00:00:05 -!- coppro has joined.
00:00:12 <ehird> You came in at exactly midnight.
00:00:40 <nooga> no, it's 1st o'clock
00:00:55 <AnMaster> ehird, are you seriously saying you can't wiggle your ears?
00:01:05 <ehird> Well, I can wiggle them if I smile/frown.
00:01:09 <GregorR> I can wiggle my ears back and forth, but not up and down.
00:01:19 <AnMaster> GregorR, I can't do back forth, only up/down
00:01:21 <GregorR> I can wiggle and lick my nose :P
00:01:41 <GregorR> And I can fold my tongue length-wise.
00:01:49 <AnMaster> GregorR, you mean back (ear -- ear) ?
00:02:04 * GregorR wonders what "ear -- ear" is supposed to mean.
00:02:09 <AnMaster> <GregorR> And I can fold my tongue length-wise. <-- can't do.
00:02:16 <GregorR> I've only met one other person who can :P
00:02:19 <AnMaster> <GregorR> I can wiggle and lick my nose :P <-- what? wiggle your nose?
00:02:26 <GregorR> Those are two separate talents :P
00:02:31 <GregorR> I can wiggle my nose, and I can also lick my nose.
00:02:38 <AnMaster> GregorR, as for lick nose. No. But I can reach far in the other direction.
00:02:58 <nooga> AND WHO'S THE BEST MUTANT HERE?!
00:03:52 <ehird> Oh Madagascar, you are so naive. Once I have you all Iw ill escape on a ship
00:04:05 <ehird> MARTIAL LAW EVERYWHERE
00:04:11 <ehird> BUT I'M INVISIBLE IN MADAGASCAR
00:04:32 <AnMaster> http://www.wikihow.com/Wiggle-Your-Ears
00:05:01 <ehird> VACCINE DEPLOYED!!
00:05:05 <ehird> I hate you, Pandemic II.
00:05:40 <nooga> thank God Poland is safe - no one comes to this shithole
00:06:17 <ehird> Oh well, I bought a load of symptoms and shit to try and kill the few people i have
00:07:31 <ehird> he knows, probably.
00:08:07 <nooga> i wonder what's the url
00:09:15 <AnMaster> and what was the url for the url
00:09:37 <ehird> meh, real url is http://www.crazymonkeygames.com/jkf6Tr/pandemic2.swf
00:09:40 <ehird> but, uh, you need flash.
00:10:21 <nooga> you need graphical terminal
00:10:22 <ehird> Because it's a flash game? :P
00:10:42 <nooga> hard to find one these days
00:25:29 <nooga> my virus is called ehird
00:25:50 <ehird> i hope you don't kill anyone
00:26:03 <ehird> im on the phone wit the president of madagascar
00:26:07 <ehird> he's shutting down everything
00:45:04 <AnMaster> I haven't tried it (of course, since it is flash)
00:45:30 <AnMaster> but from what you told me it mentioned it makes no sense from a biology point of view
00:45:37 <ehird> that's why it's a game
00:45:40 <ehird> and not a biology lesson
00:45:52 <AnMaster> ehird, why not realistic games.
00:46:05 <ehird> because a game about being an actual virus would be unbelievably difficult and not fun at all
00:46:17 <AnMaster> ehird, ok, but what about other ones
00:46:29 <ehird> AnMaster: if realism was fun, games wouldn't exist
00:46:37 <ehird> we'd be going out inventing our own viruses.
00:46:45 <ehird> AnMaster: did you know, supertux isn't realistic?
00:46:58 <AnMaster> ehird, yes. I'm saying there is place for both.
00:46:59 <ehird> penguins don't look like that, they don't jump on walking snowballs with eyes and those snowballs don't fall down
00:47:01 <ehird> also the world isn't 2d
00:47:12 <ehird> and blocks that make you big don't come out of things that you wack from below
00:47:40 <AnMaster> ehird, don't be daft. Why do you think you are still short.
00:47:57 <AnMaster> You thought it was all made up right?
00:47:58 <ehird> AnMaster: well in super mario bros it's a mushroom
00:48:07 <ehird> so what I get from that is, do mushrooms and possibly other drugs
00:52:23 <nooga> wonder if llvm would handle dynamic typing shit, the thing i'm trying to implement
00:55:55 -!- Judofyr has quit (hubbard.freenode.net irc.freenode.net).
00:55:55 -!- Ilari has quit (hubbard.freenode.net irc.freenode.net).
00:55:55 -!- GregorR has quit (hubbard.freenode.net irc.freenode.net).
00:56:16 -!- Judofyr has joined.
00:56:16 -!- Ilari has joined.
00:56:16 -!- GregorR has joined.
00:58:09 -!- Ilari has quit (Connection reset by peer).
00:58:20 -!- Ilari has joined.
01:04:43 <nooga> isnt code generated by llvm slower than code generated by gcc?
01:05:01 <ehird> clang is more or less gcc competitive afaik
01:08:31 <nooga> beh, clang is even faster
01:11:11 <pikhq> But GCC doesn't need to reap my soul to be installed.
01:12:11 <AnMaster> ./configure --perfix=$HOME/local/llvm
01:12:18 * pikhq has a notable lack of aptitude
01:12:34 <pikhq> AnMaster, does it work on x86_64?
01:12:53 <pikhq> If not, it's more of a bitch than I'm willing to deal with. Otherwise, I'm just misinformed.
01:12:57 <AnMaster> ......... pikhq only if you have x86_64_magic_pixie
01:13:40 <ehird> it generates fucking machine code
01:13:55 <pikhq> No it's not. There's code generation involved, making x86_64 involve actual *work*.
01:13:55 <AnMaster> ehird, yes but docs. And it supported it for years
01:14:22 <pikhq> ... Years. x86_64 is a rather recent architecture.
01:14:48 <AnMaster> pikhq, yes, not many years. But more than one certainly.
01:14:50 <pikhq> Linux has supported x86_64 for a few years, and it's supported x86_64 longer than it's been publicly available.
01:15:23 -!- Sgeo has joined.
01:15:41 <pikhq> Linux devs were handed free hardware and documentation.
01:15:56 <pikhq> So it could be supported by 2.6.0.
01:22:12 -!- olsner has quit ("Leaving").
01:27:56 <lifthrasiir> ehird: i was sleeping... sorry for too late answer. what was the question?
01:31:39 -!- Gracenotes has quit (Remote closed the connection).
01:57:16 -!- Sgeo has changed nick to Sgeo-Val-Hum-Fem.
01:57:37 -!- Sgeo-Val-Hum-Fem has changed nick to Sgeo.
02:02:54 <nooga> llvm src has been downloading for an hour
02:07:45 -!- pikhq has quit ("leaving").
02:20:51 -!- Gracenotes has joined.
02:41:37 -!- pikhq has joined.
03:03:29 <kerlo> Either explain why the latter half of this disjunction cannot be accomplished, or explain why the former half cannot be accomplished.
03:05:44 <kerlo> Heck, here's a simpler one:
03:05:54 <kerlo> There is a disproof of this statement.
03:09:27 <coppro> This statement is false.
03:11:10 <lifthrasiir> "this statement" is ambiguous, e.g. this statement is false: the pig can fly.
03:12:15 <Robdgreat> unto itself, however, "This statement is false." would appear to be self-referential.
03:18:12 <coppro> So is "There is a disproof of this statement."
03:19:17 <Sgeo> What's it like to know that your simple decision has killed hundrens, and might kill millions?
03:20:08 <kerlo> Sgeo: is this a hypothetical "you" or some "you" in particular?
03:20:48 <Sgeo> http://www.msnbc.msn.com/id/30461857/
03:21:23 <Sgeo> Hernandez wanted to keep him home from school so he wouldn't get sick, but her husband said, "We can't be afraid of what might or might not happen."
03:22:32 <GregorR> I wasn't going to shoot an AK-47 into a crowd of civilians, but we can't be afraid of what might or might not happen.
03:25:11 <GregorR> "But Jose Luis Martinez, a 34-year-old resident of the town, made the swine flu connection the minute he heard a description of the symptoms on the news: fever, coughing, joint aches, severe headache and, in some cases, vomiting and diarrhea."
03:25:18 <GregorR> These are the symptoms of ALL FORMS OF INFLUENZA
03:26:00 <pikhq> I find it quite funny that people are afraid of eating pork because of this.
03:26:53 <GregorR> When the mad cow disease craziness was about, I didn't stop eating raw cow brains *shrugs*
03:26:56 <kerlo> I don't think "we can't be afraid of what might or might not happen" actually means anything.
03:27:15 <pikhq> Yes, because the flu is spread by consuming muscle tissue. Right...
03:27:23 <GregorR> kerlo: Well, in particular because "might" and "might not" are so generic that in fact all events might or might not happen :P
03:27:24 <pikhq> GregorR: Uh... Eating raw cow brains?
03:27:43 <kerlo> It's meaningless in that it's a patent falsehood.
03:27:44 <pikhq> That doesn't seem like a very good idea regardless.
03:27:55 <GregorR> pikhq: But they're soooooooooooo delicious.
03:28:09 <GregorR> pikhq: How's PRP comin'? X-P
03:28:23 <kerlo> What's the flavor of a raw cow brain?
03:28:34 <GregorR> It's raw cow brain flavored.
03:28:46 <GregorR> If I had to describe the flavor in three words, those words would be "raw cow brain".
03:28:59 <kerlo> How salty, how sweet, how bitter, how sour, how that-fifth-taste-someone-realized-existed?
03:29:11 <Sgeo> I will infect anyone who I see on Twitter claiming you get swine flu from eating pigs
03:29:14 <pikhq> And are you serious about eating raw cow brain?
03:29:31 <pikhq> And "someone" is "most of Asia".
03:30:12 <kerlo> Which "someone" is this?
03:30:27 <pikhq> The someone that realised the existence of umami.
03:31:26 <pikhq> Whereas we were too dumb to realise that meat is (generally) neither salty, sweet, bitter, nor sour. ;p
03:32:25 <kerlo> Hey. Scent particles are generally none of those, I believe.
03:32:43 <kerlo> GregorR: "it" being the fifth taste or cow brains?
03:33:07 <GregorR> Although cow brains are also a little bit sweet, and have a bitterish after taste sort of like liver.
03:33:10 <pikhq> And don't ask me why we don't use savory, English is weird as fuck.
03:33:26 <pikhq> And it'll probably end up being "savory" in another decade, anyways.
03:33:42 <GregorR> pikhq: Sooo ... "savory" = "umami", and yet we act like its existence was discovered recently?
03:33:54 <Sgeo> I wish people would stop saying things like "SARS scare"
03:33:59 <kerlo> Science is stubborn.
03:34:10 <pikhq> ... We in the west were just too dumb and stubborn to call it a flavor.
03:34:36 <GregorR> Especially since people often make the claim that in general "women prefer sweet and men prefer SAVORY"
03:34:43 <pikhq> (never mind that we've got glutamate receptors on our tongue, it can't be a freaking flavor)
03:34:58 <kerlo> "I think I know the English language." "Are you SURE?"
03:35:17 <kerlo> "Maybe your ability to speak English comes with no knowledge whatsoever!"
03:38:16 <GregorR> Arguably "the English language" isn't defined well enough to "know" or "not know" it :P
03:41:55 <pikhq> Thi sprak Englisc?
04:00:17 <GregorR> In 200 years that could be English :)
04:03:37 * Sgeo sneezes and infects the channel with deathflu.
04:04:52 <pikhq> That was English about a thousand years ago. ;p
04:25:59 <kerlo> I want to advance yer English.
04:26:52 <kerlo> Articles are closed-class, you say? Nonsense, I say! I just invented one!
04:31:26 -!- pikhq has quit ("leaving").
04:52:27 -!- coppro has quit (Remote closed the connection).
05:34:37 -!- coppro has joined.
06:12:03 -!- WangZeDong has joined.
06:25:43 -!- Slereah has quit (Read error: 110 (Connection timed out)).
07:19:32 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:03:59 -!- coppro has quit (Read error: 60 (Operation timed out)).
08:49:41 <Deewiant> http://jaypinkerton.com/2005/02/03/superman-origin-comics/
08:57:32 -!- Slereah has joined.
09:33:27 -!- tombom has joined.
09:47:42 <AnMaster> <GregorR> When the mad cow disease craziness was about, I didn't stop eating raw cow brains *shrugs* <-- you ate that before?
10:42:32 -!- jix has joined.
11:19:01 -!- WangZeDong has quit (Read error: 110 (Connection timed out)).
11:19:17 -!- WangZeDong has joined.
11:38:14 -!- ais523 has joined.
11:55:07 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
11:56:18 -!- ais523 has joined.
12:34:57 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
12:47:57 -!- FireFly has joined.
12:49:23 -!- WangZeDong has quit (Read error: 110 (Connection timed out)).
12:49:37 -!- WangZeDong has joined.
13:43:06 -!- ais523 has joined.
14:11:59 -!- ais523 has quit (Remote closed the connection).
14:13:51 -!- oerjan has joined.
14:18:31 <oerjan> <AnMaster> <GregorR> When the mad cow disease craziness was about, I didn't stop eating raw cow brains *shrugs* <-- you ate that before?
14:18:43 <oerjan> I bet he didn't stop beating his wife, either
14:19:40 <AnMaster> oerjan, that is completely unrelated. And IWC!
14:20:04 <oerjan> __ ___ _ ___ ___ ____ _ _
14:20:04 <oerjan> \ \ / / | | |/ _ \ / _ \/ ___|| | | |
14:20:04 <oerjan> \ \ /\ / /| |_| | | | | | | \___ \| |_| |
14:20:04 <oerjan> \ V V / | _ | |_| | |_| |___) | _ |
14:20:06 <oerjan> \_/\_/ |_| |_|\___/ \___/|____/|_| |_|
14:20:17 <Deewiant> International Whaling Commission?
14:20:33 <oerjan> Deewiant: irregular webcomic
14:21:31 <oerjan> although the other IWC may be more relevant to foodstuffs.
14:22:23 <oerjan> (giant WHOOSH courtesy of figlet)
14:23:56 <oerjan> also, poor captain ponsonby
14:24:23 <AnMaster> mouse and everything froze, then screen went black for a few seconds, then back to normal.
14:24:28 <AnMaster> [1106272.328198] NVRM: Xid (0001:00): 13, 0002 beef5f01 0000009f 00000308 030000d8 00000010
14:24:40 <AnMaster> anyone got any clue what might have happened
14:25:13 <oerjan> AnMaster: your monitor has zaphod beeblebrox' glasses
14:25:28 <oerjan> for *WHOOSH* protection, obviously
14:25:49 <AnMaster> I was trying to start an opengl app so I suspect it is somehow related to graphic card of ndivia drivers rather
14:27:04 <oerjan> AnMaster: maybe it's cosmic rays? any supernovas lately?
14:27:59 <AnMaster> it seems like Xid with the value of 13 there is nvidia related memory fault/invalid instruction/similar.
14:29:11 -!- bsmntbombdood has quit (Read error: 60 (Operation timed out)).
14:29:23 * oerjan noted a science reddit recently about a new early warning facility for supernovas in the milky way, and found it amusing how some people in the comments looked forward to the next days...
14:30:59 <fizzie> There was that recent piece of news about the most distant gamma ray burst evar. Maybe your bits were flipped by a star that exploded some 13 short-scale-billion years ago.
14:31:37 <AnMaster> * oerjan noted a science reddit recently about a new early warning facility for supernovas in the milky way, and found it amusing how some people in the comments looked forward to the next days... <-- looking forward why...
14:32:16 <AnMaster> anyway it is reproducible with this program
14:32:37 <oerjan> apparently no one must have told them that supernovas happen only every 50 years on average, and the last one in our galaxy was seen in 1630 (iirc from wp)
14:32:55 <oerjan> *on average in a galaxy our size
14:33:58 <fizzie> For some reason my brain miscombinated ideas so that it came up with "oh, AnMaster has a reproducible supernova".
14:34:21 <fizzie> ./supernova --galaxy=ours --at=now+1h
14:36:10 <oerjan> fizzie: that _would_ be scary. *SPOILER* there was one in the recent baby-eating aliens story
14:37:06 <fizzie> --star-selection=random for the irresponsible folks, --star-selection=populated_worlds_only for the downright evil ones.
14:37:23 <oerjan> they used it to do what yudkowsky probably considers saving humanity
14:38:30 <oerjan> hm this channel _is_ +c...
14:39:46 <oerjan> logic porn? now that's scary
14:40:04 <nooga> ah yes, a woman was sentenced to 9 months of jail in Norway... for a rape
14:41:18 <oerjan> nooga: this was big news in poland? you perverts
14:42:08 * oerjan either hasn't read that or considered it nonmemorable
14:42:34 <nooga> nope, quite old actually
14:43:13 <nooga> but heh, norwegian jails are like 3 star hotels
14:43:48 <nooga> maybe it's quite benefoicial to rape a guy by sucking his cock while he's asleep
14:44:02 <nooga> and then... 9 months of free vacation in 3 star hotel
14:44:23 <oerjan> i don't think most norwegians consider that a vacation
14:44:33 <oerjan> you have to be very poor for that
14:44:34 <nooga> msut think about doing that
14:45:30 <nooga> yeah, everyone is poor compared with you :D
14:45:48 <oerjan> also i don't think you will have internet access even in norway, at least in your room.
14:46:22 <oerjan> (some of you here might consider that a deterrent ;D)
14:46:30 <fizzie> Can you get something to write code on, though? 9 distraction-free months do not sound too bad.
14:46:34 <nooga> i eat bark from trees and drink water from polluted stream
14:46:51 <nooga> i ride a bicycle made from garbage
14:47:23 <nooga> use computer stolen from one german guy that went where he shouldn't
14:48:11 <nooga> and and and, there are sheep everywhere, but they belong to rich poles
14:53:22 -!- ais523 has joined.
14:54:07 <oerjan> or it would have been better if selecting it actually made the text visible
14:54:51 <oerjan> is it impossible to do spoiler hiding with irc colors
14:55:06 <fizzie> On a color-filtering channel, certainly.
14:55:21 <ais523> clearly, we need spoiler characters in Unicode
14:55:27 <ais523> INVISIBLE UNLESS SELECTED LATIN A
14:55:32 <ais523> or maybe COMBINING SPOILER
14:56:06 <oerjan> oh irssi only reverses background/foreground when selecting
14:56:18 <oerjan> obviously that won't work for this :(
14:56:24 <fizzie> It is probably your terminal emulator doing it, not irssi.
14:56:54 <ais523> along the same line as right-to-left-override?
14:57:47 <lifthrasiir> that's unicode-only mechanism for tagging certain portion of text, for example language identification.
14:58:34 <oerjan> oh well it's not going to be portable anyway if there isn't a specific tag for it
14:58:36 <fizzie> Also the Unicode FAQ: "Should I be using those language tag characters? No. Use of the language tag characters is strongly discouraged."
14:58:39 <lifthrasiir> there is only one defined tag type: language tag. e.g. <LANGTAG><TAG E><TAG N><TAG -><TAG U><TAG S>text here<TAG CANCEL>
14:59:36 <lifthrasiir> but if there is <SPOILERTAG><TAG S><TAG P>... and so on, it might be useful
15:00:03 <nooga> i see all this Testing stuff
15:00:46 <oerjan> nooga: well you should have seen only the first one, if your client supported colors
15:01:08 <oerjan> and solid blocks for the others
15:01:27 <fizzie> Er, you know, the color-filtering mode is on.
15:01:34 <fizzie> It was "1,1Testing" there and so on.
15:02:00 <oerjan> remembering the flag backwards
15:02:49 <oerjan> i mean +c should mean _more_ colors, right? grmblgrmbl
15:03:48 <oerjan> and stupid irssi which doesn't notice on your own text
15:04:37 <oerjan> and me only doing this same stupid error at _least_ twice before :D
15:05:20 <oerjan> alzheimer. that must be it.
15:05:57 <oerjan> also, oklopolio, which leads to incessant babbling
15:06:18 <oerjan> wait, where _is_ the guy
15:28:33 <nooga> i thought it'd be bold
15:29:05 <nooga> since _underlined_ is underlined
15:29:23 <nooga> so, no ? allowed inside
15:29:50 <oerjan> indeed. also, we both use irssi, i don't know how portable this is
15:44:03 <Slereah> What is wrong with this expression :
15:44:11 <ais523> Slereah: it's only a single colon
15:44:20 <ais523> which isn't a valid expression in most languages
15:44:44 <Slereah> F[Floor[RandomInteger[BinomialDistribution[Nu, p]]*k/Nu]]
15:44:44 <Slereah> Set::setps: Particules[[71]] in the part assignment is not a symbol. \
15:45:01 <Slereah> Particules := Table[{0; En/k*i}, {i, 0, k}]
15:45:01 <Slereah> F[x_] := Particules[[x]][[0]]++
15:46:04 <Deewiant> Slereah: Lists indices start from 1
15:46:50 <Deewiant> [[0]] gives you the head of the expression, namely List
15:47:11 <Deewiant> Slereah: And what's that {0: En/k*i}
15:48:09 <Deewiant> Slereah: You used a semicolon. Did you want a comma?
15:48:26 <Deewiant> {0; En/k*i} is equivalent to {En/k*i}
15:48:34 <Deewiant> Since ; is like C's comma operator
15:50:32 -!- ais523 has set topic: http://dickensurl.com/1b34/It%E2%80%99s_my_old_girl_that_advises._She_has_the_head._But_I_never_own_to_it_before_her._Discipline_must_be_maintained..
15:50:47 <Deewiant> Beyond that it looks good to me, if something else is messed up it's probably because of Nu/En/k which you didn't give
15:50:59 <Slereah> Still Set::setps: Particules[[69]] in the part assignment is not a symbol. \
15:52:39 <Slereah> Apparently, Particules[[x]][[1]]++ can't work
15:52:43 <Deewiant> ais523: Is that link supposed to be non-broken?
15:52:48 <Deewiant> Slereah: Yeah, I was wondering about that myself.
15:53:04 <Deewiant> ais523: Ah, my auto-linkifier missed the period at the end.
15:53:20 <Slereah> How can you return the list incremented at x?
15:53:36 <Deewiant> You want to return Particules[[x]]?
15:54:11 <Deewiant> Particules[[x]][[1]]++; Particules[[x]] ?
15:54:14 -!- ais523 has set topic: <http://dickensurl.com/1b34/It%E2%80%99s_my_old_girl_that_advises._She_has_the_head._But_I_never_own_to_it_before_her._Discipline_must_be_maintained.>.
15:54:24 <Deewiant> ais523: It still misses it, FWIW. :-)
15:54:25 <ais523> beh, even with the angle brackets my autolinkifier gets it wrong too
15:54:34 <Slereah> What does ; do in said case?
15:55:07 <Deewiant> Slereah: Like ; or , in most other languages. Executes the first part then returns the second.
15:55:09 <oerjan> ais523: i guess this channel was never meant for fine art
15:55:32 <ais523> maybe I should vary the target URL slightly
15:55:46 <ais523> even then, though, it's likely to have a final full stop
15:57:55 <tombom> what language are you using there
15:58:03 <Deewiant> Slereah: Hmm, I think some pass-by-ref stuff has to be done for F
15:58:17 <Deewiant> Slereah: It works standalone, just not in a function.
15:59:35 <Deewiant> Slereah: Can't you just program functionally, like you're meant to? :-P
16:00:24 <Deewiant> Write that so that you don't try to modify the original
16:00:27 <Slereah> Also it doesn't actually work as standalone
16:00:45 <Deewiant> In[9]:= xs={{1,2},{3,4}}; xs[[1]][[1]]++; xs[[1]]
16:01:56 <Slereah> In[40]:= Particles; Particles[[1]][[1]]++; Particles
16:01:56 <Slereah> During evaluation of In[40]:= Part::partd: Part specification \
16:01:56 <Slereah> Particles[[1]] is longer than depth of object. >>
16:01:56 <Slereah> During evaluation of In[40]:= Set::setps: Particles[[1]] in the part \
16:01:56 <Slereah> assignment is not a symbol. >>
16:02:22 <Deewiant> Slereah: Particules, not Particles?
16:02:34 <nooga> what's the language?
16:02:38 <Slereah> Particules gives Set::setps: Particules[[1]] in the part assignment is not a symbol. >>
16:03:16 <Deewiant> Shrug, it works for me at toplevel scope.
16:03:36 <nooga> mean lean blatmachine by stephen wolfram ?
16:04:48 <Deewiant> Slereah: I don't know. Just code it non-imperatively :-P
16:05:18 <oerjan> nooga: it's good for your hair!
16:06:14 <oerjan> it's imperative to be objective about your functional programming
16:06:14 <nooga> let me take another one
16:07:12 <nooga> damn, the keyboard is weird
16:08:35 <ehird> 02:19 Sgeo: What's it like to know that your simple decision has killed hundrens, and might kill millions?
16:08:35 <ehird> I don't exactly fault the husband here
16:08:38 <ehird> 00:27 lifthrasiir: ehird: i was sleeping... sorry for too late answer. what was the question?
16:08:44 <ehird> lifthrasiir: how do you handle exact bounds / wrapping?
16:08:55 <AnMaster> ehird, question: where does OS X store it's font files.
16:09:05 <lifthrasiir> ehird: counting non-whitespace cells in each lines and columns, and so on.
16:09:11 <nooga> probably in OSX hell
16:09:14 <lifthrasiir> maybe same with AnMaster's algorithm, iirc.
16:09:18 <ehird> AnMaster: /Library/Fonts, ~/Library/Fonts
16:09:34 <ehird> AnMaster: to convert a .dfont use fondu
16:09:38 <ehird> lifthrasiir: what license is pyfunge again? :)
16:10:07 <nooga> how many funges are there?
16:10:35 <ehird> nooga: implementations
16:10:37 <AnMaster> ehird, but there are some in /System/Library/Fonts too
16:10:44 <ehird> AnMaster: that's as may be.
16:10:54 <ehird> lifthrasiir: yay, I can wantonly steal code from it! ;-)
16:10:58 <ehird> Or, at least, could if I knew where it was.
16:11:00 <Slereah> F[x_] := MapAt[s, Particules, x]
16:11:08 <nooga> maybe an iFunge for iPhone?
16:11:10 <Slereah> Can't go wrong with the successor function*
16:11:11 <AnMaster> ehird, ah it seems to be core GUI fonts, for menus and such
16:11:11 <ais523> AnMaster: you're using OSX?
16:11:20 <ehird> AnMaster: seems likely
16:11:34 <ehird> ais523: don't think about it, i think your/my brain might overheat
16:11:39 <AnMaster> ais523, Well I'm trying to solve some bugs with an OS X system. And I was going to copy some nice fonts while I was at it.
16:11:47 <AnMaster> but the issue in question seems LDAP related.
16:11:50 <ais523> ehird: this in combination with your comments about Windows 7 is really quite worrying
16:11:59 <nooga> my brain overheats all the time i'm using mac
16:12:17 <ais523> nooga: yep, Macs use psychic cooling
16:12:19 <ehird> guys I'm going to work for Microsoft, AnMaster's working on OS X 10.7's hardware-lockdown code
16:12:25 <AnMaster> ehird, basically: waking up from sleep or logging in takes several minutes. System logs show lots of LDAP related errors.
16:12:31 <ehird> we're being paid millions
16:12:35 <ais523> fans would be too noisy, so instead they channel excess heat into the user's brain
16:12:35 <ehird> AnMaster: How odd.
16:12:38 <ehird> AnMaster: What did the user do?
16:12:45 <ehird> Don't say they didn't do something, they did :P
16:12:46 <ais523> it drives them slightly mad, it's how the reality distortion field works
16:12:57 <AnMaster> ehird, the account is *supposed* to sync to work place sync server.
16:13:01 <ehird> ais523: Subject: Cease and desist
16:13:07 <AnMaster> ehird, except it doesn't. Due to failing to connect.
16:13:13 <ehird> You are revealing Apple patented trade secrets.
16:13:21 <ehird> We must demand you cease immediately.
16:13:22 <AnMaster> ehird, any idea where I can look up exact meaning of error codes for apple system daemons
16:13:33 <ehird> AnMaster: paste the error?
16:13:37 <ehird> This is from Console.app, right?
16:13:49 <ehird> /Applications/Utilities/Consol
16:13:51 <AnMaster> ehird, the error is found using the system log viewer thingy.
16:14:10 <ehird> But, yeah, paste the error
16:14:13 <AnMaster> ehird, the error is in English, let me find a way to transfer it
16:14:29 <ehird> Then Cmd-V in a pastebin
16:14:38 <ais523> wow, mag-heut has a massive barrier to entry
16:14:38 <ehird> AnMaster: what version is this?
16:14:52 <nooga> AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAa
16:14:53 <ehird> 10.4's netinfo/ldap stuff drives me crazy
16:15:05 <nooga> this might be even more fucked up than 10.5
16:15:20 <ehird> Slereah: you're using a trial at work?
16:15:31 <ais523> in order to register, you have to a) register via a CAPTCHA-based web form b) simultaneously, or close to it, email the admin c) reply to the activation email with actual sentences, not just a few words d) once you've passed that human-based Turing Test, you still have to post something ontopic within 24 hours
16:15:49 <Slereah> ehird: You only get some licence per day, IIRC
16:15:56 <AnMaster> ehird, lots of repeated: http://pastebin.ca/1407412
16:16:03 <AnMaster> exact same message about 200 times
16:16:07 <ehird> ais523: wow, that board is IPB 1.1.2
16:16:07 <Slereah> But fuck that shit nigger, I'll finish it at home
16:16:10 <ehird> that thing's really ancient
16:16:15 <ehird> the oldest i've seen in the wild is 1.3
16:16:24 <ehird> 1.1.2 was when IPB was free, IIRC
16:16:33 <ehird> I had its source from the internet archive a while back
16:16:35 <AnMaster> ehird, I tried google without much luck other than recommending re-configuring the ldap setup
16:16:49 <AnMaster> so it would be useful to know what the error code meant.
16:16:50 <ais523> AnMaster: why not just change settings at random?
16:17:01 <ais523> in theory, it'll start working then if you change things for long enough
16:17:16 <ehird> AnMaster: Open terminal, "man DirectoryService | grep 14002"
16:17:21 <ais523> at least, if there's a finite number of settings, each with finitely many entries
16:17:26 <ehird> tell me what the error name is
16:17:30 <ehird> eDSOpenNodeFailed = -14002
16:17:35 <ehird> AnMaster: is the server down?
16:17:48 -!- jix has quit (Read error: 113 (No route to host)).
16:17:59 <ais523> that reminds me of Windows at the University, if the server's down it freezes for 8 minutes at login
16:18:03 <AnMaster> ehird, hrrm... I don't *think* so. but I guess it is possible it is related to the network.
16:18:03 <ais523> which isn't really ideal for lectures
16:18:06 <ehird> wow, mpx3.virginska.orebro.se uses mac os x server
16:18:13 <ehird> AnMaster: http://mpx3.virginska.orebro.se/ has a stock index.html
16:18:35 <AnMaster> well I don't think $user knows.
16:18:49 <ehird> AnMaster: err why is $user syncing to a server he doesn't know
16:19:02 <Slereah> ehird : Know what's worse?
16:19:09 <Slereah> The licence is shared by everyone
16:19:22 <Slereah> So if someone wastes it, no Mathematica for you
16:20:09 <AnMaster> ehird, because workplace set it up. And the mac support guy stopped working there so no one know how to fix anything any more. $user asked me to check out why computer was so slow at login and wake up from sleep and I traced it down to this LDAP issue.
16:20:22 <nooga> they've got balls of steel
16:20:23 <AnMaster> Reason $user asked me was she knows I'm geeky.
16:20:29 <ehird> nooga: it's just unix/apache
16:20:33 <ehird> with some gui tools
16:20:41 <ehird> apple's server hardware is pretty nice, so.
16:20:55 <ehird> AnMaster: well, this syncing isn't happening, right?
16:21:01 <ehird> and nobody's yelling from the rooftops?
16:21:03 <ehird> why not just reomve it
16:21:07 <ais523> it makes about as much sense as running a Darwin server, which makes almost as much sense as running a FreeBSD server
16:21:13 <ais523> ehird: probably that'll annoy $workplace
16:21:26 <AnMaster> ehird, actually it seems to happen just now. It said "refreshed credentials from LDAPv3 server" in the log.
16:21:30 <nooga> AnMaster: you're from orebro?
16:21:56 <ehird> ais523: if it wasn't syncing it can't annoy them any more
16:22:06 <AnMaster> nooga, Something like that yes. Or close to.
16:22:10 <ais523> could be if it would have started syncing again later
16:22:18 <AnMaster> ehird, however sleep still makes it lock up.
16:22:34 <AnMaster> even though it *did* just sync
16:22:40 <nooga> AnMaster: been there, and near
16:23:07 <ehird> oh, yay, bogons accept migration authority codes
16:23:12 <ehird> their script just doesn't recognize it on our line :D
16:24:13 <AnMaster> ehird, I already considered bad wlan (there are around 8 reachable networks here...) But connecting it by ethernet doesn't help.
16:24:25 <ehird> AnMaster: scrap all ldap config, reconfigure?
16:24:38 <ehird> if you can't copy/paste a password look for a .plist in ~/Library, it'll be in plaintext
16:24:46 <AnMaster> ehird, tried that before I first mentioned it in channel. I don't have access to server side.
16:24:58 <ehird> I mean on the client
16:24:58 <AnMaster> ehird, except it isn't set up to use either ssl or auth. :(
16:25:25 <ehird> :-(... it seems my ISP is "LLU'd"
16:25:32 <ehird> local loop unbundling, go fig
16:25:34 <ehird> Deewiant: you sure?
16:26:05 <ehird> the internet archive shall settl this raging dispute
16:26:24 <ehird> ais523: do you know what local loop unbundling is?
16:26:27 <Deewiant> Or wikipedia: http://en.wikipedia.org/wiki/Invision_Power_Board
16:26:32 <ais523> AnMaster: it synched, without SSL or any authentication?
16:26:33 <Deewiant> "The last free full version is Invision Power Board 1.3.1, which is not as widespread as 1.3 because of the short available time before 2.0 replaced it."
16:26:42 <ais523> might be fun synching it from somewhere else for fun
16:26:45 <ehird> One Year License · $69.95
16:26:45 <ehird> This will give you one year license to use Invision Power Board for one installation on all current and future versions. Selecting this option will also give you discounts on many of the extras for Invision Power Board. At the end of one year your board will continue to run but will be unsupported unless you renew service. More Information...
16:26:54 <ehird> i guess 2 was out by then
16:26:54 <AnMaster> ehird, the catalog manager thingy app in Tool Apps (%¤#%&¤ l10n) says contacts are set up to sync too. Address book thing time...
16:27:08 <ehird> AnMaster: Tool Apps = Utilities
16:27:14 <ais523> almost $70? for a board?
16:27:15 <ehird> AnMaster: what do the icons look like
16:27:18 <ehird> that'll be more helpful
16:27:20 <ehird> ais523: the software
16:27:29 <ehird> ais523: IPB is targeted at enterprise lackeys :)
16:27:32 <AnMaster> ehird, Like a map, With a compass on.
16:27:33 <Deewiant> ehird: Also http://en.wikipedia.org/wiki/Local_loop_unbundling
16:27:38 <ais523> I was wondering why it was so expensive
16:27:39 <ehird> didn't used to, but us now
16:27:44 <ais523> I know the University just use phpBB
16:27:46 <ehird> ais523: vbulletin is more expensive
16:27:55 <ais523> which is awful, but I suppose cheap and good enough
16:28:01 <nooga> AnMaster: most boring area of sweden
16:28:08 <ehird> ais523: vbulletin is $100/year
16:28:25 <ehird> $180 for indefinite, though, so I don't know why people would buy a one year license
16:28:38 <nooga> AnMaster: proove it
16:28:45 <ehird> AnMaster: directory utility
16:28:54 <ais523> yay, my email's back up
16:28:57 <ehird> AnMaster: I mean remove all syncing then put it back
16:29:35 <AnMaster> ehird, yes I turned off all that. And removed the ldap server from the auth search path. That did temporarily solve the speed issue.
16:29:50 <AnMaster> and of course made errors go away.
16:29:51 -!- Dewi has quit (Read error: 113 (No route to host)).
16:29:55 <ehird> AnMaster: Problem exists between server and server imo :-P
16:30:04 <AnMaster> ehird, yes I suspect this is on server side...
16:30:45 <AnMaster> ehird, hm different errors in the log now Let me pastebin again.
16:31:12 <ehird> AnMaster: if you can't figure it out I can connect via vnc and take a look
16:31:16 <ehird> (OS X comes with a vnc server)
16:32:50 <AnMaster> ehird, http://pastebin.ca/Ha1Dr-fT (pass: blah)
16:33:01 <AnMaster> (reason for pass is I wanted to try that feature)
16:33:09 <ehird> Yeah that totally seems like a server issue
16:33:15 <ehird> But I can try and fix it if you want :P
16:33:29 <ehird> (Reason for that is I want to try OS X's vnc :-P)
16:33:42 <AnMaster> ehird, nah. I'll tell $user what to tell the incompetent admins at $workplace instead :P
16:33:57 <ehird> os x's dev tools come with such weird things
16:34:00 <AnMaster> and I wouldn't trust you with that access.
16:34:15 <ehird> apps that look like they were ported from mac os classic that let you tweak internal cpu registers and shit
16:34:40 <AnMaster> I think it had a different name on classic
16:34:52 <AnMaster> but I do remember an app like it
16:35:34 <ehird> it also comes with a few apps that are basically complete commercial quality stuff
16:35:45 <ehird> like quartz composer (max/msp for graphics) and au lab
16:35:55 <ehird> it's a load of bric-a-brac
16:36:10 <ehird> Quoth CHUD Remover.app:
16:36:10 -!- KingOfKarlsruhe has joined.
16:36:17 <ehird> You are about to remove any and all CHUD files from your system.
16:37:45 <AnMaster> this macbook has not AltGr I just noticed. And I need to write a key that needs it on Swedish keyboards.
16:37:56 <ehird> AnMaster: Use option.
16:38:12 <ais523> "You are about to remove any and all CHUD files from your system. No, this isn't a chance for you to change your mind, this is just a prediction."
16:38:47 <AnMaster> something like oprofile wasn't it?
16:38:59 <ehird> Computer Hardware Understanding Development Tools
16:38:59 <ehird> A set of software tools, collectively Computer Hardware Understanding Development Tools (CHUD Tools) measure software performance on Mac OS X, to aid in optimizing. Also provides hardware system benchmarks
16:40:01 -!- coppro has joined.
16:42:03 <AnMaster> this macbook keyboard is horribly small to type on.
16:42:14 <ehird> well tiger, I suppose 2006
16:42:25 <ehird> yeah that's gonna be a bit cramped.
16:42:33 <AnMaster> and I don't know how to find out from inside os x
16:42:45 <ehird> AnMaster: system preferences → display → examine resolution :-P
16:43:13 <AnMaster> "display" was translated to "monitors" (in plural yes)
16:44:15 <AnMaster> first item in apple menu says "2.16 GHz Core 2 Duo" btw
16:44:28 <ehird> AnMaster: stop making me feel inferior, that's the speed of mine :D
16:45:14 <AnMaster> ooh seems user have used both firefox 3 and firefox 2 at once. I wonder what that did to the profile folder...
16:45:41 <ehird> aa! firefox on mac?
16:45:47 <ehird> my unintegration senses are tingling
16:46:35 <ehird> Chipset Model:ATY,RadeonX1600 ← I really wanna know why s/I/Y/
16:46:42 <AnMaster> ehird, yes. User said "Lotus Dommino Web Access" (used at work) had issues in safari so she switched to firefox. It also had issues in firefox3... so she had to go to firefox2 when she needed to use it.
16:47:08 <Deewiant> ehird: http://support.apple.com/kb/TA26745
16:47:34 <ehird> Deewiant: that's not very helpful, I want to know why ATI did it
16:47:43 <AnMaster> "This is because ATI chose to identify themselves as ATY in the ROM."
16:48:00 <ehird> i guess Ys are cheaper :-)
16:48:34 <ehird> Array TechnologY ... Todo?
16:48:54 <AnMaster> wow user has a thunderbird "in" mailbox that is over 500 MB after selecting to compress the mail box (which took several minutes)
16:48:59 <ehird> lol, it's ATI Technologies and ATI stands for Array Technologies Incorporated
16:49:02 <ehird> Array Technologies Incorporated Technologies
16:49:10 <Deewiant> AnMaster: Mine is over a gig, I think
16:49:22 <AnMaster> Deewiant, huh. You don't move it to other directories?
16:49:26 <ehird> AnMaster: mine's 794MB
16:49:38 <ehird> i have no directories, but I do have labels :-P
16:49:45 <AnMaster> ehird, well in thunderbird I meant.
16:49:56 <AnMaster> gmail works differently of course
16:49:57 <ehird> well, my inbox is about 2k messages
16:49:59 <ehird> agora is about 10k
16:50:09 <ehird> nomicron is just a few hundred
16:50:13 <ehird> so agora+b will be the bulk
16:50:17 <ehird> modulo attachments in the inbox
16:50:21 <Deewiant> AnMaster: I used to not, but I do now so maybe it's smaller
16:50:32 <AnMaster> ehird, I have two mails in my inbox atm, the rest is in various relevant directories.
16:50:56 <ehird> AnMaster: I assume you delete messages you're done iwth
16:51:12 <Deewiant> Inbox's subdirectories are 30M
16:51:36 <AnMaster> ehird, I delete spam and old svn/cvs/whatever commit mailing list mails. But I keep all other, organized in a directory structure.
16:52:00 <ehird> Oh, you're obsessive compulsive with hierarchy because you don't have a good mail search engine
16:52:01 -!- coppro has quit ("The only thing I know is that I know nothing").
16:52:18 <AnMaster> ehird, no, I do have a good handling system. filters for mailing lists.
16:52:36 <AnMaster> 97% of the mail I get is from mailing lists.
16:52:45 <ehird> AnMaster: so if you're in 500 irc channels, how many tens of thousands of mailing lists
16:53:07 <Deewiant> ehird: I prefer hierarchical sorting to searching
16:53:08 <AnMaster> ehird, less than 1000. Just 7 mailing lists.
16:53:17 <ehird> "Experience unmatched 3D gaming performance designed and fine tuned for ultimate entertainment. Crank up the settings and get ready to frag like you’ve never fragged before. Prepare to DOMINATE!"
16:53:20 <AnMaster> ehird, I really don't like mail
16:53:33 <ais523> ehird: that would play rather badly with a defragmenter
16:53:37 -!- Slereah has quit (Remote closed the connection).
16:55:52 <Deewiant> Any ideas for what DIRF should do when rerunning them due to TRDS?
16:56:13 <AnMaster> what about HFS+. I remember it got very fragmented pre-OS X
16:56:24 <ehird> http://www.youtube.com/watch?v=wB0JodKgZ0A&eurl=http%3A%2F%2Fgame%2Eamd%2Ecom%2Fus%2Den%2Funlock%5Fphenomiiblack%2Easpx&feature=player_embedded
16:56:35 <ehird> In which we overclock a cpu to 6.5ghz and have to run it at near absolute zero
16:56:38 <AnMaster> so bad I have a copy of "Norton Utilities" that includes a defragenter... Somewhere.
16:56:39 <ehird> Because we are batshit insane
16:57:36 <ehird> "i hope you intel fanboys are embarresed amd just beat intels ass in overclocking and in 3d mark by alot i hope to see more than 8+ghz in the future from amd but hey man amd is getting back on it`s feet and starting to beat intel once again and the day amd owner kicks the intel owners ass will come "
16:57:51 <ehird> amd are better than intel because if you run amd processors at absolute zero they can have higher ghz
16:57:58 <ehird> the stunning intelligence of youtube!
16:58:09 <ehird> Deewiant: but in the most hilarious way possible
16:58:31 <Deewiant> Personally, I usually get headaches
16:58:45 <ehird> yeah but they're funny headaches
16:58:55 <AnMaster> Deewiant, Agreed. If we are going to argue about Intel and AMD lets at least do it with proper grammar. Technical relevance is an extra bonus.
16:59:19 <ehird> AnMaster: don't YOU have a liquid nitrogen/helium cooling set up at home and a wanton attitude to expensive hardware?
16:59:28 <ehird> you're not a real enthusiast!
16:59:46 <ehird> Deewiant: wuz dirf
16:59:53 <ehird> "Hope to see the 7ghz barrier gone soon."
16:59:55 <ehird> Good god you morons
17:00:05 <Deewiant> // Directory functions extension
17:00:14 <ehird> They'll be arguing for 7ghz when we're running 5 billion core computers @ 3ghz
17:00:53 <AnMaster> ehird, hm I have one of those bluetooth apple mice here. Two buttons but none visible on top. And a ball instead of a wheel. Now I just noticed that unlike most optical mice it doesn't seem to have led to light the surface. just tested by lifting it up and holding it a few mm above the surface and looking in between.
17:01:03 <ehird> AnMaster: that's a mighty mouse
17:01:09 <ehird> also, it's a different tracker technology
17:01:12 <AnMaster> ehird, right. Didn't remember the name.
17:01:13 <ehird> the wired one uses regular optical stuff
17:01:19 <AnMaster> and what is the tracker technology.
17:01:29 <fizzie> Deewiant: Do LVM snapshots after each DIRF operation so you can time-travel better.
17:01:41 <ehird> AnMaster: laser tracking
17:01:46 <ehird> as opposed to the wired's optical trackng
17:01:51 <Deewiant> fizzie: I'm not even keeping track of what came out of stdin, so no.
17:02:01 <AnMaster> ehird, so shouldn't one see a small dot from the laser on the surface?
17:02:13 <AnMaster> or am I misunderstanding something
17:02:13 <fizzie> Deewiant: You should also do snapshots of the user's mental state. And then rewrite it when TRDSing around.
17:02:26 <ehird> look above the power switch
17:02:29 <ehird> that green light is the laser, I think
17:02:39 <Deewiant> fizzie: I don't need to, TRDS affects the user's mental state heavily all by itself.
17:02:46 <AnMaster> ehird, no it is the "power is on" light I think
17:02:55 <AnMaster> it changes blinking pattern when out of range
17:03:03 <ehird> AnMaster: and the two buttons are touch-sensitive
17:03:12 <ehird> that's why you have to lift your left finger off to right click :\
17:03:40 <fizzie> I think the "laser mice" use infrared laser diodes. At least that's what infallopedia says.
17:03:47 <ehird> the wheel is nice for moving around big images
17:03:52 <AnMaster> ehird, I don't see any surface on it that would work as touch sensitive
17:04:01 <AnMaster> but it would explain why I can't right click!
17:04:11 <ehird> it takes a few tries
17:04:24 <AnMaster> ehird, the mouse is too small too.
17:04:42 <ehird> AnMaster: gawd, how big are you? do you go "fe fi fo fum"? :-D
17:05:00 <AnMaster> ehird, anyway, that is one thing Microsoft managed. Good mice. Microsoft intellimouse, usb.
17:05:12 <ehird> i haaaaaaaaate ergonomic mice
17:05:32 <ehird> i just like a little dip on the sides and a curved surface Like God Intended.
17:05:40 <AnMaster> ehird, err what. It isn't that much ergonomic. Since it is a variant that works for both left and right hand
17:05:40 <fizzie> This logitech I-forget-the-model-already is laser-based; although it always annoys me a bit when they compare it to "optical mice", given that's it's really not very non-optical, they've just changed the light source.
17:05:42 <ehird> also low scrollwheel since I rest on it :P
17:05:48 <oerjan> fizzie: there could be some testing bias there. say, if all the < infrared diodes killed the testers
17:06:07 <ehird> AnMaster: not if you rest your finger on it
17:06:10 <ehird> as opposed to switching to it
17:06:21 <ehird> because then you get issues with your hands.
17:06:23 <ehird> due to the bending
17:06:25 <AnMaster> ehird, well 3 fingers on top of the mouse does work, but it is cramped.
17:06:57 <fizzie> (Also I don't think there's any rule saying that a laser light needs to be a tightly focused beam that would form a dot; from what I've read, the point just seems to be that you get interference effects (== better surface-tracking) whenever you have coherent light coming out.
17:06:58 <ehird> lol on that 6.5ghz they used ddr2 memory
17:08:14 <AnMaster> sadly it seems like that went out of fashion currently.
17:08:18 <fizzie> There was a rather gruesome eye safety instructions text about laser-related experiments in the physics lab course manuals. It was all "this wavelength laser won't cause eye reflexes to happen so it'll cook your eyeball, whether this one just burns your retina so fast you won't have time to blink".
17:08:22 <ehird> AnMaster: what advantages?
17:08:37 <ehird> fizzie: "whereas", I assume
17:08:48 <AnMaster> http://en.wikipedia.org/wiki/Fully_Buffered_DIMM
17:08:53 <oerjan> ehird: AnMaster is actually a mountain in Skåne
17:09:13 <ehird> AnMaster: what's the advantage over ddr3
17:09:19 <ehird> In 2007 Intel Developer Forum, it was revealed that major memory manufacturers have no plans to extend FB-DIMM to support DDR3 SDRAM. Instead, only registered DIMM for DDR3 SDRAM had been demonstrated.[15]
17:09:38 <fizzie> I had an Intellimouse 3.0 which indeed was a good mouse, at least up to the point it developed a serious double-clicking-all-the-time syndrome (after years of use).
17:10:13 <AnMaster> ehird, also you could combine it with ddr3 as far as I remember.
17:10:31 <AnMaster> when it comes to the speed bit and such
17:10:34 <comex> http://thedailywtf.com/Articles/Safe-ASCII-Love.aspx
17:11:37 <ehird> http://img.thedailywtf.com/images/200904/dh.gif
17:11:40 <ehird> Oh, the embarrasment!
17:11:51 <ehird> ais523: OBSCURE REFERENCE BUDDIES *HI5*
17:12:09 <fizzie> "The color of the optical mouse's light-emitting diodes varies with each model. -- Some models' diodes even change color, cycling through colors of the rainbow for instance." Wow, that's... flashy.
17:12:33 <lifthrasiir> Deewiant: regarding TRDS, i think that argument to chdir should be stored somewhere; mkdir and rmdir should be no-op or reflect according to its original result. (i'm not good at TRDS however)
17:13:03 <fizzie> The red light in the Intellimouse was really really bright; a lot brighter than some later visible-spectrum optical mouse.
17:13:04 <Deewiant> lifthrasiir: In theory yes, if you want to be accurate
17:14:32 <ehird> "Backlit keyboard (British) & User's Guide (French)" — Apple store configuration option
17:14:39 <ehird> All my british keyboard manuals are in french.
17:14:42 <lifthrasiir> i'm considering this strategy: tag every command with unsafe flag (and by default every command is unsafe); safe command is safe to execute of course.
17:15:24 <lifthrasiir> when re-executing unsafe commands it doesn't execute them at all, it just replays recorded Funge space changes and IP/delta changes, and so on.
17:15:37 <oerjan> ehird: well obviously the french need the instructions more, seeing as they aren't used to the keyboard
17:15:44 <comex> I was always scared to look at optical mice because I thought they used lasers :/
17:16:02 <comex> except some actually do
17:16:04 <lifthrasiir> (though as i stated above this strategy is not compliant to spec. i'm not sure.)
17:16:31 <Deewiant> lifthrasiir: If you record all the changes then that should work, yes.
17:16:51 * ehird stares into optical mice
17:16:56 <lifthrasiir> and python is enough reflective to do this. :p
17:16:58 <ehird> Just try and fuckin' burn my iris! :|
17:17:22 * oerjan watches ehird get both scarred and scared.
17:17:26 <comex> python is reflective enough to burn my retina if I use a laser mouse with it
17:18:24 <AnMaster> ehird, anyway you know what fb-ram is now right?
17:19:18 <AnMaster> ehird, Intel management messup as far as I remember when I learnt about it about a year ago.
17:19:18 <ehird> "FB-DIMMs absolutely kill memory latency"
17:19:52 <ehird> Tell that to the figures.
17:20:09 <comex> they decrease memory latency?
17:20:36 <comex> that is not killing latency :<
17:20:44 <fizzie> ehird: Incidentally, I tested that Logitech Whatever Illuminated Keyboard while picking up other things; it felt scissor-switchey and reasonably pleasant to write on (although all this stuff is so user-centric), although the layout was a bit strange. At least in the fi keyboard layout the backspace wasn't completely tiny; about twice the width of a normal key, maybe a bit smaller than what's common in keyboards around here (which seems to be about the width o
17:20:44 <fizzie> f two adjacent keys, including the gap between them).
17:20:49 <ehird> there's now more of you
17:21:08 <ehird> fizzie: hokaydokey
17:21:16 <comex> fwiw, I'm not awfully pleased with my apple keyboard
17:21:19 <ehird> fizzie: did you see that auroraarara thing brushed aluminium-looking one?
17:21:26 <AnMaster> fizzie, ffs, try to keep within one line. the effect of doing what you just did is tl;dr
17:21:29 <ehird> that looked quite nice
17:21:35 <ehird> AnMaster: fu, I read it
17:21:37 <ehird> and it was directed at me
17:21:40 <ehird> and it was helpful
17:22:12 <AnMaster> ehird, yes your way of lots of linebreaks is more annoying ;P
17:24:48 <ehird> http://www.bbc.co.uk/election97/frameset.htm ← HOLY FUCK RETRO
17:25:55 <fizzie> I don't get early warning when I'm going out. But it seems I wouldn't be very twittur-compatible.
17:26:21 <ehird> ew, the main isp traffic exchange thing in the uk welcomed the pirate bay verdict
17:26:24 <ehird> go fuck yourselves
17:26:43 <ehird> lifthrasiir: where's the pyfunge src?
17:27:12 <lifthrasiir> just do hg clone http://hg.mearie.org/pyfunge-0.5/ .
17:27:52 <lifthrasiir> and i think there is PyPI package too: http://packages.python.org/PyFunge/ maybe.
17:28:07 <lifthrasiir> well, that should be http://pypi.python.org/pypi/PyFunge/
17:28:07 -!- comex has changed nick to Quazie.
17:28:20 <ehird> Quazie: THE TRUTH IS OUT
17:28:56 <ehird> lifthrasiir: do you shrink bounds as well as grow?
17:29:22 <ehird> lifthrasiir: but you only store minimum bounds for the whole space?
17:29:28 <ehird> not per-line/col or whatever
17:30:14 <lifthrasiir> well... in fact, it doesn't shrink bounds normally, it just update the information necessary to calculate correct bounds (i.e. that per-line/col things)
17:30:30 <ehird> the code looks computationally expensive.
17:30:31 <lifthrasiir> and it shrinks bounds only if requested, e.g. by y
17:31:07 <lifthrasiir> there is always a room for improvement, but i have no time to do it atm
17:31:46 <ehird> right, I pretty much want to optimize for wrapping and exactness
17:32:01 <ehird> num-of-non-space cells can be calculated at y time
17:32:50 <lifthrasiir> it can possible if right data structure is used. i don't think current one, dict from coordinate to non-space cells, cannot.
17:33:31 -!- Gracenotes has quit (Remote closed the connection).
17:33:52 <ehird> this along with the stack overflow in Data.Map is currently blocking hypha :-(
17:34:44 <Deewiant> This is not really important :-P
17:35:45 <ehird> I can't parse mycology
17:35:49 <ehird> thats pretty fuckin important :||
17:36:10 <ehird> I want a correct fungespace :< But kay, I'll make them inexact for now
17:36:52 <ehird> Deewiant: 17:36 edwardk: ehird: but Data.Map is strict except for he value
17:37:19 <Deewiant> ehird: Like I said yesterday, I suspect it's your code's fault.
17:37:29 <ehird> I'm stripping it down
17:37:38 <Deewiant> ehird: Are you forcing the map between inserts?
17:37:42 <ehird> Deewiant: also, isn't insertWith' just Map.insert !$ or $! or w/e
17:37:44 <ehird> lifthrasiir: no no no
17:37:48 <ehird> i'm trying to make it all strict
17:38:06 <oerjan> ehird: although you could still get a stack overflow just from the map if you are only looking at it after a heap of changes
17:39:13 <AnMaster> ehird, one good thing about mac hardware though. Almost silent.
17:39:23 <Deewiant> ehird: With const, it actually probably is.
17:39:27 <AnMaster> though laptops tend to be quieter so don't know how significant it is.
17:40:43 <ehird> http://www.reddit.com/r/WTF/comments/8gbhq/actually_facebook_no_no_i_didnt/c0974oj ← wow privatefreedom is an idiot
17:40:51 <AnMaster> ehird, ah I think the issue is partly due to wlan. It is slow with ethernet now. But not nearly as slow.
17:41:00 <AnMaster> ehird, it doesn't seem to connect to the wlan before login.
17:41:27 <AnMaster> I guess it is due to needing to unlock the keychain first.
17:41:43 <ehird> cat :: [Input] -> [Output]
17:41:43 <ehird> cat xs = GetChar : (let (GetCharR c : xs') = xs in PutChar c : cat xs')
17:41:44 <ehird> ↑ stream based IO is awesome (this will scare oerjan)
17:42:36 <oerjan> it's not as type safe as monads though
17:44:20 <oerjan> also no EOF handling i assume
17:44:24 <Deewiant> ehird: My guess is you're building up a thunkpile of Map.inserts which is what overflows your stack
17:44:31 -!- KingOfKarlsruhe has quit (Remote closed the connection).
17:44:54 <ehird> Any ideas as to a solution?
17:45:36 <Deewiant> ehird: Don't build up a thunkpile like that?
17:45:45 <ehird> Deewiant: No shit sherlock
17:46:05 <AnMaster> ehird, when waking up from sleep (system logs): http://pastebin.ca/1407508
17:46:12 <Deewiant> ehird: Given that you're using crap like {-# UNPACK #-} I figured you'd know how to force a thunk
17:46:15 <AnMaster> I mean for the wireless network sure, but for lo?
17:46:17 <ehird> Deewiant: I do know
17:46:45 <AnMaster> ehird, "mDNSResponder: Repeated transitions for interface lo0 (127.0.0.1); delaying packets by 5 seconds" seems like a very odd error to me.
17:47:31 -!- Dewi has joined.
17:50:14 <ehird> "EFI is an average/cheap mobo maker at best (not to say unreliable, just average in function, and performance). They are no DFI, or ASUS, not even close."
17:52:24 <Deewiant> http://img.thedailywtf.com/images/200904/mandelbrot.xml
17:52:58 <AnMaster> ehird, where was that idiot quote from
17:53:07 <Deewiant> AnMaster: http://www.anandtech.com/mac/showdoc.aspx?i=2816&cp=2
17:53:20 <ehird> AnMaster: comments of the article dissing fb-ram
17:53:46 <AnMaster> Deewiant, http://img.thedailywtf.com/images/200904/mandelbrot.xml is just some unicode stuff with #FFF and such after?
17:53:59 <ehird> AnMaster: get a decent browser
17:54:14 <ehird> make it know what xslt is?
17:54:18 <ehird> i'm sure ff3 does xslt.
17:56:12 <AnMaster> ehird, ah it needs javascript...
17:56:18 <ehird> .....................
17:56:50 <ehird> Awesome. NoScript the suck.
17:57:07 <Deewiant> Why does that make it the suck?
17:57:14 -!- pikhq has joined.
17:57:24 <Deewiant> "+ XSLT stylesheets are regarded as active content and blocked by
17:57:24 <Deewiant> default on untrusted documents and/or from untrusted origins
17:59:33 <Deewiant> For instance that Firefox crasher a month back would be blocked by NoScript.
18:07:47 <ehird> oerjan: the hard thing about implementing that streamio is, um, implementing it :D
18:09:06 <ehird> [GetChar,PutChar hello
18:09:07 <ehird> 'h',GetChar,PutChar h'e',GetChar,PutChar e'l',GetChar,PutChar l'l',GetChar,PutChar l'o',GetChar,PutChar o'\n',GetChar,PutChar
18:09:31 <ehird> nooga: crazy scary old haskell IO system implemented in modern haskell
18:10:11 <Deewiant> ehird: Did you get your Befunge parser working?
18:10:31 <ehird> Deewiant: Are you teasing me :P
18:10:42 <Deewiant> No, I'm asking if you even tried :-P
18:10:55 <ehird> Also I code in spurts.
18:11:58 <nooga> still i think befunge can be written as a linear program with jumps
18:12:14 <Deewiant> Write a befunge-to-unefunge compiler
18:13:45 <ehird> http://imgur.com/2Dqu.png
18:15:24 <fizzie> I did it wrong, too. :/
18:15:32 <ehird> i went first → lastly → then
18:15:34 <ehird> since lastly was biggar
18:18:15 -!- BeholdMyGlory has joined.
18:18:33 <ehird> Deewiant: I've ripped out all bounds code to start again
18:19:05 <ehird> Deewiant: What's the most hugest befunge file existing?
18:19:11 <ehird> actually, any big text file would be appreciated
18:20:25 <fungot> AnMaster: ' that wouldn't be very nice, i'm fnord this is always done however fnord fnord: fnord those that have feathers, and bite, and those that have feathers, and bite, and those that have whiskers, and scratch.
18:20:27 <Deewiant> If you want big files I'm sure you can find some on your own
18:20:31 <Deewiant> AnMaster: Not hardly, I don't think
18:20:32 <fungot> http://zem.fi/~fis/fungot.b98.txt
18:20:41 <Deewiant> Not even close by the looks of it
18:20:42 <fungot> ehird: ' the question is, what is the best way out of the difficulty is to place a red counter.
18:21:02 <Deewiant> Firefox sez that is 23 937 bytes
18:21:10 <ehird> my parser shall read hamlet
18:21:10 <AnMaster> ehird, what was the key combo to boot a mac into single user shell
18:21:15 <ehird> AnMaster: dunno. google it
18:21:50 <fizzie> Largest Project Gutenberg text file is puny 32 megabytes, titled "U.S. Copyright Renewals 1950 - 1977"; although it is possible I removed the Human Genome Project books from my local mirror. (Those are split in very many pieces anyway.)
18:21:51 <oerjan> to parse, or not to parse, that is the question
18:21:52 <ehird> hamlet isn't big enough
18:21:56 <ehird> fizzie: plz to link?
18:22:00 <Deewiant> Also it looks like half of fungot is comments, so it hardly counts
18:22:01 <fungot> Deewiant: 15. no children are fnord no fnord are fnord all rabbis are jews. some people are unhealthy.
18:22:13 <Deewiant> fungot: Yeah, fnord fnord to you too.
18:22:14 <fungot> Deewiant: " with this you make a kind of folk who have no horror of a joke.
18:22:20 <fungot> Available: agora alice* darwin discworld europarl ff7 fisher ic irc jargon lovecraft nethack pa speeches ss wp
18:22:28 <pikhq> You could also go ahead and use more than one Project Gutenberg text.
18:22:47 <Deewiant> ehird: If mycology already overflows your stack what do you want something bigger for?
18:23:05 <fizzie> ehird: http://www.gutenberg.org/files/11800/11800-8.txt but it is the boring.
18:23:08 <ehird> Deewiant: No, mycology overflows my stack with bounds checknig
18:23:10 <ehird> which I've removed
18:23:18 <ehird> So I want to overflow it with just this
18:23:21 <ehird> to see what the issue is
18:23:38 <Deewiant> fizzie: You froze my firefox :-(
18:24:32 <fizzie> That's rather strange; it is, after all, just 32 megs of text.
18:25:03 <AnMaster> Deewiant, konq decided to open it in kwrite and the progress dialog had a cancel in it
18:25:10 <fizzie> "31.19 MB" in fact according to the metadata in http://www.gutenberg.org/etext/11800
18:25:35 <fizzie> 32702551 bytes; I'm not sure where 35 megs would come from. Strange rounding.
18:25:51 <ehird> *Hypha.Fungespace> foo <- readFile "/Users/ehird/Code/hypha/11800-8.txt"
18:26:10 <fizzie> For the powers-of-ten megabyte, yes.
18:26:12 <ehird> Deewiant: that's rather inaccurate for 0-5...
18:26:27 <ehird> *Hypha.Fungespace> fs `seq` ()
18:26:30 <Deewiant> ehird: But it's not for 0-5, it's for 32702551.
18:26:30 <ehird> lifthrasiir: it's loading now
18:26:35 <ehird> w/ let fs = parseFungespace foo
18:26:41 <ehird> this may take a while
18:27:15 <ehird> AnMaster: dthat file—
18:27:18 <AnMaster> and I clicked cancel quite quickly.
18:27:30 <ehird> this was a mistake
18:27:36 <ehird> AnMaster: big 35mb one
18:27:40 <fizzie> Second-longest book would've been 20 megabytes, "The Entire Project Gutenberg Works of Mark Twain". Much more interesting to read.
18:27:51 <AnMaster> ehird, http://www.gutenberg.org/files/11800/11800-8.txt you mean?
18:28:00 <AnMaster> ehird, I'm not going to try it in cfunge.
18:28:18 <AnMaster> ehird, there is no "just parsing" option in cfunge.
18:28:20 <oerjan> ehird had a little ram
18:28:23 <ehird> haskell is memory intensive
18:28:27 <ehird> AnMaster: so comment out interp(foo)
18:28:53 <AnMaster> ehird, once I got this mac to work perfectly ;P
18:29:10 <ehird> is it your mac now :-P
18:30:08 <ehird> Deewiant: is there a non-shit portable hashtable for haskell
18:30:11 <AnMaster> ehird, but still. I like to try to get it to work.
18:30:11 <ehird> or not portable w/e
18:30:14 <fizzie> cfunge residential size went to something like 2.2 gigabytes to do that file.
18:30:22 <fizzie> fis@eris:~/hut/lktm/gutenberg$ (echo '@'; cat ./1/1/8/0/11800/11800-8.txt) > tmp.txt
18:30:22 <fizzie> fis@eris:~/hut/lktm/gutenberg$ time ~/inst/cfunge/cfunge/build/cfunge tmp.txt
18:30:24 <lament> is there non-shit anything for haskell?
18:30:26 <ehird> Deewiant: cuz maps are just too biggety
18:30:35 <ehird> my middle finger is a bottom.
18:31:28 <AnMaster> ehird, I mean when $user is in the group "relatives".
18:31:31 <Deewiant> ehird: Use what AnMaster does, shouldn't be hard to write an FFI for it :-P
18:31:38 <ehird> Deewiant: :||||||||||||
18:31:50 <ehird> fizzie: so got any smaller files
18:31:52 <fizzie> That's strange; /usr/bin/time's memory-usage-reporting does not seem to much work.
18:31:55 <fizzie> fis@eris:~/hut/lktm/gutenberg$ /usr/bin/time ~/inst/cfunge/cfunge/build/cfunge tmp.txt
18:31:55 <fizzie> 24.57user 1.32system 0:26.33elapsed 98%CPU (0avgtext+0avgdata 0maxresident)k
18:31:55 <fizzie> 0inputs+0outputs (0major+630781minor)pagefaults 0swaps
18:31:56 <ehird> fizzie: like, 1mb-3mb would be nice
18:31:57 <AnMaster> fizzie, it loaded it in 26 seconds? heh
18:32:05 <Deewiant> ehird: A 30-meg file is not exactly typical funge code anyway
18:32:07 <AnMaster> Deewiant, how fast does ccbi load it
18:32:44 <Deewiant> It'll take me 2 mins to download it first
18:33:25 <Deewiant> And I can't be bothered to time "just loading" properly so I'll just replace the first char with q :-P
18:33:26 <fizzie> ehird: There are quite many 1-3 MB files. http://www.gutenberg.org/files/12606/12606-8.txt is 2.91 megabytes.
18:33:32 <fizzie> "The Great Speeches and Orations of Daniel Webster by Webster and Whipple"
18:33:37 <ehird> that seems more malleable
18:33:44 <ehird> I love the word malleable
18:34:11 <fizzie> Deewiant: I added a single line with @ for that time test; I guess you saw it happening there, too.
18:34:23 <AnMaster> ehird, cat /dev/urandom | tr -Cd -- '-[:lower:][:digit:]\n\\/ ;",.+*[]{}^<>@`_|:%$#!'\'"${FPRINTINSTRS}" | tr -d 'mhlior' | head -n $(( 1024 * 1024 * 3 )) >> fuzz.tmp
18:34:40 <AnMaster> that was adapted from the fuzz test script I use
18:34:44 <ehird> *Hypha.Fungespace> butt `seq` ()
18:34:48 <AnMaster> obviously remove "${FPRINTINSTRS}
18:35:15 <AnMaster> (and the matching " at the end too yes)
18:35:34 <ehird> *Hypha.Fungespace> butt `seq` ()
18:35:34 <ehird> *** Exception: stack overflow
18:35:56 <Deewiant> cfunge eats around 2.2 gigs of RAM and takes 35.4 s user time
18:36:11 <fizzie> Deewiant: What, is my cfunge faster?
18:36:25 <Deewiant> Might be older / not optimized. Beats me.
18:36:36 <fizzie> Deewiant: Residential size was ~2.2 gigabytes here too.
18:36:36 <ehird> Deewiant: Okay, so now I'm going to strictify it.
18:36:42 <fizzie> AnMaster: No clue, really. :p
18:36:55 <fizzie> AnMaster: Not that new.
18:37:10 <AnMaster> well exact bounds does have some overhead.
18:37:34 <Deewiant> CCBI is still working on it at 900M; growing a lot slower than cfunge.
18:38:10 <fizzie> Yes, I haven't been updating this.
18:38:46 <ehird> Still overfloods in my butt.
18:39:32 <ehird> updateFungespace p v fs = fs { space = space fs `seq` Map.insertWith' const p v (space fs) }
18:40:32 <ehird> This is the oddity
18:41:30 <ehird> Maybe I'll just put $!'s EVERYWHERE.
18:42:09 -!- Gracenotes has joined.
18:42:22 <ehird> Deewiant: think maybe laziness is making my tailc alls not?
18:42:34 <ehird> go fs p@(Point x y) (z:zs) = go fs' (Point (x+1) y) zs
18:42:35 <ehird> where fs' = updateFungespace p (ord z) fs
18:42:37 <ehird> Bet that's taking stack
18:43:29 <AnMaster> ehird, is the data in it correct
18:43:39 <ehird> AnMaster: the new fungespace, not yet evaluated, was taking up space on the stack
18:43:45 <ehird> so I forced it to evaluate it before going on to the next char
18:43:50 <AnMaster> ehird, and also I had a working funge space done in less than a day in cfunge. Of course it was a lot more lines.
18:43:52 <ehird> thus throwing away the old value
18:43:55 <ehird> and making stack space constant again
18:44:01 <ehird> and I don't care :P
18:45:00 <AnMaster> ehird, for the funge space: what did you gain by using haskell? Compared to C, Lisp, Python or whatever.
18:45:09 <ehird> AnMaster: A language I like.
18:45:27 <AnMaster> ehird, will each component take this long, or is the funge-space an exception?
18:45:35 <AnMaster> and. Have you looked at hsfunge?
18:45:43 <ehird> AnMaster: Probably; the fungespace is the thing that stores a lot of data efficiently
18:45:50 <ehird> And loads very large stuff sequentially
18:45:59 <ehird> Not exactly Haskell's sweet spot
18:46:08 <ehird> funktio's site died
18:46:38 <AnMaster> I have a copy here, old one I think.
18:47:08 <AnMaster> I guess I could tar it up and upload it. Except there is NO license info. Not in the source files. Not in any separate file.
18:47:20 <AnMaster> and I never got this version of it to build.
18:48:19 <AnMaster> http://omploader.org/vMWxreg (hsfunge.tar.lzma)
18:48:42 <ehird> too lzma; didn't unpack
18:49:07 <ehird> OS X's autoexpander doesn't do lzma; I suppose I could open a terminal :-P
18:49:28 <Deewiant> CCBI is at 13 minutes and counting, using 1.9 gigs of RAM
18:49:31 <ehird> Oh god, no modules or anything
18:49:34 <ehird> Just a clump of files
18:49:40 <AnMaster> Deewiant, so less memory usage but slower
18:49:48 <ehird> FINGERPRINTS ALL IN ONE FILE
18:49:51 <ehird> INTERNAL GHC STRUCTURES
18:49:56 <Deewiant> AnMaster: Could easily go to 3 gigs, who knows :-P
18:50:10 <ehird> AnMaster: it's awful
18:50:14 <AnMaster> ehird, anyway how does one compile it
18:50:23 <ehird> AnMaster: ghc --make Main.hs -o hsfunge
18:50:41 <AnMaster> Could not find module `Data.Time.Calendar.Julian':
18:50:41 <AnMaster> Use -v to see a list of the files searched for.
18:50:48 <ehird> AnMaster: ghc --version
18:50:56 <AnMaster> The Glorious Glasgow Haskell Compilation System, version 6.8.2
18:51:05 <ehird> AnMaster: upgrade. even if it doesn't matter.
18:51:14 <Deewiant> ehird: HsFunge is older than 6.10.
18:51:33 <AnMaster> ehird, err the newer ones are *hardmasked* so I'm not going to until I find out why.
18:51:34 * ehird looks at hsfunge's parseFungeSpace; has lol attack
18:51:37 <Deewiant> I'd say 'cabal install time' but he doesn't have cabal.
18:51:45 <oerjan> maybe it's so old that Data.Time.Calendar.Julian has moved since...
18:51:46 <AnMaster> that is rather different than "testing" it usually means "broken"
18:51:48 <ehird> Deewiant: he has cabal just not cabal-install
18:51:52 <ehird> AnMaster: well, ghc 6.10 isn't broken
18:52:00 <Deewiant> ehird: No, he has Cabal but not cabal.
18:52:28 <ehird> AnMaster: I'd like to remind you that gentoo have a good history of thinking they know everything about the packages and in the process breaking them or stating incorrect things about the
18:53:01 <ehird> you don't know about gentoo's terrible relation with upstreams?
18:53:22 <ehird> a few packages refuse to fix any bugs from gentoo users because gentoo fuck up their packages
18:53:38 <Deewiant> hsfunge gets through Mycology with only one bad, in EVAR.
18:53:43 <ehird> AnMaster: no, others too.
18:53:50 <AnMaster> ehird, You haven't heard both sides of the story for ion clearly.
18:53:52 <pikhq> I only know of Ion.
18:53:54 <ehird> the ion author is a whining bastard who said he'd switch to windows
18:53:58 <Deewiant> Supports all but SOCK and TURT.
18:54:02 <ehird> i don't listen to him
18:54:06 <pikhq> And the Ion developer has about the intelligence of a fly.
18:54:13 <AnMaster> ehird, what other examples then
18:54:27 <ehird> AnMaster: i'm sorry that i don't keep a personal log of bad things i've heard about gentoo.
18:54:29 <AnMaster> since we all agree ion author is the fault in that case.
18:54:33 <Deewiant> Doesn't do input from console correctly.
18:54:35 <ehird> it's not very important to me.
18:54:45 <pikhq> ehird, this is definitely a case of [citation needed].
18:54:48 <AnMaster> ehird, so it is basically rumours only.
18:54:57 <ehird> just forgotten citations
18:55:07 <pikhq> You're the first person to have mentioned this.
18:55:17 <Deewiant> CCBI is at 2.5 gigs / 19 minutes.
18:55:25 <AnMaster> pikhq, indeed. I think he made it all up personally.
18:55:41 <Deewiant> I think the hashtable is just degenerating into a linked list.
18:55:44 <AnMaster> Deewiant, how the heck do you load into funge-space...
18:55:46 <ehird> yeah i make up shit about gentoo to satisfy my own perversive hate about it
18:55:55 <AnMaster> Deewiant, um you need to resize it then and rehash
18:56:12 <Deewiant> AnMaster: I don't resize, the array resizes itself when it feels like it
18:56:18 <AnMaster> Deewiant, the cfunge one is set up to do that once it gets too stuffed. Though it already start out pretty large.
18:56:37 <Deewiant> It's not even possible to tell a D hashtable what its size should be
18:56:47 <AnMaster> Deewiant, indeed. That is what it does too. I enabled the option to do that for the hash library that I use.
18:57:07 <AnMaster> Deewiant, I'm able to set mine to a rather large initial size.
18:57:41 <AnMaster> Deewiant, anyway, was your cfunge 32-bit or 64-bit?
18:58:46 <AnMaster> Deewiant, right, changing to 32-bit cells would reduce memory usage and increase speed a lot.
18:59:10 <AnMaster> Deewiant, how do you read the input file. Or is that not a bottle neck at all?
18:59:10 <Deewiant> I'll try hsfunge once this is done
18:59:24 <Deewiant> Not sure about this version, actually
19:00:20 <Deewiant> Tango class for reading one byte at a time, buffers internally however it wants
19:02:06 <AnMaster> Deewiant, so stream IO I guess.
19:02:14 <Deewiant> Amusing amount of overhead for 131 megs of data :-P
19:02:35 <AnMaster> Deewiant, was the original file 131 MB
19:03:35 <Deewiant> I wonder if the GC is scanning that AA...
19:04:15 <ehird> Now I need to grow bounds
19:04:18 <ehird> Exact bounds can go FUCK THEMSELVEs
19:04:26 <ehird> Deewiant: does mycology test for exact bounds
19:04:34 <ehird> put it in mycoedge
19:04:40 <ehird> i wanna pass mycology :D
19:05:08 <Deewiant> I'd rather not but I might have to to avoid it being a major pain
19:06:47 <AnMaster> Deewiant, can you see where the overhead is
19:06:48 <ehird> % runhaskell -Wall -O2 Main
19:06:55 <ehird> DOOOOOOOOOO DOOOOOOOO
19:07:07 <ehird> % runhaskell -Wall -O2 Main
19:07:07 <ehird> Point {pX = 0, pY = 0}
19:07:08 <ehird> Point {pX = 74, pY = 48944}
19:07:22 <Deewiant> AnMaster: Not from the code, certainly :-P
19:07:22 <AnMaster> Deewiant, um. I don't know. memory profiler?
19:07:37 <Deewiant> If I profiled this it'd take 30 days
19:07:46 <fizzie> valgrind + cfunge 462 + that 31-megabyte file:
19:07:48 <fizzie> ==918== Warning: set address range perms: large range [0x43f7b030, 0x4bf7b030) (undefined)
19:07:59 <AnMaster> fizzie, what the hell does that mean
19:08:04 <fizzie> (It's still running; I'm actually more interested in the malloc statistics.)
19:08:29 <AnMaster> fizzie, I think it will end up mallocing a lot in cfunge_mempool.c at least
19:08:36 <fizzie> ==918== malloc/free: in use at exit: 1,813,071,464 bytes in 8,257 blocks.
19:08:37 <fizzie> ==918== malloc/free: 16,469 allocs, 8,212 frees, 2,548,451,304 bytes allocated.
19:08:39 <ehird> Hey AnMaster, Deewiant, fizzie: I parse mycology in 0.025-0.026 seconds
19:08:55 <ehird> Including loading the file and printing out the min/max bounds
19:09:27 <ehird> Point {pX = 179, pY = 791}
19:09:28 <AnMaster> ehird, what about doing that bit in cfunge on your system. Remember my computer is slower. So hard to compare.
19:09:31 <ehird> Mycology has 796 lines
19:09:34 <ehird> Maybe the > is broken
19:09:39 <fizzie> I only have four gigabytes memory here; it's barely enough for these 32-megabyte enterprise Funge programs.
19:09:51 <ehird> AnMaster: It takes something like 6ms to RUN Mycology.
19:09:56 <AnMaster> fizzie, what was the book btw.
19:10:09 <AnMaster> ehird, differently loaded system?
19:10:13 <ehird> fizzie: Feh, my enterprise befunge programs are so big I'm upgrading to 12GB of memory!
19:10:33 <fizzie> AnMaster: "U.S. Copyright Renewals 1950 - 1977"
19:10:42 <ehird> AnMaster: It's just that cfunge is ridiculously fast and my code isn't.
19:11:10 <ehird> AnMaster: Still, that's less than 10x of your running to do my parsing
19:11:29 <Deewiant> CCBI is about 6-7x as slow as cfunge when running Mycology with fingerprints off
19:11:35 <ehird> Deewiant: So correct me if I'm wrong, but all I need to have accurate fast bounds is to keep the minimum point updated, right?
19:11:48 <fizzie> Incidentally, the university people have recently updated their main shell server to something that has a ridiculous amount of power for running a couple of IRC clients; it's a dual-processor Xeon E5450 (so 8 cores) and has 64 gigabytes of RAM. (Okay, so "couple" here means "couple hundred", but anyway.)
19:11:55 <ehird> Deewiant: e.g. involving slowdown.b98
19:12:08 <Deewiant> ehird: For precise bounds you need to know both min and max.
19:12:12 <ehird> fizzie: E5450 ... what architechture thingy
19:12:17 <fizzie> total used free shared buffers cached
19:12:18 <fizzie> Mem: 64315184 30313024 34002160 0 760520 25333160
19:12:23 <ehird> Deewiant: Right, but not per line or anything to have slowdown not slow down?
19:12:23 <Deewiant> slowdown.b98 needs precise bounds.
19:12:35 <AnMaster> ehird, cfunge on mycology here: 0.034s with clean env, output to /dev/null and non-exact bounds. For exact bounds add ~0.010s. For env and output to terminal it is a bit more complex, varies more. Generally the time then is something like 0.070s or so. Depends on what terminal emulator I use though.
19:12:35 <Deewiant> ehird: No, you just need precise min/max.
19:12:46 <ehird> That's not too hard then
19:12:51 <AnMaster> of course turning off fingerprints make it even faster
19:13:02 <fizzie> I don't know anything else except what I read from /proc/cpuinfo.
19:13:11 <Deewiant> I turn off fingerprints just because cfunge doesn't support them all, so the comparison is fair.
19:13:54 <AnMaster> $ env -i PATH=/bin:/usr/bin TERM=$TERM /bin/bash --norc --noprofile -c 'time build_fast/cfunge -F mycology/mycology.b98 > /dev/null'
19:13:59 <AnMaster> there I turned off fingerprints too
19:14:12 <AnMaster> $ env -i PATH=/bin:/usr/bin TERM=$TERM /bin/bash --norc --noprofile -c 'time build_opt/cfunge -F mycology/mycology.b98 > /dev/null'
19:15:01 <AnMaster> Deewiant, yes that is quite fair. Except I should use a profile feedback build for best results ;P
19:16:39 <ehird> Oh dear, now hypha is terribly slow.
19:17:00 <ehird> AnMaster: I made bounds growing actually work.
19:17:02 <ehird> Now it's just chugging.
19:17:05 <ehird> Maybe it's inflooping
19:17:12 <Deewiant> I could actually run out of memory before this completes :-P
19:17:12 <ehird> I think it's inflooping maybe
19:17:19 <ehird> | x < minX fs = updateFungespace p v fs{ minX = x }
19:17:19 <ehird> | x > maxX fs = updateFungespace p v fs{ maxX = x }
19:17:20 <ehird> | y < minY fs = updateFungespace p v fs{ minY = y }
19:17:22 <ehird> | y > minY fs = updateFungespace p v fs{ maxY = y }
19:17:24 <ehird> Don't see much wrong with that
19:17:29 <lifthrasiir> did you fixed the first character to q then?
19:17:49 <ehird> Yay it's instant again
19:18:02 <Deewiant> "qroject Gutenberg's Copyright Renewals"
19:18:04 <ehird> Now it thinks it has 794 lines
19:18:08 <ehird> maybe my editor is wrong
19:18:17 <ehird> Deewiant: how many fungelines does myco have?
19:18:26 <ehird> Deewiant: ok, 794 is right
19:18:27 <ehird> since I start at 0
19:18:40 <ehird> Deewiant: longest line is at col 179?
19:19:00 <ehird> Now it takes 0.026 exactly, all the time
19:19:02 <AnMaster> ehird, check with y output. It says "BAD" if it is wrong.
19:19:04 <Deewiant> ehird: You'll find it right near the start.
19:19:17 <ehird> So I've lost the 0.001s speed gain I soemtimes had :P
19:19:20 <ehird> AnMaster: Er, dude.
19:19:22 <ehird> I don't do funge yet.
19:19:28 <Deewiant> There's a comment there saying that it has to be the longest line.
19:19:29 <AnMaster> ehird, in a reference interpreter
19:19:34 <fizzie> With fungot there's extra-confusion since the actual code is loaded at (x=0,y=100) instead of the origin. So there's an offset, plus the off-by-oneness of editor line numbers.
19:19:35 <fungot> fizzie: 20. some crocodiles, when not hungry, are amiable; but some are not. " some x are not-y, and some are not.
19:19:48 <AnMaster> That the least point containing a non-space cell is ( -1 -1 )
19:19:48 <AnMaster> That the greatest point, relative to that point, is ( 180 795 )
19:19:57 <fungot> Available: agora alice* darwin discworld europarl ff7 fisher ic irc jargon lovecraft nethack pa speeches ss wp
19:20:06 <AnMaster> ehird, that is off by one since it wrote to -1,-1
19:20:11 <ehird> Anyway, hypha loads it in imperceptible time
19:20:17 <ehird> Fuck exact bounds for now.
19:20:18 <Deewiant> Grows by 8M every time it decides to
19:20:19 <AnMaster> Deewiant, how much ram do you have
19:20:27 <Deewiant> AnMaster: 7983M according to htop
19:20:33 <ehird> Deewiant: Do you think wrapping logic belongs in fungespace.hs
19:20:37 <ehird> Fungespace.hs that is
19:20:46 <ehird> AnMaster: 8 ramgbs
19:21:12 <lifthrasiir> well, it quickly starts thrashing and my mplayer process got stopped.
19:21:16 <Deewiant> cfunge had 613288 minor pagefaults, CCBI 875756
19:21:35 <AnMaster> Deewiant, that isn't such a large difference.
19:21:36 <Deewiant> AnMaster: I missed it but I guess around 3410-3450 since I last said 3407
19:21:43 <fizzie> Deewiant: Now run it under valgrind to get allocation statistics. :p
19:21:44 <ehird> I'm going to parse that fucking file
19:21:48 <ehird> Er wait Deewiant how much ram did it use
19:21:51 <AnMaster> Deewiant, and you said your cfunge was 64-bit cells though
19:22:05 <ehird> Deewiant: Right so about 1gb more than I have
19:22:08 <AnMaster> Deewiant, in that case the difference is larger in fact.
19:22:56 <ehird> I want to feel pitiful about my ram
19:23:00 <ehird> That's nice I don't use irssi :P
19:23:10 <AnMaster> Deewiant, what about a 32-bit cfunge build. Since 2.2 GB is more than what I have
19:23:10 <Deewiant> AnMaster: Well yeah, the overhead of CCBI is obviously a lot bigger.
19:23:26 <Deewiant> ehird: If your client can't, upgrade.
19:23:47 <ehird> irssi has a gui now does it? oh no it emulates a shitty one with vt terminal codes
19:23:53 <ehird> sry i don't use that kind of sw :)
19:24:14 <Deewiant> Then press pageup a few times, or whatever your client does to read lines earlier than the one you just typed
19:25:05 <AnMaster> well, not that filesystem irc one naturally. But all major ones.
19:25:21 <Deewiant> Alright, latest cfunge, 64-bit cells: 2.2 gigs / 24.4 secs
19:25:33 <AnMaster> Deewiant, and what about 32-bit cells
19:26:26 <AnMaster> turning off exact bounds would reduce it quite a bit I suspect.
19:27:10 <AnMaster> should be compared to 2.2 GB one then indeed.
19:27:26 <ehird> fizzie: what's that ipv6 tunnel that listed ipv6 isps?
19:27:46 <fizzie> ehird: Somewhere in SixXS's pages.
19:28:08 <fizzie> http://www.sixxs.net/faq/connectivity/?faq=native
19:28:39 <AnMaster> Deewiant... Could you pastebin the output from ./cfunge -f there... Because that extra memory made no sense.
19:28:43 <Deewiant> AnMaster: Inexact increases memory usage by 300M
19:29:19 <Deewiant> http://www.pastie.org/private/6kyhirehdwzelidahsyda
19:29:21 <AnMaster> Deewiant, that increase in memory usage makes no sense. Sure you didn't confuse the two options
19:29:27 <ehird> sixxs are down >_<
19:29:36 <Deewiant> AnMaster: Are you sure /you/ didn't? :-P
19:29:50 <Deewiant> Maybe an #ifndef where an #ifdef should be or vice versa
19:29:52 <AnMaster> Deewiant, it is the opposite of the results I get when I did memory profiling on cfunge here.
19:29:55 <fizzie> That sixxs url worked for me just fine.
19:30:20 <fizzie> Maybe their IPv4 side is down. :p
19:31:01 <ehird> fizzie: what's sixxs's magic ipv4 to ipv6 domain
19:31:03 <ehird> i'll view it via that :P
19:31:45 <ais523> sixxs.net = the page describing sixxs, sixxs.org = the tunnel itself
19:31:54 <fizzie> Hmm, *that* doesn't seem to answer to me.
19:32:05 <ais523> fizzie: obviously, you're on IPv6
19:32:09 <ais523> and it's an IPv6 to 4 translator
19:32:21 <fizzie> ais523: Uh, I do have IPv4 connectivity too here.
19:32:28 <ehird> http://www.sixxs.net.ipv4.sixxs.org/ works :-D
19:33:18 <AnMaster> Deewiant, doesn't match my results.
19:33:43 <AnMaster> for mycology: 3.336 peak ram usage with inexact bounds. 6.522 for exact ones.
19:34:57 <AnMaster> Deewiant, In other words: I'm unable to reproduce the bug and for now I recommend that you remove the build directory and try again on a clean checkout.
19:35:12 <AnMaster> if you are sure you didn't confuse it
19:35:42 <AnMaster> bzr branch http://rage.kuonet.org/~anmaster/bzr/cfunge/trunk
19:36:09 <Deewiant> They need to do way instain bzr>
19:36:34 <AnMaster> Deewiant, well I'm too lazy to tell you to do bzr st and revert any changed files
19:36:43 <AnMaster> Deewiant, I assume you are doing out of tree build. In tree build is unsupported as noted in the README.
19:37:25 <AnMaster> so if you are doing that it should just be removing build directory to get it clean
19:38:25 -!- pikhq has quit ("leaving").
19:38:37 <ehird> who kill thier branches
19:38:37 <Deewiant> Default CMake settings + Release build: 2203M / 24.42
19:38:48 <ehird> because these branches can't frigth back, it was on the rss this morning
19:38:59 <ehird> a committer in bzr who had kill his three patches
19:39:09 <ehird> my pary are with the programmers who lost his patches
19:39:10 <AnMaster> Deewiant, that would be with exact bounds indeed.
19:39:12 <ehird> i am truly sorry for your lots
19:39:23 <Deewiant> Toggled EXACT_BOUNDS to OFF: 2532M / 23.10
19:39:48 <AnMaster> Deewiant, do the difference go the same way for mycology?
19:40:01 <Deewiant> How can I get mem usage for something that runs so fast?
19:40:15 <Deewiant> top only samples once per second or whatever
19:40:19 <AnMaster> Deewiant, valgrind --tool=massif and do ms_print on the generated file
19:40:32 <Deewiant> valgrind changes program behaviour
19:40:48 <AnMaster> Deewiant, what about /usr/bin/time
19:41:13 -!- pikhq has joined.
19:41:28 <AnMaster> Deewiant, man page says it can with a different format string
19:43:05 <ehird> it started to look at it but then it disappeared
19:43:08 <ehird> THE HORRORS OF FAST PROGRAMS
19:43:17 <Deewiant> Probably not available on Linux.
19:43:19 <ehird> Okay so I have a fungespace YAY AM HAPPY!!
19:43:20 <AnMaster> ehird, I tried on a slow program too, so irrelevant
19:43:23 <fizzie> There is a Ubuntu bug report about it.
19:43:26 <AnMaster> Deewiant, well that leaves valgrind
19:43:29 <ehird> Now, I need to write an Interpreter data structure.
19:43:45 <fizzie> "it" being time and resource usage showing zero.
19:44:03 <fizzie> I'm sure it's doable on Linux, the /usr/bin/time people just haven't bothered.
19:44:23 <lifthrasiir> 15 minutes later, pyfunge read 80% of the file and became very slow
19:44:37 <Deewiant> AnMaster: With valgrind it's 3442744 for inexact bounds and 6785328 for exact (second column, tail -n1)
19:44:49 <ehird> Deewiant: So is the stack stack just int** then?
19:44:51 <Deewiant> So no, it doesn't go the same way as it apparently does
19:45:10 <Deewiant> ehird: Isn't it more fun to figure this kind of stuff out yourself? :-P
19:45:17 <AnMaster> Deewiant, well... you said top only updated slowly right?
19:45:25 <Deewiant> ehird: Besides, you're Haskell so it's Stack (Stack Int)
19:45:25 <ehird> Deewiant: Sure, I just don't like implementing basic concepts completely incorrectly
19:45:31 <AnMaster> Deewiant, maybe both use more and you managed to catch then in an odd way
19:45:34 <lifthrasiir> 57651 Python 0.6% 13:04.08 1 14 3183 1319M+ 188K 1311M+ 1462M
19:45:41 <ehird> Deewiant: Yeah, where stack = mutable array in IO, of course.
19:45:49 <Deewiant> AnMaster: I doubt it, for such a big difference
19:45:50 <ehird> Well. Can you get efficient arrays outside of IO in hs?
19:45:54 <fizzie> Deewiant: Just valgrindalyze the 31-megabyte file; it didn't take more than a minute or so to run it.
19:46:06 <ehird> Like, when you throw away the old value it turns into just [x] = y
19:46:23 <Deewiant> If it's pure it copies every time since it has to be, y'know, pure.
19:46:32 <fizzie> (Disclaimer: I didn't use any --tool=massif thing.)
19:47:08 <ehird> Deewiant: no, ghc optimizes pure shit
19:47:11 <AnMaster> Deewiant, replace that q with 'a,a,><
19:47:20 <ehird> if you copy & change a pure data structure and throw away the old one it optimizes it to a mutation
19:47:37 <AnMaster> Deewiant, that should help making sure the measurement is stable
19:47:37 <ehird> Deewiant: that's what i'm asking you
19:49:16 <Deewiant> ==30334== malloc/free: in use at exit: 2,140,095,696 bytes in 8,467 blocks.
19:50:02 <Deewiant> And ==30334== malloc/free: 16,941 allocs, 8,474 frees, 2,880,720,576 bytes allocated.
19:50:15 <Deewiant> fizzie: And you tricked me, it took over 2 mins.
19:50:34 <AnMaster> Deewiant, the latter being inexact ones?
19:50:40 <ehird> Deewiant: Think I should use a slowarray or an IOarray?
19:50:46 <Deewiant> AnMaster: No, they're from the same run.
19:50:49 <lifthrasiir> heck, it takes a minute per 1000 lines and i finally get tired of it.
19:50:59 <AnMaster> Deewiant, so what about inexact ones then
19:50:59 <lifthrasiir> ^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^\Quit
19:51:12 <Deewiant> AnMaster: Did you miss "took over 2 mins"?
19:51:30 <Deewiant> ==30687== malloc/free: in use at exit: 2,081,686,560 bytes in 8,203 blocks.
19:51:30 <Deewiant> ==30687== malloc/free: 16,415 allocs, 8,212 frees, 2,817,066,400 bytes allocated.
19:51:50 <AnMaster> Deewiant, that is less for inexact
19:52:02 <fizzie> lifthrasiir: There's a three-megabyte file mentioned later.
19:52:04 <AnMaster> Deewiant, so where did you get that reverse behaviour from.
19:52:10 <AnMaster> Deewiant, I can't reproduce it locally.
19:52:19 <AnMaster> Deewiant, ok, I gues it is a htop bug or something.
19:52:37 <AnMaster> Deewiant, interesting that exact bounds doesn't have a large memory overhead.
19:52:40 <pikhq> GAH! WHY DOES ALSA HATE ME?
19:52:48 <pikhq> cat /proc/asound/cards
19:52:51 <pikhq> --- no soundcards ---
19:53:12 <Deewiant> AnMaster: RES = "The non-swapped physical memory a task has used" according to man top.
19:53:13 * pikhq points at lspci, then points at lsmod, then points at his alsa configuration...
19:53:14 <fizzie> http://www.gutenberg.org/files/12606/12606-8.txt "The Great Speeches and Orations of Daniel Webster by Webster and Whipple" is 2.91 megabytes.
19:53:14 <AnMaster> pikhq, load the correct kernel driver then.
19:53:36 <Deewiant> Beats me, maybe it's just the wrong value to be looking at.
19:53:44 <AnMaster> Deewiant, unless cfunge was swapping?
19:54:08 <Deewiant> Well actually I have 256M but it's pretty much never in use.
19:54:25 <Deewiant> 14 of it is in use currently, probably due to these enterprisey funge apps.
19:54:43 <pikhq> Though I've not touched it at all.
19:54:47 <AnMaster> pikhq, cat: /etc/asound.conf: No such file or directory
19:55:14 <AnMaster> pikhq, you use gentoo right? What package installed it
19:55:28 <pikhq> Oh, installed asound.conf?
19:55:44 <pikhq> That's not installed by a package; that's only used if you want a custom ALSA configuration.
19:55:54 <AnMaster> pikhq, ah, why do you want that then!
19:55:58 <pikhq> /usr/share/alsa/alsa.conf contains the default config file, which loads /etc/asound.conf.
19:56:47 <AnMaster> pikhq, is your /etc/asound.conf empty then?
19:56:53 <AnMaster> if not try temporarily removing it?
19:57:02 <pikhq> That's one thing I tried.
19:57:03 -!- Judofyr has quit (Remote closed the connection).
19:57:15 <AnMaster> pikhq, what did you do before it stopped working
19:57:21 <lifthrasiir> rsize=185M, vsize=232M, but it takes much after reading all the file... hmm.
19:57:28 <pikhq> Upgrade my kernel, using the same config as my previous kernel.
19:57:36 <AnMaster> lifthrasiir, is that the 32-mb file in pyfunge?
19:57:52 -!- olsner has joined.
19:57:58 <lifthrasiir> no, 32MB file got very slow and i forced to terminate it.
19:58:49 <lifthrasiir> since i have less than 1GB of working, not-thrashing-proof memory it became very slow
19:58:50 * pikhq looks at his kernel config some more
19:58:56 <ehird> it's only 32 mb, wimps!
19:59:03 -!- oerjan has quit ("Good night").
19:59:24 <lifthrasiir> ehird: remember that python stores every integer as 20-byte-or-so struct in the heap... >_<
19:59:34 <ehird> wait python boxes integers?
19:59:40 <ehird> lol@python' sucking
20:00:03 <AnMaster> lifthrasiir, err doesn't it only for large ones
20:00:05 <ehird> lifthrasiir: did guido say that boxed integers were too magical or confusing or something
20:00:14 <lifthrasiir> though python 2 caches some small integers
20:00:26 <AnMaster> lifthrasiir, are you using 3.x
20:00:36 <ehird> AnMaster: Nobody uses 3.
20:00:40 <AnMaster> ehird, python 2.x has small and large integers.
20:00:48 <lifthrasiir> and that some small integers are between -5 and 100, iirc.
20:00:52 <ehird> Small integers are just cached in an array.
20:01:01 <ehird> lifthrasiir: 100 is enough for everyone
20:01:07 <lifthrasiir> and what the hell this program stucks after reading fungespace? curious.
20:01:38 <Deewiant> I don't know why but I found that particular sequence of words hilarious
20:01:43 <Deewiant> "what the hell this program stucks"
20:02:00 <fizzie> Deewiant: Well, you're such a stucker, you would.
20:02:09 <AnMaster> "is stuck" "freezes", but I never heard "stucks" as a verb
20:02:18 <ehird> -- what the hell this program stucks
20:02:21 <Deewiant> fizzie: That I found more groan-inducing than anything else.
20:02:35 <lifthrasiir> maybe i forgot english grammar for a moment.
20:02:38 <ehird> stucken: a program that stucks
20:02:38 <Deewiant> AnMaster: It's not valid English if you're wondering about that.
20:02:59 <ehird> ais523: does your program stucks?
20:03:07 <fizzie> Sticka, stickur, stuckade, har stöckit.
20:03:23 <ehird> "They surgically removed 2 feet of colon - I now have a ;"
20:03:34 <fizzie> The same thing in Swedish in an alternate dimension.
20:04:11 <AnMaster> sv:"att sticka" is en:"to knit"
20:04:25 <lifthrasiir> i have had enough fun with torturing pyfunge; now i have to do some real work.
20:04:37 <AnMaster> sv:"att sticka" is en:"to sting" (for example, a bee stinging or such)
20:04:53 <AnMaster> and then there is the slang meaning
20:05:15 <AnMaster> which is "leave quickly", not the same as "fleeing" exactly
20:06:15 <fizzie> AnMaster: Is it again overloaded so that it has a different declension for different meanings? Like "sticka, stickar, stickade, stickat" for the first and "sticka, sticker, stack, stuckit" for the other, or something.
20:07:10 <fizzie> Right. I don't understand why you people keep doing that.
20:07:40 <ais523> AnMaster: sort of leaving in a hurry?
20:08:21 <AnMaster> but there are different side-meanings in it I think.
20:08:30 <AnMaster> but then I may just be ignorant of them in English
20:08:40 <AnMaster> and I can't translate those "side-meanings" really
20:09:58 <AnMaster> ais523, lets say someone just pulled(right word?) a practical joke on you, then they might leave in a hurry before they have to flee I guess. This word would work in that situation. For example.
20:10:16 <AnMaster> ais523, flee has a lot more "panic" feeling over it.
20:10:28 <ehird> AnMaster: 23 skidoo? XD
20:10:35 <ehird> 23 skidoo (sometimes 23 skiddoo) is an American slang phrase popularized in the early twentieth century, first appearing before World War I and becoming popular in the Roaring Twenties. It generally refers to leaving quickly, being forced to leave quickly by someone else or taking advantage of a propitious opportunity to leave, that is, "getting [out] while the getting's good." The exact origin of the phrase is uncertain.
20:11:02 <AnMaster> ehird, yes, except it is wider than that
20:14:39 <AnMaster> ehird, hey, you might want to do: sudo nvram boot-args="-s"
20:14:58 <ehird> AnMaster: Starts in single user mode every time?
20:15:14 <ehird> I don't even know how to get the GUI up from there
20:17:47 <pikhq> Hmm. 'Twould seem that my SB Live! has stopped working.
20:18:01 <pikhq> And for some reason, the driver for my onboard audio didn't load.
20:18:38 <fizzie> "exit" from the single-user shell continues the normal bootup, I think.
20:21:27 <ehird> pikhq: Open up the case and examine?
20:21:38 <pikhq> ehird: I'll reseat it next time I reboot.
20:22:00 <ehird> They should make pencil sharpeners that use high-RPM fans or something, so you don't have to do any work.
20:22:49 <ehird> ais523: awesome, I want one
20:22:57 <ehird> for the sheer uselossity!
20:23:06 <ais523> I've seen them used in businesses before
20:23:25 <fizzie> I had an electronic pencil-sharpener back in third grade, actually. Very neat. I wonder what happened to it.
20:23:38 <fizzie> Maybe "electric" is more correct.
20:24:03 <fizzie> I seem to remember it being rather awfully noisy.
20:24:04 <ehird> clearly we need a bluetooth pencil sharpener
20:24:16 <ehird> (I was going to say USB but decided to jump to the next awesome step)
20:24:18 <ais523> no, a FreeBSD pencil sharpener
20:24:25 <ehird> ais523: a bluetooth pencil sharpener that runs freebsd
20:24:35 <ehird> netbsd is the toaster one
20:24:36 <ais523> and yes, I meant netbsd
20:25:07 <ehird> ais523: but fans are bad
20:25:14 <ehird> we need a passively cutting pencil sharpener
20:25:44 <fizzie> It could use some sort of controlled acid thing.
20:26:43 <ehird> hyper fan-like cutting lasers
20:27:10 <fizzie> I sense many "my pencil's on fire!" yells in the future of this channel.
20:27:25 <ehird> I knew someone who set their "pencil" on fire.
20:27:39 <ais523> well, burning the pencil into a point would be one way to sharpen it
20:27:43 <AnMaster> <ehird> I don't even know how to get the GUI up from there <-- SystemStarter
20:27:51 <AnMaster> ehird, but why do you want that
20:28:21 <ais523> ehird: also, have you not considered simply undoing the command that forced it into single user mode?
20:28:37 <ehird> i wish intel macs used openfirmware
20:28:42 <ais523> AnMaster: anyway, single user mode is a bad idea:
20:28:51 <ehird> yes, single user mode is a bad idea in general
20:28:56 <ais523> first, that means you do everything as root, which is insecure
20:29:13 <ais523> and second, single user means you can't use the user system for sandboxing
20:29:25 <ehird> ais523: you are kidding, right?
20:29:30 <ehird> you do know he was joking, right?
20:29:40 <ais523> I'm never sure, with AnMaster
20:30:27 <AnMaster> ais523, anyway there are uses for single user mode. Accurate profiling results for example.
20:30:39 <AnMaster> quite a pain to do that though
20:30:40 <ehird> On the subject of breaking "pencils": http://geekz.co.uk/lovesraymond/archive/the-redhat-package-mangler
20:30:53 <ais523> using single user mode for profiling isn't something I even thought of
20:31:04 <ais523> why not just use a per-application profile timer?
20:31:10 <AnMaster> <ais523> and second, single user means you can't use the user system for sandboxing
20:31:31 <ais523> yes, but then you're stuck as that user and can't be root as well
20:31:40 <ais523> or does single user mode have multiple terminals?
20:31:58 <ehird> ais523: sudo -s, beotch
20:31:59 <AnMaster> ais523, per-application is inaccurate if you are doing IO stuff. Which causes stuff to happen in kernel.
20:32:04 <ehird> I must be the only person who likes ELER
20:32:19 <ehird> everybody loves eric raymond
20:32:22 <ehird> eric s raymond that is
20:32:33 <ehird> http://geekz.co.uk/lovesraymond
20:32:48 <ehird> (Here's one I made [two minutes] earlier: 20:30 ehird: On the subject of breaking "pencils": http://geekz.co.uk/lovesraymond/archive/the-redhat-package-mangler )
20:33:03 <ehird> (Everybody Loves Eric Raymond is a web comic depicting the very real lives of Richard Stallman, Eric Raymond and Linus Torvalds as accurately as comedically possible.
20:33:04 <ehird> Their real lives, which include living together in a house, with dynamic dimensions, without their wives or girlfriends. Also, they always wear the same clothes. And don’t move their eyes much.)
20:33:17 <AnMaster> ehird, hm seems SystemStarter is not for modern OS X
20:33:26 <ehird> AnMaster: it may be some launchd stuff
20:33:40 <AnMaster> ehird, yes that is what I meant.
20:36:55 <AnMaster> ehird, single user mode has two processes: launchd and shell
20:37:14 <ehird> launchd = init + cron + anacron + at + ...
20:37:43 <ehird> AnMaster: no, orthogonal
20:38:18 <ehird> No manual entry for init
20:39:51 <ehird> AnMaster: have you ever heard of a cat called Public Beta?
20:41:15 <Deewiant> Encountered: 91516 instructions
20:41:18 <Deewiant> Executed: 82692 standard instructions
20:41:20 <Deewiant> Executed: 161 fingerprint instructions
20:41:29 <ehird> Deewiant: what's the io array again
20:41:30 <Deewiant> Travelled to the future: 12 IPs
20:42:02 <AnMaster> ehird, mac os x by default has a serial console set up
20:42:10 <ehird> Deewiant: what's it called :P
20:42:16 <AnMaster> Deewiant, I was going wow over serial console
20:42:26 <ehird> http://upload.wikimedia.org/wikipedia/en/9/92/Mac_OS_X_10.1_Puma_screenshot.png ← i wish current os x looked like this
20:42:53 <ehird> Deewiant: ~/.cabal/share/doc vs /usr/local/gdfgdfg/dfgdfgdfg/dfgdfgdfg :-
20:46:09 <ehird> Deewiant: /usr/local/share/doc/ghc/libraries/index.html
20:47:27 <fizzie> My /usr/share/doc seems to weigh 600 megabytes; that's quite heavy.
20:48:14 <fizzie> texlive-latex-extra-doc by itself is 100 megabytes; I wanted to look at a single manual of a single package, and, well...
20:48:26 <AnMaster> <fizzie> My /usr/share/doc seems to weigh 600 megabytes; that's quite heavy. <-- 398 MB here
20:48:33 <Deewiant> fizzie: I just use CTAN for that :-P
20:48:46 <ehird> 24M/usr/local/share/doc
20:48:47 <ehird> 65M/opt/local/share/doc
20:48:48 <fizzie> Yes, Google is what I usually use as well.
20:49:42 <AnMaster> before I said that there was one nice thing about mac hardware
20:50:03 <AnMaster> the magnetic power connector is very nice too
20:50:12 <AnMaster> I guess it is only found on laptops
20:50:28 <fizzie> My iBook G4 is too old to have that piece of niftitude. :/
20:50:51 <ehird> The power cable on my iMac is sort of welded in, I couldn't get it out if I wanted to.
20:50:59 <ehird> Without performing surgery on it. Which I might do when I get the new box.
20:51:05 <ehird> There's always shiny hardware to raid! Not RAID.
20:51:16 <AnMaster> ehird, ok found how to boot correctly from single user mode:
20:51:59 <ehird> Deewiant: UVector is sufficiently smart, apparently. Hoorj
20:52:42 <Deewiant> You can't resize it purely IIRC
20:53:55 <ais523> ooh, critical zero-day bug for adobe acrobat reader
20:54:10 <ais523> cross-platform, and can be fixed by disabling javascript
20:56:30 <Deewiant> I'm not sure, I could be wrong
20:56:38 <ehird> Deewiant: If you can resize them in IO, I'll just unsafePerformIO 'er up
20:56:46 <ehird> I should just use an io array really
21:00:41 <Deewiant> Performed: 263675 Funge-Space lookups
21:00:41 <Deewiant> Performed: 443 Funge-Space assignments
21:00:55 <Deewiant> (Mycology without fingerprints)
21:01:45 <fizzie> AnMaster: 22:18:36 [Finnish time] <fizzie> "exit" from the single-user shell continues the normal bootup, I think.
21:21:06 <AnMaster> fizzie, except bluetooth fails to start then :D
21:22:29 <AnMaster> Deewiant, you added that logging code just now?
21:23:48 <AnMaster> btw have I mentioned that disable tracing at compile time make no difference on my sempron but makes a noticeable difference on my old pentium3.
21:24:18 <AnMaster> I guess branch prediction is not very good on the p3
21:27:34 <Deewiant> No, I don't think you've mentioned that
21:49:54 <AnMaster> ehird, I just built cfunge as an universal binary.
21:50:10 <AnMaster> runs on intel at least. my old mac is too old to test it (pre-os x)
21:50:34 <AnMaster> it can dual compile for powerpc and m68k though (fat binary)
21:54:21 -!- nooga has quit (Read error: 110 (Connection timed out)).
22:03:03 <kerlo> I imagine there are many games based on Conway's Life.
22:03:24 <kerlo> "Game" meaning "game that people play against each other".
22:04:21 <kerlo> One possible mechanic: each player starts with an R-pentomino; every turn, one player kills one of their own pieces, then the Life rules are applied.
22:04:38 <kerlo> When a cell is born, it receives the color shared by two of its parents.
22:05:20 <kerlo> R-pentomino. This pattern:
22:05:33 <AnMaster> ok... so why the odd name "pentomino"
22:07:04 <Deewiant> AnMaster: pent- from Greek "five", -omino from domino.
22:10:39 * kerlo ponders what extensions should be made to a neural net to make it decent at his Conway's Life game
22:11:37 <kerlo> Say, another Life game is the "intelligent cell" game.
22:12:09 <kerlo> Which can be found at the celebrated eminently famous clickmazes.com.
22:14:42 <ehird> 20:49 AnMaster: ehird, I just built cfunge as an universal binary.
22:15:13 <ehird> i mean, ppc macs are practically dead now
22:15:17 <ehird> so it's generally a waste of space
22:15:41 <ehird> GregorR moved vps hosts again :D
22:15:41 <AnMaster> my old first generation ibook still works
22:15:50 <ehird> AnMaster: 10.6 may not support ppc macs
22:16:00 <ehird> And 'cuz I'm moving to the same host and they're cheap and look good.
22:16:10 <ehird> AnMaster: That's dead. Mac users = early adopters.
22:16:12 <AnMaster> why does everyone have to upgrade
22:17:37 <Deewiant> Tail recursion: http://g.imagehost.org/0632/snake.jpg
22:18:26 <ehird> AnMaster: that ipv6 vps you used was w/ bytemark hosting right
22:18:36 <ehird> Deewiant: Have you read your SICP today?
22:19:12 <ehird> AnMaster: what was it with
22:19:32 <AnMaster> ehird, it was with a friend who has a dedi at softlayer
22:20:00 <ehird> Bytemark seems to load quite fast from the uk (since they're a uk company) and do ipv6, so that's the main competitor to prgmr for me atm
22:20:08 <AnMaster> ehird, so how far do you get in mycology now
22:20:22 <ehird> AnMaster: I was brb→ until a few seconds ago.
22:20:35 <ehird> AnMaster: I don't get any BADs, though ;-)
22:21:34 <ehird> Deewiant: http://img230.imageshack.us/img230/9440/snakeonlysf4.png
22:23:24 <fizzie> My iBook! You're saying it's dead! You monster!
22:25:53 <ehird> stackStack :: [[Int]]
22:25:57 <ehird> In which I cave in.
22:27:16 -!- tombom has quit ("Peace and Protection 4.22.2").
22:27:30 <ehird> Deewiant: You're right though, I doubt accessing the 458743957345th element of the stack is all too common
22:32:09 <ehird> In this document, short stacks are generally notated left to right to mean bottom to top. The leftmost values listed in the documentation are the bottommost and the first to be pushed onto the stack. Long stacks are notated top to bottom, to mean precisely that, top to bottom..
22:32:11 <ehird> I am very confused.
22:32:50 <ehird> Deewiant: Is it saying that depending on the size of the stack it reverses its terminology?!
22:33:13 <Deewiant> No, it's saying that depending on the size of the stack it's horizontal/vertical
22:34:58 <AnMaster> ehird, maybe you should learn to code befunge before you write an interpreter ;P
22:35:09 <AnMaster> writing an interpreter is a good way to learn it!
22:35:18 <ehird> Nothing to do with coding it, just the spec is worded oddly
22:35:31 <Deewiant> I wrote an interpreter for DOBELA and still can't anything besides hello worlds in it
22:35:42 <ehird> You accidentally the program.
22:35:51 <ehird> Deewiant: and you had problems with even hello world no? :P
22:36:09 <Deewiant> Implementing it was hard because the specs sucked :-P
22:36:21 <ehird> data Interpreter =
22:36:22 <ehird> stackStack :: [[Int]],
22:36:30 <ehird> Lah di di daaaaaah
22:37:24 <Deewiant> It /is/ optional, you don't have to do it
22:37:38 <ehird> Deewiant: I plan on implementing all MKRY fingerprints.
22:37:55 <ehird> In fact, all fingerprints possible.
22:38:24 <AnMaster> <Deewiant> I wrote an interpreter for DOBELA and still can't anything besides hello worlds in it <-- yes. But befunge is a rather different beast.
22:38:36 <ehird> stackStack :: [[Int]],
22:38:37 <ehird> instructionPointers :: [InstructionPointer]
22:38:43 <ehird> InstructionPointer {
22:38:45 <AnMaster> ehird, so once ATHR is finished you will implement it
22:38:45 <ehird> location :: Point,
22:38:57 <ehird> I need a fungespace too
22:39:31 <AnMaster> ehird, be aware of then (for nesting purposes, not for "is best"): MVRS > ATHR > threads by t
22:39:43 <ehird> AnMaster: I can rework this code later.
22:39:44 <AnMaster> for "is best" you clearly have ATHR > t > MVRS
22:39:47 <ehird> Haskell's nice like that
22:39:56 <AnMaster> ehird, sure you can for most languages.
22:40:01 <Deewiant> MVRS's official incarnation is kind of crappy, yeah
22:40:02 <ehird> stackstack is initially [[]] right?
22:40:17 <AnMaster> ehird, you are violating GPL3!
22:40:19 <Deewiant> ehird: No, it's repeat (repeat 0)
22:40:23 <AnMaster> since that is what it looks like in efunge
22:40:32 <ehird> AnMaster: Shiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
22:40:39 <ehird> Deewiant: Isn't there a way to tell if the stack's empty?
22:40:55 <Deewiant> Yes, there is, which is why it can't be that.
22:40:55 <AnMaster> ehird, I doubt that is enough code for it to be a violation though. To general
22:40:58 <ehird> http://pastie.org/463114.txt?key=uaxd3mxfem6b43an62w47a Hum di dum
22:41:10 <Deewiant> funktio complained about not being to able to initialize to (repeat 0) because of that.
22:41:14 <AnMaster> ehird, btw n clears all item on the TOSS
22:41:19 <ehird> Functional bastard :D
22:41:34 <ehird> "There are no multicharacter instructions in Funge."
22:41:40 <ehird> Tell that to XHELLOZ
22:42:05 <AnMaster> ehird, echnically efunge has support for typed memory btw. I have plans to make a fingerprint that requires typed funge space.
22:42:22 <AnMaster> I don't plan to reveal for what yet.
22:42:39 <Deewiant> Will you implement it in cfunge?
22:42:46 <AnMaster> Deewiant, in efunge, not in cfunge.
22:43:07 <AnMaster> Deewiant, what answer did you expect ;P
22:43:18 <ehird> I wonder if I should use a map of instructions
22:43:35 <AnMaster> ehird, if you plan to implement IMAP...
22:43:40 <Deewiant> Map makes it easier to do some things
22:43:44 <ehird> AnMaster: I. Can. Restructure :-P
22:43:49 <ehird> But yeah, I'll go for a map.
22:44:19 <AnMaster> ehird, so did you get any good ideas from hsfunge
22:44:28 <ehird> No, I'm not reading it for licensing concerns ;)
22:44:39 <AnMaster> ehird, you were, saying "OMG" and such
22:45:00 <ehird> Yes, NIH. As in I'd like to learn something,.
22:45:20 <ehird> "Egypt orders slaughter of all pigs over swine flu"
22:45:30 <ehird> Let's order the slaughter of all humans over swine flu
22:45:30 <AnMaster> ehird, you will have a hard time implementing most RCS fingerprints then
22:46:06 <ehird> AnMaster: I've always had the gift of being able to interpret gibbering retards, so I don't think I'll have a problem. MKRY isn't all that vague at all IMO.
22:46:16 <ehird> (Note: ↑ does not imply I think MKRY is a gibbering retard)
22:46:20 <AnMaster> ehird, not that fingerprint no
22:46:21 <Deewiant> ehird: It's not so much interpretation as lack of information.
22:46:41 <ehird> Deewiant: Yes, but it seems easy to deduce intent from what i've read
22:46:42 <Deewiant> There are N different ways to do something and you just have to know the intended one.
22:47:13 <ehird> Deewiant: I can handle *some* ambiguity...
22:47:16 <AnMaster> Deewiant, me too in that case.
22:47:33 <AnMaster> and TURT is insanely badly speced
22:47:43 <AnMaster> it doesn't even say if 0 degrees is up or down
22:47:59 <Deewiant> H 'Set Heading' (angle in degrees, relative to 0deg, east)
22:48:12 <ehird> Deewiant: It's from the guy who brought you funge-108
22:48:27 <Deewiant> AnMaster: TURT is from Cat's Eye.
22:48:31 <ehird> he probably thinks C is underspecced
22:48:36 -!- Quazie has changed nick to comex.
22:48:42 <AnMaster> # L 'Turn Left' (angle in degrees)
22:48:42 <AnMaster> # R 'Turn Right' (angle in degrees)
22:49:02 <Deewiant> ehird: Well, the idea of writing a disambiguated Funge isn't bad in theory since the spec does kinda suck in some places.
22:49:09 <ehird> Deewiant: It's fun.
22:49:15 <AnMaster> Deewiant, ah but which way do the bot start
22:49:20 <Deewiant> In practice I don't think it's smart since Funge is small.
22:49:33 <ehird> Deewiant: I mean, it's fun because it's esoteric and this kind of shit is the only reason I'm implementing it.
22:49:46 <Deewiant> AnMaster: So don't rely on it.
22:50:11 <ehird> direction = error "Demons fly out of your window, washing the windows API"
22:50:36 <AnMaster> Deewiant, and there are various other issues are we both know
22:51:00 <AnMaster> Deewiant, and it is hard to implement correctly
22:51:19 <Deewiant> I don't think TURT has any important spec issues.
22:52:12 <Deewiant> Some things could be explicit but implicitness doesn't matter, it can't be misinterpreted IMO.
22:55:44 <ehird> "the Funge interpreter should at least provide an option for informing the user that it was told to execute an instruction that isn't implemented"
22:55:55 <ehird> What code should be used? _
22:56:03 <ehird> It's just like the good old days in '93!
22:56:52 <AnMaster> you can't implement it without looking at an existing implementation.
22:56:54 <Deewiant> AnMaster: However, at least the Cat's Eye fingerprints say that they're under development.
22:57:03 <Deewiant> As in, you're not really supposed to implement them. :-P
22:57:07 <AnMaster> Deewiant, yes, but nothing happen.
22:57:29 <AnMaster> http://catseye.tc/projects/funge98/library/TOYS.html
22:57:59 <AnMaster> B ('pair of shoes') pops two cells off the stack and pushes the result of a "butterfly" bit operation.
22:58:00 <Deewiant> The only thing that can't be done is the butterfly op.
22:58:12 <AnMaster> Deewiant, as I just said yes ;P
22:58:13 <Deewiant> And actually, even that I did find by googling.
22:58:16 <ehird> Deewiant: how do you implement that one?
22:58:19 <Deewiant> But it's so obscure that I didn't trust it.
22:58:31 <Deewiant> Maybe because it's the only one I found ;-P
22:58:51 <ehird> If I read it the license gnomes will infect my heart.
22:59:11 <AnMaster> ehird, err ccbi -p is abstract algorithm info
22:59:26 <ehird> Does ccbi2 have it
22:59:32 <AnMaster> tar.us.to[0: 88.114.245.125]: errno=Connection refused
22:59:32 <AnMaster> fatal: unable to connect a socket (Connection refused)
22:59:45 <ehird> AnMaster: License. Gnomes.
22:59:57 <ehird> AnMaster: ccbi is broken on my system
22:59:59 <ehird> because ldc sucks.
23:00:12 <Deewiant> ehird: Didn't you pull the non-regex one?
23:00:30 <AnMaster> Deewiant, can you put it up to let me pull last
23:00:37 <ehird> Deewiant: Until I execute a program, IIRC
23:00:38 <Deewiant> AnMaster: Which one do you have?
23:00:46 <Deewiant> There is no 'last', it's in 10 or so branches currently
23:00:55 <Deewiant> ehird: ccbi -p doesn't execute a program!
23:00:58 <ehird> % bin/ccbi ~/Downloads/mycology/mycology.b98 [HANG]
23:01:18 <ehird> But why is it broken
23:01:33 <Deewiant> I don't know, I can't see into the mind of your computer.
23:01:52 <AnMaster> Deewiant, I know for sure you can serve git from plain http. I can accept the lower speed.
23:01:54 <Deewiant> AnMaster: git show, for instance
23:02:02 <AnMaster> commit 732e49a3401df2a1bd36221f8d767e00f8553a85
23:02:02 <AnMaster> Author: Matti Niemenmaa <matti.niemenmaa+git@iki.fi>
23:02:02 <AnMaster> Date: Mon Apr 27 20:05:15 2009 +0300
23:02:15 <ehird> 'B' pops y, then x, and pushes x+y, then x-y. This may or may not be the
23:02:16 <ehird> the "butterfly bit operation" requested.
23:02:17 <Deewiant> Yes, well, the first line would've been enough.
23:02:20 <ehird> How did you get that defn?
23:02:44 <Deewiant> Which is when I decided that it's probably right
23:02:44 <ehird> Deewiant: It doesn't seem very...well...bit operationy
23:03:35 <AnMaster> Deewiant, still connection refused.
23:03:38 <ehird> "FRTH" 0x46525448 Some common forth [sic] commands
23:03:51 <ehird> Deewiant: It's Forth.
23:04:01 <ehird> When being pedantic it helps to be right.
23:04:02 <Deewiant> ehird: It's not forth, though.
23:04:30 <AnMaster> Deewiant, Anyway! Why is it still timeout
23:04:34 <ehird> Deewiant: Chuck Moore calls it Forth. He is right.
23:04:46 <Deewiant> AnMaster: Because you haven't told me how I export the branch you want
23:05:01 <AnMaster> Deewiant, I said what git show showed
23:05:12 <AnMaster> Deewiant, I'm happy to follow the same branch
23:05:15 <Deewiant> But I don't know how to export only the main branch.
23:05:25 <AnMaster> Deewiant, well. Do you think I do
23:05:31 <Deewiant> If I give you everything you'll ask me how to switch branches etc. and I don't want to be your Git tutorial.
23:05:47 <AnMaster> Deewiant, I actually done that before
23:05:52 <Deewiant> AnMaster: No, but if you really want it it's your onus to figure it out, not mine.
23:06:09 <AnMaster> Deewiant, won't it default to the main one by default
23:06:11 <Deewiant> Or at least, you shouldn't ping me every few minutes.
23:06:19 <Deewiant> AnMaster: There is no "main" one in Git.
23:06:30 <ehird> Deewiant: What about >1 per sec
23:06:31 <AnMaster> Deewiant, surely you have a trunk
23:06:33 <Deewiant> The default one is just called "master".
23:06:39 <Deewiant> AnMaster: Yes, but Git doesn't know that.
23:06:48 <AnMaster> Deewiant, so won't I track master by default?
23:06:52 <ehird> Deewiant: Look at your irc server console
23:07:16 <ehird> Deewiant: I /ping'd you :P
23:07:20 <fizzie> (x, y) → (x+y, x-y) is the only butterfly operation I am aware of, too; the one used in FFT. I admit the "bit" word is a bit distracting there.
23:07:46 <AnMaster> since TOYS is inspired by INTERCAL
23:07:52 <ehird> fizzie: Sqqms rqght. Hqy, Q'm vqwqllqss qgqqn!
23:08:04 <AnMaster> ais523, does "butterfly bit operation" mean anything in intercal?
23:08:33 <ais523> it has a meaning in fast fourier transforms
23:08:40 <ais523> which is the one RC/Funge tries to implement
23:08:51 <AnMaster> ais523, the "bit" bit doesn't make sense there though
23:08:51 <ehird> Deewiant: What does FBBI do?
23:09:00 <Deewiant> ehird: FBBI implements only ROMA and NULL.
23:09:05 <AnMaster> ehird, iirc FBBI doesn't implement TOYS
23:09:23 <ehird> what's null? nop fingerprint?
23:09:50 <AnMaster> ehird, NULL and ROMA are trivial
23:10:43 <ehird> GOAL: Execute sanity.bf
23:11:03 <ehird> 0-0, space, ., # and @ must be implemented.
23:12:01 -!- BeholdMyGlory has quit (Remote closed the connection).
23:12:20 <Deewiant> AnMaster: Welp, I just made a tmpdir into which I pushed, that might've worked.
23:12:51 <fizzie> NULL is not exactly "nop" in the sense that it has observable effects: it does add a r-like instruction for all [A-Z]; I used it in fungot for swapping things with FING before replacing the parts that needed that with Z instead of X.
23:12:52 <fungot> fizzie: ' i don't know what. what's the good of having it all over again?' she said to herself, that she mightn't take his breath away: but, when he had satisfied himself that the flowers are always asleep.'
23:13:20 <AnMaster> Deewiant, http://pastebin.ca/1407889
23:13:22 <Deewiant> fizzie: It adds nop to everything, which justifies calling it a nop fingerprint, I guess.
23:13:37 <ehird> AnMaster: Please read the docs.
23:13:44 <fizzie> Deewiant: It adds a reflect; that's a strange thing to call a "nop". Well, I guess it depends on your viewpoint.
23:13:49 <Deewiant> AnMaster: RTFMessage then if you need it RTFManual.
23:13:50 <ehird> AnMaster: Then you don't get ccbi2.
23:13:56 <fizzie> Deewiant: But the official "nop" is not a reflect.
23:14:09 <AnMaster> Deewiant, so I do a clean checkout Right? Easier.
23:14:11 -!- FireFly has quit ("Later").
23:14:18 <ehird> AnMaster: Congrats; you just avoided learning something.
23:14:25 <Deewiant> AnMaster: Sorry, I killed the server.
23:14:48 <AnMaster> ehird, I don't plan to learn git
23:14:53 <AnMaster> Deewiant, tell me when you restarted it
23:15:08 <Deewiant> fizzie: Originally I actually thought it added a nop and not a reflection. A no-longer-existent version of Mycology expected that
23:15:16 <AnMaster> Deewiant, you are just silly now
23:16:19 <Deewiant> AnMaster: You don't think it's worth it to read 5 lines of error message and figure it out for yourself?
23:16:32 <ehird> Deewiant: It would be giving in to the enemy (git)
23:16:51 <ehird> Deewiant: Then he shall have to do without.
23:17:05 -!- coppro has joined.
23:17:08 <Deewiant> AnMaster: You only need to read the first 5.
23:17:10 <ehird> Indeed, a libc is but a speck compared to the horrors of learning one command of git.
23:17:35 <AnMaster> try again (e.g. 'git pull <repository> <refspec>').
23:17:56 <ehird> AnMaster: Did you lose your English lobe overnight?
23:18:06 <ehird> Deewiant: does each IP have its own stack of instruction maps?
23:18:08 <Deewiant> Clearly you're on a different branch than the remote you pulled from.
23:18:15 <Deewiant> Since it asks which branch to merge with.
23:18:28 <AnMaster> Deewiant, ok. I'm used to separate directories like for svn. And bzr.
23:18:41 <AnMaster> Deewiant, so I don't get what this error mean. Nor how to correct it.
23:18:45 <ehird> "instructionMapStack". What an unwieldy name.
23:19:06 <AnMaster> ehird, opcode_stack in cfunge.
23:19:23 <AnMaster> Deewiant, and which one do I want to answer here
23:19:28 <Deewiant> AnMaster: It's saying that you're on a different branch and it wants to know what branch to merge with.
23:19:41 <AnMaster> Deewiant, and I don't know which one you recommend.
23:19:42 <ehird> Hmm, maybe a stack is bad, if you load 4 fprints it'll take 4 "tail"s to get to a core instruction.
23:20:21 <Deewiant> Not sure if that's correct, but that's what I'd try first.
23:20:24 <AnMaster> Deewiant, so git pull git://tar.us.to/ HEAD
23:20:53 <Deewiant> It's not like it'll rm -rf $HOME if you do it wrong.
23:20:57 <AnMaster> Deewiant, seems you didn't kill it at all. How nasty to claim you did it.
23:21:00 <ehird> Ooh that felt good.
23:21:20 <Deewiant> AnMaster: I did, but I restarted it when I realized that you need it to pull changes again.
23:21:39 <AnMaster> Deewiant, any reason why you don't publish it over plain http
23:21:45 <AnMaster> it seems rather closed development model
23:22:08 <ehird> AnMaster: how many non-deewiant committers are there?
23:22:11 <ehird> none? Yes, that sounds closed.
23:22:11 <Deewiant> Originally I didn't know it was possible and since then I haven't bothered to set it up
23:22:25 <Deewiant> ehird: IIRC you offered a patch but didn't bother ;-)
23:22:52 <Deewiant> AnMaster: So did the pull work?
23:24:22 <ehird> Deewiant: thoughts re: stack?
23:24:45 <Deewiant> ehird: I didn't get your thoughts. My suggestion is the YAGNI development model.
23:24:49 <AnMaster> Deewiant, why should I tell you when you act like that :P
23:24:56 <ehird> Deewiant: 23:19 ehird: Hmm, maybe a stack is bad, if you load 4 fprints it'll take 4 "tail"s to get to a core instruction.
23:25:00 <Deewiant> AnMaster: Well, it's your loss if it didn't work.
23:25:03 <AnMaster> Deewiant, anyway bzr is a lot more logical here. Acts more like you expect.
23:25:05 <ehird> Deewiant: That's not quite YAGNI as This May Slow Down Everything.
23:25:13 <ehird> AnMaster: FAMILIARITY IS NOT INTUITIVENESS!
23:25:15 <AnMaster> Deewiant, it seems to have worked
23:25:21 <ehird> By the same argument, AnMaster, Windows is more logical than Linux.
23:25:53 <AnMaster> ehird, No. bzr is closer to what you expect from a file system and from a static web server
23:25:59 <AnMaster> one resource per directory/file
23:26:09 <ehird> AnMaster: You're on crack, sir, and don't understand git at all.
23:26:17 <ehird> Because you come from the context of another VCS.
23:26:28 <ehird> This is not git's fault.
23:26:30 <AnMaster> ehird, I don't want to understand such an abomination as git no.
23:26:41 <Deewiant> I actually always expected branches to live in one directory and was surprised that VCSs don't do it
23:26:55 <ehird> Second by second you make me less likely to ever try and help you with anything ever again, AnMaster.
23:26:56 <Deewiant> SVN in particular struck me as really messed up in terms of layout
23:27:00 <AnMaster> ehird, I fully support google going for mercurial. I don't like hg. But I can live with it.
23:27:07 <AnMaster> it isn't too broken in semantics.
23:27:12 <ehird> But git eats babies after fucking them, yeah, yeah, STFU.
23:27:40 <AnMaster> Deewiant, yes. branches/tags/trunk is kind of messy in svn
23:27:53 <AnMaster> it is trying to be semi-CVS compatible
23:28:43 <ehird> I wonder how I should tell instructions what IP they're executing on.
23:28:50 <AnMaster> Deewiant, also I managed to get by with simple git pull from several kernel.org projects
23:28:54 <ehird> Interpreter -> InstructionPointer -> IO (Interpreter,InstructionPointer), maybe.
23:29:02 <ehird> AnMaster: Yes, it's HIS fault. Obviously.
23:29:16 <Deewiant> AnMaster: Well, you've been pulling from my development state which means it wasn't on the master branch.
23:29:20 <ehird> AnMaster: lolllllllllll
23:29:23 -!- Gracenotes has quit (Remote closed the connection).
23:29:28 <ehird> Deewiant: feel bad about yourself!
23:29:35 <Deewiant> AnMaster: Why on earth should I have told you?
23:29:43 <Deewiant> You can find out trivially by asking git
23:29:48 <AnMaster> Deewiant, I just followed your instructions. I would assume that to lead to a sensible state.
23:30:05 -!- Gracenotes has joined.
23:30:12 <Deewiant> I recall being badgered by people to serve CCBI2
23:30:20 <Deewiant> While I was in the middle of writing something for it
23:30:24 <AnMaster> Deewiant, what? git info? No such thing. I tried many commands that seems kind of common to many VCS, including svn, bzr, darcs, hg. None of them worked.
23:30:42 <ehird> AnMaster: what you get is what you get and it's not gonna work
23:30:46 <AnMaster> Deewiant, yes but didn't find what I wanted there.
23:30:54 <Deewiant> ehird: s/not gonna work/no good whining/
23:31:01 <ehird> Deewiant: I was guessing
23:31:07 <AnMaster> Deewiant, I prefer ehird's interpretation
23:31:09 <Deewiant> AnMaster: branch List, create, or delete branches
23:31:13 <Deewiant> AnMaster: status Show the working tree status
23:31:19 <Deewiant> AnMaster: show Show various types of objects
23:31:20 <ehird> AnMaster: Oh fuck off if you're going to be so ungrateful to Deewiant
23:31:24 <Deewiant> AnMaster: log Show commit logs
23:31:26 <ehird> he has NO OBLIGATION to give you a ccbi2 repository
23:31:27 <AnMaster> Deewiant, how do I show what branch I pulled from
23:31:51 <Deewiant> I don't know if it even remembers that
23:32:18 <AnMaster> the darcs one was similar iirc
23:32:23 <Deewiant> I think you have to do an explicit 'git remote add' to save the URL.
23:32:42 <Deewiant> Read the manpages for git-remote, I don't know about that.
23:32:43 <AnMaster> Deewiant, I just had to do git pull before without giving the url
23:33:32 <AnMaster> ah darcs was slightly different indeed:
23:33:58 <AnMaster> git may be good internally. I'm not arguing against that.
23:34:20 <AnMaster> But it need a better user interface
23:34:21 <ehird> Deewiant: The stack stack is initially filled with infinite (infinite 0s), right?
23:34:35 <ehird> AnMaster: I daresay the problem is with you.
23:34:49 <Deewiant> ehird: 2009-04-30 00:40:19 ( Deewiant) ehird: No, it's repeat (repeat 0)
23:34:49 <ehird> Me and Deewiant don't seem to have any issues with gi
23:34:51 <AnMaster> ehird, and with 95% of the other developers.
23:35:00 <ehird> AnMaster: Yes, because nobody at all uses git.
23:35:11 <ehird> It is not a very popular DVCS, if not the most popular.
23:35:17 <Deewiant> Ulrich Drepper uses git so it must be good
23:35:17 <ehird> It is not gaining large amounts of traction.
23:35:27 <ehird> In fact, everyone hates it.
23:35:36 <ehird> Birds shit on Linus just for making git.
23:35:41 <AnMaster> Deewiant, why. You are a fanboy of him?
23:35:59 <Deewiant> ehird: You broke his sarcasm detector
23:36:03 <AnMaster> Deewiant, I know he is glibc maintainer. I think glibc is bloated.
23:36:04 <ehird> Deewiant: It never worked.
23:36:20 <ehird> Deewiant: Along with his Sense of Humour Lobe and the Intelligence cortex.
23:36:22 <AnMaster> ehird, it did before I met you
23:36:23 <Deewiant> AnMaster: Maybe you know that people in general think he's batshit insane
23:36:36 <ehird> Deewiant: He's right you know, though. Everyone who likes git is inherently a fanboy.
23:36:54 <AnMaster> Deewiant, I haven't really tracked the general opinions about him. So I have no clue if that is sarcasm or not.
23:37:47 <AnMaster> Deewiant, it's like ais523 a few days ago (yesterday? or day before that?). He quoted jwz. He didn't know jwz was one of the developers of Netscape
23:38:00 <ehird> He just didn't know that jwz meant Jamie Zawinski.
23:38:09 <ehird> I was under the impression you did know
23:38:14 <ehird> It's not exactly common knowledge
23:38:18 <ehird> Unlike ulrich drepper
23:38:19 <AnMaster> ehird, read logs. Before you talk
23:38:27 <ehird> That sounds like a metal guitarist's name btw
23:38:28 <Deewiant> I'm not sure how that is relevant to anything at all?
23:38:30 <ehird> AnMaster: Oh fuck off.
23:38:44 <AnMaster> ehird, you always complain when I don't read fully or misunderstood
23:39:01 <ehird> Tip of the Day: On the list of things I care about, AnMaster whining that I'm so unfair to him comes in rock bottom.
23:39:27 <AnMaster> ehird, I'm not "whining". I'm just stating a fact.
23:39:41 <ehird> Mm. Yes. You're stating a fact in an incredibly whiney way.
23:39:58 <ehird> "NO I'M NOT BEING WHINY! HOW DARE YOU ACCUSE ME OF BEING WHINEEEEEEEEEY"
23:40:15 <Deewiant> I think this is a good opportunity to sleep for some 10 hours
23:40:47 <AnMaster> ehird, and I really regret I provided that description of how cfunge bounds work to you.
23:41:06 <ehird> AnMaster: If you didn't have I could always have read the source and wantonly violated the GPL3.
23:41:22 <AnMaster> ehird, well you could. I doubt you would have.
23:41:33 <ehird> AnMaster: You think I care about copyright law...?
23:41:49 <AnMaster> ehird, No. I think you would have reacted in same way as for hsfunge source.
23:42:03 <ehird> I've read cfunge source before.
23:42:07 <ehird> It's shit code but I can read shit code.
23:42:14 <AnMaster> ehird, not since I started using inline asm.
23:42:29 <ehird> I can read asm, albeit slowly.
23:42:30 <AnMaster> I guess it isn't your preferred indention style.
23:42:35 -!- ais523 has quit (Remote closed the connection).
23:42:50 <ehird> AnMaster: your code would give k&r a heart attack
23:43:05 <AnMaster> ehird, sure. I'm using C99 as you very well know.
23:43:21 <AnMaster> ehird, and I'm using a different indention style yes.
23:43:36 <AnMaster> I don't like the k&r indention style at all
23:50:48 <AnMaster> ehird, your use of that word all the time makes no sense.
23:51:03 <AnMaster> I guess it is some compulsive behaviour.
23:51:10 <ehird> That's funny, because I use it to exactly mark you missing the point entirely.
23:51:17 <ehird> Hardly surprising, then.
23:51:25 <ehird> You miss your missing of the point.
23:51:54 <AnMaster> ehird, seems highly non-standard. Checked urban dict.
23:52:24 <ehird> It is onomatopoeic
23:52:40 <ehird> "joke going over your head" is highly idiomatic.
23:52:54 <ehird> "Whoosh" is the internet-created shortening.
23:53:19 <AnMaster> ehird, "going over your head" I know of yes.
23:53:27 <AnMaster> But how does it end up as "whoosh"
23:53:39 <ehird> AnMaster: I see nothing large has ever gone over your head.
23:54:14 <ehird> or you were in a vacuum at the time
23:55:30 <AnMaster> ehird, you still doesn't make sense. I guess you are trying (and failing as usual) to be funny
23:55:47 <ehird> I'm forced to assume that if you
23:55:48 <AnMaster> I wonder how ais523 can stand you. It's strange.
23:55:56 <ehird> 've really never heard the "whoosh" sound of air as something goes over your haed,
23:56:01 <ehird> you live in a faraday cage.
23:56:17 <AnMaster> ehird, you mean low flying bird for example?
23:56:28 <ehird> Yes. If that's the only thing that flies in the air in Sweden.
23:56:42 <AnMaster> ehird, aircrafts usually fly a lot higher
23:57:01 <AnMaster> at least I don't stand under one a few meters above
23:58:07 <AnMaster> ehird, anyway it isn't called "whoosh" in Swedish. I don't think I ever heard a name for it.
23:58:38 <AnMaster> the wind. then the verb is "viner"
23:59:11 <AnMaster> or for the noun, "ett vinande ljud" (a whooshing (from wind) sound I guess)
23:59:47 <AnMaster> ehird, you seem to assume everyone have near native English knowledge.