00:05:16 <oerjan> NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
00:05:21 <oerjan> stay and hate us instead
00:05:56 <oklopol> if spain is as boring as it sounds, i'll probably come back tomorrow
00:06:17 -!- oklopol has quit.
00:06:17 <oerjan> well have a good trip then
00:07:12 -!- tzxn3 has quit (Read error: Connection reset by peer).
00:13:16 <ais523> the treadmill on wheels thing is ridiculous :)
00:13:33 <ais523> it took me a while to figure out if it'd go forwards or backwards
00:13:51 <ais523> (forwards if the wheels rub against the treadmill belt, backwards if they're connected to the treadmill's rollers)
00:16:31 * oerjan recalls there was a treadmill accident in terry pratchett's Eric, i think it may have been caused by the Luggage getting on it
00:16:50 <oerjan> i don't recall if it went backwards or forwards
00:19:07 -!- ais523 has quit (Remote host closed the connection).
00:29:01 -!- oerjan has set topic: Gsoc away | http://codu.org/logs/_esoteric/ | http://esolangs.org/ has moved servers! Fuck you!.
00:29:09 <oerjan> i hear it's all the rage
00:29:51 <oerjan> wait, what _was_ gosc again.
00:33:35 -!- augur has joined.
00:49:16 <Sgeo_> oerjan, thingy to VHDL translatior
01:02:06 -!- augur has quit (Remote host closed the connection).
01:23:50 -!- cswords__ has joined.
01:27:42 -!- cswords_ has quit (Ping timeout: 276 seconds).
01:31:14 -!- cheater_ has joined.
01:34:22 -!- cheater has quit (Ping timeout: 252 seconds).
01:50:50 -!- Frooxius has quit (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.17/2009122204]).
01:51:18 -!- Frooxius has joined.
02:07:34 -!- tswett has changed nick to WRISTWATCH_ZOMBI.
02:07:48 -!- WRISTWATCH_ZOMBI has changed nick to tswett.
02:08:44 <oerjan> the clock is ticking to the end
02:15:08 <Sgeo_> 0 minutes to midnight.
02:16:06 <oerjan> `frink light year / (38000 miles / hour)
02:16:24 <HackEgo> 5.5691050870714991331e+11 s (time)
02:16:38 <oerjan> `frink light year / (38000 miles / hour) -> years
02:17:23 <oerjan> voyager 1 isn't approaching anywhere new in a while
02:17:54 <oerjan> (http://www.aljazeera.com/indepth/opinion/2012/02/2012228912705553.html)
02:18:19 <Sgeo_> This C API has sort of an implicit global variable. If I replace that with a CL global variable, I think that's actually thread-safer, because each thread maintains its own copy of the variable?
02:18:59 <oerjan> i have no idea if common lisp does that, but it sounds weird
02:20:38 <Jafet> `frink c / (38000 miles / hour)
02:20:47 <HackEgo> 3747405725/212344 (approx. 17647.80603643145)
02:21:07 <HackEgo> 299792458 m s^-1 (velocity)
02:22:56 <Sgeo_> It might be implementation dependent
02:28:22 <oerjan> `frink 30 years * 38000 miles / hour -> light hours
02:33:53 * oerjan reads that article's point about inbreeding and thinks that an interstellar mission needs to bring a large sperm / egg bank
02:34:03 <Jafet> Who knows, there could be aliens RIGHT IN THE KUIPER BELT
02:34:49 <oerjan> Jafet: waiting for humanity to reach outside the solar system before they violate their prime directive, or something
02:35:54 <Jafet> http://www.terrybisson.com/page6/page6.html
02:40:29 <Jafet> Also, inbreeding amplifies latent mutant powers
02:42:08 <ion> I can move my arms with the power of my mind.
02:42:19 <oerjan> there _are_ some surprisingly healthy inbred villages, but they've presumably been noticed because they're so rare
02:42:49 <oerjan> that is, you need to have the luck _not_ to have any major diseases in your pool.
02:43:32 <oerjan> there was this one in italy i read about, don't remember exactly where
02:44:06 <oerjan> with long life spans despite not living unusually healthy
02:44:19 <ion> “Where are you in the family digraph?”
02:45:42 -!- Vorpal has quit (Ping timeout: 276 seconds).
02:47:40 -!- azaq23 has joined.
02:47:41 <hagb4rd> there is kind of an alien life form right here in my fridge. it might be too late they already demand the equality and voting right
02:47:50 -!- azaq23 has quit (Max SendQ exceeded).
02:48:43 <Jafet> You left something in your fridge for twenty-one years?
02:49:06 <hagb4rd> it might have been there before time even started!
02:49:21 <shachaf> Jafet: You can demand those things at any age.
02:49:51 <Jafet> Yes, but if they're not old enough, you can just tell them off and ground them.
02:50:34 <zzo38> Is there anything in your refrigerator capable to demand equality and voting right?
02:51:20 <hagb4rd> there is evidence of an uprising species
02:53:19 -!- cswords_ has joined.
02:57:24 -!- cswords__ has quit (Ping timeout: 276 seconds).
02:58:29 <hagb4rd> respecting the main directive of the federal starfleet, i did not mean to disturb their evolution till they invent their first warp drive
02:58:34 -!- Friendship has set topic: Is there anything in YOUR refrigerator capable to demand equality and voting right? | http://codu.org/logs/_esoteric/ | http://esolangs.org/ has moved servers! Fuck you!.
02:58:50 -!- nys has quit (Quit: quit).
02:59:43 -!- azaq23 has joined.
03:02:00 <hagb4rd> or at least find a solution for the rieman hypothesis
03:03:54 <oerjan> hagb4rd: what's wrong with fermat's theorem, really? HI ALIENS DO YOU HEAR ME WE SOLVED FERMAT'S LAST THEOREM
03:04:19 <oerjan> also the classification of finite simple groups.
03:05:25 <Jafet> A classification of finite simple groups won't fit in his fridge
03:06:01 <hagb4rd> to be honest, i found this nice shot of a captcha that might be difficult to solve -> http://oi32.tinypic.com/2co0ehh.jpg ..just in case you do not already know it
03:06:06 <MDude> Well that's just another reason to want to spead out of it, thne.
03:07:14 <oerjan> i have this nice solution but it's too small to fit in the answering form
03:07:34 <oerjan> i guess small it is, then.
03:08:41 <Jafet> Did you try to enter it in the password field
03:09:22 <Jafet> On that matter, password fields are so marginalizing
03:09:55 <Jafet> Some of them won't accept a SHA-1 digest because it doesn't contain punctuation marks
03:10:30 <zzo38> Then punctuate the SHA-1 digest.
03:16:40 -!- augur has joined.
03:26:01 <MDude> http://www.youtube.com/watch?v=jnVoKSf021k
03:31:41 -!- augur has quit (Remote host closed the connection).
03:40:14 <Sgeo_> I wonder what PSOX would be like in Common Lisp
03:44:00 -!- kallisti has joined.
03:44:46 <kallisti> guards, dynamically dispatch this putrid finger tree!
05:04:00 <hagb4rd> "There was Eru, the One, who in Arda is called Ilúvatar; and he made first the Ainur, the Holy Ones, that were the offspring of his thought, and they were with him before aught else was made. And he spoke to them, propounding to them themes of music; and they sang before him, and he was glad."
05:04:02 <hagb4rd> ~ * ~ http://www.youtube.com/watch?v=a5XMewJR3k0&fmt=18 ~ * ~
05:04:57 * oerjan pelts hagb4rd with lembas
05:06:25 <oerjan> `quote [a]is523>.*obscure.*obscure
05:06:51 <oerjan> `quote [a]is523>.*inside.*obscure
05:07:05 <oerjan> `log [a]is523>.*inside.*obscure
05:07:46 <oerjan> `log [a]is523>.*obscure.*obscure
05:08:38 <oerjan> `log [a]is523.*obscure
05:08:46 <HackEgo> 2011-07-22.txt:20:14:13: <ais523> probably a really obscure method involving libraries
05:08:51 <oerjan> `log [a]is523.*obscure
05:08:59 <HackEgo> 2009-03-03.txt:16:09:09: <ehird> ais523_: ## is not obscure!!
05:09:05 <oerjan> `pastelog [a]is523.*obscure
05:09:16 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.17281
05:09:24 <oerjan> i guess he must have used more obscure words
05:10:28 <oerjan> `log [a]is523.*reference.*obscure
05:10:37 <HackEgo> 2011-12-01.txt:21:31:28: <ais523> (on another note, I love the way that the standard way to indicate that you get a reference is to make a different obscure reference to the same thing)
05:11:50 <oerjan> that was a bit too much work.
05:20:28 <hagb4rd> somehow like in c.g.jungs synchronicity <- http://en.wikipedia.org/wiki/Synchronicity
05:26:43 -!- myndzi has quit (Ping timeout: 245 seconds).
05:27:07 <MDude> hagb4rd: That's a good music thing, but the mp3bounce site linked too seems to be a stupid spam type web page?
05:27:43 * oerjan knows about synchronicity but doesn't see how that applies here
05:27:59 -!- myndzi has joined.
05:28:18 <oerjan> well you're just saying it's analogous i guess
05:28:22 <hagb4rd> mdude.. the site isn't anymore sorry
05:30:07 <kallisti> where do I go to tell linux not to start a particular daemon?
05:30:35 <kallisti> I see etc/rc<n>.d directories but I'm afraid to touch them.
05:30:42 <hagb4rd> oerjan: thats what actually happend reflecting on the hard earn quote.. sorry for the eventually missing causality :p
05:30:44 <MDude> At least the filler page is more creative than jsut naother "what you need, when you need it" list.
05:32:16 <oerjan> hey it _is_ an acausal principle.
05:33:30 <hagb4rd> but we don't know for sure yet
05:41:37 <hagb4rd> mdude: if you liked this, make sure you find the time to check out the music of 'apparat'..and the cooperative work of modeselektor and apparat -> _moderat_
05:43:21 <hagb4rd> here one last track: http://www.youtube.com/watch?v=-R6SV-4bd-0&fmt=18
05:43:45 <hagb4rd> from the amazing album: shitkatapult
05:53:39 <MDude> It's alright, but didn't really hold my attention as well as the other one.
05:54:16 <MDude> I'll take a look at this stuff when I'm less falling asleep.
05:56:32 -!- MDude has changed nick to MSleep.
06:03:00 -!- augur has joined.
06:06:16 <kallisti> so... now emacs is basically the only thing I use that looks really really ugly in xmonad
06:06:19 <kallisti> everything else looks pretty nice.
06:06:33 <kallisti> libreoffice is kind of mediocre.
06:34:38 -!- kmc has joined.
06:45:15 -!- PiRSquared has quit (Quit: bye).
07:55:52 -!- Phantom_Hoover has joined.
08:07:07 <Sgeo_> Phantom_Hoover, kallisti monqy UPDATE
08:08:25 <Sgeo_> And Homestuck is on PUASE
08:09:16 <oerjan> it had to get stuck eventually
08:11:12 <Sgeo_> http://mspandrew.tumblr.com/post/18716755896/puase
08:25:23 -!- zzo38 has quit (Remote host closed the connection).
08:33:06 -!- pikhq has joined.
08:33:14 -!- pikhq_ has quit (Ping timeout: 255 seconds).
10:01:53 -!- Vorpal has joined.
10:09:42 -!- kallisti has quit (Remote host closed the connection).
10:16:38 -!- kallisti has joined.
10:16:39 -!- kallisti has quit (Changing host).
10:16:39 -!- kallisti has joined.
10:42:48 -!- hagb4rd has quit (Ping timeout: 252 seconds).
10:47:55 -!- hagb4rd has joined.
11:16:52 -!- tzxn3 has joined.
11:31:26 -!- oerjan has quit (Quit: Lost terminal).
11:31:34 -!- monqy has quit (Quit: hello).
11:39:39 -!- jix has quit (Ping timeout: 246 seconds).
11:41:31 -!- jix has joined.
11:53:48 -!- Friendship has quit (Quit: Coyote finally caught me).
11:54:30 -!- Friendship has joined.
11:57:36 -!- Friendship has quit (Excess Flood).
11:57:47 -!- Gregor has joined.
12:01:22 <Phantom_Hoover> Wow, I didn't realise that it's been three weeks since the last Prequel update.
12:07:52 <Phantom_Hoover> Still, Kazerad does say in the comments that he should move onto a more rapid update schedule.
12:08:05 <Phantom_Hoover> And no matter what, it's still better than Dresden Codak.
12:11:05 <Slereah> Dresden Codak suffers from the Webcomic Disease
12:11:20 <Slereah> Going from a quirky humorous webcomic to a dark and gritty tragic one
12:11:42 <Phantom_Hoover> Which is in turn just a symptom of Pretentious Creator Disease.
12:13:23 -!- Gregor has quit (Excess Flood).
12:13:28 -!- Gregor has joined.
12:15:42 -!- FireFly has quit (Changing host).
12:22:43 <fizzie> Perhaps it's the "lite" version of endless torment.
12:30:00 -!- Gregor has changed nick to Friendship.
12:32:17 -!- Phantom_Hoover has quit (Remote host closed the connection).
12:33:38 -!- Phantom_Hoover has joined.
12:53:35 -!- Taneb has joined.
13:10:15 -!- Taneb has quit (Quit: lunch).
14:06:12 -!- Taneb has joined.
14:06:33 <Taneb> Could "11" be considered a BCT Truth-machine?
14:06:59 <Phantom_Hoover> Taneb, elliott says sorry, also that it's like being force-fed chocolate so it doesn't count for Lent.
14:08:01 <Taneb> @tell elliott It's okay, I didn't click the link. But then DMM got me instead. Graaah!
14:08:39 <Taneb> But 11 will add a one to the data string if it starts with 1, otherwise leave it be.
14:08:48 <Phantom_Hoover> Of all the people who could be more dickish than elliott, DMM is... not the one I'd think of.
14:09:16 <Taneb> He disguised the link as an annotation to Darths and Droids.
14:09:24 <Taneb> So it could be any of the Comic Irregulars...
14:12:15 <Taneb> But yeah... my original question...
14:12:22 <Taneb> Could it be considered a Truth-machine?
14:12:30 <Phantom_Hoover> So in other words you're on your guard for URL shortener links and server-side redirections, but not a plain link to Wikipedia in an annotation written by people who constantly link to WP?
14:12:51 <Taneb> I clicked it automatically!
14:13:09 <Taneb> And didn't realize until I was half way through the page!
14:26:41 -!- azaq23 has quit (Quit: Leaving.).
14:40:03 -!- derdon has joined.
14:50:09 -!- Taneb has quit.
15:53:30 -!- elliott has joined.
15:53:41 <elliott> http://esolangs.org/w/index.php?title=Talk:Basic_Input/Output_Commander&diff=30876&oldid=30856
15:53:41 <lambdabot> elliott: You have 1 new message. '/msg lambdabot @messages' to read it.
15:53:59 <elliott> http://esolangs.org/w/index.php?title=Talk:Basic_Input/Output_Commander&diff=next&oldid=30876
16:04:08 <elliott> // TODO: add support for [ and ]
16:18:11 <elliott> "It's worth noting that no useful language can remove all undefined behaviour; Gödel's Incompleteness Theorem, and its applications to computer science, tell us that in general you can't determine whether a method will complete, or whether a program can run indefinitely without running out of resources."
16:18:46 <elliott> This is the absolute height of Gödel-Turing software engineer woo.
16:19:15 -!- elliott has set topic: It's worth noting that no useful language can remove all undefined behaviour; Gödel's Incompleteness Theorem, and its applications to computer science, tell us that in general you can't determine whether a method will complete, or whether a program can run indefinitely without running out of resources. | http://codu.org/logs/_esoteric/ | http://esolangs.org/ has moved servers!.
16:20:18 <elliott> WHY IS NOBODY OBJECTING TO THAT SENTENCE IN THE COMEMNTS
16:20:43 <elliott> 23:29:47: <ais523> bleh, why doesn't tab-complete work in mkdir for the name of the new directory
16:21:43 <elliott> 00:29:51: <oerjan> wait, what _was_ gosc again.
16:22:08 <elliott> @tell oerjan gosc is the compiler ais523 talks about a lot from idealised ALGOL (with lambdas) to VHDL
16:23:01 <elliott> 02:18:19: <Sgeo_> This C API has sort of an implicit global variable. If I replace that with a CL global variable, I think that's actually thread-safer, because each thread maintains its own copy of the variable?
16:23:05 <elliott> Sgeo_: Common Lisp does not have threads.
16:25:10 <olsner> one thread, one variable - then each thread has its own copy of the global variable :)
16:26:25 <olsner> probably the only way to write actually thread-safe code
16:28:17 <elliott> 03:03:54: <oerjan> hagb4rd: what's wrong with fermat's theorem, really? HI ALIENS DO YOU HEAR ME WE SOLVED FERMAT'S LAST THEOREM
16:28:18 <elliott> 03:04:19: <oerjan> also the classification of finite simple groups.
16:28:42 <elliott> yes, only Goldbach, Collatz, Riemann and [REDACTED] to go before the aliens consider us intelligent enough to make contact
16:29:21 <Jafet> I accomplish thread safety with loops and strings
16:29:37 <olsner> is it likely that another intelligent species would discover and solve these problems in the same order as us?
16:30:09 <elliott> olsner: who said anything about order?
16:30:38 <elliott> they're well-known among the intergalactic community as the most basic level of mathematical problems
16:30:45 <elliott> almost all species solve them within 10 years of starting out
16:30:52 <olsner> i.e., why would they care about goldbach, collatz and riemann, when they might not even know that those are problems worth solving yet
16:30:59 <elliott> 05:20:28: <hagb4rd> somehow like in c.g.jungs synchronicity <- http://en.wikipedia.org/wiki/Synchronicity
16:31:00 <elliott> http://upload.wikimedia.org/wikipedia/commons/e/e2/Sch%C3%A9ma_synchronicit%C3%A9_in_English.png
16:31:23 <Jafet> I think they'd be pretty disappointed to find that we have made nearly no progress in the n-body problem.
16:31:27 <elliott> olsner: of course they're not worth solving, it's like requiring that the species you contact can spell the alphabet
16:31:49 <elliott> you see, once we solve those another civilisation will adopt us for the next few billion years
16:31:57 <elliott> until we start actually being useful
16:32:26 <Jafet> Or that we don't actually use one language yet
16:32:48 <elliott> Jafet: We should really be on the zero language stage by now.
16:33:11 <Jafet> We probably were, up till a few thousand years ago
16:33:35 <elliott> Jafet: No, we grunted and stuff then! That's totally a language.
16:33:48 <Jafet> Telepathy sounds like it would have weird societal impli- you pig!
16:34:53 <elliott> Telepathy??? I don't think you understand, we're meant to start hating each other and stop communicating entirely.
16:34:57 <elliott> This is the Way of the Species.
16:38:09 <Friendship> Hm, who /topic'd this topic and how true/actually-makes-sense-at-all is it intended to be >_> <_<
16:39:19 <myndzi> ** set by elliott!~elliott@unaffiliated/elliott (Sun Mar 04 08:17:22 2012)
16:39:28 -!- Jafet has set topic: Soup of the Day: Word Salad. | http://codu.org/logs/_esoteric/ | http://esolangs.org/ has moved servers!.
16:39:36 <elliott> Friendship: <elliott> "It's worth noting that no useful language can remove all undefined behaviour; Gödel's Incompleteness Theorem, and its applications to computer science, tell us that in general you can't determine whether a method will complete, or whether a program can run indefinitely without running out of resources."
16:39:36 <elliott> <elliott> This is the absolute height of Gödel-Turing software engineer woo.
16:39:42 <elliott> Jafet: NO. It must stay. As a monument.
16:39:45 <elliott> I only set it like half an hour ago!
16:39:50 -!- elliott has set topic: It's worth noting that no useful language can remove all undefined behaviour; Gödel's Incompleteness Theorem, and its applications to computer science, tell us that in general you can't determine whether a method will complete, or whether a program can run indefinitely without running out of resources. | http://codu.org/logs/_esoteric/ | http://esolangs.org/ has moved servers!.
16:39:56 <elliott> It must fuel my hatred for humanity.
16:40:19 <Friendship> elliott: I like the conflation of "well-defined" and ... whatever the hell he's talking about, to which I can't even assign a word ...
16:40:37 <elliott> Friendship: Anyway, it clearly makes perfect sense; since the incompleteness theorem is true, it's equivalent to the Halting problem, which proves that you can't determine whether a method will complete, or whether a program can run indefinitely without running out of resources.
16:40:54 <elliott> Friendship: This is the same thing as undefined behaviour because [IRC message too small to contain proof].
16:40:55 <myndzi> ok let's change the subject
16:41:03 <myndzi> has anyone ever been so far as to...
16:41:26 <elliott> Friendship: This was in the context to a rebuttal of the C++ FQA, btw.
16:41:39 <elliott> Friendship: Objecting to the complaint that C++ has no runtime safety.
16:42:26 <elliott> Friendship: I assume this means that there's some diabolical Gödel Python program that inherently causes segfaults.
16:42:39 <elliott> On ANY implementation. Even if you implement it in the lambda calculus.
16:43:10 <Friendship> elliott: Also, in the best traditions of "undefined behavior", the lambda calculus can launch the nukes.
16:43:15 <elliott> Jafet: Why would you fuse your seatbelts?
16:43:42 <elliott> Friendship: Causing serious international side-effects.
16:43:51 <olsner> seatbelt fusion, it's the dog's bollocks
16:44:07 <elliott> (It's an SPJ quote, everybody clap in unison.)
16:44:15 * elliott sounds the SPJ quote alarm.
16:44:23 <olsner> everybody clap in comic sans
17:05:55 <Phantom_Hoover> <elliott> Friendship: I assume this means that there's some diabolical Gdel Python program that inherently causes segfaults.
17:06:20 <Phantom_Hoover> You'd think there was, given the way Hofstadter portrays it.
17:13:04 <olsner> well, how hard can it be to find a bug in Python?
17:22:38 -!- MSleep has changed nick to MDude.
18:00:04 -!- zzo38 has joined.
18:06:45 <fizzie> It's even shorter than the book.
18:09:05 <fizzie> The LOTR book, I mean.
18:09:56 <elliott> Phantom_Hoover: Wasn't it a webcomic?
18:10:38 <fizzie> That's possibly a subclass of "book". I mean, it has pages.
18:11:15 <fizzie> (Virtual memory systems are books too.)
18:11:41 <elliott> fizzie: So by the Liskov substitution principle, you can read virtual memory systems.
18:17:14 -!- PiRSquared has joined.
18:20:04 <elliott> "Well sorry the version that was on the link is not the same I was actually working on, I pushed the current version on github and that gets me the error displayed before."
18:20:18 <elliott> Hey Phantom_Hoover, can you help me fix a problem with my code? Here's some other code to help you.
18:33:21 -!- brisingr has joined.
18:50:12 -!- PiRSquared has changed nick to iambored.
19:00:54 -!- ais523 has joined.
19:09:20 * ais523 sees no need to continue the pattern
19:21:24 <ais523> I think I fixed the permissions problem in Web of Lies, anyway
19:21:43 <ais523> by soft-dropping permissions, and temporarily undropping them in the offending bit of code
19:22:02 <ais523> it's not /quite/ as secure, but the only bit that soft-drops is the control process, the other two still hard-drop
19:22:45 <ais523> should be secure against non-malicious accidents; it just increases the attack surface somewhat for people trying to exploit suid weboflies, or whatever, and who'd be mad enough to suid it?
19:26:40 <ais523> elliott: tempt you into what?
19:27:07 <Phantom_Hoover> I was attempting to imply that DMotR was the well-known book; I failed.
19:32:47 <ais523> what I really need to do is get X working inside it
19:33:01 <ais523> perhaps I could trick X into thinking it's root; I wonder what would go wrong then
19:34:00 <ais523> is there any way I can get X to just use the framebuffer for rendering?
19:34:04 <elliott> ais523: X tries to talk to the graphics hardware, so...
19:34:13 <elliott> which is part of X.org nowadays
19:34:15 <elliott> it's a configure-time option
19:34:25 <elliott> ais523: there's also Xephyr
19:34:28 <elliott> which runs an X server as an X window
19:34:32 <elliott> and you probably already have it
19:34:56 <ais523> hmm, those both seem reasonable
19:35:02 <elliott> ais523: wait, Xvfb doesn't seem to be KDrive
19:35:11 <elliott> ais523: search your distro for a package
19:35:19 <ais523> one step ahead of you there :)
19:36:07 <ais523> even better, I can't think of any reasonable reason for xvfb to need root
19:36:14 <ais523> that doesn't mean it won't, ofc
19:36:29 <elliott> well, it'll try to write to the framebuffer
19:36:33 <elliott> which is usually owned by root :)
19:36:57 <olsner> xvfb creates its own framebuffer, owned presumably by the user who started xvfb
19:37:11 <fizzie> Virtual framebuffer. It's all in memory.
19:37:19 <elliott> well, doesn't X have a normal framebuffer background too?
19:37:31 <olsner> virtual framebuffer, it doesn't actually exist at all
19:37:44 <elliott> ais523: you could VNC into Xvfb or something, anyway :P
19:37:56 <olsner> yeah, x11vnc into xvfb works
19:38:08 <ais523> olsner: it's what Wikipedia recommends
19:38:09 <fizzie> Xvnc would work equally well. :p
19:38:27 <ais523> it also suggests using a screenshot program inside xvfb to get at the screen
19:38:40 * elliott 's distro doesn't have a package for Xvnc
19:38:50 <fizzie> There is a regular framebuffer graphics driver top, though. No idea about it's non-rootability.
19:38:55 <elliott> but yes, Xvfb works as a user, I've used it
19:39:08 <olsner> might still be suid root
19:39:20 <olsner> (haven't checked if it is or not)
19:39:31 <fizzie> "Xvnc" is what I think is a X server in the vncserver package.
19:40:53 <ais523> ooh, xvfb has a command-line option to memory-map its framebuffer into a file
19:41:04 <ais523> that's exactly what I was doing with the fake framebuffer in weboflies anyway
19:41:07 <elliott> ais523: VNC sounds like less work :P
19:41:14 <ais523> so the two graphics methods should be swappable trivially
19:41:33 <ais523> elliott: using one command-line option to reduce it to a problem I've already solved is more work than VNC?
19:41:44 <fizzie> "vnc4server" is one name; and "tightvncserver".
19:41:48 <elliott> but I bet you haven't completely solved the problem yes :)
19:41:56 <elliott> fizzie: ah, there's a tightvnc package
19:42:03 <ais523> elliott: well, the specific problem that VNC would solve
19:42:13 <ais523> not the whole thing, obviously
19:42:19 -!- nys has joined.
19:42:26 <elliott> ais523: I mean, I bet the file has a different format, etc.
19:42:28 <elliott> ais523: anyway, if this saga doesn't end with you running KDE in it, you've failed
19:43:00 <Phantom_Hoover> elliott, dammit, I just remembered the Thing That Must Not Be Spoken Of About PA.
19:43:08 <elliott> ais523: well... it was still more of a hog than Gnome as of a short while ago
19:43:50 <fizzie> ais523: Someone claims that Xorg with the fbdev graphics driver is doable as non-root assuming it can access all the necessary dev nodes: http://lists.debian.org/debian-x/2010/09/msg00091.html
19:44:16 <ais523> fizzie: thanks; that's my fallback position if I can't get xvfb to work
19:44:44 <ais523> (the /dev nodes don't actually exist, but I'm faking them by intercepting system calls involving them, so I can choose permissions arbitrarily
19:44:53 <elliott> that sounds like less work than Xvfb to me :p
19:45:48 <ais523> elliott: well, it's how I got /dev/fb working
19:45:53 <ais523> it was quite a bit of work
19:46:01 <ais523> fun fact: quite a few games actually run without X, using the framebuffer
19:46:07 <ais523> but you typically get a bunch of graphical glitches
19:46:23 <ais523> I suspect the hard part's actually going to be emulating socketcall, IIRC X uses sockets to work
19:46:29 <elliott> ais523: right, I just meant you wouldn't have to change your code at all
19:47:02 <ais523> and hmm, right, I wouldn't have to change it, apart from loading X somehow
19:47:09 <fizzie> If you do have a framebuffer already handy, X+fbdev sounds reasonable.
19:47:40 <fizzie> I don't quite know how Xvfb usually does input; or do you have a fake mouse already?
19:47:43 <ais523> the framebuffer itself works fine; it's actually the keyboard I have the most trouble with, as I can't figure out what format it's in
19:47:47 <elliott> anyway, I don't see why weboflies couldn't just pretend to the running program that it's root
19:47:51 <ais523> I was working on a fake mouse at the time
19:48:08 <ais523> elliott: it can, that's what I was going to do if I couldn't get X working as non-root
19:48:22 <ais523> but permissions checks are all over the place in syscalls, it'd take a bunch of work to fake comprehensively
19:48:27 <ais523> and it'd also mean figuring out what process X was
19:48:34 <elliott> ais523: well, there's no problem to keep the checks around
19:48:47 <elliott> as long as the program never actually /needs/ its root powers, you just need to make get* tell it it's root
19:48:53 <elliott> so that its sanity-checks for being root pass
19:49:06 <fizzie> X can use /dev/input/eventN with the "evdev" format for both keyboard and mouse, that's probably the most... "orthogonal" format to do input in.
19:49:08 <ais523> I think the sanity checks are probably actually in startx
19:49:30 <ais523> what's the most common $DISPLAY value? :0.0?
19:49:52 <elliott> but :0.0 is the same i think
19:49:55 <ais523> $ sudo ./weboflies ls /dev/input
19:49:57 -!- MoALTz has quit (Quit: brb).
19:50:00 <ais523> + pty output: mice\x0d\x0a
19:50:52 <elliott> oh, that \r\n isn't part of the filename
19:50:58 <elliott> I assumed you had some internal counter severely fucked up
19:51:02 <ais523> $ sudo ./weboflies Xvfb :1
19:51:12 <ais523> + pty output: _XSERVTransmkdir: ERROR: euid != 0,directory /tmp/.X11-unix will not be created.\x0d\x0a
19:51:19 <ais523> I wonder why it wanted to be root?
19:51:29 <elliott> to create the directory /tmp/.X11-unix?
19:51:43 <ais523> but why would it want to be root to do that?
19:51:49 <elliott> to create the directory /tmp/.X11-unix
19:51:59 <elliott> ais523: that's the socket, it seems
19:52:18 <ais523> that explains why it'd want to be root for /that/
19:52:19 <elliott> maybe it sets options on that socket that only root can set?
19:52:30 <elliott> well, that doesn't explain it fully if you ask me :)
19:52:32 <fizzie> The eventN interface is I think the nicest of them, since it's all keyboards and mice and sticks unified, and you get the events in quasi-sensible structures instead of some silly protocols, like intellimouse-faked PS/2 or whatever.
19:52:35 <olsner> maybe disable unix-socket stuff and make it listen on TCP instead?
19:52:55 <ais523> yep, there's a lot of "WARNING: Unknown syscall 102" there
19:53:00 <elliott> evdev is the conventional thing nowadays methinks
19:53:05 <elliott> at least I believe my keyboard and mouse are through evdev
19:53:09 <ais523> and then a couple of "unexpected signal 11"
19:53:19 <elliott> olsner: I don't think ais523 has TCP either
19:53:27 <elliott> but yes, using a high TCP port would avoid the root requirement, perhaps
19:53:35 <ais523> olsner: TCP and unix-socket go through the same syscall, which I haven't implemented
19:53:59 <ais523> elliott: I know, I have a text file of syscals
19:54:04 <olsner> ais523: oh, I think you need to implement that syscall then :>
19:54:08 <ais523> together with timing properties, and why they have or haven't been implemented
19:54:24 <ais523> the problem is, it's mostly undocumented
19:54:33 <fizzie> There's nothing *inherently* root-requiring when it comes to Unix sockets, though.
19:55:00 <fizzie> ais523: You should be doing this on a BSD, they don't mux all socket stuff through a single syscall.
19:55:11 <elliott> ais523: why not just forward socketcall without processing to test it?
19:55:16 <elliott> and then make it secure when you know it works
19:55:18 <ais523> the man page says it's the syscall that implements accept, bind, connect, getpeername, getsockname, getsockopt, listen, recv, recvfrom, recvmsg, send, sendmsg, sendto, setsockopt, shutdown, socket, socketpair
19:55:32 <ais523> elliott: that's what weboflies does do on unknown syscalls
19:55:38 <ais523> but apparently it makes xvfb segfault
19:55:56 <fizzie> ais523: I've heard the Linux source code has been leaked, maybe you could sneak a peek.
19:55:57 <ais523> (actually, the segfault's probably unrelated, it's probably assuming something that isn't actually true)
19:56:18 <ais523> fizzie: I have also dived into kernel source quite a bit for figuring out this sort of thing
19:56:24 <ais523> but I can hardly do that in a few minutes while eating dinner
19:56:26 <elliott> <ais523> elliott: that's what weboflies does do on unknown syscalls
19:56:31 <elliott> ais523: err, it clearly doesn't forward 102
19:56:39 <elliott> given "WARNING: Unknown syscall 102"?
19:56:45 <ais523> it warns and then forwards
19:56:53 <elliott> maybe the warning is breaking it :)
19:56:53 <ais523> the warning means it's being forwarded without knowing the implications
19:59:03 <ais523> I doubt it, it's being printed to stdout of the host process, which isn't part of the anonymous fs at all
20:00:31 -!- Phantom_Hoover has quit (Ping timeout: 276 seconds).
20:06:44 <ais523> ais523@desert:~/weboflies$ ln -s Xvfb_screen0 /tmp/Xvfb_screen0.xwd
20:06:45 <ais523> ais523@desert:~/weboflies$ convert /tmp/Xvfb_screen0.xwd /tmp/t.png
20:06:47 <ais523> ais523@desert:~/weboflies$ eog /tmp/t.png
20:06:57 <ais523> so Xvfb is definitely working outside weboflies
20:08:16 <elliott> ais523: what does t.png look like?
20:08:36 <ais523> a black screen with an xterm in the top-left corner
20:08:50 <ais523> there were meant to be two of them, but I hadn't started a window manager
20:08:58 <ais523> so I guess they overlapped exactly
20:09:09 <elliott> http://wtfqrcodes.com/post/18730127220/my-throat-hurts-from-sighing-the-hardest-ive-ever
20:16:52 <olsner> in hindsight, they should've thought about that kind of error when designing the qr code
20:19:54 <elliott> In hindsight, they should not have designed the QR code.
20:21:59 <elliott> Because it's useless and stupid?
20:22:45 <olsner> it's a way to distribute URL:s across the "real world" that is much more convenient than typing the url yourself
20:23:27 <elliott> For values of "convenient" equal to "highly dependent on lighting conditions and position".
20:23:57 <elliott> Also, they're hideous, and much bigger than simply, e.g. buying a domain name and putting it on whatever it is you're putting it on.
20:23:58 -!- MoALTz has joined.
20:24:07 <elliott> Also, nobody actually scans them.
20:24:30 <Friendship> Also, the people who make them don't know how and do it wrong, and the people who scan them are disappointed 140% of the time.
20:24:52 <Friendship> I remember some company putting a QR code on a T-shirt, except it was white-on-black so it didn't actually work.
20:25:00 <elliott> Also this site has screenshots of WEBSITES AND EMAILS with QR codes in them, and at that point I think we just have to go back in time and stop that ever happening.
20:25:23 <elliott> "FINALLY they've invented something which lets you go from websites to OTHER WEBSITES!"
20:26:42 <olsner> "The moment we've all been waiting for!"
20:27:16 <Friendship> QR codes that least to gopher pages: Best idea?
20:28:00 <Friendship> QR codes that least to lmgtfy.com: Best idea?
20:29:14 <elliott> <Friendship> QR codes that least to gopher pages: Best idea?
20:29:18 <elliott> Friendship: Shhh! zzo38 is online!
20:29:59 <elliott> http://tomkingaerial.files.wordpress.com/2011/05/airplane-towing-qr-code.jpg
20:30:16 <elliott> I rate the probability of anyone having successfully scanned this as -1.
20:30:47 <zzo38> You can make QR codes of whatever you want; they can be URLs of any URI scheme; but they do not even have to be a URL at all.
20:31:02 <pikhq> Yeah, QR codes can be arbitrary data...
20:31:11 <pikhq> That said, they're really nothing *special*.
20:31:17 <elliott> Yes, I'm sure people use non-URL QR codes all the time.
20:31:29 <elliott> Just to make them even more bewildering.
20:31:31 <elliott> God, the same moron who wrote that Haskell-vs-Python article is on proggit again.
20:31:34 <pikhq> elliott: I've not seen it in *common* use.
20:31:53 <pikhq> e.g. "Print out QR codes for hardcopy backup"
20:32:54 <pikhq> And you're a god-damned idiot if you put a QR code on a website.
20:33:02 <zzo38> "Print out QR codes for hardcopy backup" is actually something I have thought of too; possibly even to send data by postal mail you can use this
20:33:16 <pikhq> zzo38: Slightly silly and inefficient, but yeah, works just fine.
20:33:35 <pikhq> A gig would be a nice set of shelving.
20:33:38 <elliott> http://ollydbg.de/Paperbak/
20:33:46 <elliott> Wayyy more compact than QR code, I would assume.
20:34:31 <zzo38> Possibly the program to print out HTML documents could have an option to either print out URLs linked to as text, or as QR codes, or both
20:34:46 <olsner> just remember not to use it for piracy or for government criticism, because all pages you print are watermarked by the printer
20:35:02 <Friendship> I can't find an Android gopher client :(
20:35:13 <elliott> olsner: Isn't that US-only?
20:35:21 <pikhq> The QR code backup scheme I've seen is ~24k per page. Which is much less efficient than PaperBak.
20:35:32 <pikhq> elliott: Not really.
20:35:45 <pikhq> olsner: Only on color printers.
20:35:48 <olsner> I assume it goes for all models of printers that are sold in the US, why would they spend money removing that feature from the rest of the world?
20:36:00 <olsner> right, only printers that can print fake money
20:36:19 <pikhq> Even though you're not going to pull off convincing fakes with a color printer anyways...
20:36:32 <olsner> ... but the FAKE MONEY
20:36:38 <zzo38> Friendship: It is not on the Android Market, but it does exist.
20:37:09 <elliott> <olsner> I assume it goes for all models of printers that are sold in the US, why would they spend money removing that feature from the rest of the world?
20:37:20 <elliott> the EU is rather stricter about that stuff than the US in general, I would say
20:37:34 <elliott> "EU: Secret printer watermarks may violate human rights"
20:37:42 <elliott> EU: Secret printer watermarks may violate human rights
20:37:48 <elliott> http://arstechnica.com/security/news/2008/02/eu-commissioner-printer-tracking-may-be-human-rights-violation.ars
20:38:04 <olsner> they they ever go somewhere with that though?
20:38:09 <elliott> oh, https://www.eff.org/deeplinks/2008/02/eu-printer-tracking-dots-may-violate-human-rights is the source
20:38:10 <pikhq> Sorry, I think I meant 2.4k?
20:38:20 <zzo38> I think secret printer watermarks are only on color printers as far as I know? Still, you should turn them off (it should be a physical switch)
20:39:05 <pikhq> zzo38: Given that they work by printing yellow-on-white, I don't think they could work otherwise.
20:39:58 <fizzie> Can't say I've ever really had any lighting-or-angle-or-whatever problems actually scanning a QR code; admittedly my sample set is not very large.
20:40:26 <olsner> depends on how feasible it is to sprinkle some black dots here and there, but I guess that would be too noticable
20:41:30 <zzo38> Can PaperBack codes be faxed? I am just curious.
20:42:25 <zzo38> Tracking codes on printouts might sometimes be useful; but it should be off by default and only enabled when specifically enabled by the user. If you don't want to install such a switch, make them off all the time and have no tracking codes at all, is better.
20:42:37 <elliott> I suspect fax machines have a more limited resolution than the printers it's designed for... but I may be wrong.
20:42:47 <fizzie> Presumably, if the "dot density" drop-down goes low enough.
20:42:59 <elliott> But yes, you could always just dial down the density.
20:43:06 <olsner> looks like you can select the dpi in paperback, so just set that below whatever resolution fax has
20:43:09 -!- Phantom_Hoover has joined.
20:43:09 -!- Phantom_Hoover has quit (Client Quit).
20:43:40 <fizzie> olsner: It's a drop-down box, not a combo-editbox-whatever.
20:43:46 <zzo38> (Of course, tracking dots are never useful if the user cannot decode it themself)
20:44:00 -!- Phantom_Hoover has joined.
20:45:28 <Friendship> http://qkies.de/ Best worst idea in history.
20:45:37 <fizzie> 200 dpi is mentioned in the text, which sounds slightly optimistic for fax.
20:46:24 <elliott> Friendship: OK, who says we buy a batch linking to $shocksite and give them out in public?
20:46:30 <elliott> Problem 1: Nobody will scan them.
20:46:59 <Friendship> Problem 2: Almost assuredly they taste like sandpaper and dextrose.
20:47:08 <fizzie> I scan most QR codes I encounter, situation permitting.
20:47:40 <fizzie> Normally they're just boring URLs that I don't bother following, though.
20:47:51 <fizzie> I keep hoping maybe there'll be a surprise.
20:51:22 <Phantom_Hoover> <Friendship> Problem 2: Almost assuredly they taste like sandpaper and dextrose.
20:51:45 <Phantom_Hoover> Also we've established that you are not allowed to make judgements on what things taste like.
20:53:40 <Friendship> Phantom_Hoover: Dextrose is the solid name for glucose.
20:54:20 <Phantom_Hoover> That makes ~no sense, especially since I suspect it's the d-enantiomer.
20:54:54 <Friendship> What's the unicode character people use for fake censorship?
20:55:05 -!- zzo38 has quit (Remote host closed the connection).
20:55:39 <fizzie> I've seen just the regular full block been used.
20:55:41 <fungot> ████████████████████████████████████████████████████████████ ...too much output!
20:56:05 <olsner> hmm, "qkies", that's approximately cockies in swedish
21:00:29 <elliott> Friendship: It's U+[DATA EXPUNGED].
21:10:59 <Friendship> Is there a space block of the same width >_>
21:12:46 <Sgeo_> You could always just look on the SCP wiki
21:13:27 <Friendship> Welp, that was all wasted effort since my amazing plan doesn't actually work :(
21:14:27 <olsner> oh, SCP, I love that place
21:17:01 <elliott> olsner: I referenced SCP before Sgeo_ :'(
21:17:42 <Friendship> So anyway, imagine this spam works then be amazed by my ingenuity:
21:18:22 <elliott> fizzie: You know what to do.
21:18:35 <elliott> Someone screenshot that and try and use one of the online scanners :P
21:18:36 <pikhq> Friendship: So close.
21:18:58 <elliott> fizzie: Except you'll have to invert it >_>
21:19:01 <elliott> Friendship: I don't see why that wouldn't work with correctly-sized fonts
21:19:15 <Friendship> elliott: Does it show breaks in between lines for you?
21:19:24 <Sgeo_> "Enjoy being locked in your matrix of solidity."
21:19:28 <ais523> Friendship: the spaces are two narrow for me
21:19:32 <Sgeo_> It worked when I turned my phone on its side
21:20:09 <elliott> Friendship: Well, yeah, but lots of terminals don't have those.
21:20:58 <Sgeo_> I can't seem to replicate getting it
21:21:39 <elliott> Friendship: How did you convert that?
21:22:13 <Friendship> elliott: Found an online QR-code converter, opened it in GIMP, converted it to ASCII PBM, wrote a little script to convert the ASCII PBM to Unicode.
21:22:17 <ais523> a/ha/, Xvfb wants /tmp/X11-unix to exist already
21:23:33 <ion> friendship: The http://code.google.com/p/google-authenticator/ PAM module can also display QR codes like that.
21:24:27 <Friendship> ion: But can it tell you to enjoy being locked in your matrix of solidity?
21:26:40 <fizzie> Friendship: Well done. I had to "-fg black -bg white" a terminal for it to happen.
21:27:27 <elliott> <elliott> fizzie: Except you'll have to invert it >_>
21:27:37 <elliott> QR codes should totally be inversion-resisetnt.
21:28:17 <fizzie> ▖▖▗ ▐ ▐ ▌▌▛▘▌ ▌ ▞▖ ▟▖▌ ▗ ▖▖▗ ▗▖▗ ▐ ▐ ▗ ▖▖ ▛▖▌ ▞▖▞▖▌▌ ▛▖▛▖▞▖▌▌▜▘▙▌▞▘
21:28:20 <fizzie> ▙▌▛▘▐ ▐ ▛▌▛▘▌ ▌ ▌▌ ▐ ▛▖▛▘▛ ▛▘ ▙ ▛▘▐ ▐ ▌▌▙▌ ▛▖▌ ▌▌▌▖▛▖ ▌▌▛▖▛▌█▌▐ █▌▌▌
21:28:24 <fizzie> ▀▘▝▘ ▘ ▘ ▘▘▀▘▀▘▀▘▝ ▘▘▘▝▘▘ ▝▘ ▘ ▝▘ ▘ ▘▝ ▀▘ ▀ ▀▘▝ ▝ ▘▘ ▀ ▘▘▘▘▘▘▀▘▘▘▝
21:28:51 <elliott> I had to go to the raw logs to read that.
21:28:56 <elliott> It deteriorated as the line went on with this font.
21:29:08 <fizzie> It possibly has the wrong sort of space.
21:29:11 <elliott> Is that the rf86k thing font?
21:29:59 <fizzie> Maybe em-space works better, is that what Friendship used?
21:30:52 <fizzie> ▄ ▄ ▖▖ ▌ ▗ ▟▖▟▖▗ ▖▖ ▐ ▝ ▌ ▗ ▟▖▌ ▝ ▗▖▀▖
21:30:55 <fizzie> ▄▌▌▌▚▌ ▛▖▛▘▐ ▐ ▛▘▛ ▐ ▐ ▙▘▛▘ ▐ ▛▖▐ ▘▖▝
21:30:59 <fizzie> ▀▘▘▘▄▘ ▀ ▝▘ ▘ ▘▝▘▘ ▘▝ ▘▘▝▘ ▘▘▘▝ ▀ ▝
21:31:18 <Friendship> Those were both readable to me *shrugs*
21:31:41 <fizzie> Oh well. Everything presumably works with monospace fonts.
21:31:47 <fizzie> olsner: Did The Code work for you?
21:31:57 <olsner> I have a monospace font, Friendship worked, but fizzie didn't
21:32:44 <fizzie> Friendship: Oh, the code is done with half-blocks? Mine were with quarter-blocks.
21:32:52 <olsner> could be a not-quite-monospace font I suppose, or falling back to the wrong non-monospace characters or something... but then gregor's should be broken too?
21:33:17 <Friendship> Yeah, the code uses only em spaces, half blocks and full blocks.
21:33:21 <fizzie> It seems to use different characters.
21:33:43 <fizzie> Mine uses the quarter-blocks too.
21:33:50 <ais523> !c printf("%x %x", '', '‹');
21:34:29 <ais523> !c printf("%d", 0xe2);
21:34:48 <ais523> !c printf("%d", 0213);
21:36:28 <fizzie> Given the usual short of character cell shape, the half-blocks are probably more rectangular pixels than the quarter-blocks.
21:36:55 <Friendship> Each character is two cells, stacked vertically.
21:38:04 <fizzie> All the block elements are completely rectangular in the Unicode code chart examples. But, of course, "[t]he shapes of the reference glyphs used in these code charts are not prescriptive."
21:40:19 -!- monqy has joined.
21:40:29 <elliott> <fizzie> ▄ ▄ ▖▖ ▌ ▗ ▟▖▟▖▗ ▖▖ ▐ ▝ ▌ ▗ ▟▖▌ ▝ ▗▖▀▖
21:40:29 <elliott> <fizzie> ▄▌▌▌▚▌ ▛▖▛▘▐ ▐ ▛▘▛ ▐ ▐ ▙▘▛▘ ▐ ▛▖▐ ▘▖▝
21:40:29 <elliott> <fizzie> ▀▘▘▘▄▘ ▀ ▝▘ ▘ ▘▝▘▘ ▘▝ ▘▘▝▘ ▘▘▘▝ ▀ ▝
21:40:33 <Friendship> elliott: There, I did the same for the logo
21:40:37 <fizzie> They could use one 64k plane for a set where there's all the 4x4 characters; and one more that would have the same 2x2 elements but at 16 different levels of shading. I mean, planes 3-13 are just *sitting* there.
21:41:04 <elliott> <ais523> !c printf("%x %x", '', '‹');
21:41:14 <elliott> what character is that underline?
21:41:57 <ais523> elliott: someone's got a bunch of binary data they're trying to decote
21:42:06 <ais523> fizzie: underlined for me
21:42:08 <fizzie> Yeah, it's ctrl-underscore in here.
21:42:33 <ais523> I suspect it's a zipfile that's been run through a bunch of re-encodings between UTF-8, and some 8-bit encoding that isn't latin-1
21:42:35 <ais523> but can't tell for sure
21:43:03 <ais523> the first byte is correct, but the second isn't
21:43:09 <fizzie> Décote. Sounds French.
21:43:18 <fizzie> "Decote (do francês décolletage ou décolleté) --" pt.wiki.
21:43:43 <elliott> !c int algol = 42; printf("%d\n", algol);
21:43:52 <ais523> hey, perhaps it's Windows-12452
21:44:03 <elliott> (Yes, I know, that's inverse to Algol.)
21:44:14 <olsner> it's probably not EBCDIC
21:45:13 <ais523> olsner: it was copy-pasted into a forum post here: http://pnewman.net/shelter/viewtopic.php?p=11249&f=2#p11249
21:45:19 <EgoBot> .ACTION hugs elliott.
21:45:21 <ais523> I asked the author if they have the original
21:45:25 <ais523> because phpBB is not a good way to send binary files
21:45:30 <olsner> copy-pasted? oh damnit
21:45:47 <ais523> olsner: see why I'm having problems decoding it?
21:45:53 <ais523> I think it was just gzipped
21:45:59 <ais523> before it went through the copy-paste
21:46:06 <olsner> yeah, no wonder you can't read that, it's gibberish!
21:47:23 <elliott> Friendship: Please, never remove the CTCP filtering.
21:48:08 <olsner> wtf, running less on anything called ".bin" gives you the message "Install mkisofs to view ISO images"
21:51:14 <fizzie> Since it starts "1f e2 80 b9 08 ...", and most gzip files are compressed with deflate (== 0x08), one presumes you just have to find a transformation that maps e2 80 b9 into 8b.
21:51:26 -!- Phantom__Hoover has joined.
21:51:47 <olsner> maybe latin (or something) to utf8 twice?
21:54:50 <fizzie> e2 80 b9 looks very UTF8y for the "single left-pointing angle quotation mark" ‹, and that's exactly 0x8B in CP1252.
21:54:55 -!- Phantom_Hoover has quit (Ping timeout: 276 seconds).
21:55:04 <fizzie> But there's something else going on, since iconv complained about illegal UTF-8 at some point.
21:55:30 <fizzie> $ iconv -f utf-8 -t cp1252 < bin.txt > bin2.txt
21:55:31 <fizzie> iconv: illegal input sequence at position 121
21:55:34 <fizzie> bin2.txt: gzip compressed data, ASCII, has CRC, last modified: Fri Jan 31 17:14:52 2014
21:55:54 <fizzie> Well. It's gzippy enough for 'file' after that, but position 121 is pretty soon.
21:56:51 -!- tzxn3 has quit (Quit: Leaving).
21:58:11 <elliott> http://eprint.iacr.org/2012/064
21:59:35 <ais523> fizzie: I'm guessing that phpBB interpreted it as windows-1252 and converted it to utf-8, except possibly where it was legal utf-8 already
22:00:01 <olsner> Ignoring the encoding errors when converting to cp1252, I get "mystery.cp1252: Minix filesystem, V3, 63802 zones" instead
22:00:36 <olsner> and "gzip: mystery.cp1252 is a multi-part gzip file -- not supported"
22:01:04 <fizzie> c2 8f c3 88 54 .. is the point where iconv doesn't like it any more, which is slightly weird, since C2 8F => 11000010 10001111 looks perfectly good UTF-8 for U+008F... okay, which CP1252 doesn't actually have.
22:01:30 <ais523> I guess it should be treated as a literal 8f, then?
22:02:24 <fizzie> Yes, possibly, but I'm not sure how to tell iconv that.
22:03:22 <ais523> we need a windows-1252-or-otherwise-literally encoding
22:04:13 <olsner> not sure what literally can mean when the source is unicode and the target only has 8 bits available :)
22:06:08 <ais523> oh dear, I doubt we're going to be able to get iconv expanded to handle this case, either
22:06:11 <fizzie> gzip: bin2.txt is a multi-part gzip file -- not supported
22:06:14 <ais523> it was written by Ulrich Drepper
22:06:50 <fizzie> That was with Perl's "Encode", which by default seems to replace the bad things with a "?" when encoding to cp1252.
22:07:20 <olsner> that's barely better than iconv's //ignore option :)
22:08:15 <fizzie> Well, I think this one PERLQQ mode can be hacked to put in the byte itself.
22:08:21 <ais523> As of Encode 2.12 CHECK can also be a code reference which takes the
22:08:22 <ais523> ord value of unmapped caharacter as an argument and returns a string
22:08:24 <ais523> that represents the fallback character. For instance,
22:08:31 <olsner> but with any tool that can tell you when it's broken, you just take 1/2/3 bytes of input, see if it decodes as utf-8, and then if it encodes as either 1252 or latin1
22:08:35 <fizzie> Oh, it can take a coderef directly.
22:08:38 <ais523> fizzie: try using sub {@_[0]} as your argument
22:08:41 <fizzie> I was going to s/// the things.
22:09:30 <fizzie> Let's see what gzip says.
22:10:18 <fizzie> gzip: bin2.gz is a multi-part gzip file -- not supported
22:10:23 <fizzie> That's still not good.
22:10:38 <fizzie> The last-modified date in 2014 is a bit strange too.
22:10:43 <ais523> but it seems likely to be
22:10:52 <ais523> it was originally from a web browser cache
22:10:55 <fizzie> The first three bytes look very gzippy, that's about it.
22:11:01 <ais523> and it has the right magic number
22:11:26 <ais523> oh, huh, that's indeed possible
22:11:42 <elliott> thats content-encoding: deflate
22:12:41 <ais523> anyway, is this likely to be fixed soon, or shall I go home? I need to go home
22:13:06 <fizzie> I need to go sleep, so probably not by me.
22:13:29 <fizzie> Anyway, a DEFLATE stream wouldn't (except by chance, of course) have the gzip headers at the start.
22:13:41 <fizzie> (Admittedly it's just two fixed bytes and a very likely third.)
22:16:25 -!- ais523 has quit (Remote host closed the connection).
22:16:27 <olsner> hmm, starting with the modification time it looks broken... XFL=0x23, OS=0xc9 according to my reading
22:18:42 <fizzie> Hmm. FLG = 0x03 -> FTEXT, FHCRC; then a four-byte little-endian mtime of "ec bd eb 52"; indeed, 23 C9 .. doesn't seem very likely for XFL and OS.
22:18:54 <fizzie> Or is that bigendian? I guess it is.
22:19:04 <fizzie> Well, it doesn't look all *that* likely.
22:19:43 <olsner> both xfl and os are single-byte fields, endianness shouldn't matter there?
22:20:56 <fizzie> No, I just meant the mtime.
22:21:57 <fizzie> Anyway, most .gz files I can make seem to start with "1f 8b 08 00 ...", so just the 03-as-FLG is a bit suspicious, even if the flags aren't *that* weird.
22:22:26 -!- brisingr has left.
22:23:10 -!- augur has changed nick to etaexpansion.
22:23:17 -!- etaexpansion has changed nick to augur.
22:23:24 -!- oklopol has joined.
22:23:26 <olsner> the date displayed by `file` is what I get from date -d @$[0x52ebbdec], i.e. little endian
22:25:00 <olsner> hmm, has there been some special funky encoding of null characters?
22:25:51 <fizzie> olsner: Or just omission of them. :p
22:27:39 <olsner> but if you just insert a 0 before the three I think it'll be an even weirder mtime
22:28:51 <fizzie> That much is true, and there doesn't really seem to be a reasonable XFL value (04 or 02) or an OS value (0x or FF) in the vicinity.
22:31:01 <fizzie> (But it is also true that there are no 00s in the whole thing. Maybe it would be good if the author would use some other sort of a method to show the binary file to people than just "paste it into a webforum".)
22:31:20 <fizzie> Okay, I think this concludes my involvement in this thing. ->
22:31:39 <oklopol> but what is "show", really
22:31:47 <oklopol> can we really ever "show" anything
22:32:54 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
22:33:17 <elliott> <olsner> hmm, has there been some special funky encoding of null characters?
22:33:21 <elliott> olsner: there is a "standard" alt. encoding of them
22:34:40 -!- Deewiant has quit (Ping timeout: 245 seconds).
22:36:09 <fizzie> The "standard" one is just the two-byte version of U+0000, and I'm not seeing any of those either. But I could easily imagine that some point during the copy-paste process would drop embedded raw nuls.
22:36:10 <olsner> fizzie: "I saved that straight from the cache and the [code] tags should stop any formatting. I'll upload it when I have a chance, but I'm pretty sure it's identical."
22:37:10 <fizzie> What's that whole thing even all about?
22:37:27 <fizzie> "mafiascum.net Fallout Shelter: A temporary home during Moving Day"
22:37:32 <fizzie> I don't quite know what any of that means.
22:37:38 <fizzie> Well, I've heard about "home".
22:37:43 <elliott> fizzie: mafiascum.net is the most popular forum for playing [[Mafia (game)]].
22:37:51 <elliott> aka <a whole bunch of things>
22:37:56 <olsner> from what I gather, some place where lots of people had "games" uploaded, and now the whole thing crashed
22:38:11 <olsner> and people are trying to restore their savegames from other peoples caches, or something
22:38:26 <elliott> olsner: "savegames" = forum threads.
22:39:04 <elliott> You know that game where everyone votes to eliminate the members of the [mafia|werewolves|any number of names]? No?
22:43:44 <Sgeo_> elliott, have you ever played on EpicMafia?
22:44:06 <oklopol> is that one of those Virtual Games that you love so much?
22:44:25 <oklopol> also can anything really ever be a "thing"
22:44:59 <Sgeo_> It's just Mafia more real-time. I guess like it might be played on IRC, but with some help
22:45:54 <Sgeo_> EpicMafia appears to be down :(:(:(
22:47:13 <elliott> Sgeo_: I have never played mafia on any forums dedicated to the purpose.
22:54:44 <oklopol> is there ever really any real "purpose"
22:56:06 <olsner> oklopol seems to be in a philosophical mood
22:58:16 <olsner> does Friendship really exist?
22:58:48 <Friendship> From pseudophilosophy to just plain stupid in one leap!
22:59:21 <oklopol> do these questions even really ask anything
22:59:38 <oklopol> what is the purpose of love?
22:59:54 <olsner> Has anyone really been far even as decided to use even go want to do look more like?
23:01:09 <oklopol> this hotel room is way bigger than my apartment
23:02:15 <fizzie> Sounds like you are in the LAP of LUXURY there.
23:03:14 <oklopol> pekka haavisto arrived at the airport after baggage claim was already closed
23:03:22 <oklopol> they let him in the plane because of me
23:03:33 <oklopol> (to non-finns, pekka haavisto is gay)
23:04:49 <fizzie> Isn't he pretty gay to most Finns too?
23:04:52 <oklopol> i will not go into details because this may have been more about me being retarded than me being helpful.
23:08:20 <elliott> "In" the plane? Isn't it usually "on"?
23:09:28 <fizzie> Maybe an omitted word? "They let him <do something unspeakable> in the plane."
23:09:31 <olsner> elliott: he was going to ride on the plane, oklo got him a place in the plane
23:10:40 <olsner> I dunno, but I guess that's a good place if you want some heating (it's cold in space, after all)
23:12:04 <elliott> I think olsner is severely confused about the nature of aeroplanes.
23:14:47 <Friendship> He's deathly afraid of them. When he needs to fly, he takes a sedative and has his assistant handle everything. He just wakes up at his destination.
23:14:50 <oklopol> elliott: doesn't in the plane just mean actually inside the plane without taking a stand on whether he'll actually get to fly with it
23:15:37 <oklopol> well then i guess i meant inside the plane
23:15:48 <oklopol> that you certainly can't have a problem with
23:16:21 <oklopol> (i don't know if they let him fly, i just know they let him inside the plane)
23:18:21 <elliott> Friendship: Maybe his fears would be lessened if he didn't think they went to space.
23:18:39 <oklopol> wait are you still talking about something
23:19:28 <Friendship> OK peoples votin' time: Friendship or Gregor?
23:19:53 <Friendship> Only <name> can sail the mighty Gregship.
23:19:55 <oklopol> i find myself thinking about dying a lot when flying. and then i'm like lol, i'm much more likely to die in a bus. and then i get on a bus and i'm like wow i guess i should've hidden my chinese sex slaves better in case i die.
23:20:09 <elliott> (that's intermingling gregor and friendship except omitting all the letters that only friendship has lengthwise)
23:21:21 <oklopol> and then i'm like do i really care if people find my chinese sex slaves, i'm dead so nothing matters. and then i think a few other obvious further thoughts and then i realize shit this is philosophy, and then i get back to math.
23:21:43 <elliott> Friendship: Yes, agreed, use that.
23:22:00 <Friendship> (Most of which are living, breathing humans)
23:22:10 <oklopol> also my tunnels of face are full of snot because of the snot disease i had this week
23:22:16 <elliott> Friendship: Anyway, just rotate them. Though I'm used to Friendship now. Have you considered adopting Friendship as a fully-fledged alias?
23:22:34 <elliott> Heck, just change your name to Friendship.
23:22:35 <Friendship> I'm considering it. It's pretty cocky though, even for me.
23:22:36 <oklopol> my ears don't be hearing even still.
23:22:46 <elliott> Nonono, by "fully-fldged alias" I mean "beyond IRC".
23:22:55 <elliott> IDEALLY you should get it on your employment contract.
23:23:10 <elliott> (That is the official Final Stage of Aliasation.)
23:23:14 <Friendship> Yeaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaah Idonno about that.
23:23:24 <elliott> Friendship: You're so antialiased.
23:23:41 <Friendship> I mean, since we've already established by #esoteric psuedocanon that my middle name is Friendship anyway, it is on contracts.
23:23:56 <elliott> #esoteric pseudocanon affects reality now???
23:24:47 <elliott> Yeah, I say stick with it, if only for gems like that :P
23:25:15 <olsner> ooh, that's Friendship's full name right there
23:25:17 -!- oklopol has changed nick to nods.
23:25:30 <elliott> olsner: That's what Friendship said :P
23:25:39 <olsner> just repeating it so it sticks
23:25:41 <Friendship> elliott: The best "gem" thusfar though was: <B-ZaR> "Friendship is now known as Gregor" who decided this
23:25:43 <elliott> Ooh, Tim Schafer is doing an AmA!
23:25:55 -!- nods has changed nick to nads.
23:26:16 <nads> apparently these are both registered
23:27:04 <elliott> I bet 90% of people here don't even know who Tim Schafer is. :(
23:27:26 <olsner> I suspect you are right
23:27:43 <Friendship> elliott: The name makes me think Monkey Island.
23:27:46 <elliott> `Timothy Schafer (born July 26, 1967) is an American computer game designer. He founded Double Fine Productions in January 2000, after having spent over a decade at LucasArts. Schafer is best known as the designer of critically acclaimed games Full Throttle, Grim Fandango, Psychonauts, and Brütal Legend, and co-designer of the early classics The Secret of Monkey Island, Monkey Island 2: LeChuck's Revenge and Day of the Tentacle.
23:27:54 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: Timothy: not found
23:27:56 <pikhq> Tim Schafer, game designer, responsible for Grim Fandango, Psychonauts, and some of Monkey Island?
23:27:59 <elliott> Friendship: Yes, co-designer (Ron Gilbert being the main designer).
23:28:21 <elliott> It will take three more people to prove me wrong.
23:28:51 <elliott> JIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIX
23:28:56 <olsner> I didn't recognize the name
23:29:05 <Friendship> <Nisstyre> I also never speak in this channel.
23:29:08 <elliott> ^ul (He's that Grim Fandango guy, right?)S
23:29:09 <fungot> He's that Grim Fandango guy, right?
23:29:39 <elliott> How great would it be if clog piped up just to prove me wrong, and it turns out there's an ACTUAL PERSON there who reads the channel all the time.
23:29:40 <nads> i don't know the name but i can feel his spirit
23:29:43 <elliott> They just have their client set to log to the web, that's all.
23:30:47 <elliott> ghlflzd;kgjfh'skl;fj;ol AAAAAAAAAAH
23:31:21 -!- nads has changed nick to oklopol.
23:31:30 <oklopol> hey who is this nads guy, he seems smart
23:31:51 <oklopol> too confusing to have two smart ppl on the samme chanel
23:32:05 <elliott> oklofok is way smarter than you
23:32:27 -!- Vorpal has quit (Ping timeout: 260 seconds).
23:32:40 <oklopol> yeah but i keep him tied up in my bed
23:33:14 <elliott> I would totally try and play Grim Fandango again except Wine is apparently determined to confound my efforts in playing games.
23:47:31 -!- CHeReP has joined.
23:47:54 <HackEgo> CHeReP: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page
23:48:48 <elliott> http://esolangs.org/wiki/Talk:Basic_Input/Output_Commander WHY WOULD YOU ARCHIVE THE PAGE LITERALLY DAYS AFTER IT WAS CREATED
23:50:00 -!- azaq23 has joined.
23:53:35 <oklopol> can anything ever really be "created"
23:54:26 <oklopol> haven't you read pluton's book about how he's not a planet and how things have ideal representations in the world of superphilosophy and everything's just copy paste from there
23:56:35 <oklopol> pumping air into my ear helps temporarily
23:56:58 <oklopol> but at least it's fun to do