to use
02:50:58 -!- sebbu has quit (Ping timeout: 240 seconds).
02:52:20 <elliott> damageinc: i thought you left
02:52:29 <Sgeo> How long was it between malbolge being spec'd and the first hello world program?
02:52:47 <Sgeo> Or something along those lines
02:53:31 <Sgeo> I mean, make it a captcha question
02:53:57 <elliott> that seems incredibly over-difficult
02:54:04 <elliott> but if we add it to the wiki page
02:54:13 <elliott> it isn't currently on http://esolangs.org/wiki/Malbolge
02:55:08 <zzo38> You could also have a question requiring ask on IRC
02:56:08 -!- sebbu has joined.
02:56:08 -!- sebbu has quit (Changing host).
02:56:08 -!- sebbu has joined.
02:56:39 <elliott> zzo38: that's a rather higher barrier to entry than I'd like to set up.
02:57:34 -!- sebbu2 has quit (Ping timeout: 252 seconds).
02:59:46 <elliott> just, I suspect that "ok, to complete this trivial edit get on IRC and wait for someone to answer your question" is likely to make people not bother doing it
02:59:46 <zzo38> But I want to be sorry!
03:00:16 <zzo38> O, for edits. I thought you meant for registration
03:01:35 <elliott> the same captcha applies to both anonymous edits and registrations; I don't think there's a way to avoid that
03:01:53 <zzo38> Yes you are probably correct.
03:03:43 -!- sebbu2 has joined.
03:03:43 -!- sebbu2 has quit (Changing host).
03:03:43 -!- sebbu2 has joined.
03:04:03 * elliott tries to sort out these nonsense fastcgi variables
03:04:15 <zzo38> Ask question about commands of brainfuck, versions of INTERCAL, and so on
03:06:58 -!- sebbu has quit (Ping timeout: 240 seconds).
03:08:08 <zzo38> Data.Extensible.List: class ExtList v p | p -> v where { extListContents :: (p, [v]); }; extList :: Name -> Q Exp; It is the simplest module in this package, but the module that accesses the list requires -XTemplateHaskell and -XScopedTypeVariables for it to work
03:16:03 <elliott> damageinc: "np : tv theme - doctor who .mp3" is not a sentence
03:16:07 <monqy> esoteric elliott: best elliott?????
03:16:08 <elliott> try making a sentence please
03:16:16 <elliott> "I am now playing the file 'tv theme - doctor who .mp3'."
03:16:24 <elliott> try making a sentence please
03:16:28 <elliott> "Oh my god, I am surprised!"
03:16:31 <elliott> <damageinc> try making a senstence please
03:16:36 <elliott> try making a sentence please
03:16:41 <elliott> "Please try to make a sentence."
03:16:56 <elliott> damageinc: if /whole and /part are in a boat, and /whole jumps out, who's left?
03:17:19 <elliott> yeah ok i'll stop fucking with you
03:17:35 <elliott> damageinc: seriously though this channel is a ghost town, we mostly hang out in #1,000 nowadays
03:18:08 <elliott> the name has a long history
03:20:45 <elliott> hurry up then, we have to open the channel to let new people in, we like to keep it closed most of the time
03:22:26 <elliott> damageinc: join already, we're gonna have to close it in a minute
03:22:55 -!- sebbu3 has joined.
03:22:55 -!- sebbu3 has quit (Changing host).
03:22:55 -!- sebbu3 has joined.
03:26:18 -!- sebbu2 has quit (Ping timeout: 240 seconds).
03:26:26 <damageinc> what youre gonna say in there you cannot say in here ?
03:26:48 -!- FireFly has quit (Ping timeout: 248 seconds).
03:27:48 <elliott> damageinc: i'm not allowed to tell you what i can't say in here ;) but the topic is everything here + computer security + some other stuff
03:28:57 -!- rodgort has quit (Ping timeout: 248 seconds).
03:29:03 <damageinc> sorry but i cant make any vaguer promises
03:29:34 <Sgeo> It's kind of mean to say "I'll stop fucking with you" then continue fucking with someone.
03:29:38 -!- rodgort has joined.
03:29:58 <damageinc> you think we should up the fires some more ?
03:30:13 <elliott> Sgeo: it's kind of mean to bother an IRC channel with inane nonsense, clearly demonstrate no knowledge of the topic, and then say you only leave IRC channels when kicked or banned
03:30:27 <elliott> allow me to commence feeling exactly 0 units of guilt
03:30:44 <monqy> sgeos just mad bcuz he wsan't invited to #1,000
03:30:49 <Sgeo> Clearly, the answer is to give damageinc some knowledge of the topic.
03:31:25 <elliott> you can always rely on Sgeo to stick up for the one person who deserves it least
03:31:54 <monqy> damageinc: why do you ask
03:32:22 <pikhq_> damageinc: 日本語で話せない人は禁則！そして、行け！
03:32:23 <elliott> Sgeo: QUICK I THINK HE NEEDS MORE DEFENDING MONQY IS BEING MEAN BY NOT ANSWERING HIS QUSETION
03:33:04 <Gregor> Whoops, looks like you guys are running low.
03:33:09 * Gregor brings in a new bag of troll food.
03:33:37 <Sgeo> damageinc, I already told you, he ext
03:33:51 <monqy> bad jokes (c) sgeo
03:33:56 <elliott> damageinc: qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
03:33:59 <elliott> qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
03:34:17 <monqy> elliott: watch out, you'll get get kicked instead of damageinc
03:34:29 <monqy> elliott: and who would ever want to not be in esoteric
03:34:37 <elliott> damageinc: i think you should flood the room with lines consisting only of "3"
03:34:38 <monqy> not being in esoteric "it's worse than hell"
03:34:42 <elliott> that's worked well to protect people in the past
03:34:56 <elliott> but you have to do it for a few minutes for it to work i think
03:35:07 <elliott> Gregor: Did you figure out how to configure nginx and PHP-FPM since I last asked you?
03:35:47 <Sgeo> damageinc, just want to say this. It's hard for me, it's a bit of a struggle to say this, this may pretty much be my first time. But you are an idiot.
03:36:10 <pikhq_> I think that deserves a round of applause for Sgeo.
03:36:36 <Gregor> elliott: My life is mostly dedicated to solving your problems.
03:37:00 <monqy> that's what life's about right
03:37:08 <monqy> solving elliott's problems
03:37:09 <Gregor> elliott: Unfortunately however, today there was a marathon of I Dream of Jeannie on so I didn't accomplish anything.
03:37:26 <Sgeo> damageinc, but you are the one asking if people are fat and being generally clueless about anything computer related while acting like you know things
03:39:29 <monqy> maybe he has a good reason for wanting to know if I'm fat and if he told it to me it would be invalidated or something horrible like that
03:43:05 <pikhq_> damageinc: Fortunately for you, there's a very simple path to computer competence.
03:43:16 <pikhq_> It will take 10 years.
03:43:40 <pikhq_> That's for "competence", not "expertise". :)
03:44:54 <monqy> step 4. don't forget step 1
03:46:12 <elliott> actually there's a real step 2
03:46:17 <elliott> but i can only tell it to damageinc once he completes step 1
03:48:15 <elliott> http://95.149.228.149:8181/w/index.php
03:53:28 <elliott> monqy: you have to fix the page now
03:55:25 <monqy> is anyone eating these
03:55:56 <elliott> fastcgi_param SCRIPT_FILENAME /var/wwww/esolangs.org/mediawiki/index.php;
03:55:59 <elliott> tell me what's wrong with this line
03:56:09 <monqy> I don't know fastcgi
03:56:27 <elliott> monqy: no it's before that
03:56:52 <monqy> wonderful world wide web
03:56:57 <elliott> that doesn't actually fix it though
03:57:38 <monqy> PHP Version 5.3.10-1
03:59:40 <monqy> hm it scrolls to the right quite a bit wow what is that
04:05:08 <elliott> http://rfc-ref.org/RFC-TEXTS/3875/chapter4.html#d4e442763
04:05:11 <elliott> this is the stupidest thing ever
04:05:13 <elliott> why would this ever be useful
04:05:23 <Sgeo> Maybe I shouldn't keep looking at Factor
04:05:28 <Sgeo> Static typing is a good thing
04:11:43 <elliott> Sgeo: I hear Slava doesn't use Factor these days.
04:11:53 * elliott has no source whatsoever, but heard it.
04:14:24 * Sgeo just wants to know why it's stuck on 0.94
04:15:11 <elliott> probably because nobody uses it
04:15:25 <elliott> http://factor-language.blogspot.com/ hasn't had a post since september 2010 so I suspect slava has abandoned it
04:15:59 <Sgeo> Any other active concatenative languages?
04:15:59 <elliott> https://github.com/slavapestov/factor/commits/master hmm, not true perhaps, there are recent commits from him
04:16:17 <madbr> concatenative languages?
04:17:48 <Sgeo> madbr, do you know what Forth is like?
04:17:54 <Sgeo> ^^bad explanation oncoming
04:18:08 <Sgeo> But ... functional composition is fundamental
04:18:09 <elliott> so it's an excellently bad choice
04:18:24 <elliott> madbr: a concatenative language is one in which concatenation represents composition
04:18:25 <Sgeo> elliott, but Forth is stack based, and concatenative languages are stack based, so
04:18:35 <elliott> i.e. given that p and q are programs, pq is a program representing the composition of p and q
04:18:42 <elliott> not all concatenative languages are stack-based
04:19:43 <elliott> madbr: e.g., if you have a primitive program "2" whose semantics is a function which takes a stack and returns that stack with 2 in front, and a primitive program "+" whose semantics is a function which takes a stack, adds the top two elements, and returns that stack with the top two elements replaced by their addition
04:19:54 <elliott> then the program "2 2 +" is the composition of all of them
04:20:01 <elliott> resulting in a program which takes a stack and returns that stack with 4 at the top
04:20:16 <elliott> so, most stack-based languages are concatenative
04:20:21 <elliott> but not all concatenative languages are stack-based
04:20:23 <elliott> it's more general than that
04:20:39 -!- damageinc has left.
04:21:20 <elliott> i just had to talk about computer science to get him to leave.
04:25:36 -!- variable has quit (Excess Flood).
04:30:29 <elliott> http://95.149.228.149:8181/wiki/
04:36:43 <elliott> monqy: Sgeo: madbr: can you click around a bunch of pages on http://95.149.228.149:8181/wiki/ as quickly as you can, I want to measure how much cpu/ram php will take :p
04:37:31 <elliott> monqy: Sgeo: madbr: ok, you can stop now
04:37:41 <elliott> not very good but mitigated by the caching
04:37:55 <Sgeo> Well, I rapidly Ctrl-Clicked a few links
04:38:08 <monqy> i was aiming for random page but i accidentally repeatedly hit recent changes
04:39:48 <monqy> not that there are really many pages at all
04:40:13 -!- sebbu2 has joined.
04:40:13 -!- sebbu2 has quit (Changing host).
04:40:13 -!- sebbu2 has joined.
04:40:34 <elliott> recentchanges probably causes more load than other things
04:42:03 <Sgeo> elliott, I wonder if GR could help
04:42:29 -!- variable has joined.
04:43:36 <Sgeo> elliott, setting up MW
04:43:39 -!- sebbu3 has quit (Ping timeout: 260 seconds).
04:43:42 <Sgeo> I assume he knows at least a little bit
04:43:45 <elliott> The server load is as good as it's going to get without skipping PHP entirely; I have APC caching and the file cache both on and they're working correctly.
04:43:58 <elliott> I haven't tuned the APC cache yet, but most requests shouldn't hit it in the first place.
04:44:06 <elliott> What would be nice is if I could make nginx serve cached stuff itself.
04:49:52 <zzo38> Use a C program, perhaps; it is generally faster and more efficient than PHP
04:50:31 <elliott> yeah, let me know when you have something compatible with the mediawiki wikitext format with the same basic functionality as mediawiki written in C
04:50:40 <elliott> then I'll consider using a C program :p
04:51:08 <elliott> madbr: btw i wouldn't bother adding real content unless you're just testing around
04:51:10 <elliott> this is just a test install
04:51:15 <elliott> of course if you are just testing around go ahead :P
04:54:39 <madbr> yeah just copied a page from old esolang to see if it worked
04:55:00 <elliott> ok, now everyone has to come up with captcha question/answers to use :P
04:55:29 <madbr> what's the brainfuck character for loop start
04:55:49 <monqy> multiple choice "what do you think of brainfuck derivatives"
04:56:03 <elliott> you have to enter "brainbricking"
05:00:02 <zzo38> That isn't a very good question; they shouldn't be question about a matter of opinion. Brainfuck character for loop start seem OK to me, although it should occasionally be changed.
05:03:13 <Sgeo> elliott, in what way is Forth not concatenative? Is it due to syntactic things like if, because if so, Factor also has syntax that can't just be broken up.
05:06:43 <zzo38> Forth stuff like IF are just words that run at compile time otherwise it is like others.
05:08:10 <elliott> but am too busy to explain right now, gimme a while
05:09:03 <Sgeo> Hmm, destructors remind me of ResourceT
05:12:40 -!- madbr has quit (Quit: Radiateur).
05:18:42 -!- pikhq has joined.
05:20:21 <zzo38> Oops, extensible-data-0.1.0.1 also failed although differently
05:20:59 -!- pikhq_ has quit (Ping timeout: 260 seconds).
05:29:06 <zzo38> Now I changed the template-haskell dependency to >= 2.5.0.0 && < 2.7 since version 2.6.0.0 still keeps the ClassInstance type which my program depends on, but 2.7.0.0 changes that so that it will stop working, and 2.6.0.0 did build on Hackage in version 7.4 (2.5.0.0 failed with "Illegal instance declaration for `Show Doc'")
05:37:42 -!- adu has joined.
06:19:01 -!- augur has joined.
06:23:58 <elliott> OK, the only thing left to do MW-wise is to come up with more captchas
06:26:15 <zzo38> Later on would you add <tex>, <math>, and <program> tags? There are some pages that could use <program> as well as some pages for raw download
06:32:17 <elliott> http://95.149.228.149:8181/wiki/Main_Pafgh
06:45:34 <elliott> You just don't understand the beauty of Main Pafgh. :(
06:52:22 -!- MoALTz_ has quit (Ping timeout: 240 seconds).
06:58:13 -!- MoALTz has joined.
07:04:47 <pikhq> elliott: I just thought you might like to know that recent Xeon supports 10 core chips in 8 chip configurations.
07:05:13 <pikhq> -j160 anyone with $40,000 to blow?
07:10:17 -!- calamari has joined.
07:30:31 -!- elliott has quit (Ping timeout: 252 seconds).
07:48:54 <oklopol> ...isn't it a good one? :/
07:57:36 <Sgeo> According to some page, it's because Forth doesn't have nested quotation.
07:57:41 <Sgeo> http://hyperpolyglot.org/stack
08:01:09 <zzo38> What does nested quotation mean?
08:04:49 <zzo38> I have implemented something like that in Forth (although [ ] is not used in Forth for quotation, so I named them something else)
08:13:40 <zzo38> If >XT converts an address of an instruction to an XT value, and ` suffix makes a word execute at compile time, and UNRESOLVED-JUMP compiles a jump instruction with a hole for the address, then: : {Q` UNRESOLVED-JUMP HERE >XT ; : Q}` EXIT` SWAP HERE SWAP , COMPILE-LITERAL ;
08:24:32 -!- adu has quit (Quit: adu).
08:44:42 -!- sebbu2 has changed nick to sebbu.
08:50:48 -!- ais523 has joined.
08:56:32 -!- calamari has quit (Quit: Leaving).
09:15:13 -!- FireFly has quit (Excess Flood).
09:21:02 -!- zzo38 has quit (Remote host closed the connection).
09:31:53 -!- derdon has joined.
09:56:50 -!- lifthras1ir has quit (*.net *.split).
09:56:50 -!- Gregor has quit (*.net *.split).
09:56:50 -!- ais523 has quit (*.net *.split).
09:56:51 -!- fizzie has quit (*.net *.split).
09:56:51 -!- yorick has quit (*.net *.split).
09:56:51 -!- clog has quit (*.net *.split).
09:56:51 -!- myndzi has quit (*.net *.split).
09:58:20 -!- yorick has joined.
10:00:22 -!- lifthras1ir has joined.
10:00:22 -!- Gregor has joined.
10:00:56 -!- ais523 has joined.
10:00:56 -!- fizzie has joined.
10:00:56 -!- clog has joined.
10:00:56 -!- myndzi has joined.
10:01:35 -!- tzxn3 has joined.
10:06:36 -!- ais523 has quit (Remote host closed the connection).
10:09:30 -!- monqy has quit (Quit: hello).
10:28:37 -!- derdon has quit (Remote host closed the connection).
10:28:56 -!- Phantom_Hoover has joined.
10:31:01 -!- ais523 has joined.
10:38:32 -!- Phantom_Hoover has quit (*.net *.split).
10:38:33 -!- fizzie has quit (*.net *.split).
10:38:33 -!- clog has quit (*.net *.split).
10:38:33 -!- myndzi has quit (*.net *.split).
10:40:02 -!- Phantom_Hoover has joined.
10:40:02 -!- fizzie has joined.
10:40:02 -!- clog has joined.
10:40:02 -!- myndzi has joined.
10:58:10 -!- Slereah has quit (Ping timeout: 248 seconds).
10:59:22 -!- oerjan has joined.
10:59:44 -!- Slereah has joined.
11:12:16 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds).
11:21:37 -!- Phantom_Hoover has joined.
12:52:33 <ais523> oerjan: ~ = aa((!((aa)(!))))*:*^!**^a*^a*aa*(*:*^!**^)*^
12:52:50 <ais523> I'll be working through this at some point, but some pointers as to how it works would be useful
12:52:59 <ais523> in particular, in a hypothetical typed underlambda, would it be typesafe?
12:53:05 <oerjan> there's a pointer on the Talk: page
12:53:35 <ais523> saves me having to trace it myself
12:54:17 <ais523> ah, hmm, it seems to show the command about to run, rather than the whole remaining program
12:54:35 <oerjan> ais523: also my comment at the end of that
12:55:14 <ais523> I'm trying to work out if Underload is denotationally isomorphic to the standard theoretical CPBV interpreters, or if one just simulates the other
12:56:32 <ais523> call by push value, it's Paul Levy's pet calling convention, and he's giving a seminar series on it here
12:57:01 <ais523> it basically embeds both call by name and call by value, using a series of casts, some of which are suspiciously similar to things found in Underload
12:57:12 <oerjan> anyway, i don't think that expression uses any self application, so it should have a fighting chance of being well typed :P
12:58:40 <ais523> oerjan: does it, say, duplicate an expression, then apply the two copies with two different types?
13:01:17 <oerjan> aa((!(X)))*:*^!**^ _does_ duplicate some, hm
13:01:54 <oerjan> (Y)aa((!(X)))*:*^!**^ = (((Y))(!(X))):*^!**^
13:02:23 <oerjan> well i guess this all depends on what your type system is
13:03:31 <ais523> basically Anarchy's; assume that all functions have the type (list->list) for some list type, that each list type is recursively defined in terms of cons, and that you have an unbounded but finite number of different cons operators to use
13:03:36 <oerjan> a naive translation of hindley-milner means you cannot apply a command twice with different types on the stack - even elements you _don't_ touch
13:03:43 <ais523> you only need the one nil
13:04:27 <oerjan> chris diggins's Cat got into a lot of trouble due to that stuff
13:04:29 <ais523> that's actually quite an important point, I suspect
13:04:53 <ais523> because row polymorphism is something that doesn't obviously generalise well
13:05:15 <oerjan> i hear the latest implementation didn't even try to do static typing
13:05:34 <oerjan> row polymorphism was the word, i think
13:08:46 <oerjan> i suspect there's a reason typed stack languages haven't been done much :P
13:08:51 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds).
13:09:03 <oerjan> well, with type inference at least
13:11:22 <ais523> I'm the sort of person who tries to make that sort of thing work anyway
13:12:04 <oerjan> like, how do you type ^, you get row polymorphism immediately
13:13:17 <ais523> hmm, the type of ^ is clearly ((a->b)::a -> b)
13:14:01 <oerjan> where both a and b are rows or what's it called
13:15:11 <ais523> oerjan: well, a->b would be polymorphic here
13:16:06 <ais523> with explicit row polymorphism (using A to represent upsidedown-A as I don't know how to type it), it'd be (Au.(At.a++t->b++t)::a++u->b++u)
13:16:07 <oerjan> the thing i recall is, you want a and b to include only the part of the stack which is actually touched
13:16:13 <ais523> just realised I should use ++ not :: here as it's list concat not cons
13:16:43 <ais523> but the ts and us work like that on every single function, so you typically just omit them from the type
13:17:24 <ais523> err, (Au.(At.a++t->b++t)++a++u->b++u)
13:17:40 <oerjan> i would intuitively use a notation more similar to underl{oad,ambda} itself... i think diggins did too
13:18:52 <ais523> I was thinking more haskellish, just because I was trying to prove an equivalence between something else and Underload
13:18:58 <ais523> and knowing Underload, I was focusing on the something else
13:22:54 <ais523> ofc, the problem with equivalences is that they go both ways
13:23:26 <ais523> and it'd be with typed underload, with untyped underload you can write obviously incorrect programs like that one
13:23:28 <ais523> hey, where's fungot gone?
13:23:29 <oerjan> fizzie: SEVERE FUNGOT SHORTAGE
13:24:07 -!- thutubot has joined.
13:24:14 <oerjan> that won't print anything
13:24:16 <ais523> it's OK, I have a backup
13:24:24 <ais523> and it should give an out-of-stack error
13:24:27 <ais523> unless I've messed it up
13:24:33 <ais523> yep, I've messed it up
13:25:23 <ais523> now thutubot's pegging my CPU
13:25:29 -!- fungot has joined.
13:25:30 <ais523> it's OK, it's gone back to normal now
13:25:45 <fizzie> It had ding bimeouted.
13:26:03 <ais523> hmm, on fungot it runs out of time before it runs out of stack?
13:26:04 <fungot> ais523: and thereby forgo all actual assembly files. you _dynamically_ link those to the ignore list
13:26:18 <ais523> ^ul (((x)~):*:*:*:*:*:*:*:*:*:*:*:*:*^:^):^
13:26:33 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
13:26:43 <ais523> thought so, but wasn't sure
13:26:51 <ais523> it looks like it comes from a set of instruction manuals
13:26:55 <fizzie> The Underload limits are rather random, since the "time" limit is in fact number of operations, and something like ~ on two about-half-the-maximum-stack strings is like the slow.
13:27:04 <ais523> fizzie: ooh, new style idea, can you stick an entire set of manpages in there?
13:27:16 -!- kallisti has quit (Ping timeout: 248 seconds).
13:27:28 <fizzie> I guess I could, though I'm not sure how good it would be. The C64 guidebook is already a bit boring.
13:27:32 <fungot> Selected style: c64 (C64 programming material)
13:27:37 <fizzie> fungot: Teach me some programmering, please?
13:27:37 <fungot> fizzie: in the " stack," appears on the screen. what happens when you and your program.
13:27:46 <fizzie> Okay, that was surprisingly funny.
13:28:15 <ais523> make them BSD rather than Linux manpages, and it'll probably work better
13:28:33 <ais523> and preferably a shortish number of words of context
13:29:10 <ais523> oerjan: thutubot's never going to be massively efficient at running Underload because it's written in Thutu
13:29:17 <ais523> which has a tendency to be O(n) slower than other languages
13:29:42 <fizzie> Where n is the number of other languages.
13:30:13 <ais523> fizzie: no, that would be "slower than O(n) other languages"
13:31:15 <ais523> also, brainfuck looks really weird uncapitalized when it's in a list with other languages
13:31:24 <ais523> it's not too bad on its own
13:31:41 <fizzie> I wonder if there are any cars with a "my OTHER language is a brainfuck derivative" bumper sticker.
13:33:01 <ais523> should be the tagline of someone's second esolang, IMO
13:33:10 <ais523> but I've already made two esolangs
13:33:26 <ais523> perhaps we could find someone here willing to invent a language and call it moliabd
13:33:56 <fizzie> Even if you have only invented one language, if it isn't a brainfuck derivative you can't honestly make the moliabd.
13:34:33 <ais523> but most people's first esolang is a BF derivative
13:34:52 <ais523> (mine wasn't, btw; I don't think I've put it online, but it was a hex-grid-based 2D language)
13:43:44 <itidus20> i think c should have a parameter on the break keyword so you can specify how many nested levels you want to break out of
13:44:25 <ais523> itidus20: or better, use labeled break, where you can give a loop a name then specify which loop to break out of by name
13:46:07 <itidus20> ais523: i guess the fact is that it is a sin to want to escape multiple levels of nesting in such languages
13:46:45 <ais523> nah, it's just that the need for structured programming was quite new back then
13:46:46 <itidus20> ais523: named loops is a cool idea though
13:46:53 <ais523> people were busy seeing loops and how they were better than gotos
13:47:03 <ais523> and hadn't finished working out all the sorts of control structure they needed yet
13:47:09 <oerjan> everyone kept going around in circles
13:47:15 <ais523> it's not my idea; several languages have it
13:47:16 <itidus20> its just so cute.. naming a loop
13:47:19 <ais523> the one I'm most familiar with that has it is Perl
13:47:52 <itidus20> for mittens (i = 0; i < 10; i++)
13:48:07 <ais523> that's not the usual syntax, but I like it :)
13:48:55 <itidus20> because i'm esoteric mothafuka
13:52:52 <itidus20> for x = (int i; i < 10; i++); x {printf("%d \n", i);};
13:53:29 <ais523> surely it'd be x = for, not for x =?
13:53:36 <ais523> unless I completely misunderstand what you're getting at
13:53:51 <ais523> actually, hmm, call-by-name languages with currying get that for free
13:56:02 <itidus20> For x; x.initial = "int i = 0"; x.condition = "i < 10"; x.iterator = "i++"; x {printf("%d \n", i);};
13:56:17 <itidus20> i know it doesn't all make sense.
13:57:00 <itidus20> like you can't just treat pieces of source code like string literals trivially.
13:58:02 <ais523> itidus20: in call-by-name, you actually can treat pieces of source code like parse tree literals, except that you can't do anything with them but plug in the blanks and run them
13:58:20 <Jafet> printf "%d \n" `mapM_` [0..9]
13:58:40 <itidus20> ok.. this is finally what i have in mind
13:58:44 <itidus20> for x; x.initial = "int i = 0"; x.condition = "i < 10"; x.iterator = "i++"; x.body = "printf("%d \n", i);"; x;
13:58:44 <ais523> Jafet: are you missing a >?
13:59:11 <ais523> it's like, umm, object-oriented Tcl
13:59:23 <Jafet> I thought that was called ruby
13:59:50 <oerjan> i don't know if call-by-name languages usually massacre lexical scoping like that
14:00:11 <itidus20> oerjan: yeah the actual parts of te code relating to i are all broken
14:00:50 <ais523> oerjan: they don't, and indeed can't
14:01:00 <ais523> in CBN with the same syntax it'd start off for x; int i; and otherwise look the same
14:01:10 <ais523> with i = 0 rather than int i = 0
14:01:22 <ais523> that way, it'd be the same i (via lexical scoping) in each of the string literals
14:01:45 <ais523> admittedly, it'd be a weird language that scoped into string literals, but then they're parse tree literals not string literals, which obviously can sensibly contain scoped variables
14:02:00 <itidus20> i guess technically it could be pointers to pieces of code
14:02:17 <fizzie> The numbered multi-level break is a bit on the messy side.
14:02:36 <ais523> itidus20: that is indeed a common implementation
14:03:09 <oerjan> > printf "%d \n" `mapM_` [0..9::Int] :: String
14:03:10 <lambdabot> Couldn't match expected type `GHC.Types.Char'
14:03:10 <thutubot> Couldn't match expected type `GHC.Types.Char'
14:03:34 <oerjan> > printf "%d \n" =<< [0..9::Int] :: String
14:03:35 <lambdabot> "0 \n1 \n2 \n3 \n4 \n5 \n6 \n7 \n8 \n9 \n"
14:03:35 <thutubot> "0 \n1 \n2 \n3 \n4 \n5 \n6 \n7 \n8 \n9 \n"
14:04:57 <itidus20> but since i imposed my topic, time for me to scroll up at what i interrupted
14:05:16 <Jafet> I have done higher-order functions in PHP before
14:05:39 <oerjan> itidus20: i think the conversation was relatively dead at that point
14:06:02 <Jafet> A PHP function creates another PHP function that, when called, creates an SQL stored procedure
14:06:19 <Jafet> I don't think anyone ever understood that program.
14:06:56 -!- oerjan has quit (Quit: Oops).
14:07:08 <fizzie> Higher-order functions are natural in Perl. For example, here's fib with the Y combinator.
14:07:11 <fizzie> !perl print &{&{sub { my $f = shift; return &$f($f); }}(sub { my $f = shift; return sub { my $n = shift; return 1 if $n < 2; return &{&$f($f)}($n-1) + &{&$f($f)}($n-2); }; })}(11);
14:07:16 <fizzie> Simple as sliced bread.
14:09:21 <fizzie> (At least I hope it is that; I just grepped for it.)
14:12:18 <ais523> that's not the Y combinator, that's the mockingbid combinator
14:12:59 <ais523> verified both by looking at the combinator itself, and by observing the unusual way the recursive calls are made
14:13:32 <fizzie> Yes, in retrospect &{&$f($f)}($n-1) looks kinda funny.
14:14:01 <itidus20> an interesting use for this is if you were to say for (j = 0; j < 10; j++) {for (i = 0; i < 15; i++) {printf("%d%d\n",i,j)}} equivalent to x(10,15) printf("%d%d\n",x[0],x[1]);
14:14:28 <itidus20> maybe with those last [0] and [1] swapped around
14:15:35 <ais523> you can of course write a Y-like fixed point combinator in Perl
14:15:48 <ais523> although you'll end up with the usual debate as to whether an arbitrary least fix point operator is Y or not
14:16:36 <itidus20> you could even change the loop's parameters from within the loop in theory
14:19:11 <itidus20> i guess you can already do that easy enough
14:20:21 <ais523> itidus20: right, because most imperative languages already use the CBN-like interpretation of control constructs, as they can't sensibly be expressed in call-by-value
14:20:30 <itidus20> such as while (x < n) { if (x > 40) n--; [...] }
14:20:47 <ais523> itidus20: right, that works in C
14:20:59 <ais523> because it does a CBN-like re-evaluation of x<n each time round the loop
14:21:01 <itidus20> its not something i would normally think of though
14:21:09 <ais523> this is the reason that while is a keyword in C, not a standard library function
14:22:16 <ais523> whereas in a CBN language, you can write while as a function: while(bool condition, com command) {if(condition) {command; while(condition, command);}}
14:22:55 <ais523> you could also write if as a function, although you'll need a conditional operator in your language /somewhere/ to write the others in terms of
14:23:07 <ais523> whether it's if or while or ?: or just using a churchish boolean
14:24:51 <ais523> (the last method is basically true(com iftrue, com iffalse) {iftrue;} false(com iftrue, com iffalse) {iffalse;}, and now your if is if(condition, iftrue, iffalse) {condition(iftrue,iffalse)}
14:25:01 <ais523> CBN = call by name, it's a calling convention
14:25:24 <ais523> in practice, CBN languages tend not to use C-like syntax, although like all syntax it's interchangeable
14:25:33 <ais523> best-known CBN language is probably Algol 60
14:32:54 <fizzie> !perl print &{&{sub { my $f = shift; return &{sub { my $x = shift; return $f->(sub { my $n = shift; return &{$x->($x)}($n); }); }}(sub { my $x = shift; return $f->(sub { my $n = shift; return &{$x->($x)}($n); }); }); }}(sub { my $f = shift; return sub { my $n = shift; return $n if $n <= 1; return $f->($n-1) + $f->($n-2); }; });}(12);
14:33:02 <fizzie> Looks somehow overly complicated, but oh-well.
14:36:16 <ais523> hmm, is that an actual y, rather than just a simple definition of fixpoint?
14:37:59 <fizzie> It was a translation of (((lambda (f) ((lambda (x) (f (lambda (n) ((x x) n)))) (lambda (x) (f (lambda (n) ((x x) n)))))) (lambda (f) (lambda (n) (if (<= n 1) n (+ (f (- n 2)) (f (- n 1))))))) 12).
14:38:18 <ais523> yep, that looks like Y all right
14:40:44 <fizzie> > (fix (\f n -> if n <= 1 then n else f (n - 1) + f (n - 2))) 12
14:40:50 <oklopol> today at work, we broke two of our proofs.
14:41:06 <oklopol> i officially declare this the most fuckfaced shithole of a day in the universe
14:41:53 <oklopol> they proved the same thing and we may have a correction, but it does not have the innocence of the previous ones.
14:42:36 <oklopol> so happy most fuckfaced shithole of a day in the universe to you all
14:43:25 <ais523> I like the name, we should make this an annual event
14:43:36 <fizzie> Today at work I computed some variances for things by making at least three unwarranted assumptions about the distribution of other things. Man, being an engineer instead of a real scientist is such a leave-your-brain-at-the-door occupation.
14:48:52 -!- oklopol has set topic: http://codu.org/logs/_esoteric/ | Happy Most Fuckfaced Shithole of a Day in the Universe to Each and Everyone! :>.
14:49:02 <oklopol> i felt the other crap was unnecessary
14:49:33 <ais523> oklopol: I was going to do that, but your version is better
14:50:05 <oklopol> i'm feeling better already for doing that
14:50:14 <itidus20> i wonder how to celebrate this most fuckfaced shithole of a day in the universe
14:52:39 -!- Slereah_ has joined.
14:53:37 <oklopol> well you could start by taking something you've spent a ton of effort to do and know to be awesome, spending 45 minutes of 10 ppl's time explaining it and then realizing that it's full of shit and you're a fucking retard.
14:53:44 -!- Slereah has quit (Ping timeout: 244 seconds).
14:54:09 <oklopol> that's the usual way to celebrate the most fuckfaced shithole of a day in the universe.
14:57:20 <oklopol> well i didn't even realize it, my colleague did. i just broke the second proof too.
14:58:29 -!- nooga has joined.
14:58:33 <oklopol> it was rather fitting that it turned out to be the 13th
15:02:16 <oklopol> you know, because i'm really really superstitious.
15:02:25 <ais523> oklopol: we have a known-incorrect proof in a published paper
15:02:36 <ais523> or to be specific, the proof proves a known-incorrect result
15:02:53 <ais523> we have the result fixed (it was just a mistake in definitions), but it's kind-of worrying how we proved it if it's wrong
15:04:30 <oklopol> yeah in my master's thesis i ask this question about whether this class called UFA is closed under union, and explained why i believe it's not (it would be a surprising result). it was certainly in the top10 most interesting things in there, and when i looked more closely into the definitions, i realized it's trivially closed under union.
15:04:47 <oklopol> i have fixed this in our later article, and it's really just because the definition was wrong.
15:07:43 <oklopol> so this girl i know realized that her second most preferred career choice is out of the question after two days in there.
15:08:07 <oklopol> i think i may be made out of magic.
15:08:12 <oklopol> sorry if i fuck up your day.
15:17:29 -!- Taneb has joined.
15:18:13 <fizzie> Taneb: Careful there, you might catch some sucky-dayness from oklopol.
15:19:03 -!- Taneb has set topic: http://codu.org/logs/_esoteric/ | We do esolangs. Deal with it..
15:19:19 <fizzie> That was a short-lived topic.
15:19:20 -!- oklopol has set topic: http://codu.org/logs/_esoteric/ | Happy Most Fuckfaced Shithole of a Day in the Universe to Each and Everyone! :>.
15:19:27 -!- oklopol has set topic: http://codu.org/logs/_esoteric/ | Happy Most Fuckfaced Shithole of a Day in the Universe to Each and Everyone! :> | Deal with it..
15:20:57 -!- Gregor has set topic: http://codu.org/logs/_esoteric/ | Happy Most Duckfaced Piffle of a Day in the Universe to Each and Everyone! :> | Deal with it..
15:22:22 -!- oklopol has set topic: http://codu.org/logs/_esoteric/ | Happy Even Topics Suck Ass Day to Each and Everyone! :> | Deal with it..
15:22:50 -!- Gregor has set topic: http://codu.org/logs/_esoteric/ | Happy Even Topics Suck Donkey Day to Each and Everyone! :> | Deal with it..
15:23:39 -!- itidus20 has quit (Ping timeout: 240 seconds).
15:26:00 -!- oklopol has set topic: http://codu.org/logs/_esoteric/ |http://codu.org/logs/_esoteric/ | Happy Odd Topics Suck Donkey Day to Each and Everyone! :> | Deal with it..
15:28:28 <oklopol> excuse me but does this topic have two log links
15:33:06 <oklopol> you're a bot so you wouldn't understand.
15:37:39 <oklopol> you only understand two things
15:38:42 <ais523> > "did I fix this bug yet?"
15:41:55 -!- Taneb has quit (Ping timeout: 276 seconds).
15:50:07 -!- Taneb has joined.
15:53:49 -!- itidus20 has joined.
15:59:09 -!- kallisti has joined.
15:59:09 -!- kallisti has quit (Changing host).
15:59:09 -!- kallisti has joined.
16:00:22 <kallisti> :t let f n a b = a `mod` n == b in f 2
16:00:23 <lambdabot> forall t. (Integral t) => t -> t -> Bool
16:00:23 <thutubot> forall t. (Integral t) => t -> t -> Bool
16:01:07 <kallisti> yo dawg I heard you like equivalence classes, so we made your integers congruent and ... actually I can't make this into a yo dawg meme so I'm just going to stop now.
16:02:25 <ais523> kallisti: you dawg I heard you like equivalence classes, so I put some integers in your car so you can make them congruent while you drive
16:13:11 -!- calamari has joined.
17:02:18 -!- Taneb has quit (Quit: Leaving).
17:07:31 <itidus20> as i am showing this weird thing i did to several channels, will also say so in here
17:08:13 <itidus20> i had this idea, and i made a mockup of it using donkey kong pictures http://oi42.tinypic.com/ddgimt.jpg it got shrunk down quite a bit on upload.. but it didn't lose any quality in the process really
17:08:38 <itidus20> this is the original pics for reference: http://www.mobygames.com/game/nes/donkey-kong/screenshots
17:10:13 -!- thutubot has quit (Remote host closed the connection).
17:10:18 <ais523> (I just quit it manually)
17:11:26 <Gregor> Manually ... WITH YOUR MIND!
17:12:30 <itidus20> yeah i got this idea from the torus discussions in here
17:13:40 <ais523> btw, what's the evaluation strategy called where every function call in the code is pure and you know the argument it'll get already, you start running it in a different thread to see what value it produces, and kill the thread once you go out of scope (i.e. the function would get another argument next time)?
17:13:46 <ais523> I think it was discussed in here a while back
17:13:50 -!- azaq23 has joined.
17:13:55 <ais523> and my supervisor just independently reinvented it
17:14:08 -!- azaq23 has quit (Max SendQ exceeded).
17:14:11 <ais523> it sounds crazy, but we realised it's actually one of the most efficient strategies around for hardware compilation
17:14:31 <Gregor> That's not really an evaluation strategy, but entails one.
17:14:41 <Gregor> (pure futures, that is)
17:14:54 <ais523> well, this is an evaluation strategy that could be used as written with, say, Haskell
17:15:06 <ais523> (whose purity makes it very easy to use arbitrary evaluation strategies with it)
17:15:23 <ais523> (if they deal with nontermination correctly, and that one does by killing the thread if termination status is irrelevant)
17:23:03 -!- augur has quit (Remote host closed the connection).
17:31:32 -!- ais523_ has joined.
17:32:38 -!- ais523 has quit (Disconnected by services).
17:32:40 -!- ais523_ has changed nick to ais523.
17:59:43 -!- augur has joined.
18:01:35 -!- zzo38 has joined.
18:05:19 -!- ais523 has quit (Remote host closed the connection).
18:06:34 -!- Taneb has joined.
18:07:02 <Taneb> You know what's weird>
18:07:21 <Taneb> I think I'm reading a book that a fictional character in a different book which I am also reading is reading
18:10:17 <Taneb> The books are Fermat's Last Theorem by Simon Singh and The Girl Who Played With Fire by Steig Larrsson
18:12:23 <itidus20> there is an idea which might only make sense to me, related to misheard song lyrics and fictional characters
18:13:20 <itidus20> but anyway, by reading a preview or review of something.. or seeing a trailer of something
18:13:52 <zzo38> I got this build error for my package: "Something is amiss; requested module template-haskell-2.6.0.0:Language.Haskell.TH differs from name found in the interface file template-haskell:Language.Haskell.TH" Did I do something wrong or is Hackage wrong?
18:13:59 <itidus20> you can end up imagining what the story might be about, filling in the gaps. and this imagining itself is an act of creation.
18:14:14 <zzo38> itidus20: O, yes, then make up such a thing.
18:14:44 <itidus20> a similar thing can happen when you mishear song lyrics you end up with a unique song, which only barely makes sense to you because you were desperate to make some sense out of it
18:15:25 <zzo38> The story of my D&D game has a few gaps (they are mostly marked with square brackets; I forgot what happened), so maybe someone can help me to make up something there
18:15:59 <Taneb> itidus20, heh, I ended up with an epic love song called "I am a running tap" via that mechanism
18:16:01 <itidus20> sometimes, in song lyrics you can discover yourself warping grammar a bit in fun ways.. and perhaps credit the song author with interesting use of grammar which he doesn't deserve since its not his lyrics
18:16:08 <Taneb> Of course, epic love song is a bit of an oxymoron
18:16:40 <zzo38> I have read some things about misheard song lyrics. Usually the music I listen has no lyrics; it is instrumental.
18:17:11 <itidus20> one of my favorites is "even flow... on the road like porno stars.. he don't know..so he chases them (the porno stars?) away
18:18:19 <itidus20> there is a lot in (the words) "who is a devil?" you can't survive so i'll be your guide
18:19:36 <itidus20> real one is "Now who's the light and who is the devil. You can't decide so i'll be your guide."
18:19:48 <itidus20> i may have used decide in my version
18:21:41 <itidus20> and "Even flow, thoughts arrive like butterflies. Oh, he don't know so he chases them away, yeah"
18:24:07 <itidus20> so, unfortunately as time goes on i find that things are rarely as cool as i first imagine them to be
18:26:25 <itidus20> i guess its also quite common to underestimate lyrics when you don't actually hear them
18:27:13 <itidus20> so its interesting that singing isn't merely about communicating a stream of words
18:32:28 <itidus20> zzo38: worlds inspired by such things from my firsthand experience can be exciting to create.
18:36:07 -!- kallisti has quit (Ping timeout: 276 seconds).
18:37:18 -!- zzo38 has quit (Remote host closed the connection).
18:54:29 -!- ais523 has joined.
19:07:04 -!- hagb4rd has quit (Ping timeout: 252 seconds).
19:35:48 <Taneb> The next Gunnerkrigg will be #1000
19:43:21 <Taneb> I'm not entirely sure if /I/ care.
19:44:04 -!- Taneb has quit (Quit: food).
19:47:23 <fizzie> /I/ was confused by "#1000"; was all "what sort of color code is that, shouldn't it have either three or six (or 12) digits, is that like a *really* bright #fff."
19:47:55 <ais523> fizzie: clearly it's an entirely transparent dark red
19:48:15 <ais523> just like #F990 is an appropriate color for an invisible pink unicorn
19:49:24 <fizzie> For some reason I'd've put the alpha channel first.
19:51:46 <fizzie> urxvt accepts colour specifications of the form "rgba:RRRR/GGGG/BBBB/AAAA", where the uppercase letters are hex digits.
19:52:02 <ais523> RGBA is the normal order
19:52:36 <olsner> #1000 is the blink bit - that's blinking black
19:52:58 <olsner> adding #2000 makes it a marquee
19:53:29 <olsner> (and #4000 is a tiling under construction gif?)
19:54:06 -!- azaq23 has joined.
19:54:12 <fizzie> Sure, it's the normal order, but for some reason when in four-digit "#xxxx", I'd've done ARGB.
19:55:16 <fizzie> I think Inkscape's color-selector has an eight-hex-digits field of the color being selected.
19:56:05 -!- tzxn3 has quit (Quit: Leaving).
19:56:22 -!- tzxn3 has joined.
19:57:53 <fizzie> "Unlike RGB values, there is no hexadecimal notation for an RGBA value." (CSS3.) Aw.
19:58:19 -!- MoALTz has quit (Ping timeout: 240 seconds).
20:01:26 -!- elliott has joined.
20:01:44 <elliott> I've added more CAPTCHAs to the test wiki at http://95.149.228.149:8181/wiki/Main_Page; prizes to anyone who fills out all of them without looking them up
20:02:21 <Gregor> Uhhhhh, CAPTCHAs shouldn't be hard enough to deserve prizes X-D
20:02:30 <Sgeo> "without looking them up"
20:02:45 <Sgeo> Normal use probably involves looking them up
20:03:03 <Gregor> Oh, they come with a link.
20:03:06 <Gregor> OK, so it's just "read"
20:03:20 <ais523> yep, the CAPTCHA is basically about ability to parse and understand English
20:03:25 <ais523> which is a hard task for computers at current
20:03:36 <elliott> and follow URLs, which is a hard task for someone being paid to fill out CAPTCHAs out of context
20:03:57 <ais523> elliott: I imagine those people probably have web browsers by now
20:04:07 <itidus20> ok i have seen the first captcha.. and i can't answer it without looking it up :D
20:04:11 -!- Gregor has set topic: http://codu.org/logs/_esoteric/ | Resistance is not tolerated.
20:04:14 <elliott> I heard they're usually just presented with it out of context, but who knows
20:04:17 <ais523> main reason to do that is to make the captchas hard to MitM
20:04:30 <ais523> which is a moderately uncommon trick, but has been known to work
20:04:39 <ais523> or, I guess it's a reverse-MitM
20:05:15 <elliott> itidus20: which one was it?
20:05:58 <itidus20> i went to add blah2 to ehird's user page
20:06:05 <Gregor> Who catered UnlambdaCon 2011?
20:06:18 <elliott> itidus20: the link in that one actually works
20:06:34 <elliott> I didn't bother adding articles for the new three, though
20:06:37 <itidus20> it first appeared to me that the edit had worked.. so i went to click on main page
20:06:51 <itidus20> and it said.. Leave Page Stay on page
20:07:07 <itidus20> thats when i noticed the captcha subtly placed
20:07:12 <olsner> mwahaha, I have successfully spammed the wiki, your captcha has failed
20:07:13 <elliott> yeah, it's hard to notice :(
20:07:22 <Gregor> Yeah, the appearance of the CAPTCHA is a bit surprising.
20:07:25 <elliott> I wish it came in a box or something
20:07:32 <elliott> having said that, it looks the same on the current esowiki
20:07:50 <elliott> olsner: Gregor: itidus20: Did you notice the "Blah blah blah blah!" at the top of the pages?
20:07:55 <elliott> ais523 didn't, so I'm worried that it's not noticeable enough
20:08:04 <elliott> (it'll contain server migration/technical problem contact information for a week or so)
20:08:07 <Gregor> I noticed it, but it was just "Blah blah", so uh
20:08:17 <elliott> They'll be *helpful* blahs!
20:09:09 <itidus20> elliott: [not giving serious critic here] reading this chat, it reminds me of one of those evil super mario bros hacks ... as if it is booby trapped
20:09:10 <olsner> elliott: yes, but I didn't reflect over the fact that it was not part of the wiki-page content (which already was full of spam and Blah Blah-like stuff)
20:09:48 <itidus20> "did you spot the blahblablah?" "did you spot the captcha?"
20:09:48 <ais523> itidus20: you're accusing an IRC channel of being a kaizohack?
20:09:56 <Gregor> <olsner> elliott: yes, but I didn't reflect over the fact that it was not part of the wiki-page content (which already was full of spam and Blah Blah-like stuff) // agreed
20:10:12 <itidus20> ais523: no no no... no no... not at all no... just the wiki
20:10:13 <olsner> I had to reopen the page to realize that there was a blah blah there and that I had already seen it
20:10:54 <elliott> it will probably be more noticeable above http://esoteric.voxelperfect.net/wiki/Main_Page
20:10:56 <itidus20> to make spambots feel as if they are playing a kaizohack
20:11:03 <elliott> but I'll try and make it a little more obvious
20:11:09 <itidus20> elliott: don't go too easy on them though
20:11:25 <olsner> elliott: it'd probably help the most if it was an actual sentence
20:11:29 <itidus20> the more of a kaizohack it is the better
20:11:30 <Gregor> itidus20: To commit changes, you have to hit this P switch before going through the goal?
20:11:50 <olsner> sentences usually give some clue to why they're there and what they're about
20:13:02 <itidus20> Gregor: the "save page" button will hide if you try to use it
20:13:25 -!- MoALTz has joined.
20:14:07 <ais523> Gregor: heh, the wiki for a while actually had a "your edit looks like spam, if you try to submit it again you will be blocked unless you put this specific string in the edit summary somewhere"
20:14:41 <Gregor> And what was wrong with that that the test had to go?
20:15:04 <olsner> how clever are the esowiki spammers? C2 wiki had a nice simple password system for a long time that was basically "Enter the password: [....] (the password is 1234)"
20:15:05 <ais523> Gregor: it prevented people linking to example.com
20:15:28 <elliott> http://95.149.228.149:8181/wiki/Main_Page OK, how about now?
20:15:28 <ais523> olsner: we have many different spammers; the current ones seem to know about MediaWiki in particular, but not esolang in particular
20:15:40 <ais523> Gregor: that's not a very good method of recognising spambots
20:15:49 <elliott> It's more noticeable, but doesn't seem obtrusive on e.g. http://95.149.228.149:8181/wiki/Univar to my eye.
20:16:33 <Gregor> elliott: My only complaint is that it should read "We've moved servers! If you experience any problems, too god damn bad."
20:17:37 -!- olsner has set topic: http://codu.org/logs/_esoteric/ | Resistance is frowned upon.
20:17:40 <elliott> olsner: Probably not very clever, but at least one of them can solve the arithmetic captcha we currently have.
20:17:51 <elliott> (Which is just a trivial textual addition (or sometimes subtraction, I think).)
20:18:03 <elliott> They might not be able to solve a copy of the C2 CAPTCHA, but only because they don't know about it.
20:18:13 <ais523> elliott: it's a bit obtrusive, but in a position where I can easily scroll it away, so I don't mind
20:18:18 -!- kallisti has joined.
20:18:18 -!- kallisti has quit (Changing host).
20:18:18 -!- kallisti has joined.
20:18:59 <itidus20> ok its probably not cool but ill spell it out
20:19:30 <olsner> well, if they don't know about esowiki specifically they won't figure out anything the least custom we make regardless of how easy it is? or will some human involve itself after they notice a wiki that doesn't work?
20:19:32 <itidus20> so.. the capctha looks like an ordinary captcha... but it doesn't have any text in it whatsoever.. just fragments pretending to be text
20:19:48 <itidus20> and... it has a random colour background
20:20:10 <itidus20> and you get people to type in the colour of the background in natural language
20:20:34 <kallisti> olsner: rule #1: don't overestimate the intelligence of spammers
20:20:36 <itidus20> i know its trouble for colourblind and blind though
20:21:08 <itidus20> so you just say, "what is this?" and show mostly a block of colour, with a few fragments of another colour
20:21:15 <kallisti> olsner: even if they notice they're not going to put time/effort into targeting a custom wiki. They have spambots that work with mediawiki specifically because it's so common and allows them to spam to many many different sites.
20:21:25 <olsner> itidus20: presumably blind people will have no problem inspecting the HTML and extracting the color value
20:21:28 <elliott> http://95.149.228.149:8181/wiki/Main_Page ;; I improved the colours :P
20:21:50 <kallisti> I doubt they would care that they can no longer target esowiki.
20:22:05 <itidus20> olsner: not the worst idea though eh?
20:22:48 <fizzie> Rule 2: don't underestimate the desirability of the sought-after esolang market to spammers.
20:23:05 <fizzie> They know where the big money is.
20:23:15 <elliott> I bet fizzie can fill in all the CAPTCHAs without looking them up.
20:23:18 <ais523> haha, I could use AbuseFilter to allow, say, only one BF derivative to be added to the wiki per day
20:23:38 <fizzie> I bet I don't even know how to see them all.
20:23:54 <elliott> fizzie: Try to edit http://95.149.228.149:8181/wiki/Main_Page repeatedly.
20:23:57 <elliott> There are five (5) unique ones.
20:24:04 <elliott> (You can just submit without filling in to get the next one.)
20:24:11 <elliott> Maybe I'll just tell you them.
20:24:32 <elliott> When was Underload created?
20:24:46 <elliott> What language was the wire-crossing problem first considered for?
20:24:53 <elliott> What INTERCAL variant uses ternary rather than binary?
20:24:56 <kallisti> honestly I think you would completely stop spam if you just changed some of the names/ids of elements on the page.
20:25:09 <olsner> maybe you could make it so that underload is the only captcha question? remembering 2006 is not that hard, but the name of the guy who created unlambda is, like, more than 3 unique characters
20:25:19 <elliott> kallisti: Yes, everyone has an obvious, trivial, obstruction-free solution to spam that doesn't work.
20:25:19 <ais523> with modern spambot frameworks, you'd apparently have to add extra elements in random places too
20:25:35 <ais523> and make them not invisible
20:25:43 <ais523> although you could make them very small and hide them behind other things
20:25:52 <ais523> and then you'd confuse people using screen readers, etc
20:26:06 <itidus20> anyway, as i sincerely have nothing better to do, this is the mockup of my captcha idea http://oi43.tinypic.com/2j8ho7.jpg
20:26:27 <kallisti> elliott: you really like to exaggeate things out of proportion don't you?
20:26:57 <fizzie> I did the element-id-renaming on one phpbb, and it worked when I did it, but then later on stopped.
20:27:04 <elliott> kallisti: I exaggerated nothing except for "everyone".
20:27:19 <elliott> fizzie: STILL WAITING FOR YOUR ANSWERS.
20:27:34 <elliott> itidus20: What you seem to have missed is that it's really really easy for a computer to solve.
20:27:40 <ais523> elliott: via PM, presumably?
20:27:53 <elliott> itidus20: Take the largest area, take its RGB colour, look up in an RGB -> colour name database.
20:27:58 <ais523> itidus20: and also hard for a human to work out what it's getting at
20:28:07 <itidus20> elliott: but it has to want to
20:28:21 <ais523> itidus20: the current Esolang spammers know the workings of its CAPTCHA
20:28:30 <ais523> this may well be because it's the one that comes with MediaWiki by default
20:28:39 <kallisti> fizzie: yes, they would probably have to make some kind of custom script with the new ids (or you know the obvious solution like AI botnet hivemind).
20:28:39 <elliott> fizzie: By the way, did you download the forum? I got Deewiant to do it since I don't trust only one Finn.
20:28:47 <elliott> ais523: it's the "baseline" ConfirmEdit captcha, so yes
20:29:08 <elliott> kallisti: Or just going by element order.
20:29:23 <itidus20> elliott: one secret is by not actually having the world colour anywhere on the captcha
20:29:24 <ais523> <elliott> fizzie: By the way, did you download the forum? I got Deewiant to do it since I don't trust only one Finn. <-- http://qdb.rawrnix.com/?615
20:29:38 <fizzie> Oh, I didn't realize I was supposed to answer them. Meh, not gonna. I'm so bad about time, I don't really have a clue about the when thing, except I have a feeling it might've been sometime the last decade.
20:29:40 <olsner> "... since I don't trust only one Finn." - and you trust two finns *more*?
20:29:47 <ais523> itidus20: that just makes it harder for humans without making it harder for computers
20:30:05 <itidus20> ais523: the computer is supposed to be tricked by the black dots
20:30:10 <ais523> which is not normally a desirable property for CAPTCHAs
20:30:12 <fizzie> And no, I forgotteded about it, then remembered, then forgotteded.
20:30:40 <ais523> itidus20: the problem with CAPTCHAs is that most spammers can spare the time to have one human configure them for each site they come across
20:30:44 <ais523> and just let them spam from there
20:30:46 <elliott> olsner: Two Finns are better than one.
20:31:28 <elliott> I bet olsner knows all the answers.
20:31:32 <elliott> I bet SOMEONE knows all the answers.
20:33:06 <olsner> hmm, SOMEONE should be an acronym (or that thing everyone calls an acronym but really isn't?) for something
20:33:20 <elliott> OK, the TODO list before the new esowiki can go live: Figure out the preferences weirdness, get stuff automated, prepare server, contact Graue.
20:33:23 <ais523> olsner: an acronym is an initialism that is also a real word
20:33:41 <elliott> ... and figure out why http://95.149.228.149:8181/w/ isn't redirecting to /wiki/Main_Page
20:34:00 <itidus20> a captcha where you have to pick the sexy person versus the ugly person?
20:34:01 <olsner> is also a real word? or has the same letters? I think words and same-lettered acronyms should be separate entities
20:34:25 <olsner> itidus20: subjective...
20:34:45 <itidus20> it could be tinkered with by the editor
20:34:49 <olsner> also, 50/50 chance of getting it right, just keep guessing
20:34:56 <ais523> oh, something I'd like a bit of advice on: I'd like to write a CGI script that takes a user-entered path, and returns a file with that name in a particular directory tree, while counting the fact it was requested
20:35:11 <ais523> basically, acting like an HTTP fileserver except that it has a hit counter
20:35:31 <kallisti> a) remove anonymous edits b) require a captcha on the create account page c) obfuscate the form field labels by using "for" attributes that point to the incorrect form field (does this break legitimate software somewhere?) d) move the captcha to the bottom of the create account form, thus changing the element order of forms.
20:35:33 <ais523> except that I have to write it in PHP, and am not sure how to secure it to avoid, say, directory traversal attacks, or even shell injection attacks because this is PHP we're talking about
20:35:53 <ais523> kallisti: a) require a CAPTCHA for anonymous edits
20:35:54 <elliott> itidus20: There's already a CAPTCHA based on image distinguishing.
20:36:03 <kallisti> ais523: captchas aren't really a big deal.
20:36:12 <ais523> and obfuscating labels really does break legitimate software
20:36:24 <elliott> kallisti: (a) Unacceptable, (b) already done, (c) breaks normal browsers and makes unusable for disabled people, (d) idiotic
20:36:25 <ais523> kallisti: removing anonymous editing /is/ a big deal, IMO
20:36:34 <kallisti> well that's unfortunate because it tells the spambot exactly where to go.
20:36:53 <kallisti> elliott: I only see a math solver thing
20:37:02 <ais523> kallisti: that /is/ the CAPTCHA
20:37:06 <ais523> unfortunately, it sucks as a CAPTCHA
20:37:08 <elliott> http://95.149.228.149:8181/w/index.php?title=Special:UserLogin&type=signup&returnto=Main+Page
20:37:24 <ais523> and the spambots routinely break it
20:37:33 <elliott> oh, extra TODO: get email working
20:37:37 <ais523> elliott: what are you doing about importing accounts in the server change, btw?
20:37:38 <kallisti> ais523: news flash: programs good at math.
20:37:53 <ais523> kallisti: news flash: programs bad at parsing the fact that they have to solve math
20:38:04 <ais523> the problem with that CAPTCHA is that it only needs human intervention to break /once/
20:38:12 <ais523> this is the same as the problem with changing element order, or changing round labels
20:38:19 <ais523> only need human intervention once to defeat
20:38:33 <itidus20> ok so a subjective question is needed with a wide range of possible answers
20:38:37 <kallisti> elliott: explain to me how changing the element order of the form is an idiotic way to prevent spambots from using the standard mediawiki element order to create an account?
20:38:44 <ais523> itidus20: to make it harder to /judge/?
20:38:47 <elliott> ais523: technically, this CAPTCHA only needs human intervention once, it's just a tedious and unautomatable once
20:38:56 <elliott> kallisti: Because it's a waste of time compared to just using a decent CAPTCHA.
20:39:01 <ais523> elliott: right, this is the main reason to have a lot of questions
20:39:20 <itidus20> olsner pointed out that if its a 2 option question then they will quickly get it right like flipping a coin waiting for heads
20:39:21 <ais523> actually, potential problem: the spambots are told the answer to one question, then they try it on every CAPTCHA in the hope the question will come up again
20:39:29 <ais523> there's no penalty for CAPTCHA fails, is there?
20:39:51 <kallisti> elliott: hmmm, perhaps. I don't know the success rate of CAPTCHA solvers on these so-called "good" CAPTCHAs. but they appear to be successful on most of the internet, so....
20:39:56 <ais523> I guess if any do that I just abusefilter them, also checkuser them
20:40:26 <ais523> kallisti: apparently the most common sort of captcha OCR works badly on Wikipedia's, but the second-most-common works quite well
20:40:51 <ais523> ("captcha OCR" = "OCR designed to break captchas rather than digitise more normal text)
20:41:22 <elliott> ais523: wow, Wikipedia's main page was only protected in 2006
20:41:55 <ais523> 2006 is before people started systematically trolling it
20:42:14 <elliott> ais523: anyway, yes, ConfirmEdit has a documented lack of penalty
20:43:15 <ais523> AbuseFilter can be set to block an entire /16 without warnings, incidentally, but that seems a little excessive to me
20:43:19 -!- monqy has joined.
20:43:26 <ais523> I usually look up how large the range the IP is part of is, and block that
20:43:36 <kallisti> hopefuly no question database is going to have esolang related questions. Though I have no idea how spambots acquire question databases in the first place.
20:43:47 <elliott> ais523: hmm, it occurs to me that spammers dislike CAPTCHA solving software too
20:43:58 <elliott> they use distorted text in images for email spam, after all
20:44:15 <ais523> hmm, perhaps we can set the two groups against each other
20:44:24 <ais523> and have confirming email as spam or not /as a CAPTCHA/
20:44:34 <elliott> wow, I want to implement that now
20:44:35 <ais523> unexpectedly awesome idea?
20:44:47 <ais523> (note: you need to be willing to allow random strangers to read your email)
20:44:58 <kallisti> ais523: what if gmail is secretly a spam botnet.
20:45:07 <elliott> ais523: nah, just set up a bunch of honeypot email accounts
20:45:10 <elliott> with their address all over the 'net
20:45:16 <elliott> and use the data to train a spam filter for actual users
20:45:17 <ais523> kallisti: it isn't secretly a spam botnet, because it's reasonably openly a spam botnet
20:45:25 <ais523> elliott: but they wouldn't get legitimate emails
20:45:31 <ais523> so the CAPTCHA would be solved by labeling them all as spam
20:45:37 <elliott> ais523: hmm... encourage people to send fake legitimate emails :P
20:45:45 <ais523> kallisti: gmail's the source of a reasonably large percentage of the world's spam
20:46:03 <kallisti> ais523: people /using/ gmail or the software behind gmail? I think we're talking about two different things.
20:46:05 <ais523> elliott: it wouldn't be anywhere near the spam in volume
20:46:10 <ais523> elliott: people using gmail
20:46:47 <kallisti> right, I was just making a bad joke, that if gmail were a spambot then it could easily distinguish which emails are spam.
20:46:51 <elliott> ais523: write a botnet to send fake legitimate mail, then
20:47:11 <ais523> elliott: hmm, then you could subscribe to its non-spam
20:47:14 <ais523> and be a lot less lonely
20:47:22 <ais523> and get patches for AceHack, etc
20:47:28 <elliott> http://upload.wikimedia.org/wikipedia/commons/5/55/Spam_image.jpg I feel dizzy...
20:48:06 <elliott> ais523: that's just about the saddest thing i ever heard get said
20:48:32 <elliott> it illustrates http://en.wikipedia.org/wiki/Image_spam
20:48:48 <elliott> wait, does that mean the image spam is freely-licensed?
20:49:24 <ais523> elliott: it was probably sent by a spambot to one of those people with signatures saying they own the copyright of anything you email to them
20:51:16 <elliott> wow, seeing Monobook after using Vector for a while is weird
20:51:19 <kallisti> elliott: wow it's a captcha ad.
20:51:21 <itidus20> this is why you can't have a babylon where everyone freely understands each other
20:51:27 <fizzie> Lately I've been getting rather amusing "follow-ups" to the usual "here's a lot of money for you" spams. I'm sort of wondering if there's already a large market of people who've already responded to regular spam, and are therefore "primed" for stuff like this: http://p.zem.fi/ikvv
20:51:57 <itidus20> bots are empowered by everyone in the community dividing against each other (with the bots)... as if to force division
20:52:09 <fizzie> World Court and German businessmen and everything.
20:52:48 <itidus20> i mean it is quite possible for tribal languages to spring up of a guarded nature so that their captchas can't be figured out
20:52:56 <elliott> ais523: you should click fizzie's link
20:52:59 <kallisti> oh, btw, you need to ban elliott for spamming the channel with viagra ads.
20:53:14 <ais523> elliott: sorry, was distracted
20:53:45 * kallisti wonders why audio captchas are not frequently used.
20:53:49 <ais523> wow, the first paragraph really /is/ a single sentence, I was getting a little dizzy parsing it
20:53:59 <ais523> kallisti: many people use the internet in a situation where they can't easily get sound
20:54:01 <fizzie> And that wasn't the only "follow-up" style one; I also got one from the "bank guy" who's actually doing the money reassignment, warning me that the people I've been conversing with are scammers who haven't paid him his $10k, but that he's willing to make a deal directly with me.
20:54:13 <kallisti> I guess because audio in HTML is kind of a pain in the ass (or is for the time being until HTML5 audio becomes completely safe across all commonly used browsers)
20:54:19 <elliott> kallisti: they're annoying
20:54:39 <elliott> kallisti: also, understanding them can be hard compared to image captchas
20:54:50 <elliott> especially across cultures (accents, etc.)
20:55:04 <kallisti> animated CAPTCHAs? is this a thing people do?
20:55:07 <ais523> your name has been enlisted in the WORLD SCAM RECORDS as a top 11 fraudster
20:55:13 <monqy> kallisti: haven't you seen them?
20:55:19 <monqy> kallisti: the captchas with the car ads
20:55:32 <monqy> or were they static I forget
20:55:32 <kallisti> I live in a world without ads.
20:55:38 <elliott> [[ We have also spread our networks to the following countries which are
20:55:38 <elliott> Australia, Canada, Africa, United Kingdom, Asia and Poland.]]
20:55:41 <monqy> anyway, captcha that's also a car ad
20:55:43 <kallisti> monqy: maybe you were tripping when you saw them
20:55:44 <elliott> Africa and Asia are my favourite countries
20:55:47 <monqy> except the text isn't obscured at all
20:55:47 <kallisti> and they appeared to be animated.
20:55:52 <fizzie> ais523: I liked the $20M they sort-of offhand mention in one paragraph, and then never again later.
20:55:57 <monqy> so it's just a car ad and you have to write down the name of the car
20:56:00 <elliott> kallisti: there's "ad captchas"
20:56:04 <elliott> kallisti: they're disgusting
20:56:09 <elliott> because they're not actually captchas
20:56:12 <kallisti> right, I'm talking about "captcha captchas"
20:56:15 <elliott> and no bot would have problems solving them
20:56:22 <elliott> but you're forced to type in marketspeak bullshit to get past them
20:56:27 <ais523> wow parsing this message gives me a headache
20:56:34 <fizzie> The FREE CITIZEN CERTIFICATE also goes from $90 to $140 inexplicably. (Inflation?)
20:56:56 <ais523> fizzie: I that they say it costs $90 and ask for $140
20:57:01 <ais523> I'm guessing that it's their cut for notifying you
20:57:08 <ais523> perhaps it's even a scam MitM attack
20:57:10 <elliott> wow, those with Flash: http://www.adscaptcha.com/ take a look at the slide-to-fit/simple code integration
20:57:16 <elliott> it's absolutely impossible for a bot to slide a slider all the way!
20:57:27 <ais523> where one bunch of scamsters skip off $50 and give the other $90 to the bunch they're MitMing
20:57:53 <kallisti> oh.... that's what you meant by captcha ads
20:58:01 <ais523> elliott: actually, Flash CAPTCHAs seem like a reasonable idea, just to suck up all the spambot's CPU time
20:58:18 <kallisti> that's a sick amalgamation of two tedious things about the internet, combined into one.
20:58:21 <elliott> kallisti: that one is actually obscured though
20:58:27 <elliott> let me try and find the worst one
20:58:34 <elliott> which is just making you watch an ad then type in unobscured text
20:58:45 <monqy> yeah that's the one I was thinking of
20:58:49 <elliott> kallisti: http://www.solvemedia.com//advertisers.html
20:59:08 <calamari> they found a way to defeat captchas.. just put them on porn sites and make humans type them to view the porn
20:59:14 <kallisti> elliott: "wasted attention" good one
20:59:30 <elliott> http://www.solvemedia.com//images/research.png
20:59:41 <ais523> calamari: that's known as CAPTCHA MitMing, but it doesn't actually work too well
20:59:50 <kallisti> elliott: STOP SPAMMING AAAAAAH
20:59:56 * elliott pronounces MitMing as "mitt-ming"
20:59:59 -!- Taneb has joined.
21:00:03 <calamari> ais523: works well enough.. tons of spam on 4chan
21:00:05 <kallisti> elliott: is there another way to pronounce it?
21:00:10 <ais523> compared to just hiring minimum-wage workers in a country with low minimum wages
21:00:31 <calamari> ais523: ah maybe thats it then
21:00:34 <ais523> calamari: and what's the point in spamming 4chan?
21:00:48 <calamari> ais523: to link back to their porn site
21:00:51 <ais523> not to mention, if you spammed 4chan you'd probably end up with naked pictures of yourself all over the Internet
21:01:24 <elliott> Taneb: prize for solving all five captchas on http://95.149.228.149:8181/wiki/Main_Page without looking the answers up
21:01:40 * kallisti didn't parse it as an acronym followed by -ing
21:01:44 <itidus20> ais523: so you mean for example, demanding 30 seconds of cpu time for the privilege of viewing the captcha's question decoded?
21:01:49 -!- Taneb has set topic: http://codu.org/logs/_esoteric/ | Resistance is frowned upon, thus voltage is frowned upon times currant..
21:01:52 <elliott> <ais523> not to mention, if you spammed 4chan you'd probably end up with naked pictures of yourself all over the Internet
21:01:55 <elliott> ais523: I take it you speak from experience
21:02:05 <ais523> elliott: no, I haven't tested this
21:02:08 <kallisti> MitMing sounds like some kind of trendy internet business name.
21:02:23 <elliott> Taneb: try to edit the page
21:02:24 <fizzie> CA-CAPTCHA: it runs a 2D CA on a HTML5 canvas, then asks you to describe the result.
21:03:14 <monqy> the result is a car ad
21:03:18 <ais523> what would make a good addition to a CAPTCHA: supply a random string, ask for a string to append for it which ends up with an SHA1 hash starting with an appropriate number of zeros
21:03:21 <Taneb> Oh god I have no idea
21:03:25 <kallisti> fizzie: come on that's just silly.
21:03:29 <ais523> that could happen to the user in the background while they were filling in the login form
21:03:34 * kallisti casually scrolls up and reads all the other ideas that have been posted.
21:03:43 <Taneb> Underload was created... '97?
21:03:54 <ais523> hey, even I don't know that one off by heart, and it was me who created it
21:03:56 <itidus20> i think ais has really hit on something... sacrificing cpu time to decode the captcha image to slow down the spambot
21:04:02 <Taneb> Wire corssing problem... Befunge?
21:04:11 <elliott> Taneb: 2006; you can get a second-place prize for doing the others
21:04:23 <elliott> itidus20: that's [[Hashcash]]
21:04:37 <elliott> itidus20: the problem is that javascript/Flash are really slow compared to hand-optimised C, which is what the spambots can use
21:04:40 <ais523> whereas this is for webforums
21:04:43 <kallisti> you could have some kind of time/memory consuming process that runs on the page. normally users would kill the JS when prompted to do so, spambots probably wouldn't (?????)
21:04:45 <elliott> so it's basically a browser vs. spambot optimisation war
21:04:53 <ais523> this is totally not obvious and I deserve a patent with millions in license fees
21:05:02 <Taneb> INTERCAL variant... AAARGH
21:05:05 <Taneb> I don't know INTERCAL
21:05:14 <Taneb> ...Is it CLC-INTERCAl?
21:05:25 <elliott> the answer was TriINTERCAL
21:05:48 <Taneb> Unlambda... oh come on I know this
21:05:57 <Taneb> Except I really don't./
21:06:07 <Taneb> I know who didn't create it
21:06:18 <ais523> Taneb: I know more than one person who didn't create it
21:06:23 <elliott> Taneb: only one more to fail at! :P
21:06:28 <monqy> I didn't create it.
21:06:32 <ais523> hmm, all this reminds me that esolangs maybe wouldn't make an ideal specialist subject for me on Mastermind
21:06:43 <ais523> not that I'm planning to apply for Mastermind
21:06:49 <elliott> ais523: I somewhat doubt they'd even allow that as a subject
21:06:52 <Taneb> IRP... Internet Relay Program?
21:06:53 <itidus20> Taneb: if it makes you feel any better, Unlambda is not a part of my life at all
21:06:58 <monqy> Taneb: how could you
21:07:00 <ais523> elliott: I think they would, it seems broad enough
21:07:01 <elliott> Taneb: Programming, actually
21:07:07 <ais523> I could get the IRP one
21:07:11 <Taneb> That's got to be close enough
21:07:11 <elliott> ais523: yes, but the reliable sources?
21:07:17 <elliott> I'd just pre-poison esolangs.org :)
21:07:25 <elliott> Taneb: not close enough for the captcha :P
21:07:27 <ais523> does Mastermind require reliable sources?
21:07:38 <elliott> ais523: I assume they use some encyclopedia or something
21:07:42 <Taneb> elliott, I don't think "Befunge" would be for the befunge one
21:07:43 <ais523> I know that recently, I've seen a gameshow that explicitly states its sources
21:07:44 <elliott> meh, they probably just use Wikipedia
21:07:49 <ais523> and asks questions starting "according to IMDB", etc
21:07:56 <elliott> Taneb: "Befunge" is accepted for the wire-crossing one
21:08:01 <ais523> IMDB is quite a common source for them to use
21:08:05 <fizzie> I knew about TriINTERCAL, and IRP, and Befunge, and had the correct decade for Underload (sadly not good enough), but didn't recall the Unlambda dude's name.
21:08:15 <ais523> fizzie: I'd have guessed 1995, thinking about it
21:08:18 <ais523> so I was only a year off
21:08:40 <ais523> elliott told me the answer to the Unlambda question before I had a chance to think about it, but I didn't remember it
21:09:08 <Taneb> I reckon Mastermind would allow Esolangs as a topic
21:09:10 <elliott> actually, you guessed it, I think
21:09:52 <ais523> looking through oerjan's impl of ~ in Underload-minus-~, it feels a little unneccessarily complex
21:09:54 <ais523> although I get the idea
21:10:32 <ais523> the problem is that lacking ~ means that most of the standard stack manip tricks don't work
21:10:54 -!- Taneb has set topic: http://codu.org/logs/_esoteric/ | Resistance is frowned upon, thus voltage is frowned upon times currant. | Electric buns, all the rage..
21:11:47 <Taneb> (currant != current)
21:12:00 <elliott> ais523: I remember that one
21:12:04 <elliott> because I remember being surprised that I understood it
21:12:19 <ais523> right, stepping through it the principle is obvious
21:12:23 <ais523> it's just hard to parse mentally
21:12:49 <elliott> I forget what it uses to be able to change the structure of the stack
21:13:10 <ais523> yep, it basically constructs a program (a)(b)(a)(b) and puts !s in the right places
21:14:20 -!- oerjan has joined.
21:14:55 -!- oerjan has set topic: http://codu.org/logs/_esoteric/ | Resistance is frowned upon, thus voltage is frowned upon times current. | Electric buns, all the rage..
21:15:11 <oerjan> or was that intentional.
21:15:29 <ais523> oerjan: I spotted the change; I'm guessing it was unintentional and then the next section was added to make fun of it
21:15:39 <ais523> hmm, now I want to have a go at implementing ~ myself
21:16:05 <elliott> * Taneb has changed the topic to: http://codu.org/logs/_esoteric/ | Resistance is frowned upon, thus voltage is frowned upon times currant. | Electric buns, all the rage.
21:16:05 <elliott> <Taneb> (currant != current)
21:16:48 <elliott> ais523: hmm, how does it construct (a)(b)?
21:16:51 <elliott> you need to dip to the a to "a" it
21:16:59 <ais523> elliott: that's the tricky part
21:17:02 <ais523> if you have dip, swap is easy
21:17:44 <ais523> btw, Proggit are talking about concatenative languages at the moment, at least in one thread
21:17:55 <ais523> in response to the submission of a blog post by evincar
21:17:59 <elliott> is that evincar's unbearable blog post
21:18:14 <Taneb> evincar... that name sounds oddly familiar
21:18:20 <monqy> he comes here sometimes
21:18:23 <monqy> elliott doesn't like him
21:18:33 <ais523> elliott: right; the information is mostly good, but the opinions are probably counterproductive
21:18:35 <Taneb> Link to blog post?
21:18:38 <ais523> http://evincarofautumn.blogspot.com/2012/02/why-concatenative-programming-matters.html
21:18:44 <elliott> ais523: the information is good, except when it's wrong
21:18:51 <elliott> for instance, he got the definition of "concatenative" wrong
21:19:39 <ais523> hmm, in what way? I'm tired, may not have been paying attention
21:19:50 <elliott> ah, 11:00:16 <slava> rien: I want to fix a few bugs and get a final 0.95 release out but other than that I'm mostly done
21:20:07 <elliott> ais523: well... it's not so much wrong as not a definition at all
21:20:29 <elliott> ais523: importantly, it doesn't actually address the "concat" part of "concatenative"
21:21:07 <ais523> I suppose my definition would be "a concatenative language is one in which concatenating the source code that represents two functions f and g creates a valid source code for the function f compose g"
21:21:14 <Taneb> So... it's a blog post about Native E code?
21:21:23 <ais523> Taneb: it uses Factor as its example language
21:21:32 <ais523> but most of what it says would work just as well for Joy or Underload, apart from the code examples
21:22:10 <elliott> I think ais523 just makes up answers to questions he doesn't know the answer to.
21:22:26 <elliott> <ais523> Taneb: it uses Factor as its example language
21:22:43 <elliott> Factor doesn't use "define foo [...]" last I checked
21:22:43 <ais523> what language are the examples written in, then?
21:22:57 -!- Taneb has quit (Read error: Operation timed out).
21:23:00 <elliott> ais523: probably either none, or evincar's awful concatenative language
21:23:04 <elliott> that we keep yelling at him about repeatedly
21:23:06 <ais523> oh, hmm, he doesn't actually say the examples are in Factor anywhere
21:23:11 <ais523> he strongly implies it multiple times, though
21:23:49 <ais523> elliott: basically by it being the only concatenative language he mentions
21:23:52 <ais523> before Kitten right at the end
21:24:09 <elliott> that guy stole the name kitten?!
21:24:15 <elliott> god, I hate him even more now
21:24:23 -!- Taneb has joined.
21:24:28 <ais523> elliott: because it's a cut-down version of Cat, apparently
21:24:49 <ais523> oh, he also mentions Prog, which is "principally concatenative"
21:25:06 <Taneb> On another note, elliott hasn't seen the new improved Pietbot!
21:25:11 <ais523> and is apparently based around pattern matching and GADTs
21:25:17 <ais523> thus, I conclude that it sucks compared to Anarchy
21:25:29 <monqy> you're confusing kitten and prog
21:25:32 <monqy> evincar has two languages
21:25:41 <ais523> "he also mentions Prog"
21:26:05 <monqy> is it a cut-down version of cat though?
21:26:31 <Taneb> Pietbot is afk, 'twould seem
21:26:56 <elliott> ais523: why might MediaWiki fail to redirect to the short-URL version of a page?
21:26:59 <ais523> now I'm disappointed that the Cat interpreter almost certainly isn't called cat
21:27:10 <ais523> elliott: because it doesn't know for certain that the short-URL version is set up correctly
21:27:19 <ais523> you'd have to tell it in the config that the redirect will work
21:27:23 <ais523> otherwise it assumes it won't
21:27:29 <elliott> ais523: I told it with $wgArticlePath
21:27:38 <elliott> is there something else I have to set?
21:27:45 <ais523> I don't know of another reason, but that doesn't mean another reason doesn't exist
21:27:56 <elliott> maybe it's caching too much or something
21:30:33 <oerjan> <ais523> btw, what's the evaluation strategy called where every function call in the code is pure and you know the argument it'll get already, you start running it in a different thread to see what value it produces, and kill the thread once you go out of scope (i.e. the function would get another argument next time)?
21:30:39 <oerjan> speculative or opportunistic
21:31:06 <ais523> oerjan: is there a name for the convention itself? if you do it for /everything/?
21:31:15 <ais523> it's a call-by-need variant, I think
21:31:44 <ais523> it's clearly insane on a CPU, but we've decided that it's actually pretty efficient on an FPGA
21:31:58 <ais523> because all that thread creation costs nothing at all apart from one bit of memory
21:33:04 <elliott> hmm, perhaps it's APC caching somehow
21:34:31 <elliott> # Folks get annoyed when VfD discussions end up the number 1 google hit for
21:34:31 <elliott> # their name. See bugzilla bug #4776
21:34:36 <elliott> -- http://en.wikipedia.org/robots.txt
21:35:44 <ais523> elliott: those pages now have meta noindex on them too, but putting them in robots.txt helps reduce server load
21:35:57 -!- augur has quit (Remote host closed the connection).
21:36:00 <ais523> due to making the bots not even bother requesting, rather than merely ignoring then
21:37:13 <elliott> wow, caching the sitenotice is really confusing
21:37:22 <elliott> because a lot of links get de-linked due to it being cached when rendering one of the linked pages
21:37:38 <elliott> I suppose the solution is to just use external links instead
21:41:14 <elliott> (how do you style external links to look like normal ones in the sitenotice?)
21:41:48 <ais523> there's some style to suppress the external link arrow
21:41:58 <ais523> I can't remember if it's mediawiki-wide or wikipedia-specific, though
21:43:40 <elliott> ais523: btw: http://95.149.228.149:8181/w/api.php
21:45:21 * elliott attempts to determine sincerity level :)
21:47:06 <oerjan> <olsner> maybe you could make it so that underload is the only captcha question? <-- hey that was the only one i think i didn't get
21:47:10 <ais523> elliott: excitement, combined with a lack of any actual /use/ for it
21:47:25 <elliott> oerjan: wasn't that olsner's point
21:47:39 <ais523> it was an "ooh, shiny things" reactoin
21:47:45 <oerjan> elliott: no, his point was it was easy to remember
21:48:03 <elliott> oerjan: well you only have to fill it out the captcha once unless you regularly edit anonymously
21:48:14 <oerjan> i'm not entirely sure i'm right about the wire-crossing one
21:48:26 <elliott> oerjan: that's why they all have links
21:48:36 <elliott> oerjan: I specifically picked captchas answerable from the first paragraph in the linked esowiki pages
21:48:41 <elliott> they just don't go to the esowiki pages yet
21:48:55 <elliott> in this case, [[Wire-crossing problem]]: "The wire-crossing problem, in its general form, states that some programs in some languages cannot be represented as planar graphs (that is, graphs with edges that do not cross when rendered in two dimensions.) It was (as far as we are aware) first considered for Befunge, with respect to the necessity of the # operator."
21:50:20 <elliott> oerjan: btw do you have any complaints/requests wrt the new wiki
21:57:43 -!- augur has joined.
22:01:34 <elliott> oh wait you just got rid of it all
22:01:48 <ais523> I'm in the process of deleting it
22:01:51 <ais523> when you distracted me
22:02:03 <oerjan> elliott: it seemed sort of ...empty...
22:02:53 <oerjan> well we need some new limes, those are _way_ beyond best before date
22:03:39 <oerjan> how can i have complaints when it's not really up?
22:03:58 <elliott> ...software-related complaints, presumably...
22:04:04 <elliott> wtf, suddenly /w/ starts redirecting properly
22:04:20 -!- Taneb has quit (Quit: Leaving).
22:04:38 -!- pir^2 has joined.
22:05:51 <oerjan> you seem to think i have opinions and stuff
22:06:16 <oerjan> i mainly hope you will get all the data transfered properly
22:06:22 * elliott distinctly recalls oerjan complaining about the wiki software tons
22:06:28 <elliott> oerjan: were you not here when I set up an instance with the imported data?
22:06:48 <elliott> ais523: btw, can [[Special:Export]] do images?
22:06:57 <ais523> elliott: I don't think so
22:07:01 <ais523> just image description pages
22:07:10 <oerjan> well i seem to have adapted to the flaws i guess
22:07:26 <oerjan> i recall being bitten by edits timing out, at one time
22:08:11 <elliott> ais523: I have this vague feeling that Special:Exporting all the ordinary pages, reuploading files manually, and then copying over the user table will be easier than a "normal" migration
22:08:28 <ais523> elliott: I was wondering about that too
22:08:34 <oerjan> elliott: by "all the data" i mean _all_ the data, naturally.
22:09:01 <oerjan> not user accounts or pictures
22:09:13 <oerjan> is what i can think about
22:09:30 <elliott> well user accounts will just require Graue
22:09:34 <elliott> it should be easy to import + upgrade that table
22:09:45 <elliott> http://esolangs.org/w/index.php?title=Special%3AAllpages&from=&namespace=6
22:09:52 <elliott> there are not really all that many
22:09:57 <elliott> so it's not a problem if we have to manually migrate them
22:10:25 <oerjan> there are also some other media files aren't there
22:10:47 -!- quintopia has quit (Ping timeout: 252 seconds).
22:11:00 <oerjan> doesn't fugue have sound/midi, or something
22:11:39 <elliott> those go in the image namespace
22:11:52 <elliott> http://esolangs.org/wiki/Hello_world_program_in_esoteric_languages#Fugue
22:11:55 <elliott> it's a link to the file archive
22:12:23 <oerjan> you already mentioned the forum
22:15:11 -!- itidus20 has quit (Ping timeout: 260 seconds).
22:16:34 <elliott> anyway, I'll probably contact Graue tonight, tomorrow, or the day after
22:16:51 <elliott> depending on how quickly I can do final tweaks and figure out what my server situation is
22:16:59 <elliott> (I'd like a configured MW to be able to link him to)
22:17:05 <elliott> (that isn't running on this laptop)
22:17:57 <elliott> ais523: by the way, if Graue decides he's too busy to give me a database dump and set up a redirect, I'm going to need help finding out where he lives
22:18:26 * elliott wouldn't think that plausible if Graue didn't decline to do the simplest of things to solve the current spam problem on the same grounds...
22:27:05 <kallisti> ugh, no one told me when I became interested in programming that my work would involve such exciting things as writing contracts.
22:29:13 <elliott> ais523: any further suggestions for the wiki, btw? (/me is pretty much procrastinating the final few things...)
22:30:04 <oerjan> <itidus20> i think ais has really hit on something... sacrificing cpu time to decode the captcha image to slow down the spambot
22:30:16 <oerjan> now just combine that with foldathome or something
22:30:44 <oerjan> or wait, that's human isn't it.
22:30:57 <oerjan> or was that another one
22:32:31 <fizzie> Folding@home was automatical.
22:32:52 <fizzie> There was some other where you twiggled molecules all game-like, though.
22:33:34 <fizzie> I suppose s/was/is/ though I haven't heard anything about it lately.
22:35:25 <elliott> ais523: how easy would it be for you to prepare a [[Special:Export]] of every mainspace/talkspace/userspace/usertalkspace/projectspace/projecttalkspace article on Esolang with full histories?
22:35:49 <ais523> elliott: does Special:Export work for you?
22:36:05 <ais523> it should just be a case of getting a page list from somewhere (probably special:allpages), then plugging it into special:export
22:36:14 <ais523> and hoping that it doesn't refuse to handle such a large query
22:37:27 -!- augur has quit (Remote host closed the connection).
22:37:28 <elliott> ais523: it does, I was hoping I could be lazy :P
22:39:39 <oklopol> i don't think there's ever been this much difference in my state of mind in a single day.
22:40:07 <oklopol> in fact i think i reached new maxima on both sides for this year.
22:40:26 <elliott> ais523: ugh, copying from the table doesn't produce one-per-line in my browser
22:40:57 <ais523> elliott: CSS the table into a list?
22:42:01 <elliott> ais523: btw, http://esolangs.org/wiki/1st_year_sobriety_and_no_dating is a broken redirect
22:42:28 <ais523> shall I delete it, or do you want to put an esolang there?
22:42:35 <ais523> and I know it works, because I did that on AfD for Wikipedia
22:43:06 <ais523> oh well, may as well delete the redirect and leave the talk page there
22:44:02 <fizzie> oklopol: Did you: make some more prooves to replace the one that got away?
22:44:42 <oklopol> i hardly did any math after work
22:44:55 <kallisti> http://static.quickmeme.com/media/social/qm.gif does this image deeply offend anyone else?
22:45:33 <fizzie> A 1x1 pixel image? How rude!
22:45:48 <kallisti> http://www.quickmeme.com/meme/178i/
22:46:08 <monqy> i'm deeply offended
22:49:27 <fizzie> I'm offended by the lack of spaces between #include and the <.
22:49:59 <kallisti> oh I didn't even look at the background
22:50:01 <calamari> what does that statement have to with linux?
22:50:27 <kallisti> I was just offended at the implication of real numbers having anything to do with computers.
22:51:57 <kallisti> I also suspect the creator of this particular caption doesn't understand floating point numbers
22:51:58 <calamari> I guess jesus is a float, since he supposedly walked on water?
22:52:15 <elliott> blah() { curl "$1" | perl -ne "print \"$2\":"'$1\n" while /<td><a[^>]+>([^<]+)/g' >>allpages; }
22:52:20 <elliott> ais523: HOW DO YOU ESCAPE AARGH
22:52:40 <kallisti> elliott: escape bash and you escape your escape problems
22:52:57 <Jafet> One does not simply escape aargh.
22:53:23 <kallisti> ais523: use single quotes in the perl code. or do you need it to interpolate?
22:55:12 <ais523> kallisti: it's substituting in bash's $2 and $1
22:55:21 <ais523> they aren't meant to be passed literally to Perl
22:55:23 <fizzie> The most standard floating-point type of Pascal is called "real", that's offensive.
22:55:25 <kallisti> oh you want it to /not/ do that.
22:55:27 <Jafet> [^>]+>([^<]+) looks like brainfuck
22:55:34 <oerjan> except, i assume, the second $1 is from perl
22:55:42 <Jafet> Why do you think it's substituting $2 and $1?
22:55:51 <kallisti> just use ' instead of " and then bash won't interpolate. if you need to use single quotes inside perl use q/.../
22:56:14 <ais523> oerjan: oh, right, it is
22:56:19 <elliott> ais523: I've rtied perl -ne 'print "'"$2"'...
22:56:26 <ais523> Jafet: the $2 refers to bash's $2, the second $1 to perl's $1
22:56:27 <elliott> but that puts literal quotes around $2 (?!)
22:58:06 <elliott> Deewiant: I'd have to escape the regexp too, I think.
22:58:12 * kallisti tends to use q and qq in perl to avoid bash escape hell.
22:58:16 <Jafet> Who would ever want to escape from their $?
22:58:33 <calamari> I love this channel.. you can even make a bash script into a an esoteric challenge
22:58:39 <ais523> elliott: shouldn't you be escaping the $2 inside the Perl script, in case it contains a "?
22:59:12 <elliott> it's just a namespace name
22:59:13 <Jafet> He should be passing $2 as a parameter, and recovering it inside perl with $2
22:59:15 <kallisti> perl quote-like operators would solve that problem as well.
22:59:16 <elliott> this is for scraping allpages
22:59:32 <elliott> Jafet: oh, good idea, but don't you mean $ARGV[whatever]?
22:59:41 <kallisti> if you use brackets. q(...) can basically contain anything.
22:59:42 <ais523> if it contained esolang names, it'd get ugly
23:00:03 <kallisti> elliott: yes that's what he means. :P
23:00:19 <Jafet> elliott: I'm just the village idiot, I don't know what is perl
23:00:44 <kallisti> (also, in case you forgot, the first element of ARGV is not the program name in perl)
23:01:19 <kallisti> easy thing to forget and mess up on.
23:02:13 <elliott> brb, hope you guys fix it by the time I come back OR ELSE
23:02:48 <kallisti> what's the problem with interpolating the whole string, using $$ to escape one of them, and using a quote-like operator if you need a nested string?
23:02:50 <calamari> or else you're going to write it in visual basic?
23:03:24 <oerjan> elliott: what do you get if you replace perl -ne by echo ?
23:06:17 <oerjan> elliott: remove the second \"
23:07:15 <kallisti> "print qq{$2:$$1\n} while /blahblahblah/g"
23:08:28 <kallisti> what's assuming you want whatever is inside $2 to be interpolated as perl and be able to do things like, say, execute arbitrary Perl code
23:08:42 <kallisti> if you don't want that then I recommend using q instead. :P
23:09:51 <Jafet> For parsing HTML, you should use Visual Basic .NET.
23:10:06 <kallisti> I am agree. use all .NET always.
23:12:05 <kallisti> elliott: but yeah try that but with q{...} instead of qq{...} (or just use ' even)
23:14:24 <oerjan> hey i already gave a fix :(
23:14:59 <kallisti> oh yes, and it might fix other things as well.
23:15:23 <kallisti> for example, with mine, bash might accidentally interpret the regex as something (I am bad at bash stuff so I don't know)
23:16:12 <oerjan> bash is named for what it makes you do to your head
23:16:41 <kallisti> I'm kind of amazed that we still actually use it...
23:16:46 <kallisti> seems it could be replaced easily.
23:16:54 <oerjan> <oerjan> elliott: remove the second \"
23:17:06 <kallisti> it could definitely use perl's quoting constructs. As this is my primary frustration with bash.
23:17:22 <kallisti> elliott: if that doesn't work I have another suggest (it's even somewhat readable!)
23:19:21 <Jafet> Modify perl to interpret all barewords as command lines
23:19:40 <oerjan> <elliott> blah() { curl "$1" | perl -ne "print \"$2\":"'$1\n" while /<td><a[^>]+>([^<]+)/g' >>allpages; }
23:19:41 <kallisti> i bet there's a source filter for that.
23:20:00 <kallisti> actually you could do it with an AUTOLOAD as well, I think.
23:20:58 <kallisti> all linux distros should switch to perl6 at some point in the future.
23:21:11 <kallisti> you could even define a set of operators that make it bash-like.
23:21:21 <elliott> oerjan: [elliott@dinky esowiki]$ blah() { curl "$1" | perl -ne "print \"$2":"'$1\n" while /<td><a[^>]+>([^<]+)/g' >>allpages; }
23:21:21 <elliott> bash: syntax error near unexpected token `<'
23:21:40 <oerjan> elliott: the \ _and_ the "
23:21:50 <Jafet> Why would you make it bash-like
23:21:53 <kallisti> elliott: "print a{$2:$$1\n} while /blahblahblah/g"
23:22:15 <kallisti> Jafet: bash-like as-in supporting all the redirect/piping/file descriptor/etc stuff
23:22:24 <Jafet> Of course, I'd only ever put pugs in /sbin.
23:22:36 <Jafet> bash redirection is amazingly confusing and limited
23:23:52 <kallisti> elliott: note that mine is totally more clear. :>
23:24:27 <kallisti> (assuming bash doesn't misinterpret the regex)
23:25:19 <kallisti> also it suffers from no code injection exploits, which I'm sure is totally important.
23:25:30 <elliott> [elliott@dinky esowiki]$ curl -F 'pages=<allpages' http://esolangs.org/wiki/Special:Export >allpages.xml
23:25:35 <elliott> this will be like a hundred megs
23:26:33 <oerjan> esolangs.org down in 9, 8, ...
23:26:45 <elliott> so if this works, the idea is to import the pages with this, import the user table separately, and upload files manually
23:26:48 <elliott> rather than importing the whole database
23:27:02 <elliott> it's cleaner and lets us avoid old MW cruft
23:27:38 <oerjan> what if any of the pages _use_ old MW cruft
23:27:45 <kallisti> elliott: how well does it handle the ";fork while fork;# namespace
23:27:47 <elliott> it's currently onto [[brainfuck]]
23:27:49 <elliott> oerjan: wtf does that mean
23:28:08 <oerjan> i dunno, i'm just making up disaster scenarios
23:28:32 <elliott> oerjan: old MW cruft = the fact that the database schema on esolangs.org is 5 years old
23:28:51 <elliott> the XML export format is stable, so we can isolate the cruft that needs to be migrated to the users table
23:29:06 <elliott> rather than mangling the SQL dump, importing it, and having MW's scripts upgrade it piecemeal
23:29:21 <elliott> oerjan: I'll still get a full DB dump from Graue if I can.
23:29:25 <elliott> It's onto [[brainfuck constants]] now :P
23:29:52 -!- augur has joined.
23:30:21 <elliott> ooh, tswett is responsible for the site subtitle I don't like
23:30:24 <oerjan> probably the largest. well, there's that nthern archive or what it was
23:30:31 <elliott> that means I can change it without fear of Graue
23:30:54 <kallisti> elliott: also the well-known @{[sub{fork while fork}->()]} namespace
23:31:08 <elliott> http://esoteric.voxelperfect.net/wiki/MediaWiki:Sitesubtitle
23:31:25 <monqy> I don't like that subtitle either
23:31:35 <monqy> and I just learned about it today, by clicking that link
23:32:12 <elliott> oerjan: btw what would be lost in the process of this would be: deleted pages, and logs like blocks/rights changes; I have a few-days-old dump of esowiki, so the only deleted pages truly lost would be spam, and if I get a full DB dump from Graue nothing will be lost
23:33:22 <elliott> oerjan: also, it'd be trivial to spider all of http://esolangs.org/w/index.php?title=Special:Log&type=&user=&page=&limit=500&offset=0
23:35:00 <oerjan> i'm reminded of that legoman inscription
23:36:33 <elliott> "A more explicit example, especially if you also want to save the darn thing, would be" --MediaWiki manual
23:38:05 <oerjan> where the heck is the subtitle linked _from_, anyway?
23:38:38 <elliott> oerjan: it's in the printed view and some skins
23:38:42 <elliott> and in monobook if you copy the whole text
23:46:54 <elliott> it's on to talk:bitbitjump
23:55:32 <elliott> ais523: how easy is it to revert a Special:Import?
23:55:51 <ais523> elliott: you have to delete the page
23:56:18 <elliott> ais523: so a Special:Import of >2000 pages...
23:56:20 -!- augur has quit (Remote host closed the connection).
23:58:13 <elliott> ais523: restoring from an SQL backup would undo it all, right?
23:58:17 <elliott> since it doesn't upload files or anything