00:20:33 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | From when they take effect..
00:21:05 -!- FireFly has quit (Quit: swatted to death).
00:26:22 * Sgeo is watching Firefly
00:33:54 <zzo38> A chess variant about "Nobody expects the Spanish Inquisition" should be invented.
00:35:01 <coppro> the spanish inquisition is kept in pocket
00:35:04 <coppro> it has the moves of a knight
00:35:24 <coppro> you can places it only if it attacks two or more non-pawn pieces or delivers check
00:36:12 <zzo38> Yes, OK. I like that idea.
00:36:35 <zzo38> My idea too was it is kept in pocket initially
00:37:14 <zzo38> Then it is something like Pocket Knight chess.
00:45:45 <Sgeo> Whatever happened to Fourplay?
00:46:07 <zzo38> Sgeo: What is Fourplay?
00:46:13 <Sgeo> I think I linked to that idiot's blog who mentioned Fourplay a while ago
00:46:20 <Sgeo> Fourplay itself however is nt by idiots
00:46:26 <Sgeo> It's a nomicchess
00:46:35 <Sgeo> http://www.nomic.net/deadgames/fourplay/fourplay.html
00:46:41 <Sgeo> Probably not balanced, really
00:50:39 <tswett> zzo38: LaTeX and Microsoft Word can be compared?
00:51:38 <zzo38> tswett: Are you at Stanford University?
00:52:08 <tswett> No. I just know how to look the look.
00:58:26 <elliott> zzo38: He's in Stanford IN THE FUTURE.
01:06:51 -!- quintopia has quit (Ping timeout: 255 seconds).
01:07:49 -!- elliott has quit (Read error: Connection reset by peer).
01:07:56 -!- elliott has joined.
01:15:40 <elliott> Progress(TM): gcc have removed the option to inhibit warnings about #import without using -w, but not #import itself (despite wanting to deprecate it since 2003).
01:15:49 -!- pikhq_ has quit (Read error: Operation timed out).
01:16:14 -!- pikhq has joined.
01:17:38 <elliott> (Oh well, -Wno-deprecated works.)
01:25:43 -!- quintopia has joined.
01:25:43 -!- quintopia has quit (Changing host).
01:25:43 -!- quintopia has joined.
01:25:48 <zzo38> My TeX chess program is not only for FIDE chess. So after I finished I can post to Chess Variants, too, since you can also use it for other games, not only for FIDE chess.
01:38:29 -!- quintopia has quit (Quit: Lost terminal).
01:38:49 -!- quintopia has joined.
01:39:27 -!- jcp has quit (Ping timeout: 252 seconds).
01:46:41 -!- jcp has joined.
01:55:43 -!- augur has quit (Remote host closed the connection).
02:05:37 -!- cheater00 has joined.
02:13:10 -!- augur has joined.
02:43:27 <elliott> Vorpal: you know you ported c-intercal to mac os 9?
02:43:41 -!- copumpkin has quit (Ping timeout: 240 seconds).
02:43:47 <Sgeo> It came up again
02:43:51 <Sgeo> The mysterious error
02:44:08 -!- copumpkin has joined.
02:44:22 * Sgeo clicks Cancel in the hopes that whatever debugger comes up will give him a hint
02:44:38 <elliott> Vorpal: I was thinking that it would be a lot easier if one simply wrote Cygwin for Mac OS.
02:44:40 <elliott> Vorpal: SO CLEARLY IT MUST BE DONE.
02:45:02 <elliott> And it must work on System Software 6, obviously. To become the best operating system!
02:45:04 <Sgeo> Isn't OS X a UNIX?
02:45:19 <Sgeo> Derivative of BSD, iirc?
02:45:38 <Sgeo> Or am I flat out mistaken? (Or somewhere in the middle)
02:46:04 <Sgeo> No, I am not trying to make that into a list containing merely data and placing the ' on the wrong side
02:47:42 <Sgeo> Nlo debugger came up
02:49:29 <elliott> zzo38: But older versions are not.
02:49:31 <zzo38> It even has a few GNU programs.
02:49:41 <elliott> It comes with gcc and Emacs, for one.
02:49:49 <elliott> And of course it's quite easy to compile your own.
02:49:55 <elliott> Well, comes with; gcc is part of Xcode.
02:50:08 <zzo38> Yes, I know those things.
02:50:37 <Sgeo> Is elliott ignoring me again?
02:51:17 <variable> Sgeo: mac is a derivative of mach and BSD
02:51:18 <zzo38> Sgeo: Maybe. I read your messages though.
02:51:36 <variable> and it has been certified as UNIX
02:51:40 <elliott> variable: Heavens, it's much more fun than that!
02:52:06 <variable> elliott: well yeah - it has its own stuff as well. It was mainly BSD + Mach + Cool looking shell
02:52:09 <elliott> variable: Mac OS X's kernel is XNU. XNU is 4.3BSD updated to FreeBSD, *running on top of Mach*.
02:52:49 <elliott> variable: But then drivers are written in a subset of C++ separately... for no apparent reason :)
02:53:02 <pikhq> Except not quite, as Win32 is merely an API implemented on top of NT.
02:53:09 <elliott> pikhq: No; Win32 is a subsystem.
02:53:18 <pikhq> Whereas XNU has an entire BSD kernel running on top of Mach.
02:53:20 <elliott> OK, so it's the Windows subsystem with the Win32 API.
02:53:23 <variable> elliott: because c++ has saner type safety and allows one to use nicer features like classes without exposing implementation
02:53:45 <elliott> variable: (1) I mean, why not use the BSD driver layer, and (2) I don't buy OOP propaganda :)
02:53:56 <elliott> Rather: I think that the goal of abstraction is obviously an admirable one, but I don't believe OOP is the best way to do it.
02:54:00 <variable> elliott: I wasn't think OOP (which has little to do with classes)
02:54:01 <pikhq> elliott: The OS X setup is more analogous to User Mode Linux than anything else, really.
02:54:15 <Sgeo> variable, ...saner than C, I assume. I assume that other languages are out of the question, otherwise "C++ is sane" is.. silly
02:54:24 <elliott> pikhq: Not quite. Mach is *designed* to have kernels run underneath it.
02:54:33 <elliott> pikhq: I mean, what do you think Hurd runs on top of?
02:54:54 <pikhq> Yes, but it's also designed to have those kernels actually *use* Mach.
02:55:11 -!- augur has quit (Remote host closed the connection).
02:55:17 <pikhq> Instead of, uh, using it as a pointless abstraction layer.
02:55:38 <elliott> variable: It doesn't even support USB! :)
02:55:59 <elliott> pikhq: It isn't *quite* pointless.
02:56:05 <Sgeo> Surely if Linux never existed, Hurd would be further along...
02:56:06 <elliott> pikhq: Mach gives the OS soft real-time support, for one.
02:56:16 <elliott> pikhq: And more importantly, the drivers run on top of Mach instead of BSD.
02:56:23 <elliott> = SPEEEEEEEEEEEEEEEEEEEEEEEEEEEED.
02:56:40 <elliott> Also, a lot of OS X drivers run in user-space, which would be "fun" to do with BSD.
02:57:12 <pikhq> Well, Linux actually has quite a few user-space drivers by now...
02:58:04 <elliott> Not in the 90s when nextstep was created :)
02:58:37 <elliott> pikhq: Anyway, RATE MY IDEA OF WRITING CYGWIN FOR MAC.
02:59:47 <elliott> pikhq: AFAIK microcosm can't actually run anything yet :P
03:00:13 <pikhq> It can run a few trivial things...
03:00:36 <elliott> pikhq: And I VERY MUCH DOUBT microcosm would work on a platform without stdin/out, with no more than ANSI C (unless you wrote Mac OS Classic-specific code!), ...
03:00:46 <elliott> pikhq: ... I doubt it will yield to other processes, ...
03:00:58 <pikhq> Oh, Mac OS Classic?
03:01:12 <pikhq> Yeah, that'd need to be a full kernel.
03:01:15 <elliott> pikhq: Mac OS X needs no Unix :P
03:01:23 <elliott> pikhq: What does Microcosm target, anyway?
03:01:27 <elliott> If it only works on other Unices it's a bit useless.
03:01:37 <pikhq> elliott: Any vaguely modern OS.
03:01:54 <elliott> Not far away from Windows :trollface:
03:01:56 <elliott> pikhq: Also, not necessarily a kernel; all Unix functions could yield.
03:02:02 <elliott> pikhq: Then only tight loops would lock up.
03:02:11 <elliott> pikhq: DOS has this problem, for instance :)
03:02:28 <elliott> pikhq: It'd be NICE to multitask with tight loops, but...
03:02:41 <elliott> pikhq: And I think you could hook a "stop it!" key right to the OS to terminate them.
03:05:16 <elliott> pikhq: Also, a Unix with only 8 megs of ram: FUN?
03:05:50 <pikhq> You can manage that without any real effort.
03:06:00 <pikhq> It just isn't going to be doing anything too fancy.
03:06:05 <elliott> pikhq: I WANT TO RUN GCC ON IT
03:07:14 <elliott> pikhq: djgpp ran with only 2 megs of ram at the start...
03:07:26 <elliott> pikhq: I could just port gcc 2 :)
03:07:57 <pikhq> Okay, you could probably get older GCC working fine.
03:08:14 -!- BeholdMyGlory has quit (Remote host closed the connection).
03:08:18 <pikhq> *Modern* GCC can suck up gigs if you look at it wrong.
03:09:26 <elliott> pikhq: Seriously though, Mac OS is PERFECT for this.
03:09:35 <elliott> pikhq: You basically run as your own OS with a large graphical API.
03:13:19 <elliott> pikhq: In fact, if I implemented my own timer, I could do proper multithreading...
03:13:45 <elliott> pikhq: Awesomely though, every Unix application would stop if you tabbed away.
03:14:03 <pikhq> Hah, forcing Mac OS into proper multithreading.
03:14:57 <elliott> pikhq: I dunno if it's worth it though :p
03:28:30 <lambdabot> [3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,...
03:41:22 <Ilari> Hmm... Did British Secondary Prevention Trial inadvertently end up reproducing Lyon Diet-Heart stydy (one of the most successful dietary trials ever) in reverse?
03:53:12 -!- augur has joined.
03:54:08 -!- augur has quit (Remote host closed the connection).
03:58:10 <zzo38> Ilari: What is that?
04:00:38 -!- oerjan has joined.
04:05:24 -!- augur has joined.
04:09:53 -!- variable has quit (Ping timeout: 260 seconds).
04:12:44 <oerjan> elliott: i see you have discussed optbot more. i'd like to repeat/point out that my basic executive decision in this case is that if optbot is to change the topic, it _must_ look at the topic that is already set, and preserve what others have put there. i suggest changing only the part after the last |.
04:12:44 <optbot> oerjan: What does `ls' do on VMS
04:12:56 <elliott> oerjan: when was that your executive decision?
04:13:02 <elliott> I don't recall you ever saying that, not in 2008 and not in 2011
04:13:12 <elliott> and that was never optbot's behaviour
04:13:12 <optbot> elliott: what do you think about that?
04:13:25 <elliott> not according to herobrine
04:13:36 <oerjan> well it's what i _thought_.
04:14:20 <elliott> well, i'll just take it down then, since it'd be a huge pain to make the code do that, and also generate really clipped topics because of the limit
04:15:13 <oerjan> ok i may have also allowed an actual optbot command that other people than you can use.
04:15:26 <elliott> oerjan: that'd still clip topics
04:15:57 <oerjan> i'd point out that both options are considerably more lenient than what some other people suggested.
04:16:14 <zzo38> Another option is just remove the topic changing.
04:16:15 <elliott> some other people = Gregor
04:16:40 -!- azaq23 has joined.
04:16:45 <elliott> where's your sense of fun gone i ask!
04:17:37 <zzo38> Another another option is that it changes immediately if the log URL is missing, otherwise it changes it seven days after the previous change (including changes by other users).
04:18:23 <oerjan> zzo38: it appears that elliott does not think actually paying attention to TOPIC commands is a reasonable change.
04:18:33 <elliott> It's not about reasonability, it's about feasibility.
04:18:54 <elliott> The topic limit <<< the message limit, and optbot does messages, + the overhead it has now is about max before it sarts chopping unreasonably
04:18:54 <optbot> elliott: http://www.google.com/search?q=x-directory&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-GB:unofficial&client=iceweasel-a#hl=en&client=iceweasel-a&rls=org.mozilla:en-GB:unofficial&&sa=X&ei=b6bZTJbjCOaV4gbn2qS8CA&ved=0CBEQvgUoAA&q=%2B%22application/x-directory%22&nfpr=1&fp=8ec2bc6eb6ca6ae0
04:19:01 <zzo38> Then, just make it never change the topic message.
04:19:12 <elliott> that's 50% of the fun of optbot, so i'd rather take it down.
04:19:12 <optbot> elliott: i meant to run it in the Integer -> Program direction, fwiw
04:19:27 <oerjan> elliott: well _most_ irc messages aren't that long.
04:19:38 <elliott> oerjan: no, but the topic limit is quite a bit less.
04:19:46 <elliott> and it has a hundred or so overhead.
04:20:19 <oerjan> i think the message limit including nick prefix is 510 or so
04:20:41 <zzo38> Or change the message only when "PRIVMSG optbot :TOPIC" is received, is another way.
04:20:46 <optbot> zzo38: But seriously, nobody would pay that much for a drive.
04:21:03 <zzo38> Do it for free then.
04:22:06 <oerjan> elliott: hm a suggestion, put a link to tunes in your own logs, then we can remove tunes from the topic.
04:22:32 <elliott> oerjan: that might work, but does not solve the issue that user-added topics will be far longer anyway.
04:22:32 <oerjan> (you may have already done so afaik, i only visit individual dates)
04:23:16 <zzo38> No I don't like it, I think tunes log should always be available there it is the official policy.
04:23:53 <oerjan> zzo38: um the official policy surely is that _some_ logs must be there if there are any...
04:24:12 <zzo38> You could delete the word "logs:" and write "http://tunes.org/~nef/logs/esoteric/?C=M;O=D | http://208.78.103.223/esoteric/ | " as the prefix.
04:24:19 <zzo38> Since "logs" is in the URL.
04:24:25 <elliott> it doesn't need logs in the URL.
04:26:16 <zzo38> elliott: It doesn't, but since "logs" is already in the URL, the prefix "logs: " can be removed. However, I (and others) still think you should make it optbot not changing topic message.
04:40:19 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | I KNOW.
04:42:41 <oerjan> ok current compromise is a 12 hour timeout after anyone changes the topic.
04:42:49 -!- zzo38 has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | I DON'T KNOW.
04:43:25 <zzo38> oerjan: That might work. Keeping track of the last change of *anyone* is important.
04:44:12 <oerjan> if/when something longer is needed, either elliott or an op should be able to take action in that time.
04:44:35 <elliott> compromise for when i'm not lazy, mind you.
04:44:57 <oerjan> (i'm just summarizing mine and elliott's private conversation so it's official)
04:45:05 <zzo38> elliott: OK, then for now just disable that feature until you get less lazy to be able to correct it.
04:45:06 <elliott> we don't have private conversations
04:45:07 <elliott> what are you talking about?
04:45:24 <elliott> zzo38: nah; if someone really has something desperately urgent to put in the topic in the next few days, oerjan or fizzie can +t
04:45:32 <elliott> it's a mythical situation, but just in case :)
04:45:36 -!- augur has quit (Read error: Connection reset by peer).
04:45:36 <elliott> then I'll fix it, because I'll be less lazy.
04:45:41 <oerjan> elliott: it was done by telepathy. your consciousness may not have perceived it yet.
04:45:53 -!- augur has joined.
04:46:08 <oerjan> expect strange dreams.
04:46:20 * elliott backs slowly away from oerjan
04:46:24 <elliott> you weren't joking about the wet furry porn
04:46:38 <zzo38> I often have a strange dream
04:47:07 <zzo38> A lot (not all) of them, is ones with no words (or picture) to explain.
04:48:14 <zzo38> And what kind of reason is there in such dreams why Achilles cannot catch the Tortoise?
04:48:36 <quintopia> where you have to get on a conveyor belt and lie down
04:48:57 <quintopia> and it just does all the airporty stuff while you lay there and wait for it to take you to your plane
04:49:55 <quintopia> it was poorly designed in the dream, but the basic premise would be nice
04:50:47 <zzo38> It would work for tired people with no packages, no money, and no hunger.
04:51:12 <oerjan> no packages are fine, they can also go on the belt.
04:51:24 <zzo38> It might work a little bit in some other cases too, possibly, but I don't know quite.
04:52:00 <zzo38> Do they need to put the separator bars so you know whose package it is?
04:52:03 <elliott> `addquote <zzo38> It would work for tired people with no packages, no money, and no hunger.
04:52:58 <oerjan> also you could solve the hunger part by having some of those things from manufacturing belts that they use to fill food packages
04:53:03 <zzo38> Separator bars might help a little bit in current airport, but then barcodes would also be required. Otherwise you always lose all packages
04:53:21 <oerjan> you better be positioned right on the belt when that happens
04:53:42 <elliott> quintopia: do the tsa just run up to you and grope you?
04:53:48 <elliott> "this won't take a second."
04:53:49 <zzo38> oerjan: Still, if you are tired you might prefer to be lying on the conveyor belt. Other people who are not tired might prefer to stand up.
04:54:20 <oerjan> elliott: hey they could give an actual MRI scan then. would solve that bombs in your cavities problem.
04:54:30 <zzo38> Or else, sit down on the chair and go in the plane from walking outside.
04:54:34 <elliott> I've got a bomb in my cavity hur hur I don't even know what that would mean
04:54:37 <Sgeo> I posted a dream of mine online a while ago
04:54:41 <oerjan> CLEARLY THIS IS THE FUTURE
04:54:44 <Sgeo> Ok, so that's 3 dreams
04:54:47 <zzo38> But lying down on conveyor belt would help if you are tired.
04:54:48 <Sgeo> http://www.dreamviews.com/f107/sgeos-dream-journal-53892/
04:54:48 <Sgeo> And http://www.dreamviews.com/blogs/sgeo/alternate-universe-elevator-1827/
04:55:04 <oerjan> elliott: disturbingly, it can mean exactly what it sounds like.
04:55:16 <elliott> oerjan: you mean a thing that can blow up in my cavity?!?!?!?!
04:55:41 <elliott> oh wait, just got another telepathic communication from you. jesus christ. well, now i know what "i've got a bomb in my cavity" could mean.
04:55:46 <elliott> not sure i want to though.
04:56:16 <quintopia> elliott: ROBOTS DO THE GROPING. THEY HAVE NO FEELING. THEY WON'T POP A BONER.\
04:56:44 <elliott> quintopia: but do they handle the recipient ``popping a boner'' to use your crude American terminology?
04:56:49 <oerjan> mind you an MRI scan would give new meaning to the "please remove metal objects" part
04:57:02 <zzo38> Is HackEgo broken again?
04:57:19 <elliott> maybe the airport could just have a total nudity rule.
04:57:35 <quintopia> elliott: i'll leave it to you to design a robot that can tell the difference between a pipe bomb and
04:58:17 <oerjan> elliott: i'm sure i've seen approximations to that joke (well not the goatse) in several comics, including dilbert.
04:58:31 <oerjan> actually, maybe even the goatse.
04:58:44 <zzo38> Regardless of how they do it, I no longer go on airplanes anyways; I have already for a few years, decided to stop using airport.
04:58:46 * oerjan is not quite sure about that.
04:59:51 -!- azaq23 has quit (Read error: Operation timed out).
05:00:50 <elliott> zzo38: have you ever been on an aeroplane?
05:02:55 <zzo38> elliott: Yes I have once. But even then, and even more now, they do all stupid things with the surveillance, lost packages, prohibition of items on planes, and various other things.
05:03:10 <elliott> zzo38: i think they shouldn't prohibit bombs on planes. What do you think?
05:03:22 <zzo38> So, just go by boat if you need to go across the ocean.
05:04:20 <elliott> zzo38: i prefer to go overseas by train
05:04:25 <zzo38> elliott: I'm not sure, but I meant other things, such as nail files, toner cartridges, pencil and paper, and a lot of others.
05:04:34 <zzo38> elliott: Is there a train in the water?
05:04:47 <elliott> It exists, in your imagination
05:05:27 * oerjan wonders if the dilbert one may even have been pre-2001
05:05:57 <oerjan> i mean it's precisely the kind of dystopian idea scott adams _would_ get
05:06:12 <elliott> trying to remember why i hate scott sdams
05:06:49 <oerjan> elliott: i assume his comics have degraded like most long-time runners, and now he sucks?
05:06:52 <zzo38> I had a dream once, there was various clubs in a building, for the pokemon element types, and also the administration club. The administration club was not permitted to have any members.
05:06:55 <elliott> oerjan: no no unrelated to his comics
05:07:06 <elliott> oerjan: he posted some ridiculously stupid anti-scientific thing to his blog once so i started hating him
05:07:11 <oerjan> ah. some absurd opinion...
05:07:30 <elliott> oerjan: is that you going throough the list of things i mate hate people for?
05:08:08 <elliott> http://dilbertblog.typepad.com/the_dilbert_blog/2007/08/poster-child-fo.html ;; well here he misinterprets "atheist" and is stupid about it.
05:08:22 <oerjan> <elliott> zzo38: i prefer to go overseas by train <-- underseas, you mean
05:08:56 <zzo38> elliott: It probably is not a train, even if they call it a train, it isn't.
05:09:10 <elliott> there's railtracks under the water, it grows legs to connect to them
05:09:43 <oerjan> zzo38: i am now going to annoy elliott by pointing out that he is joking.
05:09:44 <zzo38> Each pokemon element type club has two ways of being a member (and no payment is required).
05:10:13 <elliott> oerjan: <elliott> It exists, in your imagination
05:10:28 <oerjan> elliott: also not a list, i vaguely recall you mention that adams thing before
05:10:45 <elliott> oerjan: not interesting enough to donate brain cpu cycles to :D
05:10:57 <oerjan> elliott: this is not cpu, it's memory
05:11:10 <elliott> oerjan: it hogs cpu cycles to access memory and process it
05:11:13 <zzo38> O, in your imagination, you can make it whatever you want including illogical things in seventeen thousand dimensional space with one is not both the same.
05:11:18 <elliott> thinking is much easier than remembering :)
05:11:30 <elliott> remembering is less "work".
05:11:43 <elliott> oerjan: what i'm saying is, use Checkout for your brain, bro
05:13:19 <elliott> spirited away is kinda one big gob in my head.
05:13:20 <oerjan> zzo38: seventeen thousand dimensional space is not illogical, mind you. although maybe a little inconvenient.
05:14:10 <zzo38> Sgeo: I also had some dream about not push the buttons in elevator door.
05:14:22 <oerjan> well actually for linear algebra it may even be convenient.
05:14:26 -!- jix has quit (Ping timeout: 250 seconds).
05:14:37 <zzo38> oerjan: I did not mean that 17000 dimension space is illogical, I meant you have illogical imagination things in it.
05:15:35 -!- jix has joined.
05:15:52 <zzo38> Here is URL about my dream (look for the line with "*I was looking at programs for different programming languages." for the one about the elevator button)
05:15:55 <zzo38> http://zzo38computer.cjb.net/misc/weird_dream/dream.txt
05:16:22 <zzo38> The one with @ instead of * is somebody else, I just collected the information. The one with * is my own.
05:17:37 -!- tswett has quit (Ping timeout: 276 seconds).
05:18:16 -!- tswett has joined.
05:18:23 <Sgeo> Undefined things happened.
05:18:47 -!- azaq23 has joined.
05:19:34 <zzo38> Sgeo: In a dream, such things is possible as undefined things happened.
05:19:45 * Sgeo goes to press a random button in zzo38's elevator
05:21:41 <oerjan> beware of the button saying "outer space"
05:22:21 <oerjan> you will _not_ enjoy it when the doors open
05:22:55 <quintopia> zzo38: have your dreams ever solved major problems in real life?
05:23:14 <quintopia> zzo38: for instance, the answer to what's the difference between a duck?
05:23:30 * oerjan swats quintopia -----###
05:23:40 <zzo38> Not as far as I can remember. However, I cannot remember, so I don't know.
05:23:46 <oerjan> btw did you finish your scoring system
05:24:05 <zzo38> What scoring system? Whose scoring system?
05:25:48 <quintopia> because, i'm too lazy to reimplement linear algebra, and i've heard gregor will not allow lapack, so i have to wait on him to tell me with linear algebra library i'm allowed to use.
05:25:54 -!- elliott has quit (Ping timeout: 240 seconds).
05:26:21 <oerjan> i thought that might be a stumbling block :/
05:26:30 <Gregor> I didn't say I wouldn't ALLOW lapack, just that I wasn't going to learn it myself just to write a friggin' BFJoust scoreboard calculator :P
05:27:15 <quintopia> oic. so if i learn it, you'll allow it?
05:27:33 <Gregor> Well, so long as it's reasonably easy to install :P
05:27:39 <Gregor> I assume it's in Debian.
05:29:32 <oerjan> http://packages.qa.debian.org/l/lapack.html
05:45:59 -!- aloril has quit (*.net *.split).
05:48:41 -!- asiekierka has joined.
05:51:05 -!- zzo38 has quit (Remote host closed the connection).
05:52:33 -!- SimonRC has quit (Ping timeout: 250 seconds).
05:52:37 -!- aloril has joined.
05:52:52 -!- SimonRC has joined.
05:54:59 -!- cheater00 has quit (*.net *.split).
05:54:59 -!- shachaf has quit (*.net *.split).
05:54:59 -!- Sgeo has quit (*.net *.split).
05:54:59 -!- Leonidas has quit (*.net *.split).
05:54:59 -!- mycroftiv has quit (*.net *.split).
06:08:21 -!- asiekierka has quit (Excess Flood).
06:08:25 -!- asiekierka has joined.
06:10:41 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
06:11:43 -!- copumpkin has joined.
06:15:52 -!- cheater00 has joined.
06:15:54 -!- shachaf has joined.
06:15:54 -!- Sgeo has joined.
06:15:54 -!- Leonidas has joined.
06:15:54 -!- mycroftiv has joined.
06:20:22 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | It's just a few vetical pixels at the top..
06:20:34 -!- asiekierka has quit (Ping timeout: 240 seconds).
06:22:30 <Vorpal> <elliott> Vorpal: I was thinking that it would be a lot easier if one simply wrote Cygwin for Mac OS. <elliott> Vorpal: SO CLEARLY IT MUST BE DONE. <-- oh my
06:26:22 <Vorpal> <elliott> pikhq: And I think you could hook a "stop it!" key right to the OS to terminate them. <-- MacBugs?
06:27:31 <Vorpal> <elliott> pikhq: I dunno if it's worth it though :p <-- depends on what standard. #esoteric? Hell yeah!
06:36:11 -!- augur has quit (Remote host closed the connection).
06:39:20 -!- SimonRC has quit (Ping timeout: 250 seconds).
06:39:27 -!- SimonRC has joined.
06:57:37 -!- augur has joined.
07:04:47 <coppro> man, my respect for Al Gore just went up
07:08:29 <coppro> I read about how he, as president of the senate, correctly enforced a rule that ultimately led to him losing the election
07:10:42 -!- TLUL has joined.
07:12:23 <coppro> the specific rule was that challenging a state's electoral result requires the signature of a representative and a senator from that state, and there was no senator signed on
07:20:18 <pikhq> Shame, too. Al Gore damned well should've been President.
07:20:28 <pikhq> Aaaand may or may not have actually won.
07:22:34 <coppro> but he upheld the law and denied himself the opportunity to win; which is very honorable
07:24:46 -!- elliott has joined.
07:24:53 -!- elliott has quit (Changing host).
07:24:53 -!- elliott has joined.
07:25:20 <coppro> also, fun legal fact: the constitution of canada is not a set whose elements are all known
07:27:00 <pikhq> I thought Canada actually had a well-defined legal system?
07:27:08 <coppro> We have a constitution act which says
07:27:14 <coppro> (2) The Constitution of Canada includes
07:27:15 <coppro> (a) the Canada Act 1982, including this Act;
07:27:15 <coppro> (b) the Acts and orders referred to in the schedule; and
07:27:15 <coppro> (c) any amendment to any Act or order referred to in paragraph (a) or (b).
07:27:26 <coppro> however, it also includes a bunch of other things
07:27:32 <coppro> like parliamentary privelege
07:27:53 <coppro> which has in particular been confirmed by the supreme court to actually be a part of the constitution
07:28:06 <pikhq> Unlike the UK, where even the premise of the government being divine right isn't well-defined...
07:28:31 <pikhq> coppro: ... Oh, great, your supreme court is as fucking nuts as ours.
07:28:50 <coppro> actually, I think I have to agree
07:29:11 <coppro> privelege is mentioned in the constitution act but never explicitly defined
07:29:21 <elliott> wonder how long humans can actually keep up this only sleeping every other day thing
07:29:25 <pikhq> I see nothing in clause 2 of the Constitution Act there that would include parliamentary privilege...
07:29:47 <coppro> 18. The privileges, immunities, and powers to be held, enjoyed, and exercised by the Senate and by the House of Commons, and by the members thereof respectively, shall be such as are from time to time defined by Act of the Parliament of Canada, but so that any Act of the Parliament of Canada defining such privileges, immunities, and powers shall not confer any privileges, immunities, or powers exceeding those at the passing of such Act held,
07:29:59 <pikhq> elliott: Depends: how long are you sleeping?
07:29:59 <coppro> from a different part of the constitution act
07:30:13 <elliott> pikhq: too long... like 11 hrs, but would be 12 if not for alarms andsuch
07:30:25 <pikhq> coppro: Yup, definitely not in the Canadian constitution. Merely means of defining such.
07:32:01 <elliott> but it's better than waking up at 5pm.
07:32:01 <elliott> i hate my circadian rhythm :(
07:32:01 <pikhq> elliott: You almost certainly have a sleeping disorder.
07:32:01 <coppro> in this particular case, the privilege was the long-standing parliamentary privilege of excluding strangers; a TV company wanted to film the proceedings of the Nova Scotian legislature, but the legislature wouldn't let the cameras in
07:32:01 <elliott> feeling tired already, definitely a record, probably should have slept this time, oh well, coffee
07:32:01 <pikhq> Of course, that much we already knew.
07:32:01 <coppro> the TV company cited freedom of speech, of course
07:32:01 <elliott> pikhq: i'm fucking buying melatonin today
07:32:01 <pikhq> coppro: Strange that they'd cite the constitution for that.
07:32:07 <pikhq> coppro: Surely freedom of speech doesn't mandate that they have the right to record everything, everywhere?
07:32:34 <elliott> 06:06:41 <Vorpal> <elliott> Vorpal: I was thinking that it would be a lot easier if one simply wrote Cygwin for Mac OS. <elliott> Vorpal: SO CLEARLY IT MUST BE DONE. <-- oh my
07:32:35 <elliott> 06:10:34 <Vorpal> <elliott> pikhq: And I think you could hook a "stop it!" key right to the OS to terminate them. <-- MacBugs?
07:32:35 <elliott> 06:11:43 <Vorpal> <elliott> pikhq: I dunno if it's worth it though :p <-- depends on what standard. #esoteric? Hell yeah!
07:32:43 <elliott> (3) i mean the "make tight loops not hang" thing
07:32:44 <pikhq> coppro: Also strange that you don't have your proceedings recorded already... I know here that the federal government *itself* actually does video recording of the entire proceedings of Congress.
07:32:44 <coppro> pikhq: No, but the SCC decided that the privilege is itself constitutional and thus not subject to obeying the rights set out in the Charter
07:33:01 <elliott> pikhq: can i have so much coffee
07:33:06 <pikhq> coppro: How very strange; I'd think that human rights would be irrelevant.
07:33:38 <coppro> pikhq: it's gets muddled when it's government business, because there are some very deep rights there
07:33:51 <coppro> a citizen's right to know what the government is doing, and the government's right not to tell them
07:34:14 <coppro> The federal government here records the House of Commons and committees of both chambers; the Senate itself has not authorized video recordings
07:34:26 <coppro> transcripts are available in both French and English by the next day online
07:34:38 <coppro> (which is, incidentally, absolutely insane)
07:35:04 <coppro> (they also do live translation of proceedings so that anyone can participate in either language)
07:35:25 <pikhq> "Aaaah", multilingual society.
07:35:38 <pikhq> Though if our government was *sane*, they'd probably have to do the same.
07:35:48 <pikhq> Remember, the US has *no* official language at all.
07:36:11 <pikhq> That is but *one* important language in the US.
07:36:43 <pikhq> Perhaps the one most likely to be relevant...
07:36:55 <elliott> pikhq: i should make cygmac work on system software 1 :D
07:37:21 <pikhq> But there's rather a *lot* of languages spoken here, even if you discount those languages which have mostly bilingual native speakers.
07:37:48 <coppro> popping the stack a little, the issue that is really sticky about the scope of the constitution is that the constitution says that an amendment to the constitution that changes the composition of the Supreme Court requires unanimous consent of the provinces.
07:37:55 <coppro> Sounds great, except for one minor detail
07:38:46 <coppro> the composition of the Supreme Court is defined in the Supreme Court of Canada Act, which is not in the schedule
07:39:16 <coppro> some people hold this to mean that this requirement on amendment would apply only if the SCC were explicitly moved into the constitution
07:39:27 <coppro> others hold it to mean that the SCCA is implicitly part of the constitution
07:40:32 <pikhq> Pushing to the stack again. There's 35 million people who speak primarily Spanish in the US. *Damn*.
07:40:55 <pikhq> Making the USA have the second-largest Spanish-speaking community in the world.
07:41:06 <coppro> moreover, the SCA predates that part of the constitution by a long time
07:41:12 <elliott> thati sn't pushing to the stack
07:41:15 <elliott> thats calling the continuation
07:41:24 <pikhq> elliott: Fuck you and your "accurate semantics".
07:41:24 <elliott> your conversational control structures are spaghetti
07:41:37 <coppro> moreover, this is potentially relevant as there is a bill passing through the Senate right now to change the selection requirements for Supreme Court justices, although I expect it has about a snowball's chance in hell of actually passing
07:41:38 <elliott> man wyh do yyou have to type the keys in order
07:41:41 <elliott> that's a bad thing about keyboards.
07:41:53 <elliott> should just press them all at once, and it uses a dictionary :/
07:42:09 <pikhq> coppro: T3h groans.
07:42:12 -!- sftp has quit (Read error: Connection reset by peer).
07:42:21 <coppro> (and it would fail if the supreme court is found to be part of the constitution as it would then require a bunch of provinces to agree)
07:42:42 <coppro> (although, oddly enough, it would make it irrelevant that the Conservatives control the Senate)
07:43:22 <elliott> you're currently in the spanish stack frame
07:43:55 <coppro> elliott: no, I stayed behind; it's a cactus stack
07:44:17 <elliott> coppro: your conversational system is so inelegant. horrible hacky whore language on top of spaghetti control.
07:44:25 <coppro> elliott: I speak in Perl
07:44:27 <elliott> from now on I will talk only in Has<THUNK>
07:45:47 <coppro> uh oh, elliott's not currying properly
07:45:52 <coppro> where's the nearest indian restaurant?
07:46:24 <elliott> you forced evaluati** Exception: undefined
07:46:46 -!- pikhq_ has joined.
07:46:54 -!- pikhq has quit (Ping timeout: 250 seconds).
07:48:04 <coppro> pikhq_: what did you miss?
07:48:25 <pikhq_> 01:26 < coppro> (and it would fail if the supreme court is found to be part of the constitution as it would then require a bunch of provinces to agree)
07:48:34 <elliott> <coppro> (although, oddly enough, it would make it irrelevant that the Conservatives control the Senate)
07:48:39 <coppro> 03:26 < coppro> (although, oddly enough, it would make it irrelevant that the Conservatives control the Senate)
07:48:42 <coppro> 03:27 < elliott> you're currently in the spanish stack frame
07:48:45 <elliott> pikhq_: use an irc bouncer or something srsly :P
07:48:45 <coppro> 03:27 < elliott> thoguht you should know
07:48:47 <coppro> 03:28 < coppro> elliott: no, I stayed behind; it's a cactus stack
07:48:50 <coppro> 03:28 < elliott> coppro: your conversational system is so inelegant. horrible hacky whore language on top of spaghetti control.
07:48:51 -!- elliott has left (?).
07:48:53 <coppro> 03:28 < coppro> elliott: I speak in Perl
07:48:53 -!- elliott has joined.
07:48:55 <coppro> 03:28 < elliott> from now on I will talk only in Has<THUNK>
07:48:58 <coppro> 03:30 < coppro> uh oh, elliott's not currying properly
07:49:00 <coppro> 03:30 < coppro> where's the nearest indian restaurant?
07:49:03 <elliott> i should have to hodl down every modifier I have to close the window
07:49:05 <coppro> 03:30 < elliott> you forced evaluati** Exception: undefined
07:49:19 <elliott> or with coppro maybe the opposite
07:49:24 <elliott> if his message matches ".win" it just does it
07:50:21 <coppro> I need to hack my irssi so as to warn me every time I send a message that looks anything like a window switch
07:50:26 <elliott> http://www.emaculation.com/doku.php/basilisk_ii So does anyone see a source link on this page?
07:50:41 <elliott> unless they just compile the old jit release they call a historic archive
07:50:52 <elliott> coppro: do what i do, only be in two channels, so much less confusing
07:50:56 <elliott> still manage to part them on a regular basis tho
07:51:26 <coppro> elliott: I'm at about 25
07:51:43 <coppro> your method sounds lame
07:51:53 <elliott> i think my natural hatred of people helps me stay out of too many channels
07:52:21 <coppro> ah, yeah, people are dumb
07:52:55 <elliott> only sleeps once every two days
07:52:57 <elliott> what kind of moron does that
07:53:37 <elliott> you know, usually i even fix typos
07:53:43 <elliott> but that seems pretty hard at this point
07:53:44 <coppro> my faculty is very efficient
07:53:51 <coppro> we have an entire combinatorics and optimization department
07:54:00 <elliott> can you just never sleepif you drink enough coffee, i think this might be practical for me
07:54:29 <elliott> as in mathematical. saying "oh a bunch of people did this and died" doesn't count.
07:54:32 <elliott> start from peano arithmetic
07:54:46 <coppro> I time traveled to the future and brought back your corpse after you went insane, thought yourself an automobile, and got run over on the motorway
07:54:52 <elliott> you may assume as an axiom that things exist
07:55:00 <elliott> that is the one additional axiom you may assume
07:55:18 <coppro> that axiom is inconsistent with the coffee axiom
07:55:25 <elliott> coffee makes things not exist?
07:55:33 <coppro> in sufficiently large quantities, yes
07:55:34 <elliott> almost as deep as the sleep i'd like to be in.
07:55:45 <elliott> yeah unfortunately i haven't hallucinated yet with this sleep dep
07:55:50 <elliott> mostly i'm just irritable and tired
07:56:03 <elliott> things would be unicorns and stuff, hallucinations are obviously just like nethack
07:56:06 <elliott> coppro: omg, but short is only 16 bits.
07:56:13 <coppro> sleep dep gives boring hallucinations
07:56:19 <coppro> elliott: everything you know is wrong
07:56:22 <coppro> just forget the words and sing along
07:56:30 <elliott> <coppro> sleep dep gives boring hallucinations
07:59:59 -!- clog has quit (ended).
08:01:55 -!- clog has joined.
08:01:55 <elliott> checking for gcc... gcc-4.5
08:01:55 <elliott> checking whether we are using the GNU C compiler... no
08:01:55 <elliott> checking whether gcc-4.5 accepts -g... no
08:02:38 <elliott> configure:4076: checking whether we are using the GNU C compiler
08:02:39 <elliott> configure:4095: gcc-4.5 -c conftest.c >&5
08:02:39 <elliott> configure: failed program was:
08:02:44 <elliott> $? = 0 means it worked, autoconf.
08:03:30 <elliott> it keeps assuming that because it worked and $? was 0 that everything failed after that :)
08:03:38 <elliott> which explains why it's all "oh no, you don't have any headers basically".
08:03:53 <elliott> configure:4255: gcc-4.5 -qlanglvl=extc89 -c conftest.c >&5
08:03:53 <elliott> gcc-4.5: unrecognized option '-qlanglvl=extc89'
08:03:53 <elliott> configure: failed program was:
08:23:39 -!- TLUL has quit (Quit: *disappears in a puff of orange smoke*).
08:29:47 <elliott> what's the best archive format to distribute code for a 68k mac emulator in
08:29:52 <elliott> obviously the author of mini vmac
08:29:58 <elliott> decided "a macintosh disk image"
08:30:16 <elliott> i swear to god, it's a bootstrapped mac emulator
08:33:29 <pikhq_> To play devil's advocate here, can't OS X mount those old disk images?
08:36:45 <elliott> pikhq_: but more importantl
08:36:53 <elliott> pikhq_: the build system is a mac os <=6 application
08:37:17 <elliott> http://minivmac.sourceforge.net/doc/build.html
08:37:20 <elliott> First download the source code archive from the download page, a file with the name “minivmac-3.1.3.src.zip”. Extract from this zip file a disk image (named “minivmac-3.1.3.src.dsk”).
08:37:20 <elliott> Now launch Mini vMac (version 3.0.0 or later), booting from a disk image containing a system folder. (The source code disk image doesn't contain a system folder.) (See the Start page for information about getting started with Mini vMac.)
08:37:20 <elliott> Mount the source code disk image in Mini vMac. At the top level of this disk is an application named "Build". Launch this application. A text editing window will open in which to type in the desired options.
08:37:27 <elliott> pikhq_: it's literally a bootstrapped macintosh emulator.
08:37:39 <elliott> source distributed as a macintosh boot image, compileable with a macintosh, or a prebuilt verison of itself.
08:37:59 <elliott> pikhq_: you will note that this is insane. It does CROSS COMPILING
08:38:16 <elliott> OK, it... extracts into the host system or something.
08:38:21 <elliott> So its configure is just a mac app I DON'T KNOW
08:39:43 <pikhq_> I should sleep. It's 02:23.
08:40:22 <pikhq_> And I intend to do my diff eq. homework before class tomorrow.
09:11:02 <elliott> pikhq_: echo hello >:foo.c; cp foo.c ::; mv ::foo.c :bar.c
09:11:09 <elliott> BEHOLD: MAC PATHNAMES IN UNIX
09:27:26 -!- Guest1055 has quit (Ping timeout: 240 seconds).
09:32:58 -!- Guest1055 has joined.
09:45:24 <elliott> olsner: in my mind!! with magic!!!
09:45:37 <elliott> olsner: i want to implement cygwin for classic mac os. and even have a vague idea how.
09:45:54 <olsner> wow, that's .. suitably crazy
09:46:03 <elliott> olsner: not cygwin itself, but, you know
09:46:17 <elliott> olsner: oh, and i want it to work on System Software 6 too. :)
09:46:27 <olsner> generally broken layer of brokenness on top of a broken os, check
09:46:29 <elliott> which looks like this. http://www.operating-system.org/betriebssystem/bsgfx/apple/system6-scr-03.jpg
09:46:43 <elliott> olsner: old classic mac os isn't broken, just... selective... as to what it wants to acheive
09:46:57 <olsner> nice redefinition of not broken there
09:47:20 <elliott> says the person using the broken os linux
09:54:58 -!- oerjan has quit (Quit: leaving).
10:11:24 -!- pikhq has joined.
10:11:57 -!- pikhq_ has quit (Ping timeout: 260 seconds).
10:21:43 <elliott> Come on little mac, load the page, load the page!
10:29:20 -!- cheater99 has joined.
10:30:51 -!- cheater00 has quit (Ping timeout: 255 seconds).
10:35:23 -!- copumpkin has quit (Ping timeout: 252 seconds).
10:35:53 -!- copumpkin has joined.
10:43:02 -!- Kufiyeh has joined.
10:44:22 -!- Kufiyeh has quit (Client Quit).
10:45:09 -!- Kufiyeh has joined.
10:53:37 -!- Kufiyeh has left (?).
11:06:46 -!- copumpkin has quit (Ping timeout: 252 seconds).
11:12:26 <cheater99> http://morgsatlarge.wordpress.com/2011/03/13/why-i-am-not-worried-about-japans-nuclear-reactors/
11:12:57 -!- copumpkin has joined.
11:31:22 <elliott> 12:36:04 <ais523> AnMaster: that doesn't solve the issue, in fact it introduces an issue, that being that the tabstops are no longer 8
11:31:22 <elliott> this log has so much circular logic it hurts
11:45:19 <elliott> unless it's a really persistent troll that he's kept up for years and strongly abides by :)
11:45:49 <elliott> (the "TABS=8 AND EVERYONE WHO THINKS TABS!=8 EVER IS WRONG" thing; not that he ever gives a decent argument for it, but this was basically like 3 hours of circular argumentation from him in the logs)
11:45:49 -!- Vorpal has quit (Ping timeout: 260 seconds).
11:46:09 -!- Vorpal has joined.
11:47:01 <elliott> hey Vorpal, does that ick on mac classic thing actually work :)
11:48:56 -!- elliott has quit (Remote host closed the connection).
11:49:00 -!- elliott_ has joined.
11:49:07 <elliott_> Vorpal: where did you download mpw from anyway
11:50:27 <fizzie> It used to be on Apple's FTP, at least.
11:50:41 <fizzie> The FTP links at http://developer.apple.com/tools/mpw-tools/ seem dead at the moment, though.
11:51:39 <fizzie> Well, it wasn't long ago (less than three years, I'd guess) that it still worked.
11:51:46 <elliott_> yes but the port thing was 2009.
11:52:01 <elliott_> i pretty much just want mpw 3 so i can write a unix kernel for system 6 :D
11:52:05 <elliott_> tbh i might want to develop on a later version
11:54:03 <fizzie> How about http://mirrors.vanadac.com/ftp.apple.com/developer/Tool_Chest/Core_Mac_OS_Tools/MPW_etc./ then?
11:54:38 <elliott_> fizzie: whoaa. it's reflecting into my screen-space-visor-vision.
11:55:01 <elliott_> to to figure out what GM vs PR is
11:55:05 <fizzie> Still haven't slept, eh?
11:55:53 <fizzie> "The MPW-GM folder contains a complete Golden Master version of the MPW development environment including the MPW Shell, tools, interfaces and libraries. The software in this folder is considered to be "final" quality. The MPW-PR folder contains Pre-Release versions of some of the software components that make up the MPW development environment."
11:56:24 <fizzie> The Game Master version.
11:56:41 <elliott_> isn't this just the newest version?
11:57:01 <elliott_> i kind of require the version that works on system software 6 which i think is 3 and erliar :)
11:57:25 <fizzie> Ahm'k; that might be more problemostic, yes.
11:57:54 <fizzie> The FTP'd one does require 7.5.
11:58:04 <elliott_> fizzie: whoa. this is how you sneak your bad spellungs into my mindbrane!!
11:58:11 <elliott_> you do it, when i am not conscious, in the fullest degree!!!
11:58:16 <elliott_> that's sneaky and you're bad! i have noticed this!
11:58:25 <elliott_> wonder if i could stick to the ceiling
11:59:29 <olsner> apply enough duck/duct/gaffer tape and I bet you could
12:00:32 <fizzie> Curiously enough, "jesus tape" is the most common Finnish nickname for it. Or even "jesse" or some-such.
12:00:36 -!- variable has joined.
12:01:13 <elliott_> jesus tpae. that's nice i like that
12:14:06 <elliott_> my operationg system is based on it
12:20:11 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | Uh, I had a question but I've almost forgotten it..
12:38:53 -!- ais523 has joined.
12:43:26 <elliott_> ais523: inspired by the port of C-INTERCAL to classic Mac OS, I have become highly tempted to try writing Cygwin for Classic Mac OS
12:43:38 <elliott_> because, perfect match, right?
12:43:45 <elliott_> (not actual Cygwin, just... a unix layer)
12:43:45 <ais523> wow, that's a thought that my mind can't actually finish thinking
12:43:48 <ais523> it gets stuck halfway through
12:43:53 <ais523> so it must be a really incredibly bad idea
12:44:06 <elliott_> ais523: yes, and really incredibly bad ideas are also known as really good ideas!
12:44:08 <ais523> and I will now attempt to forget what it was to save my sanity
12:44:44 <elliott_> ais523: Well hey, C-INTERCAL supports m86k-macos-unknown-mpw (or whatever the thing would be), get ready to support m86k-macos-unknown-gnu!
12:44:50 <elliott_> ais523: Did that patch ever get committed, btw?
12:45:05 <ais523> not by me because I didn't review it
12:45:10 <ais523> not by esr because it didn't have timestamps
12:45:16 <ais523> I'm not sure which reason is more spurious
12:45:28 <ais523> as in, the changes weren't dated with when they were made
12:45:35 <elliott_> also, you could just say you were looking for an excuse to reject it :P
12:46:49 <elliott_> ais523: anyway, C-INTERCAL bugreport: the configure script doesn't run on i386-win32-unknown-gnu
12:47:02 <elliott_> ais523: that is, it doesn't run in a native Windows version of bash, equipped with all the standard GNU tools
12:47:02 <ais523> can you give a more specific bug report?
12:47:08 <ais523> as in, what goes wrong when you try?
12:47:09 <elliott_> doesn't run = garbage errors from bash
12:47:14 <elliott_> I forget exactly, it was a while ago I tried
12:47:17 <ais523> what if you try converting newlines first?
12:47:21 <elliott_> I tried two ports, none worked
12:47:38 <elliott_> ais523: Note that the port versions of bash were 1.something and 2.something respectively
12:47:43 <elliott_> I sort of doubt configure supports bash 1
12:47:57 <ais523> I wouldn't actually be surprised if it did
12:48:03 <elliott_> ais523: but it's obviously a bug anyway, because it has a shell and gnu tools, so obviously it should be able to run the build system
12:48:05 <ais523> there are all the sections on writing portable sh, and working around bugs
12:48:08 <elliott_> (if I downloaded mingw to be a cc)
12:48:18 <ais523> although nowadays they tend to abandon support for really old stuff, which is annoying
12:48:29 <ais523> that defeats half the point of configure, which is to be stupidly backwards-compatible
12:48:38 <elliott_> ais523: all the bloat, none of the benefit!
12:49:14 <elliott_> ais523: you should write a configure program in strict, avoiding-anything-that-anyone-has-fucked-up-implementing-ever K&R C, then it's _guaranteed_ to be able to configure wherever it can compile
12:49:41 <ais523> then all it needs is a configure script to configure the compiler correctly for running it
12:49:50 <ais523> hmm, IIRC jettyplay has a configure script, because I'm insane
12:50:16 <elliott_> <ais523> then all it needs is a configure script to configure the compiler correctly for running it
12:50:23 <elliott_> you design the configure script so it works with just "cc configure.c" on any compiler
12:50:29 <elliott_> and use system() to do the rest, somehow
12:50:38 <ais523> on Windows, the compiler is called cl
12:50:44 <elliott_> ais523: substitute cc with the compiler name
12:50:50 <ais523> hmm: AC_PATH_PROG([JAVAC], [javac])
12:50:56 <elliott_> I mean that any user that can type ./configure should be able to just give it to the compiler and have it work :)
12:51:03 <ais523> actually in jettyplay source
12:51:28 <ais523> you'll be glad to hear I wrote the makefile.in by hand rather than using autoconf, though
12:51:45 <ais523> (it's not in the repo, btw; it's in a separate directory)
12:52:06 <ais523> # This fails on unusual names for /usr/lib that contain ', &, \ or %. That
12:52:07 <ais523> # shouldn't be a problem, but you never know...
12:52:18 <ais523> I'm trying to figure out if this is a major failing or something completely irrelevant
12:52:26 <elliott_> my name for /usr/lib is "Tordy Complat Infisiant tumbly Hork"
12:52:30 <elliott_> my name for /usr/lib is "Tordy Complat Infisiant tumbly Hork%"
12:52:50 <ais523> normally failing on weird punctuation in filenames is bad (see: clc-intercal and mandb), but it seems vaguely unlikely to find it in a localized name for /usr/lib
12:52:53 <elliott_> ais523: you see in @, you don't have to worry about escaping like that!
12:52:54 <ais523> especially as most people just hardcode /usr/lib
12:53:03 <elliott_> ais523: Lojban uses ', at least
12:53:32 <ais523> ' is the one it's very hard to fix
12:53:42 <ais523> due to there being no way to tell make(1) to shell-escape variables it substitutes
12:53:45 <elliott_> <elliott_> ais523: you see in @, you don't have to worry about escaping like that!
12:53:54 <ais523> I noticed, just didn't have a reaction
12:54:01 <elliott_> ais523: but it's TOTALLY RELEVANT >:D
12:54:49 <elliott_> ais523: because there's no shell, mostly
12:55:01 <ais523> then how do you run your makefiles?
12:55:14 <ais523> note that I was trying to substitute something into the replacement side of a sed s/// regexp
12:55:17 <elliott_> ais523: you don't; if you _did_ have a build system, it'd run code in @lang, which serves as the shell
12:55:39 <elliott_> ais523: but even just using Python as the shell in Unix doesn't compare, because @lang actions don't just take strings and return strings
12:55:40 <ais523> but then it doesn't run standard GNU configure files
12:55:49 <elliott_> I've fixed your bug at its source!
12:55:52 <ais523> also, seriously, people use Python as a shell?
12:55:52 <elliott_> and it turns out, the source is everything
12:55:58 <elliott_> ais523: no, but I was preempting a rebuttal
12:56:16 <ais523> I interpreted as being more like Powershell than sh
12:56:19 <ais523> only probably less insane
12:56:31 <elliott_> hmm, might want a beta oberon, this hasn't been updated since 1999...
12:56:32 <ais523> actually, what you've described sounds more or less like the entire rationale for Powershell
12:56:44 <elliott_> ais523: yep, except powershell makes a lousy programming language
12:56:48 <ais523> although you can ignore it as Microsoft almost certainly screwed up the implementaiton
12:56:51 <elliott_> (it also makes a lousy shell, but that's not inherent)
12:57:54 <ais523> indeed, languages can be good at one and bad at the other, or vice versa, or both or neither
12:58:05 <ais523> wow, that last sentence of mine was worryingly zzo38ish
12:58:15 <elliott_> ais523: A C compiler would probably look like - GCC compile(my-bytestring)
12:58:23 <elliott_> It has to be a bytestring (or a string, I suppose), because of cpp
12:58:39 <elliott_> If you had some already-preprocessed-and-now-a-C-code-object code, it'd look more like GCC compile(that) :P
12:59:02 <elliott_> ais523: I was actually thinking of what a "Unix-like" shell in @ would look like
12:59:31 <elliott_> would produce a box with "Native function" at the top of it, and some assembly code below, plus links to a bunch of inspectors
12:59:42 <elliott_> and "$ cc foo.c > foo" would introduce the alias "foo" in the current environment for that object
12:59:45 <elliott_> which could be called with ./foo
12:59:55 <elliott_> (foo.c is of course just an alias for another object, as @ has no filesystem)
13:00:24 <ais523> wow, if you do that lazily, you could have lazy debug info
13:00:48 <elliott_> ais523: although, all standard @lang objects already have debugging, by virtue of being tied to their source
13:01:16 <elliott_> ais523: (AFAICT prebuilt binaries would be completely useless in @ as the compiler would be as fast as machinely possible)
13:01:16 <ais523> I mean more extensive debug info could be generated on-demand
13:01:28 <elliott_> well, it sort of can, if the machine code is tied to the source
13:01:32 <elliott_> because you can always re-scan that area
13:02:00 <ais523> hmm, I was going to say "you could even add things like debug printf statements to a program and rerun it in the environment it started running in", but then realised that things were getting dangerously Feather
13:02:22 <elliott_> ais523: nope, that's still pretty solidly @
13:02:29 <elliott_> although, obviously the outside world wouldn't change
13:02:49 <elliott_> ais523: oh, and the debugger would be entirely source-based, and let you change expressions in the program on the fly :)
13:02:54 <elliott_> and have them recompile into the right place
13:03:09 <elliott_> so you could even add debug printfs to a running program just by adding them in the debugger
13:03:16 <elliott_> and then save the changes if you want to keep them
13:03:34 <ais523> hmm, why are there not enbuggers?
13:03:46 <ais523> as in, programs that help you figure out good places to put hard-to-find and subtle bugs in programs?
13:03:47 <elliott_> ais523: because the debugger works perfectly well for that job
13:03:58 <elliott_> oh, and of _course_, you can debug a running debugger.
13:04:08 <elliott_> just in case you've ever wanted to do something completely pointless.
13:04:37 <ais523> elliott_: some of the TAS people have been known to run debuggers around emulators that didn't have memory watch, to debug the program running inside the emulator by debugging the emulator itself
13:04:55 <ais523> which to me seems a lot more work than just changing the emulator to have the features you need
13:05:48 <elliott_> ais523: well, in @, you could have a really long-running debugging session, so long-running that you never want to exit it ever
13:05:52 <elliott_> but you want to do something the debugger can't do
13:05:56 <elliott_> so you open a debugger on the debugger
13:05:59 <elliott_> patch the feature in to the debugger
13:06:10 <elliott_> leaving the long-lived debugger with the new feature you need!
13:06:17 <elliott_> and then you could even save it to the standard debugger for future use
13:06:40 <elliott_> ais523: I hope to replace all development with successive applications of nested debugging by 2020
13:08:53 <elliott_> argh, oberon's mouse acceleration is far too high
13:09:55 <ais523> I thought it was a programming language, not an OS?
13:10:01 <ais523> do programming languages have mouse acceleration?
13:10:06 <ais523> I suppose their standard libraries might
13:10:19 <ais523> elliott_: you mean it's a programming language and also an OS, Smalltalk-style?
13:10:33 <ais523> (even then, I tend to mentally separate the two, and do not consider gst to be an abomination, but that's maybe just me)
13:11:03 <elliott_> ais523: I am referring to {Oberon (operating system)}, which is based on an implementation of {Oberon (programming language)}; *but* {Oberon (operating system)} is the only existing embodiment of the "Oberon system" that the {Oberon (programming language)} specifies and is designed to be used in
13:11:17 <elliott_> specifically, I'm referring to the {Native Oberon} flavour of {Oberon (operating system)}
13:11:32 <elliott_> but, really, there's little point distinguishing them
13:11:45 <elliott_> just because Unix pretends to be language-agnostic (it's not, see: C), doesn't mean everything does
13:12:04 <elliott_> (similarly, just because something isn't language-agnostic, doesn't mean it's not language-hostile (see Lisp Machines, which had C compilers))
13:12:09 <ais523> also, does {} introduce an FFI to Wikipedia-style disambig parens?
13:12:50 <elliott_> hmm, so right-click drag with insanely high mouse acceleration is the most difficult operation possible on a macbook
13:13:04 <ais523> are you using the touchpad or an external mouse?
13:13:14 <ais523> also, do you mean mouse acceleration, or mouse speed?
13:13:21 <ais523> very high levels of acceleration give you lots of precision on a touchpad
13:13:25 <elliott_> it's hard to tell when it's so fast
13:13:30 <elliott_> ais523: this is a 640x480 window :)
13:13:45 <ais523> well, very high mouse acceleration would mean that if you moved the touchpad slowly, the cursor would move very slowly indeed
13:13:49 <ais523> compared to moving it quickly
13:14:46 <elliott_> gah, better create a partition before trying to isntall oberon
13:14:53 <Sgeo> The wikipedia page reads like an advertisement
13:15:02 <elliott_> ais523: or, it could go insanely fast when i moved it slowly, and SUPER INSANELY FAST hen i went faster
13:18:18 <ais523> but then you couldn't tell if it was accelerated or not
13:18:56 <elliott_> ais523: what's the standard size of a cylinder these days? :-P
13:19:12 <ais523> oh, you mean a hard drive cylinder?
13:19:22 <ais523> I thought that question was along the same lines as "how long is a piece of string?"
13:19:46 <elliott_> no no let's pretend i meant it in the other sense
13:22:48 <elliott_> i would like to see a self-modifying paintfuck
13:22:51 <fizzie> All CHS addresses are utter lies nowadays anyway, so it's probably not a very relevant number any more.
13:22:55 <elliott_> as in, a paintfuck that interprets the graphical space as a program somehow
13:23:00 <elliott_> so that you set the space up and run it
13:23:07 <elliott_> fizzie: yes but fdisk wants me to set one
13:23:13 <elliott_> Number of cylinders (1-1048576):
13:23:16 <elliott_> to let me paaaartition this fake disk
13:23:34 <ais523> elliott_: it should OCR the graphical space and interpret it as a PF program
13:23:39 <ais523> the interpreter should, I mean
13:23:53 <elliott_> ais523: i didn't mean just for initial loading though :)
13:23:53 <ais523> it's the only way to stay within the spirit of PF
13:23:58 <elliott_> or does it reverse-OCR it every step?
13:24:11 <ais523> I mean, every step it OCRs the current playfield, and interprets it as the program
13:24:23 <ais523> the playfield's initialised with a reverse-OCR (i.e. printout) of the program you give it
13:25:20 <elliott_> because that sounds terrifying :)
13:26:06 <ais523> it surely isn't that bad compared to some of the things #esoteric comes up with?
13:26:30 <elliott_> ais523: well, no, but it's still perverse in a unique way of its own
13:26:56 <fizzie> Which sort of fdisk is this that it can't read the disk geometry from somewhere?
13:27:07 <ais523> fizzie: one working on an imaginary drive
13:27:18 <elliott_> fizzie: one working on a file that's a bunch of 0s
13:27:21 <ais523> that doesn't actually have any geometry until you define what it is
13:27:22 <elliott_> it's for a qemuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
13:27:39 <fizzie> Well. A "standard" fake-cylinder in pc-ishnowadays is 63*255 sectors (255 fake-heads and 63 sectors per fake-track; about 8 megabytes).
13:27:54 <fizzie> s/ishnow/ish systems is now/
13:28:17 <elliott_> So 130-or-so cylinders in this.
13:28:59 <fizzie> That's how many bytes you'd get.
13:29:07 <fizzie> (With the largest possible cylinders.)
13:31:22 <ais523> gah, fizzie's nick length is almost exactly the right length (out by one pixel) to make the > to call lambdabot line up with the > to delimit elliott_'s nick
13:31:35 <ais523> and it made me confused as to why lambdabot responded to one but not the other
13:32:00 <ais523> oh, the one pixel difference doesn't bother me, it's just that it's such a small difference I missed it was there
13:36:49 <ais523> hmm, I was reading a page on c2
13:36:55 <ais523> and noticed it mentioned the LGPL
13:37:06 <ais523> and thought "hmm, that's a bit modern for c2 to be referencing, isn't it?"
13:37:18 <ais523> now I'm trying to work out why I had such an illogical thought
13:38:20 <Sgeo> ais523, since much of c2 hasn't been touched in a decade? *exaggeration*
13:38:41 <ais523> Sgeo: for some reason I assume it captures computer knowledge from Before Time Began, or something like that
13:38:43 <elliott_> this bitz stuff is out of control
13:39:43 <olsner> I remember a time when there were occasional changes being done on C2 wiki
13:41:02 <Sgeo> Well, time to get to school
13:47:45 <ais523> elliott_: can you think of any reason why I'd get 504 errors on the vast majority (but not all) of attempts to submit HTTP POST requests, from one laptop but on more than one Internet connection, and how repeatedly trying until it works makes it work eventually?
13:47:55 <ais523> the set of symptoms seems to make no sense to me
13:48:10 <elliott_> ais523: it's making demon fly out of your nose, washing the windows api, and the ISS servers are getting all soapy
13:48:18 -!- Sgeo has quit (Ping timeout: 255 seconds).
13:48:20 <elliott_> and servers tend to perform worse when soapy
13:48:35 <elliott_> (even if they're running on apache, they tend to get bits of soap from them as the iss content disseminates throughout the internet)
13:48:37 <ais523> indeed, SOAP is famously bad
13:50:14 <ais523> anyone else have ideas that aren't DS9K-related?
13:51:00 -!- augur has quit (Remote host closed the connection).
13:56:19 -!- sftp has joined.
13:56:35 <Ilari> Hmm... I think I should actually design an esolang based on fractal circuits.
13:58:21 <elliott_> Ilari: How many billions of unpublished esolangs do you have?
14:00:08 <Ilari> 2 esolangs that are "unpublished". I like esolangs that aren't just minor variations of Brainfuck.
14:00:26 <elliott_> You just seem to talk about random esolangs you've made that nobody seems to have ever heard of a lot :-P
14:01:10 <ais523> Ilari: strangely, we have a section on that in the paper we're writing right now
14:01:13 <Ilari> I have designed 3 so far (this fractal-circuit based esolang isn't in that).
14:01:13 <ais523> fractal circuits, that is
14:01:32 <elliott_> fractal circuits is something that sounds nice but is probably useless :D
14:01:42 <ais523> it's the usual way to implement a sort in hardware
14:02:05 <ais523> why is that so worthy of damnation?
14:02:11 <Ilari> As far as I know, all three are turing complete. And this fourth would be too.
14:02:17 <elliott_> ais523: my prediction was wrong!
14:02:27 <elliott_> also, "damn" is an incredibly mild utterance no matter what you say :)
14:02:30 <ais523> it's hard to not be TC if a language is at all interesting
14:02:52 <elliott_> I still don't know whether Wapr is turing complete or not.
14:03:02 <elliott_> sorry, *jumping to -1 is exciting
14:03:03 <ais523> I still don't know whether Dupdog is turing complete or not
14:03:07 <Vorpal> <elliott> this log has so much circular logic it hurts <-- haha
14:03:17 <ais523> elliott_: was that proved?
14:03:19 <elliott_> Vorpal: It hurts because it has so much circular logic.
14:03:26 <ais523> it's rare to have a language that's non-TC, but not trivially non-TC
14:03:33 <ais523> as ()^: Underload demonstrated
14:03:34 <elliott_> ais523: hmm, not on the wikipedia page, but IIRC oerjan pretty much slam-dunked it once
14:03:53 <ais523> Xigxag's another lang with similar properties
14:03:55 <elliott_> http://esolangs.org/wiki/Jumping_to_-1_is_exciting Quick! Is my badly-specified old esolang TC???
14:03:57 <Vorpal> <elliott> (the "TABS=8 AND EVERYONE WHO THINKS TABS!=8 EVER IS WRONG" thing; not that he ever gives a decent argument for it, but this was basically like 3 hours of circular argumentation from him in the logs) <-- I think his argument boiled down to it being traditional
14:04:14 <elliott_> those things after the command chars
14:04:15 <ais523> wait, was someone ohter than me arguing that?
14:04:31 <elliott_> ais523: no, I just read the log where you spent a few hours dedicated to arguing it :D
14:04:39 <elliott_> nothing personal, it just hurt me inside
14:04:46 <Vorpal> <elliott> hey Vorpal, does that ick on mac classic thing actually work :) <-- it works but the MPW C compiler fails on some generated programs. IIRC it was due to bugs in the MPW C compiler
14:04:47 -!- cheater00 has joined.
14:04:49 <elliott_> but a true logreader never abandons his log.
14:05:11 <Vorpal> <elliott_> Vorpal: where did you download mpw from anyway <-- apple ftp. As a set of 14 or so segmented .img (no not .dmg, that is OS X only)
14:05:19 <elliott_> Vorpal: yeah, that ftp is down now.
14:05:25 <elliott_> also, MrC is that buggy? aargh
14:05:34 <elliott_> that means I'll have to try and get gcc 2 compiled...
14:05:43 <elliott_> which will be so unfun it's hard to even articulate
14:06:21 <Vorpal> elliott_, I forgot what it failed on
14:06:39 <Vorpal> <elliott_> Vorpal: It hurts because it has so much circular logic. <-- :)
14:06:45 -!- cheater99 has quit (Ping timeout: 255 seconds).
14:06:57 <Vorpal> elliott_, anyway, I did not port it for 68k
14:07:07 <Vorpal> elliott_, oh and sheepshaver is too buggy to compile one of the object files
14:07:13 <elliott_> isn't the toolchain semi-agnostic
14:07:29 <Vorpal> elliott_, iirc the compiler switches have rather different names
14:07:40 <Vorpal> anyway ick generates command lines for things
14:07:54 <elliott_> well, whatever, I have to get gcc working SOMETIME!
14:07:58 <Vorpal> elliott_, I needed to compile that one object file on my old ibook, then copy it across.
14:08:11 <Vorpal> sheepshaver crashed on it
14:08:13 <elliott_> Vorpal: what i can't figure out how to do is how to provide an fopen() with filename translation.
14:08:24 <elliott_> without dlopen and friends this sounds impossible
14:08:26 <Vorpal> elliott_, I believe I did filename translation in ick
14:08:34 <elliott_> Vorpal: but not at fopen() level
14:08:40 <elliott_> think cygwin, i have to give this to other programs instead of libc
14:08:53 <Vorpal> elliott_, anyway iirc ick had something like #define PATH_SEP '/'
14:09:11 <Vorpal> elliott_, I had to rewrite some function to not add non-required path separators
14:09:39 <Vorpal> elliott_, foo//bar is just verbose. but foo::bar is plain wrong
14:10:00 <elliott_> Vorpal: amusingly, since translation must handle <user inputted mac path><//../ from program>, you can mix and match!
14:10:22 <Vorpal> elliott_, anyway where do you plan to run MrC?
14:10:36 <ais523> elliott_: the same as x:\ from WIndows, only backwards
14:10:47 <elliott_> I explicitly want to make sure it works on System Software 6
14:10:48 <Vorpal> elliott_, oh so SC then? not MrC?
14:10:52 <ais523> so great you used a symmetrical letter there, or the joke wouldn't have worked!
14:10:53 <Vorpal> SC is the 68k compiler iirc
14:10:54 <elliott_> oh, is there another compiler?
14:11:02 <elliott_> Vorpal: really, that's just a stopgap measure until I get gcc bootstrapped
14:11:02 <Vorpal> elliott_, yes MrC is PPC, SC is 68k
14:11:10 <elliott_> NetBSD can run on them I think
14:11:14 <elliott_> so gcc is feasible, fsvo feasible
14:11:30 <Vorpal> elliott_, didn't apple use to cross compile Mac OS from some unix system or such?
14:11:45 <elliott_> If I were sane I'd just use A/UX.
14:11:56 <Vorpal> elliott_, isn't A/UX apple's?
14:12:17 <elliott_> Vorpal: Anyway, if this actually gets working I'm sure I can buy a Macintosh Plus just to see it run :P
14:12:25 <elliott_> Memory1 MB, expandable to 4 MB (150 ns 30-pin SIMM)
14:12:31 -!- Guest1055 has quit (Ping timeout: 252 seconds).
14:12:36 <Vorpal> elliott_, anyway there was this site with "abandonware" for classic mac
14:12:39 <Vorpal> trying to remember the name
14:12:42 <elliott_> there's several aren't there :D
14:12:42 <fizzie> GNU ld has a "--wrap" thing for that sort of library-wrapping, which lets you override library functions in such a way that the overriding code can call the original. (I can't be bothered to read all the context to see if that's relevant.)
14:13:00 <Vorpal> elliott_, ah http://www.macintoshgarden.org/
14:13:13 <Vorpal> elliott_, that site has various mpw versions iirc
14:13:17 <fizzie> Yes, but you keep talking about GCC too.
14:13:18 <Vorpal> elliott_, and sdks too
14:13:33 <Vorpal> elliott_, are you aware of how the MPW shell works?
14:13:52 <elliott_> Vorpal: Maybe I can avoid MPW entirely by CROSS-COMPILING.
14:13:57 <elliott_> Except I doubt cross-compiling would work at all.
14:14:00 <Vorpal> elliott_, and why ick can't call the compiler (I made it an mpw tool)
14:14:06 <Vorpal> (because that solved stdio issues)
14:14:14 <elliott_> Vorpal: Indeed; thankfully, whateverit'scalled solves that!
14:14:22 <elliott_> Vorpal: Who needs the MPW shell when you have Real Authentic Make?
14:14:24 <ais523> whateverit'scalled is a great name
14:14:34 <ais523> mostly because of the apostrophe
14:14:36 <Vorpal> elliott_, what are you trying to compile for it?
14:14:41 <Ilari> Of course, to use --wrap, one must relink the program.
14:14:53 <ais523> elliott_: I think you may need to implement stdin/out/err by hand
14:14:57 <Vorpal> elliott_, what is your *goal* here
14:14:59 <elliott_> ais523: My current naming scheme has gone something like Unix for Mac -> Uforma -> Euphorma, but I'm not sure where to go from there
14:15:13 <ais523> you could change the m to an i and get a real word
14:15:19 <ais523> which may or may not be used atm
14:15:21 <Vorpal> elliott_, right. You need to write something like a shell for it
14:15:23 <elliott_> well, what's more euphoric than Unix!
14:15:32 <elliott_> Vorpal: Just like Cygwin does.
14:15:36 <Vorpal> elliott_, what do you run bash in
14:15:53 <Vorpal> elliott_, you will need to write a terminal emulator
14:15:53 <elliott_> Vorpal: I want to be able to double click "Start Euphorma" on my System Software 6 desktop, have it start up a vt100 emulator program, and point it at a pty, on which it starts bash.
14:16:05 <elliott_> Vorpal: I could maybe even reuse an existing one for the time being *shrug* i.e. telnet client
14:16:26 <elliott_> Vorpal: But that's a trivial problem compared to all the huge ones :P
14:16:29 <elliott_> Vorpal: For instance: cooperative multitasking.
14:16:47 <elliott_> Vorpal: Current plan is just "whenever we get control from the program, YIELD, and if it does for (;;); that's not our problem".
14:17:00 <elliott_> But if you could set up some kind of timer interrupt and then cause it to yield or something, that would be an option too.
14:17:15 <elliott_> Of course in old System Software versions all Unix software will stop when you tab away.
14:17:40 <elliott_> Vorpal: In a way, it'll be a hosted microkernel plus a POSIX libc.
14:17:43 -!- FireFly has joined.
14:17:52 <elliott_> Hosted microkernel to handle taskswitching, POSIX libc to wrap the existing libc and add Unix functions.
14:18:00 <Vorpal> elliott_, how long before you give this project up do you think?
14:18:07 <elliott_> Not long! But it's fun to think about now.
14:18:16 <elliott_> I suppose I should learn the Macintosh Toolbox.
14:18:25 -!- Guest1055 has joined.
14:18:33 <Vorpal> elliott_, I have no idea about that. I know it exists... That is about all
14:18:39 <Vorpal> elliott_, wasn't it 68k only?
14:18:43 <Vorpal> as in PPC used something else
14:19:14 <Vorpal> elliott_, in the version of MPW I have, Pascal seems to be treated like a second class citizen.
14:19:24 <elliott_> Vorpal: Yah, but the APIs are all Pascal.
14:19:30 <elliott_> That's why you can say "\pxyz" in MPW to get a Pascal string.
14:19:44 <Vorpal> elliott_, I'm fairly sure you need an older one to make it run on 68k. As in last MPW won't
14:20:41 <Vorpal> elliott_, also did 68k have MMUs?
14:20:45 <elliott_> Vorpal: Of course, MPW 3 was the last I think.
14:20:52 <elliott_> But like I said, I can't rely on MPW for too long :P
14:20:58 <fizzie> Vorpal: They used to cross-develop Mac OS stuff from Lisa.
14:21:09 <ais523> elliott_: how are you planning to implement alarm()?
14:21:24 <ais523> if you can't even do non-coop multitasking, then that should be near-impossible
14:21:28 <elliott_> ais523: Does the 68k have timer interrupts? :p
14:21:40 <elliott_> It must do, since Mac OS 9 did mostly preemptive multithreading.
14:21:44 <ais523> I don't know, nor whether you can hook them without doing inline asm or whatever
14:22:06 <ais523> hmm, I think my new esolang I'm mulling over atm will have both pre-emptive and cooperative multithreading, because
14:22:40 <Vorpal> <elliott_> It must do, since Mac OS 9 did mostly preemptive multithreading. <-- wasn't OS 8.something the last to run on 68k
14:22:41 <fizzie> Processors in general tend to have just "interrupts", it's up to the surrounding hardware to cause them using timers.
14:23:01 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
14:23:16 <Vorpal> elliott_, actually OS 9 did preemptive but with one task doing most of the job. That task internally doing cooperative
14:23:17 <elliott_> maybe i could get the fpu to divide by zero at JUST the right moment
14:23:28 <elliott_> i thought it was like the other way around
14:23:35 <elliott_> cooperative at the top layer, and main task doing preempting
14:23:35 <Vorpal> elliott_, you had to call this main task for stuff like file system and what not even
14:23:51 <Vorpal> elliott_, so in practise it wasn't preemptive in any useful sense
14:23:52 <elliott_> but yeah, i need to run on 68k.
14:23:55 <elliott_> it's just not impressive if I can't!
14:24:26 <Vorpal> <elliott_> cooperative at the top layer, and main task doing preempting <-- except the top layer being thicker than the main layer
14:24:47 -!- augur has joined.
14:25:04 <elliott_> <elliott_> maybe i could get the fpu to divide by zero at JUST the right moment
14:25:06 <ais523> `addquote <elliott_> maybe i could get the fpu to divide by zero at JUST the right moment
14:25:15 <ais523> elliott_: I was addquoting it even before you requoted it yourself
14:25:15 <Vorpal> elliott_, as far as I remember about the only useful API for "native preemptive tasks" (called "blue tasks" iirc!) that didn't call the cooperative code was memory allocation
14:25:40 <ais523> I think insightful's better than funny for qdbs
14:25:41 <fizzie> I'm pretty sure a 68k mac would have timer interrupts, but hooking into them would most likely be rather undocumented and kludgy.
14:25:58 <elliott_> fizzie: this whole idea is pretty much undocumented and kludgy.
14:26:16 <elliott_> main priority is getting MPW 3 anyway :P
14:26:23 <Vorpal> wait, wouldn't the CPU have internal timers?
14:26:46 <elliott_> i imagine 68k's inability to support good preemptive multitasking drove the switch away partly.
14:26:55 <Vorpal> I just realised I never done really low level coding for anything except SOCs.
14:26:57 <elliott_> thankfully, _I_ am not constrained by requirements of goodness
14:27:12 <Vorpal> elliott_, also the memory limit I bet
14:27:14 <elliott_> Vorpal: write a boot sector, it's easy and you can say you've done it :P
14:27:25 <elliott_> but 8 megs is plenty, and that's just 24-bit system 6 limit!
14:27:45 <Vorpal> elliott_, fun thing about Mac OS. The OS could move memory around under you
14:27:46 <fizzie> 68k can run Linux just fine (for some values of "fine", anyway...), so there's nothing unfixably wrong in the hardware.
14:27:49 <elliott_> http://www.macintoshgarden.org/apps/macintosh-common-lisp-20 :D MCL
14:27:56 <elliott_> fizzie: Right, and also NetBSD.
14:27:59 <Vorpal> elliott_, you had these kind of handles, and you needed to go through them
14:28:04 <fizzie> Of *course* it runs NetBSD.
14:28:14 <elliott_> fizzie: And thankfully old Mac OS pretty much let you take over everything, so...
14:28:21 <lifthrasiir> http://craftbook.sk89q.com/wiki/Perlstone seems to qualify as a "practical" esolang
14:28:25 <Vorpal> elliott_, since there was not MMU it couldn't use virtual memory to patch up address space, so it had to move stuff around under you
14:28:34 <Vorpal> elliott_, yes kind of gc yes
14:28:42 <elliott_> lifthrasiir: lol, "The language was designed and the interpreter written by Lymia."
14:28:45 <Vorpal> elliott_, but more kind of ghetto virtual memory :P
14:28:53 <lifthrasiir> (and yeah, I just started to play Minecraft very recently)
14:29:18 <elliott_> oh no, the koreans have found another game ending in "craft"
14:29:22 <lifthrasiir> so the esoteric nature is intentional, right? :p
14:29:25 <elliott_> now nothing can stop a global economic meltdown :(
14:29:36 <fizzie> http://p.zem.fi/68k-virtual-memory ← also one of the best solutions ever.
14:29:45 <lifthrasiir> elliott_: no, MC was popular since 2010 in Korea :p
14:29:46 <ais523> I thought Korean MMO crazes mostly stayed confined to (South) Korea
14:29:49 <elliott_> Vorpal: Dem Koreans are known for their... fondness of Starcraft :P
14:29:57 <elliott_> ais523: lifthrasiir is Korean, unless I'm seriously mistaken
14:30:04 <Vorpal> fizzie, did macs use the original 68000 or?
14:30:12 <ais523> I mean, it's not the whole world that's doomed, just South Korea
14:30:33 <ais523> I've heard that people voluntarily install rootkits over there to "prove" they aren't cheating in online games...
14:30:40 <fizzie> Vorpal: Early models, yes.
14:30:58 <ais523> which I suppose is actually a sensible idea if you trust the game manufacturer
14:31:11 <Vorpal> fizzie, I heard that story you linked before, but what I never understood why they went to that length to solve it. Why not just use a different CPU?
14:31:24 <lifthrasiir> ais523: afaik it is rather involuntarily, but that sounds plausible
14:31:25 <fizzie> Mac II seems to be 68020-based already.
14:31:45 <ais523> lifthrasiir: well, when the alternative is not running the game
14:31:57 <lifthrasiir> as an account forgery is quite a problem nowadays
14:32:13 <elliott_> <ais523> I mean, it's not the whole world that's doomed, just South Korea
14:32:33 <Vorpal> elliott_, true but I remember having to reboot mac os due to buggy programs many times
14:32:41 <elliott_> Vorpal: yeah but gnu coders are perfect.
14:33:07 <Vorpal> elliott_, anyway, I had MacBugs (or whatever the spelling was... MacsBug?)
14:33:30 <Vorpal> elliott_, and I seen it say "Unable to Find System Heap. Heap list corrupted" or some such
14:33:36 <Vorpal> elliott_, that is why you want an MMU :P
14:34:11 <Vorpal> elliott_, remember the FS wasn't journaling either :P
14:34:27 <Vorpal> elliott_, oh I could debug. I just couldn't kill the app. Had to reboot.
14:35:11 <elliott_> never mind, euphorma cancelled, perfect os discovered
14:35:16 <Vorpal> elliott_, well that was OS 9, so not quite.
14:35:47 <elliott_> Temporarily unavailable due to high bandwidth costs
14:35:57 <Vorpal> elliott_, well anyway System 7 didn't boot instantly on hardware available when System 7 was new
14:36:12 <Vorpal> elliott_, possibly. Don't remember that.
14:36:21 <Vorpal> elliott_, what OS did Classic run?
14:36:28 <Vorpal> as in Macintosh Classic
14:37:05 <Vorpal> elliott_, when I was small my dad had one of these: http://en.wikipedia.org/wiki/Macintosh_Classic
14:37:20 <elliott_> The Classic used the System 6.0.7 operating system with support for all versions up to System 7.5.5. A hidden Hierarchical File System (HFS) disk volume contained in the read-only memory (ROM) included System 6.0.3.[17] The Mac Classic could be booted into System 6.0.3 by holding down the Command + Option + X + O keys during boot.[17]
14:38:07 <Vorpal> elliott_, why not use the OS included originally then (6.0.7)
14:38:10 <ais523> recovery was my guess when I saw that
14:38:19 <elliott_> Vorpal: god knows; not available when that part was being designed?
14:38:20 <ais523> and presumably because 6.0.3 was known not-fatally-buggy, 6.0.7 wasn't
14:39:00 <elliott_> "I'm learning Objective-C and my friend have a real Macintosh IIci, that uses a Mac System 7(specifically 7.5.5 with a 68k processor) and I've installed Metrowerks C/C++ IDE(I think it's the version 1, but I don't know), but i didn't tested it, then i want to know one thing: It's possible to develop in Objective-C using NSObjects/Objects and AppKit or something like this on it? Thanks.
14:39:38 <elliott_> ais523: hmm, you've used compilers that support 16-bit archs, right?
14:39:38 <Vorpal> elliott_, also Apple likes to reuse product names
14:39:55 <Vorpal> "The Classic featured several improvements over the Macintosh Plus, which it replaced as Apple's low-end Mac computer: it was up to 25 percent faster than the Plus and included an Apple SuperDrive 3.5" floppy disk drive as standard.[16] The SuperDrive could read and write to Macintosh, MS-DOS, OS/2, and ProDOS disks."
14:40:01 <ais523> despite having used them, I didn't really understand how they worked
14:40:14 <Vorpal> elliott_, SuperDrive. Wasn't that Apple's DVD-RAM drive during the G3/G4 era?
14:40:28 <ais523> it was more on the "run the command and hope" level, like I'm currently at with Marst
14:40:48 <Vorpal> elliott_, they reused the trademark is what I'm saying
14:41:34 <Vorpal> elliott_, lack of in this case?
14:41:49 <Vorpal> elliott_, wait you mean *that* key repeat?
14:41:50 <elliott_> "Installing binutils and GCC as cross-compiler for the Motorolla 68000"
14:41:55 <Vorpal> people use it to type?
14:42:08 <elliott_> i just sortal inger on the e key
14:42:10 <Vorpal> oh you are tired. Okay
14:42:22 <elliott_> so wait since when does m68k be supported by...
14:42:33 <ais523> elliott_: I didn't realise it was possible to be so tired that you couldn't press a given key twice in a row
14:42:39 <ais523> also, I recommend sleeping, in such cases
14:42:47 <elliott_> ais523: at 2 pm? that would be unwise
14:42:56 <elliott_> also, my brain's still awake, mostly, it's just everything else that's being stupid
14:43:02 <elliott_> (note: this is almost always true)
14:43:06 <fizzie> GCC is what Linux/m68k runs on.
14:43:18 <Vorpal> elliott_, you need sleep then
14:43:20 <elliott_> fizzie: Yeeees, but I'm trying to produce Mack-in-tosh boonaries here.
14:43:26 <ais523> elliott_: I tend to be sufficiently bad at sleeping at will, that when I do sleep, it's often at arbitrary times of day
14:43:28 <Vorpal> <fizzie> GCC is what Linux/m68k runs on. <-- wait what
14:43:33 <elliott_> I am not sure what the executable format is. :p
14:43:35 <Vorpal> Linux/m68k runs on GCC?
14:43:44 <Vorpal> since when was GCC an OS or such
14:43:46 <fizzie> elliott_: That's just a binutils issue.
14:44:05 <Vorpal> fizzie, "is compiled with"?
14:44:14 <ais523> hmm, is gcc turing-complete?
14:44:21 <ais523> as in, the compiler itself, rather than the executables it produces
14:44:30 <ais523> (adjust for infinite memory)
14:44:40 <ais523> oh, obviously, C++ templates
14:44:43 <elliott_> anyone know what format 68k binaries are? :P
14:44:44 <ais523> what about just the C part?
14:44:45 <Vorpal> ais523, then we first need to define what it executes. Command line options?
14:44:46 <fizzie> The m68k is also what you might call "moderately 32-bit"; the address bus is just 24 bits (at least in 680N0 with small N).
14:44:50 <Guest1055> I got a captcha with "pi-calculus"
14:44:53 <elliott_> Guest1055: hello slereheareareareareareah
14:45:00 <elliott_> reearearafreaeereearreeearaerarareera
14:45:02 -!- Guest1055 has changed nick to Slereah.
14:45:15 <ais523> Slereah: did you type it?
14:45:24 <Slereah> I just switched to another one
14:45:27 <Vorpal> fizzie, by that token x86-64 is "moderately 64-bit". 48 bits isn't it?
14:45:36 <fizzie> Slereah: http://zem.fi/~fis/faircaptcha.png
14:45:41 <Vorpal> fizzie, and physical bus tends to be between 36 and 42 or so
14:45:51 <Vorpal> (probably more for server CPUs)
14:46:04 <elliott_> fizzie: it is inglip's commad.
14:46:06 <Slereah> That's the problem of using captchas as OCRs
14:46:33 <elliott_> fizzie: well system 7 had more than 24-bit
14:46:50 <Vorpal> <fizzie> Slereah: http://zem.fi/~fis/faircaptcha.png <-- did you enter the TeX code?
14:47:43 <Slereah> Is cyrilic even in basic latex?
14:47:50 <elliott_> what executable format did 68k macintosh system software use
14:48:02 <elliott_> i think the answer is obvious, and it is yes
14:48:10 <fizzie> elliott_: Seems that 68020 and later ones had a full 32-bit address bus. (Discounting the el-cheapo model 68EC020.)
14:48:24 <fizzie> Vorpal: It was a comment form and I didn't have anything to say, so no.
14:48:58 <elliott_> "Oops"; accidentally typed #minecraft instead.
14:48:59 <fizzie> They had those 68k/ppc "fat binaries" even back then, like they have ppc/x86 now.
14:49:12 <elliott_> fizzie: "Now"; not really any more :P
14:50:15 -!- elliott_ has changed nick to elliott.
14:50:17 -!- elliott has quit (Changing host).
14:50:17 -!- elliott has joined.
14:52:01 <elliott> ha ha i have bamboozled them iwth the complexiti of my questionssodifj
14:52:17 <fizzie> The old executable format seems to be called PEF, and according to http://en.wikipedia.org/wiki/Preferred_Executable_Format it's co-used by BeOS on PPC.
14:52:35 <fizzie> Well, at least the "fat binary" m68k/ppc one.
14:52:39 <Vorpal> <elliott_> what executable format did 68k macintosh system software use <-- what one did it use?
14:52:47 <fizzie> Might well be that thin binaries used something completely different.
14:52:48 <elliott> <elliott> So... anybody happen to know what executable format 68k Mac OS used?
14:52:48 <elliott> <mikeash> I don't recall what its name was, but it involved pieces of code stored as individual resources in the app's resource fork
14:52:48 <elliott> <elliott> Well, that sounds pleasant. I was hoping it'd be something I could coerce binutils into generating with only mild pain. Naive of me.
14:52:53 <elliott> Vorpal: Something terrifying.
14:53:01 <Vorpal> oh right PEF. I think I saw that somewhere.
14:53:06 <Vorpal> elliott, yes the CODE resource I think
14:53:33 <Vorpal> elliott, hey there was a "plugin" to resedit that added a disassembler for CODE resources
14:53:48 <elliott> Yeah, now tell me how to make them from outside Macq :P
14:53:59 <Vorpal> elliott, you need to handle resource fork
14:54:21 <Vorpal> elliott, I suggest generating a MacBinary or .hqx file as compiler output, then unpacking that on a mac
14:54:43 <Vorpal> elliott, anyway even PPC apps relied quite a bit on the resource fork to work
14:54:55 <Vorpal> though the actual code was IIRC in the data fork then
14:55:42 <elliott> technically i don't even wanna generate apps i think
14:55:46 <elliott> because i don't want them to spawn gui crap
14:55:47 <Vorpal> elliott, the resedit disasemmbler was quite nice. It could follow jumps when you clicked on jump or call instructions
14:55:54 <Vorpal> and also backtrack any jumps to a specific line
14:56:03 <elliott> <mikeash> it was a fun setup, code would be split into many resources and this was used as a primitive form of swap
14:56:23 <Vorpal> elliott, anyway you need one app that you run to start this thingy
14:56:44 <elliott> Vorpal: of course. i was thinking that one would be the EuphormaKernel
14:56:51 <elliott> and the other would just be EuphormaTerm
14:56:56 <elliott> and "Start Euphorma" would just open both
14:57:16 <elliott> except the terminal would have to pass control to the kernel.
14:57:28 <fizzie> Here, have some file format documentation (disclaimer: it might only have the m68k XCOFF object file format docs): http://mirrors.vanadac.com/ftp.apple.com/developer/Tool_Chest/Core_Mac_OS_Tools/MPW_etc./Documentation/MPW_Reference/File_Formats.sit.hqx
14:57:29 <elliott> Vorpal: err, because all the terminal does is a pty thing?
14:57:34 <elliott> i suppose i could make the unix thing a library
14:57:37 <Vorpal> elliott, oh and remember you can only start one instance of an app
14:57:38 <elliott> and compile it into the terminal
14:57:47 <Vorpal> so yeah you need to do the programs inside differently
14:57:50 <elliott> fizzie: that XCOFF thing, can i use it without worrying about all these forks?
14:58:20 <fizzie> I would think the object files are rather forkless, but not really sure.
14:58:22 <Vorpal> elliott, you *need* resource fork for the kernel/terminal bit though. Or at least for some type of launcher
14:58:33 <elliott> fizzie: So the only problem there is STARTING them :)
14:58:34 <Vorpal> fizzie, also you can't run an object file
14:58:44 <elliott> I need _some_ kind of linkage
14:58:50 <fizzie> Vorpal: The only "application" could be a MPW-built thing that just jumps into otherwise-generated code.
14:59:07 <elliott> fizzie: This is getting EXTRAORDINARILY like a kernel :P
14:59:09 <Vorpal> elliott, anyway even mpw tools won't work without their resource fork
14:59:14 <elliott> "So yeah, you open object file sand jump to them and ..."
14:59:18 <Vorpal> which is why the ick mac port was such a pita
14:59:24 <Vorpal> elliott, you know the .img format?
14:59:39 <Vorpal> elliott, they have resource forks that are vital for the image to be mountable
14:59:43 <Vorpal> so you need to .bin them
14:59:56 <Vorpal> still .img.bin is fairly reliable
14:59:59 <Vorpal> better than .sit anyway
15:00:24 <elliott> Vorpal: btw mac os 8 in basilisk ii = so crashy
15:00:34 <Vorpal> elliott, I can imagine. OS 7 is crashy too
15:00:37 <elliott> i just want a stable emulator that lets me have a big resolution :(
15:00:40 <elliott> Vorpal: it was emulator bugs
15:01:26 <fizzie> There's even some XCOFF support in binutils. (It was used in AIX by IBM too.)
15:01:35 <Vorpal> elliott, anyway I decided to port ick (instead of something else) for a good reason
15:01:39 <elliott> fizzie: So I should compile a gcc for XCOFF, yah?
15:01:45 <Vorpal> elliott, I already knew ick was portable to hell and back
15:02:00 <Vorpal> thus figuring it wouldn't be too much work (still was quite a bit)
15:02:03 <elliott> I have a feeling I should jump right into gcc 2 so I know what horrors await me in the future.
15:02:19 <elliott> fizzie: Alsoalso, that vandac thing isn't loading now.
15:02:40 <Vorpal> what was the PPC executable format called now again
15:02:48 <fizzie> Well, it might be borderline possible, anyway; I wouldn't expect it to *work*. And I would be rather surprised if MPW would bother linking with gcc-generated xcoff objects.
15:03:17 -!- BeholdMyGlory has joined.
15:03:20 <elliott> fizzie: AFAICT MPW will be the hacky-libc and the kernel, and gcc will be all the apps.
15:03:23 <Vorpal> fizzie, wasn't that for 68k?
15:03:28 <elliott> Hey, gcc 1.21 is available.
15:03:32 <Vorpal> fizzie, PPC used a different one that 68k I know
15:03:41 <fizzie> Vorpal: No, it's mostly a PPC format, though it also does m68k/PPC fat binaries.
15:03:41 <elliott> ftp://ftp.mirrorservice.org/sites/sourceware.org/pub/gcc/old-releases/gcc-1/gcc-1.21.tar.bz2
15:03:51 <Vorpal> fizzie, yes fat binaries is yet another thing
15:04:07 <Vorpal> fizzie, but I'm 99% sure that PPC and 68k used different binary formats for "applications"
15:04:15 <Vorpal> as in they had different names
15:04:55 <fizzie> Vorpal: Well, PEF is what OS X still runs on PPC Macs, "However, PEF is still supported on PowerPC-based Macintoshes and is used by some Carbon applications ported from earlier Mac OS versions", so the older x86-only format probably is something else.
15:04:58 <fizzie> I don't quite know what.
15:05:16 <elliott> has a configure and everything
15:05:41 <fizzie> Bah, and I don't even have elliott's no-sleep excuse here.
15:05:46 <elliott> --build=BUILD configure for building on BUILD [BUILD=HOST]
15:05:46 <elliott> --host=HOST configure for HOST [determined via config.guess]
15:06:08 <elliott> Configuring for a x86_64-unknown-linux-gnu host.
15:06:08 <elliott> Invalid configuration `x86_64-unknown-linux-gnu': machine `x86_64-unknown' not recognized
15:06:08 <elliott> Invalid configuration `x86_64-unknown-linux-gnu': machine `x86_64-unknown' not recognized
15:06:08 <elliott> Unrecognized host system name x86_64-unknown-linux-gnu.
15:06:20 <elliott> but i doubt recent 4 versions support m68k-xcoff :)
15:06:27 <fizzie> Amiga's m68k executable format was "Hunk", and of course OS-X uses "Mach-O"... is it just me, or are all these formats pretty masculine? (Well, discounting ELF...)
15:07:56 <Vorpal> fizzie, what about COFF? PE?
15:08:16 <fizzie> elliott: The binary format support probably depends more on binutils/libbfd-or-what-was-it-called; might be possible to mix-n-match versions.
15:08:32 <elliott> fizzie: gcc 2 gave up when it saw that I was this strange "x86_64" beast.
15:08:34 <Vorpal> fizzie, *up to a point*
15:08:35 <Gregor> But ELF is the best format :P
15:08:37 <elliott> I think gcc 3 will do just fine. :p
15:08:47 <elliott> But I draw the line at 4, no way am I going to use 4.
15:08:55 <elliott> That's waaaaaaaay too detached from anything I could ever run on the actual machine ever :P
15:09:00 <Gregor> elliott: At some point there were ports of GCC2 to x86_64, they're just wildly nonstandard.
15:09:08 <Vorpal> ELF is fairly sane for a *nix like system.
15:09:21 <elliott> Gregor: It's kinda irrelevant because this is just a bootstrapping anyway :P
15:09:27 <Vorpal> too much overhead for old limited machines though
15:09:32 <elliott> Gregor: A really complicated bootstrapping involving multiple compilers and writing a userspace Unix kernel.
15:09:47 <Vorpal> Gregor, he is going to run on a MMU-less 68k system. ELF is just wasted there
15:09:50 <elliott> Gregor: Unless you think Microcosm could be made to work on a 68k Macintosh ...
15:10:05 <elliott> But yeah, ELF would be a tremendous waste of time here :P
15:10:19 <elliott> Heck, COM++ (like .COM, but with moar space!) would do fine X-D
15:11:25 <elliott> hmm, --target=m68k-xcoff is what I want, right?
15:11:30 <elliott> crosscompiling lingo confuses me
15:11:33 <cheater00> you know that COFF and PE are mutually exclusive?
15:11:40 <Vorpal> elliott, I would suggest a custom format unless that is too much work. Something like a header with a magic byte and some offsets then just two sections following: code, data. Header should contain a .bss style segment size as well (but that won't be in binary of course)
15:11:52 <cheater00> there's a physiological mechanism in our bodies that disables us from coughing/sneezing, and at the same time peeing.
15:11:57 -!- Tritonio has joined.
15:12:14 <elliott> Vorpal: ...I would still like it if Unix programs could theoretically call the Toolbox, though :P
15:12:17 <Vorpal> elliott, since you don't have MMU you want relocatable code btw. Otherwise running two copies of the same program would be painful
15:12:20 <elliott> I want something closer to Cygwin, not coLinux.
15:12:38 <Vorpal> elliott, and uh, self modifying code would be a pain
15:12:46 <elliott> Here's my relocatable code format.
15:12:50 <elliott> At the start, there's a list of memory locations.
15:13:05 <elliott> Whenever you load the file, you increment the values in all those memory locations of the program image by the base you loaded it at.
15:13:14 <elliott> The assembler handles the hard part of figuring out what's an address or not
15:13:18 <Gregor> Yeah, ELF is kinda useless for that, just use a.out.
15:13:19 <Vorpal> elliott, sounds like .dll
15:13:27 <elliott> Gregor: a.out? Even that's a bit much
15:13:29 <Gregor> a.out is just .COM but less stupid.
15:13:40 <elliott> Gregor: It has segments and relocations and tables :P
15:13:42 <Vorpal> elliott, hell you need some way to handle conflicting addresses. With no mmu, loading things that want to be at the same place would be a pain
15:13:52 <elliott> Vorpal: thus why relocatable code
15:13:59 <Gregor> elliott: a.out is just .COM without wasting space for zeros.
15:14:08 <Vorpal> elliott, I don't know if you can do PC relative addressing on that thing but I doubt it
15:14:14 <elliott> JUST GONNA ASSUME THAT --TARGET= IS THE RIGHT THING
15:14:17 <elliott> Vorpal: Thus load-time code rewriting.
15:14:27 <Vorpal> elliott, so you are doing PE basically?
15:14:34 <elliott> Vorpal: You could do it without a big table of addresses.
15:14:44 <elliott> Vorpal: If I know which opcodes have addresses in their Nth position,
15:14:48 <elliott> then I can just do it to the whole code segment.
15:15:08 <elliott> Bleh, I need to compile binutils, don't I
15:15:11 <Vorpal> elliott, won't work without a separate .rodata segment. Otherwise you can't tell code from embedded read only code
15:15:27 <elliott> So don't allow embedded read-only data!
15:15:37 <Vorpal> elliott, which would break stuff :P
15:16:06 <Vorpal> elliott, anyway you want to share read only code/data, your address space is tiny remember
15:16:13 <elliott> Vorpal: 8 MiB is not tiny.
15:16:23 <fizzie> I can't quite make out what form that xcoff supports in current binutils takes, it seems a bit AIX-specific... anyway, binutils 2.20 build with --target=m68k-coff starts all right but then dies with "This target is no longer supported in gas"; so at least that's not a good version to use.
15:16:23 <ais523> gah, tail recursion is so hard in call-by-name
15:16:31 <elliott> ais523: what about in call-by-push-value?
15:16:34 <ais523> well, implementing it is trivial, but writing programs that are actually tail-recursive is tricky
15:16:49 <ais523> elliott: I haven't looked into the exact details of that, but I think it's just Underload-style
15:16:52 <elliott> fizzie: Nobody likes this, do they
15:16:52 <fizzie> (2.20 is of course extra-new.)
15:17:02 <ais523> and so transforms to and from call-by-name with about equal difficulty both ways
15:17:13 <ais523> as long as you have first-class functions, which you always do in computer science
15:17:38 <elliott> fizzie: Think 2.15 might work? May 2004. You have the inchoowishion for these things.
15:18:14 <fizzie> No clue whatsoever; binutils documentation is horrible in that it doesn't really tell anything about supported formats.
15:18:35 <fizzie> The only thing I can glean from the manual is that at least 68k in general is still there. :p
15:19:23 <Vorpal> fizzie, doesn't change log list dropped targets?
15:20:47 <ais523> haha, Deewiant's antislowpoke attempt also beats allegro
15:20:55 <elliott> <mikeash> that is impressively crazy
15:21:05 <fizzie> There's an uClinux build of m68k-coff-gcc-2.7.2.3; that's at least right processor architecture and object file format family.
15:21:25 <elliott> <elliott> Which involves some sort of saner-than-MPW compiler for compiling the actual programs, so...
15:21:25 <elliott> <elliott> But hey, how hard can loading a.out be?
15:21:25 <elliott> <mikeash> this hard: |------------------O---|
15:21:32 <ais523> pity it doesn't beat FFPSG too, that would have been hilarious
15:21:39 <Deewiant> ais523: Yes, I complained about that when I submitted it :-P
15:21:45 -!- asiekierka has joined.
15:22:00 <fizzie> Then you can just translate coff to xcoff; it's just one character more, how hard could it be?
15:22:13 <Vorpal> elliott, did A/UX run under MacOS?
15:22:22 <Vorpal> elliott, or separately?
15:22:22 <elliott> Vorpal: It was a separate Os that had Mac OS inside it sort of.
15:22:48 <Gregor> I would have to read up on it to remember, but IIRC base a.out has no relocations ... there were extensions in various BSDs that did.
15:22:56 <elliott> fizzie: So er what *is* this xcoff thing like?
15:23:05 <elliott> It says it's improved and expanded.
15:23:08 <elliott> That sounds like bloat to me.
15:23:08 <elliott> [[XCOFF, for "eXtended COFF", is an improved and expanded version of the COFF object file format defined by IBM and used in AIX. Early versions of the PowerPC Macintosh also supported XCOFF, as did BeOS.]]
15:23:14 <elliott> fizzie: OK, so 68k doesn't actually use it :P
15:23:33 <fizzie> So I've heard, anyway.
15:23:35 <elliott> Mind you, a.out is pretty damn tempting at this point...
15:23:39 <elliott> fizzie: MPW does PPC too, though.
15:23:49 <Vorpal> elliott, don't ask me. I only really looked at the PPC stuff
15:23:52 <Gregor> It does have segments though, yeah.
15:24:17 <Vorpal> elliott, but yeah I think it says XCOFF on the file icon for ppc objects
15:24:23 <Vorpal> elliott, but something else on the 68k ones
15:24:36 <fizzie> elliott: Well, it *is* the "Common Object File Format", mostly platform-agnostic.
15:24:48 -!- Sgeo has joined.
15:25:05 <elliott> So if I step back and just get COFF and then make it relocatable... hooray, PE :-P
15:25:14 <elliott> And I'm like 90% sure a.out is simpler than PE.
15:25:22 <fizzie> "Where is the 68K object file format documented?
15:25:22 <fizzie> In the "MPW 68K Object File Format" document (download)."
15:25:32 <Gregor> Ohyeah, it has relocations only for .o files :P
15:25:41 <Gregor> a.out is way simpler than PE.
15:25:42 <fizzie> The "download" link is to the File_Formats.sit.hqx.
15:25:59 <fizzie> So maybe they used their own custom format for m68k objects, then.
15:26:10 <elliott> Gregor: Programs are .o with all their library functions mixed in.
15:26:21 <Vorpal> fizzie, if the link works (and you paste it) I could fire up sheepshaver to unpack it
15:26:22 <ais523> is .o. some sort of smiley?
15:26:22 <elliott> Gregor: Alternatively: Programs are .o, and I give them a pointer to the shared libc space X-D
15:26:51 <fizzie> Vorpal: The mirror worked earlier today, but now it seems to be gone. Probably due all this excitement.
15:27:06 <elliott> I'm still disappointed that MPW used file extensions.
15:27:17 <elliott> Classic Mac OS is meant to be UNCOMPROMISINGLY Maccy!
15:27:34 <elliott> Honestly, they should have renamed stdio.h to "Standard I/O" in the Headers folder.
15:28:14 <ais523> and just called it stdio.h in the program itself
15:28:23 <elliott> ais523: No, no, you should have to write it
15:28:42 <ais523> (neither C89 nor C99 requires any sensible mapping from what you write in #include lines to what the file is called, or even if it's implemented via header file at all)
15:28:44 <cheater00> and also, you don't want a hash.. you would want something more self-explanatory
15:28:50 <elliott> ais523: And of course you can't have a header and a source file named the same in the same directory, so you end up having Headers and Sources folders in every project.
15:29:02 <elliott> And the filetype codes are lovingly set.
15:29:10 <cheater00> Dear Compiler, please include the library called ,,Standard I/O´´
15:31:46 <Vorpal> elliott, wait idea! Do this in HyperCard
15:32:36 <fizzie> That's funny, binutils-2.11.2 has an include/mpw/ directory with a README saying "This is a collection of include files that help imitate Posix in MPW."
15:33:18 <elliott> fizzie: Anything interesting in there?
15:33:20 <fizzie> I wonder what that's for; perhaps for building binutils with MPW, then.
15:33:59 <Vorpal> fizzie, is there any files other than that README in there?
15:33:59 <fizzie> Seems to be mostly about reading and writing files.
15:34:13 <fizzie> Though it does "#define LOSING_TOTALLY", accurately enough.
15:35:12 <Vorpal> elliott, which type of 68k object files are you interested in
15:35:16 <Vorpal> there seem to be a lot
15:35:25 <elliott> Vorpal: Whatever's easiest to relocate and jump to.
15:35:29 <Vorpal> I wonder what CFM64KLibraries is
15:35:33 <elliott> Vorpal: Also it'd be nice if it worked with PPC too.
15:35:38 <elliott> But "most" object formats are architecture-independent.
15:35:45 <elliott> I don't need fat binraies.
15:36:08 <Vorpal> elliott, the icon for these look like a document symbol with ones and zeros on it
15:37:15 <Vorpal> elliott, then there is the folder PLibraries. Since the other were in CLibraries I guess PLibraries is pascal
15:37:33 <elliott> fizzie: So did you see any Toolbox docs on that ftp that's now down?
15:37:37 <Vorpal> elliott, anyway the PPC one says 100\nXCOFF\n011 on them
15:37:41 <elliott> Those will probably be invaluable. I don't even know how to spawn a process!
15:37:46 <Vorpal> elliott, with the middle line in red and the first and last in blue
15:37:49 <elliott> Vorpal: Maybe I'll just use a.out for both.
15:37:57 <elliott> Vorpal: This thing is becoming SUCH a kernel :P
15:38:30 <fizzie> I got sidetracked looking at this binutils-2.11.2 tarball. Back then it seems to have this whole thing for supporting building the whole set with MPW. Mostly in order to use a PPC-targeting GCC there.
15:38:38 <elliott> Vorpal: In that you can't do anything with its files without starting and running the whole kernel.
15:38:42 <elliott> And the kernel is intimately tied to the formats.
15:39:01 <Vorpal> elliott, anyway I'm not sure MPW uses file extension that much. Both foo.o and foo.c.x seem style naming of XCOFF files seems to be in use
15:39:21 <Vorpal> also .c.o seems to be 68k object file in that case
15:39:42 <elliott> $ ./configure --target=m68k-aout --prefix=/opt/mac
15:40:42 <fizzie> elliott: At least you can then use the raw-binary formats (and maybe a custom linker script) to extract raw m68k code you can then wrap in whatever you like.
15:40:53 <elliott> fizzie: Ye-es, but I could do that with the Totally Native Fork format, too.
15:40:57 <elliott> I'm trying to avoid MORE work here :P
15:41:03 <Vorpal> fizzie, where does that dump data segment?
15:41:31 <Vorpal> elliott, how are you going to transport a resource fork to a mac after cross compiling?
15:41:44 <elliott> I'm going to use a.out. :p
15:41:58 <Vorpal> is the fork data format even documented?
15:42:22 <Gregor> My department's web page added a "featured faculty member" to the front page.
15:42:25 <Gregor> It chooses a random faculty member and shows their name and face.
15:42:47 <fizzie> Vorpal: It dumps the sections in whichever addresses you link them. I don't know if it actually generates huge files if you link things sparsely.
15:43:12 <elliott> gcc -DHAVE_CONFIG_H -I. -I. -I. -D_GNU_SOURCE -I. -I. -I../bfd -I./config -I./../include -I./.. -I./../bfd -I./../intl -I../intl -DLOCALEDIR="\"/opt/mac/share/locale\"" -W -Wall -Wstrict-prototypes -Wmissing-prototypes -g -O2 -c app.c
15:43:12 <elliott> In file included from ./targ-cpu.h:1,
15:43:12 <elliott> from ./config/obj-aout.h:25,
15:43:13 <elliott> from ./config/te-generic.h:19,
15:43:21 <elliott> ./config/tc-m68k.h:212: error: array type has incomplete element type
15:43:41 <elliott> --target= is right, right? :-P
15:43:44 <Vorpal> elliott, hey you can click a line in the resedit disassembler and find any references to that address
15:43:55 <Vorpal> and it shows with nicely coloured arrows
15:43:57 <elliott> As in, I'm on A, I want to build a compiler that runs on A, and I want it to produce binaries for B
15:44:05 <Vorpal> for stuff like conditional/unconditional jumps
15:44:07 <fizzie> Yes, that sounds correct.
15:44:33 <elliott> fizzie: What was that nice version you mentioned?
15:44:36 <fizzie> (Time to go catch the bus homewards.)
15:44:38 <elliott> With all the suppooooooort.
15:44:59 <fizzie> m68k-coff-gcc-2.7.2.3 in the context of uClinux I did see.
15:45:37 <fizzie> 2.11.2, yes; but it was support for compiling binutils with MPW.
15:45:42 <Vorpal> elliott, he is going to a bus
15:45:45 <elliott> Yes, but that sounds suitable 2001-vintage.
15:45:58 <fizzie> Well, you can try; but it seemed quite powerpc-oriented in the docs.
15:46:12 <elliott> fizzie: It'll be just the same, it's just tahat older = nicerrrr!!!11
15:46:17 <elliott> I dont wanna build it with mpw no heavens i am not the crazies.
15:46:25 <fizzie> Anyway, if gcc-2.7.2.3 compiles as m68k-coff, the same age might work for m68k-aout.
15:47:25 <Vorpal> now you made me want to replay Avernum
15:47:31 <Vorpal> elliott, and I don't have time for that
15:48:00 <Vorpal> elliott, I will refrain from it. That would be several days of playing.
15:48:31 <Vorpal> elliott, that doesn't make sense
15:48:38 <Vorpal> elliott, it runs under mac os, not unix
15:48:44 <elliott> Euphoma can invoke native programs.
15:48:46 <Vorpal> elliott, also it is PPC
15:48:55 <elliott> Euphoma is meant to be portable :P
15:49:04 <elliott> Things that work on 68k should work on PPC, I'm just going lowest-common-denominator here to start with.
15:50:06 <elliott> bucomm.o: In function `make_tempname':
15:50:06 <elliott> /opt/mac/src/binutils-2.11.2/binutils/bucomm.c:246: warning: the use of `mktemp' is dangerous, better use `mkstemp' or `mkdtemp'
15:50:50 <Gregor> <elliott> Things that work on 68k should work on PPC, I'm just going lowest-common-denominator here to start with. <-- best logic ever
15:51:06 <elliott> Gregor: It's basically true, the m68k is retarded, PPC is not :-P
15:51:19 <elliott> Every technique or format I choose for m68k out of simplicity will work on PPC, as wasteful or stupid as it might be.
15:51:37 <elliott> extern struct relax_type md_relax_table[];
15:51:47 <elliott> OK so, this shit doesn't work with gcc 4 X-D
15:52:11 <Vorpal> Gregor, apple added emulation for 68k basically :P
15:52:25 <Vorpal> Gregor, so yes it should work on ppc macs
15:52:48 <Gregor> Vorpal: I assumed he meant work as in be convertible, not work as in "I can run it in a VM herp durp"
15:52:57 <Gregor> Since unlike him, I assume people aren't idiots.
15:53:24 <elliott> Gregor: An assumption that has cost you a great amount.
15:53:25 <Vorpal> Gregor, classic MacOS is retarded because they didn't want to break compatibility :P
15:53:30 -!- oerjan has joined.
15:53:35 <Vorpal> Gregor, so bad design choices stayed
15:53:42 <elliott> Gregor: For instance when Bush won, that must have been quite surprising.
15:54:17 <elliott> <Gregor> Since unlike him, I assume people aren't idiots.
15:54:55 <Gregor> OK, let me put that differently :P
15:55:02 <Gregor> I address people under the assumption that they are not idiots.
15:55:16 <Gregor> In spite of whatever my personal assumptions or guesses may be, contrary or otherwise.
15:55:33 <fizzie> If m68k is retarded, 8086 is some sort of ultra-retarded then. (In other words, diss on the OS/systems, not on the cpu arch.)
15:55:41 <elliott> Gregor: When I do that, usually I spend half an hour backwards-and-forthing before I realise that they did something stupid, and I tell them, and they're all "why didn't you tell me".
15:55:48 <elliott> And it's because I assumed they weren't fucking idiots
15:55:52 <elliott> So it's for your own good.
15:56:44 <oerjan> one might consider that it would be an idea to make one's assumptions have a finer gradation than "idiot" and "never says anything idiotic"
15:57:03 <elliott> oerjan: sure, people earn that granularity
15:57:12 <elliott> it's called intelligent cpu cycle allocation
15:57:21 <elliott> Gregor had not yet earned such granularity in the context of minecraft :P
15:57:31 <Vorpal> fizzie, x86 is retarded though :P
15:58:08 <oerjan> elliott: i am pointing out that it is not a good idea have those two as the default options
15:58:18 <elliott> oerjan: got a better selection?
15:58:22 <Gregor> Vorpal only uses ... Alpha.
15:58:43 <Vorpal> Gregor, har. I don't know that arch well enough to comment on retardedness of it
15:59:04 <Gregor> Vorpal only uses ... MIPS?
15:59:16 <elliott> itanium's instruction set looks pretty cool, am i weird for thinking this?
15:59:27 <Gregor> VORPAL ONLY USES HARDWARE IMPLEMENTATIONS OF JAVA
15:59:29 <Vorpal> Gregor, I'm pragmatic though. I use x86-6.
16:00:05 <elliott> Vorpal secretly uses a Reduceron.
16:00:07 <oerjan> elliott: not that i can explain in words
16:00:15 <Gregor> Who needs numbers greater than 64 anyway.
16:00:31 <elliott> Gregor: It can express the number of bits in the more common variant of x86!
16:00:38 <elliott> But not, however, the original 8086 model number.
16:00:49 <elliott> Wait, it's actually 0 to 63.
16:00:57 <elliott> IT CAN'T EVEN REPRESENT THE CONCEPT OF 64 BITS
16:01:06 <Gregor> Yeah, I should have said "greater than or equal to" :P
16:01:18 <Vorpal> elliott, btw you said eww about Harvard arches but there is actually quite a good reason to use them in SOC. You can use flash that you can execute directly from. No need to copy program to RAM
16:01:23 <Gregor> Where are our bit-addressible archs :P
16:01:32 <Vorpal> elliott, saves silicon space
16:01:39 <elliott> Vorpal: von neumann computers are kinda like cellular automata if you squint tho
16:01:46 <elliott> Gregor: Bit-addressable would rock :P
16:01:55 <Vorpal> elliott, not sure how that addresses my statement :P
16:01:57 -!- Sgeo_ has joined.
16:02:14 <Vorpal> elliott, yes and it doesn't address what I just said
16:02:32 <elliott> Gregor: How about this: Everything is 1 bit, there are 256 registers, memory is bit-addressed by passing 64 values.
16:02:33 <Vorpal> also I don't see how "von neumann computers are kinda like cellular automata if you squint tho"
16:03:23 <elliott> Gregor: If the base registers are a[0-256], then bytes are h[0-32].
16:03:48 <elliott> Gregor: And, uhh, wait, we need more than 256 registers by far :P
16:03:56 <oerjan> <elliott_> ais523: hmm, not on the wikipedia page, but IIRC oerjan pretty much slam-dunked it once
16:04:01 <elliott> Gregor: Try 1,024 registers.
16:04:09 <oerjan> i certainly cannot recall proving dupdog definitely non-TC
16:04:36 <elliott> Gregor: And while you can say "deref a b c d [...]", it's more convenient to say "deref b64.r3" or something :-P
16:04:41 <elliott> COME ON, BEST ARCHITECTURE EVER
16:05:02 <ais523> that should be enough for everyone
16:05:15 <elliott> "(2^64) bytes = 16 384 petabytes"
16:05:19 <elliott> "Showing results for 2^64 bytes in petabytes. Search instead for 2^64 bits in petabyes"
16:05:20 -!- Sgeo has quit (Ping timeout: 276 seconds).
16:05:57 <elliott> if we're going bit-addressable
16:06:05 <Gregor> Kaksikymmentneljtuntiaikakausitmnhetkinen
16:06:10 <oerjan> i would say my current best guess is that it isn't TC, but then my initial best guess was non-TC for every non-trivial step of my recent underload endeavour, and i was wrong every time :D
16:06:27 <elliott> So yes, 2,048 registers :-P
16:06:31 <ais523> oerjan: well, I thought the 2,3 machine was non-TC to start off with
16:06:33 <Gregor> Wow, translate.google.com actually understood that ... "Twenty-Four Hour Period As One moment"
16:06:38 <elliott> Which gives you slightly less than 32 full addresses in registers.
16:06:46 <elliott> Gregor: It's the word I meticulously constructed with oklopol's help in Finnish.
16:06:54 <elliott> It means "day", if you're insane.
16:06:54 <Zwaarddijk> Gregor: that doesn't look like proper Finnish, though
16:07:07 <elliott> oklopol understood it, that's good enough for me.
16:07:48 <elliott> some finn pronounce it and upload the wav
16:07:51 <elliott> i can only imagine it will sound amazing
16:08:08 <elliott> that sounds even funnier, do it
16:08:14 <elliott> say bork bork bork at the end
16:08:24 <elliott> otherwise how can i take you seriously, you might not be a real swede.
16:08:36 <Vorpal> elliott, that isn't even Swedish in any sense
16:08:36 <oerjan> ais523: btw for that one, is there a definite proof that it isn't TC with finite initialization?
16:08:48 <elliott> Vorpal: Your mom isn't Swedish in any sense.
16:09:09 <ais523> there is a proof that it's non-TC unless you have infinitely many white cells, but that isn't particularly useful
16:09:23 <elliott> Gregor: Quick, add !dupdog so that oerjan can start proving.
16:09:38 <Vorpal> ais523, isn't that just saying that it needs infinite memory to be TC?
16:09:40 -!- Sgeo has joined.
16:09:59 <ais523> "infinitely many cells" is obviously needed
16:10:06 <ais523> but nothing in TCness implies that an infinite subset of them have to be white
16:10:14 <lambdabot> forall a. (Integral a) => a -> Bool
16:10:19 <Vorpal> ais523, what is the set of possible colours?
16:10:31 <ais523> white, yellow, orange in most of the Wolfram drawings of the machine
16:10:34 <ais523> although I just call them 0,1,2
16:10:52 <ais523> (I don't think anyone's tried to draw the machine but the Wolfram people and me)
16:11:12 <Zwaarddijk> http://miekko.infa.fi/kaksikymment.ogg
16:11:26 <Zwaarddijk> whoops, I should've added a bork bork bork there :|
16:11:27 <Vorpal> ais523, so is it a CA or TM what? I think I need to read up on dupdog
16:11:53 <elliott> Vorpal: another guy in here
16:11:54 <ais523> it's one of those joke esolangs that's not quite obviously stupid
16:12:05 <Gregor> QUICK, someone upload it to the wiki as a pronunciation key!
16:12:29 <Vorpal> <ais523> what, dupdog? <-- I thought you were talking about dupdog above
16:12:34 <Vorpal> ais523, or was that just elliott?
16:12:37 <elliott> Zwaarddijk: MAKE IT NON-403 SO QUICKLY
16:12:44 <Vorpal> ais523, in that case which language *were* you talking about
16:12:51 <elliott> ais523: we can't have tha fnacy pronunciation icon right?
16:12:52 -!- Sgeo__ has joined.
16:12:58 <ais523> the conversation was about more than one language
16:13:00 <Sgeo__> Dear connection: What the hell?
16:13:08 -!- Sgeo_ has quit (Ping timeout: 276 seconds).
16:13:11 <ais523> dupdog was one mentioned, but so were Xigxag and the 2,3 machine
16:13:28 <Vorpal> elliott, how is it even related to our wiki (that sound)
16:13:34 <ais523> also, minimized Underload, although that doesn't fit the question
16:13:56 <Zwaarddijk> that's random - usually, stuff I scp there gets the right permissions immediately
16:14:05 -!- HackEgo has quit (Remote host closed the connection).
16:14:12 <Zwaarddijk> apparently this time stuff didn't work out like that :|
16:14:34 <elliott> Vorpal: it's the pronunciation of what's on the homepage
16:15:00 <elliott> Zwaarddijk: Permission to upload that to the wiki as public domain? :-P
16:15:05 -!- Sgeo has quit (Ping timeout: 276 seconds).
16:15:07 <oerjan> ais523: heh now i wonder if you could remove something more from underload if you allowed an infinite program :D
16:15:27 <Vorpal> elliott, I don't know what to search for
16:16:06 <Zwaarddijk> altho' as I said, I have a bit of an accent, altho' that accent is sort of recursive
16:16:20 <Zwaarddijk> my Swedish is basically Swedish as influenced 500 years by Finnish
16:16:42 -!- Phantom_Hoover has joined.
16:16:50 <Zwaarddijk> and Finnish is influenced by germanic anyway, so we end up with ... some kind of converging shit
16:16:56 <elliott> ".ogg" is not a recommended image file format.
16:17:14 <elliott> "Warning"; I don't see any way to IGNORE that warning :P
16:17:17 <elliott> ais523: ojgoidfjgiodfogd haven't slept
16:17:27 <elliott> ais523: Can FANCY ADMINS LIKE YOU upload oggs?
16:17:29 <ais523> elliott: there's normally an "ignore warnings" checkbox
16:17:31 <oerjan> ais523: yeah it would seem that you could then _never_ get information out of a stack
16:17:55 <ais523> oerjan: is a one-way information transfer enough?
16:17:59 <Phantom_Hoover> elliott, allowed upload types are defined in the MW config, so I'm guessing no.
16:18:05 <ais523> elliott: oh, if even with the override it doesn't work, I can't override it either
16:18:12 <ais523> as it must be being done at the PHP level
16:18:16 * elliott just makes a fake page for the pronunciation
16:18:20 -!- Sgeo__ has quit (Ping timeout: 276 seconds).
16:18:21 -!- MigoMipo has joined.
16:18:55 <ais523> oh, I'd be surprised if graue didn't respond to emails
16:19:07 -!- Phantom__Hoover has joined.
16:19:43 <elliott> Zwaarddijk: Thank you for your valuable service to the community.
16:20:23 <elliott> I think we can all agree that the world of esolangs was advanced today.
16:20:52 <Zwaarddijk> would've been advanced even more if I had an esoteric finnish dialect.
16:21:00 <Zwaarddijk> something like, uh, ingrian or somesuch
16:21:08 <elliott> it's ok, we have an esoteric word to make up for it
16:21:17 <Zwaarddijk> I think I can pull that off, though, as they all are knee-deep in Russians
16:21:20 <elliott> now someone make an esolang based on it so nobody can complain
16:21:22 <Zwaarddijk> and therefore sound, unsurprisingly, like Russians?
16:21:26 <elliott> preferably kaksikymmentäneljätuntiaikakausitämänhetkinen should be the cat program
16:23:02 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds).
16:24:26 -!- Slereah has quit (Ping timeout: 252 seconds).
16:25:25 <elliott> Zwaarddijk: I have decoded your hostile language.
16:26:04 <oerjan> lessee without :, (A)(^B) is equivalent to (AB)
16:26:51 <elliott> Phantom__Hoover: Based on randomly decomposing fake Finnish words, that is :P
16:27:24 <oerjan> ((A1)...(An)B) is equivalent to (A1)...(An)(B)
16:27:43 <elliott> actually tuntiai can decompose into tunti (null), ai (equals)
16:28:21 <Ilari> APNIC down 0.01: 4k to India, 64k to China, 256 to Hong Kong. Slow day.
16:28:55 <oerjan> (A)^ is of course equivalent to A
16:29:13 <ais523> Ilari: aren't slow days a /good/ thing in this context?
16:29:35 <oerjan> i think this allows you to recude everything down to an at most 1 level deep expression
16:29:45 <elliott> Zwaarddijk: män is now end of block :P
16:30:11 <elliott> Zwaarddijk: http://sprunge.us/KdMF
16:30:42 <oerjan> hm or wait you can have (^(^(^...)))
16:31:25 <elliott> Zwaarddijk: Do any of those decomposed elements have meaning in Finnish? :P
16:31:46 -!- Slereah has joined.
16:35:14 <Zwaarddijk> -ki = colloquial variation on -kin, a suffix meaning "even, too, also"
16:38:26 <oerjan> also, every single one of those probably means something in japanese too >:)
16:39:13 <elliott> Vorpal: whas good n64 emulator cant typ
16:39:17 <Zwaarddijk> if you cut Finnish up randomly it often looks very japanese
16:39:29 <Zwaarddijk> but if you just let it be like it is, it's distinct enough
16:39:35 <elliott> iirc pikhq said that with simple substutition the bug word became nonsense japanese
16:39:46 <elliott> Zwaarddijk: i bet letter-based markov bots are fun in finnish
16:40:32 <Zwaarddijk> never tried one, but .. uh, afaict markov-bots can't maintain vowel harmony?
16:40:48 <Zwaarddijk> yeah but, vowel harmony is v. important for Finnish
16:41:11 <Vorpal> <elliott> Vorpal: whas good n64 emulator cant typ <-- for linux mupen64plus is your best bet
16:41:17 <Vorpal> heard there were better ones for windows
16:41:18 <Zwaarddijk> (and estonian chicks, my god, some of them are crazily beautiful)
16:41:24 <Vorpal> elliott, anyway how goes the toolchain?
16:41:25 <elliott> Vorpal: assume i have infinitely powerful hardware
16:41:34 <elliott> toolchain i'm holding off on until i decide on object format/versions
16:41:42 <oerjan> Zwaarddijk: why couldn't they maintain vowel harmony, it's a simple finite state thing...
16:41:46 <Vorpal> elliott, I don't know any cycle accurate one if that is what you mean
16:41:52 <elliott> Vorpal: heck no, that's excessive
16:42:07 <oerjan> you might need to adjust them a little for it
16:42:08 <Vorpal> elliott, for linux mupen64plus is the only one I know of
16:42:10 <elliott> yeah looks like mupen is the only one
16:42:19 <Vorpal> elliott, the "Plus" bit is important
16:42:29 <elliott> Vorpal: any emulation pitfalls?
16:42:34 <elliott> i'm gonna try out sm64 finally :P
16:43:49 <ais523> elliott: that's the only game that mupen emulates correctly pretty much no matter what the settings are
16:43:58 <Vorpal> elliott, and yes the emulation is not perfect. No noticeable issues in mario64 though
16:43:59 <ais523> it's incredibly picky on every other game for the console
16:44:08 <elliott> Vorpal: i heard something about a beam of light
16:44:12 <Vorpal> ais523, zelda64 oot works fairly well too
16:44:23 <ais523> anyway, why don't you just play the DS version? it's still on sale
16:44:26 <elliott> maybe we _do_ need a cycle-accurate nintendo 64 emulator :)
16:44:34 <elliott> if we can't trust people to write a decent higher-level one
16:44:51 <elliott> ais523: I find the DS uncomfortably small to play for long periods of time i.e. any longer game
16:44:57 <elliott> ais523: and prefer originals to remakes
16:45:10 <Vorpal> elliott, you can/could get most games to work. But the last mupen64plus version lacks a GUI. They rewrote the large bits but never ported the GUI over
16:45:16 -!- Tritonio has quit (Quit: Leaving).
16:45:17 <Vorpal> so you need to fiddle with command line args
16:45:22 <Vorpal> should be easy for mario64
16:45:28 <Vorpal> but for most other games *shudder*
16:45:28 <ais523> oh, I find computers uncomfortably large
16:45:41 <elliott> ais523: it's about the display for me
16:45:51 <elliott> ais523: I'm fine with a tiny gamepad, but I can't play on a small screen
16:45:54 <Vorpal> elliott, for mario64 it is probably just ./mupen64plus mario.v64 (or whatever extension it is)
16:46:04 <elliott> Vorpal: yes, once I rip the ROM from my cartridge
16:46:11 <elliott> therefore no more needs to be said
16:46:14 <oerjan> Zwaarddijk: i guess the problem is that you could have arbitrary long chains without a vowel of specific harmony, so you cannot use a finite length chain ... although i think fizzie looked at markov chains with adjustable length
16:47:10 <oerjan> (for fungot's babbling)
16:47:10 <fungot> oerjan: http://www.youtube.com/ fnord but rather violent... nothing for sensitive people
16:47:23 <Zwaarddijk> i don't actually know too much about markov chains, I should probably read up
16:47:44 <elliott> Zwaarddijk: they're trivial
16:48:01 <oerjan> fizzie: ok you need to wrap up a letter-based, vowel harmony preserving finnish style for fungot ;D
16:48:02 <fungot> oerjan: time for lesson 2? what about godel is also gödel is undecidable. of course. but that view won't help anything. shivers wrote it years ago; it's there.
16:48:06 <elliott> Zwaarddijk: for babble generation: you have a map (last N units) => {set of (following unit, probability)}
16:48:17 <elliott> you start with a START token, pick a random next one weighted by probability
16:48:23 <elliott> repeat taking the last N from your output
16:49:33 <elliott> Zwaarddijk: units can be words or vowels or whatever ofc
16:50:51 <Ilari> Of course, to preserve specific harmony, one could put the current harmony (not known, front, back) among those last N units.
16:51:16 <Zwaarddijk> ah, I thought they were defined as last units => set of following units (altho' that's equivalent, really)
16:51:37 <elliott> Zwaarddijk: yep, but the probability is of course important
16:51:41 <elliott> to generate even vaguely coherent things
16:52:02 <elliott> you can basically just insert by frequency and then turn that into weighting to make things simpelr from an implementation POV of course
16:52:02 <Zwaarddijk> yes, well, I left that sort of implicit there
16:52:11 <elliott> well that's how they are defined then
16:52:24 <Zwaarddijk> I meant "last unit => set of following units"
16:52:32 <elliott> Zwaarddijk: that's just an order-1 markov chain
16:52:39 <elliott> Zwaarddijk: which...works, but won't produce anything coherent at all
16:52:51 <elliott> Zwaarddijk: "what comes after 'a'" "oh, occasionally 'x'" "axoiajdioasjoisdfksngdfbslkjerbg"
16:53:16 <oerjan> which is the name of a small mountain in georgian.
16:53:34 <Zwaarddijk> you should go for something in salishan languages instead
16:53:36 <oerjan> (note: not _entirely_ accurate)
16:53:52 <Zwaarddijk> then you could do things like qsqvkdmllqklitsk
16:54:06 <Ilari> Of course, the more units are considered, the larger the tables (and more difficult to estimate the probabilties).
16:54:08 <Zwaarddijk> oh i accidentally almost typed clit in there
16:54:19 <elliott> Ilari: you just need a large corpus
16:54:55 <Zwaarddijk> salishan languages actually challenge the idea that syllables are a linguistic universal :)
16:55:51 <elliott> Zwaarddijk: That's a load of bullshdifogrjknydnmdfhjkgslvfkdghksjdkdjgjksletapw;rlmvo*£$&@*(!@)~{P_@)(~~
16:56:01 <elliott> "They are characterised by agglutinativity and astonishing consonant clusters — for instance the Nuxálk word xłp̓x̣ʷłtłpłłskʷc̓ (IPA: [xɬpʼχʷɬtʰɬpʰɬːskʷʰt͡sʼ]) meaning ‘he had had [in his possession] a bunchberry plant’ has thirteen obstruent consonants in a row with no vowels."
16:56:05 <elliott> that's the sexiest word i've ever seen.
16:56:13 <elliott> it has no pronunciation link though, i want a pronunciation link
16:57:01 <Phantom__Hoover> Huh, the chemical symbol for tungsten comes from "wolfram".
16:57:17 <oerjan> someone tried to upload a pronunciation, but he choked on his tongue
16:57:51 <oerjan> Phantom__Hoover: it's called wolfram in norwegian
16:58:09 <oerjan> it's a new kind of element
17:00:31 <elliott> Wow, the Mario 64 title screen is ... disturbing.
17:01:13 -!- asiekierka has quit (Read error: Operation timed out).
17:01:48 <elliott> Vorpal: Does mupen64 no longer support configuring KEYS? :P
17:02:16 <Vorpal> elliott, through a config file iirc...
17:02:22 <elliott> Vorpal: It has a GUI here...
17:02:59 <Vorpal> elliott, oh then you can find it under settings. Look for the configuration of which input plugin to use
17:03:07 <elliott> yes, the settings don't pop up anything
17:03:07 <Vorpal> elliott, I assume you use sdl?
17:03:14 <elliott> sdl input. yes. the only available one.
17:03:23 <Vorpal> elliott, then click config it or such
17:03:27 <Vorpal> if that doesn't work I don't know
17:07:27 <elliott> ais523: do you know of any compilers that are truly single-pass?
17:07:55 <ais523> as in, never ever go back to anything they've previously covered for any reason?
17:08:00 <ais523> it'd have to be not AST-based
17:08:08 <ais523> I imagine some stupid BF compilers work like that
17:08:10 <oerjan> wasn't original pascal single-pass
17:08:19 <oerjan> ok maybe not that strictly...
17:08:32 <ais523> oerjan: I was thinking about that, but not sure if it matched the restriction
17:09:24 <elliott> i like how mupen64plus doesn't trust you with its build system
17:09:29 <elliott> it just gives you an optionless shell script to use insetad.
17:09:30 <oerjan> if remembering the position of a previous element counts as going back to it, then even bf looks hard...
17:09:39 <Phantom__Hoover> Oh god, there's more tau stuff on the front page of Reddit.
17:09:58 <ais523> oerjan: you can compile it into while { and }
17:10:01 <elliott> oerjan: ais523: Let's say: You can keep track of state as you see it, but there must be NO backtracking in the input stream.
17:10:04 <ais523> without tracking the nesting at all yourself
17:10:09 <elliott> And you must never go back an arbitrary distance?
17:10:17 <ais523> elliott: in the input stream? then it's easy, you can just load the file into memory
17:10:29 <elliott> ais523: nope, because that involves going back arbitrarily
17:10:41 <elliott> (if you're using that to subvert the restriction)
17:10:52 <elliott> i.e., in (a + b), + only has to look one back
17:11:04 <elliott> being able to say (... a b c + d) would not be OK
17:11:10 <oerjan> elliott: ok so it's LR(k) parsing-wise, at least
17:11:28 <elliott> oerjan: hmm, maybe i should restrict it more :)
17:11:33 <oerjan> but that's not a very big restriction
17:13:20 <oerjan> elliott: one thing i thought of is that you can only write a bounded amount of computer result for a bounded amount of input, while always going forward in both input and compiled stream
17:14:25 <oerjan> this means of course that compiling a [ cannot look at the corresponding ]...
17:15:04 <oerjan> but then it becomes very dependent on the branching power of the language you are compiling _into_
17:15:52 <elliott> oerjan: assume you can define labels, at least
17:16:47 <elliott> Vorpal: so, uh... nintendo 64 without a gamepad or joystick
17:17:02 <oerjan> yeah then you could do it by assigning the label name for the ] when you hit the [
17:17:50 <oerjan> (you also need the label for [ of course)
17:18:16 -!- cheater- has joined.
17:18:20 <oerjan> (then the question becomes whether using the label for [ counts as going backwards or not)
17:19:27 -!- ais523 has quit (Remote host closed the connection).
17:19:30 <elliott> oerjan: no, because you can use a stack
17:19:33 <elliott> and you only have to look back one
17:19:42 <Vorpal> <elliott> Vorpal: so, uh... nintendo 64 without a gamepad or joystick <-- I used keyboard since an aircraft joystick doesn't work well with games assuming a gamepad
17:19:54 <Vorpal> the range of movement is so very different
17:19:56 <oerjan> hm you could make [ push its label at runtime instead, so you didn't have to let ] look at the [ label while compiling
17:20:01 <elliott> I really need to buy a solid PS2 controller and use it forever.
17:20:23 <elliott> oerjan: you could also compile all langs into source + interp that way and bypass all restrictions...
17:21:03 -!- cheater00 has quit (Ping timeout: 255 seconds).
17:24:23 <elliott> Vorpal: this castle is a bit dramatic.
17:25:42 <elliott> I sure hope you can just say "A" and "B" and the like here.
17:26:07 <Vorpal> elliott, not that I know of
17:26:30 <Vorpal> elliott, I think they are sdl keycodes anyway
17:30:21 <elliott> Vorpal: THIS GAME DOES NOT HAVE ALL THE NICE MOVING STUFF GALAXY HAS
17:30:35 <elliott> I miss crouch-backflipping, catching on to a wall, kicking off, then spinning.
17:30:47 <Vorpal> elliott, you can do a wall jump if that is what you mean
17:31:15 <Vorpal> elliott, you can do long jumps by crouching (Z iirc?) and then jumpng
17:31:27 <Vorpal> elliott, yes pretty sure
17:31:31 <elliott> Can you follow a backflip with a wall jump?
17:32:10 <Lymia> http://www.bbc.co.uk/news/uk-england-lancashire-12249172
17:32:12 <Lymia> What could go wrong?
17:32:23 <Vorpal> elliott, anyway changes in later games. What else would you expect
17:32:44 <Vorpal> elliott, next you will demand these kind of things in SMB!
17:33:14 <elliott> not sure how i feel about this game anthropomorphising bombs as happy, carefree creatures
17:33:20 <Phantom__Hoover> Lymia, UNLESS THE PAEDOS USE IT TO TRACK OUR CHILDREN!!!!!!
17:35:49 * oerjan recalls once upon a time he was in a role-playing game with a happy, carefree missile
17:37:38 <oerjan> (we sold it to the local mafia. it seemed like the least dangerous option.)
17:38:18 <oerjan> hm actually the missile wasn't the main thing we sold, just something that happened to be stored on the premises.
17:40:51 <elliott> oerjan: ok hm so what's LR(1) for this one-pass thing
17:41:05 <elliott> oerjan: you can look at the last production generated?
17:41:15 <elliott> that'd mean you could not do nested loops
17:41:25 <oerjan> LR(1) uses a stack and a finite state
17:41:36 <Lymia> What are you talking about.
17:43:11 -!- cheater00 has joined.
17:43:23 <oerjan> which reminds me somehow of that right bracket language
17:43:56 <oerjan> (i don't think going to the next _non-matched_ bracket can do nesting either)
17:44:12 <elliott> so is LR(!) actually sufficient for bf here?
17:44:49 <oerjan> bf is certainly LR(1) in the usual sense, and i think LL(1) too
17:45:03 <elliott> oerjan: right, but is that actually relevant to this compilery thing
17:45:06 <elliott> because it has to keep its state like that too
17:45:08 <oerjan> might even be LR(0) if you look at it the right way
17:45:20 -!- cheater- has quit (Ping timeout: 255 seconds).
17:46:15 <oerjan> elliott: well i think the labels can be put on the same stack if that's what you mean
17:46:30 <elliott> oerjan: i'm just figuring with my "can only access a bounded amount back"
17:46:37 <elliott> how small it can get while still being able to compile Pascal-- ;D
17:47:01 <oerjan> well pascal more or less requires a dictionary, doesn't it
17:47:33 <oerjan> there is no requirement to only use variables in stack order of definition :D
17:47:55 <elliott> oerjan: that would be an awesome requirement :D
17:48:30 <oerjan> hm what would lambda calculus look like with that
17:48:56 <elliott> Pascal --> lambda calculus
17:49:11 <oerjan> they _do_ share the lexical scoping
17:50:57 <oerjan> inside a \x -> ..., all accesses to x must come before all accesses to outer variables
17:51:46 <oerjan> so say \x y -> y (\z -> z y) x is legal
17:53:00 <elliott> it turns out it's equivalent to the superstrict lambda calculus, i deduced this because i'm a genius
17:53:08 <elliott> don't have any reasoning, just told my past self (i.e. my present self) it was so
17:53:13 <elliott> so i guess i figure it out in a few years, stay tuned
17:54:01 <oerjan> you have i = \x -> x and k = \x y -> x both legal
17:54:23 <Phantom__Hoover> There is an idiot on BBC news claiming that the Fukushima plant has basically gone Chernobyl and it's being covered up.
17:54:47 <oerjan> s = \x y z -> x z (y z) is not, at least in that form
17:55:13 <elliott> or can you not mention it twice?
17:56:04 <elliott> oerjan: \x y z -> (\a c b d -> a b (c d)) x y z z
17:57:44 <oerjan> um are you saying that gives s
17:57:58 <elliott> oerjan: i'm... hypothesising that that gives s, if you make it queue order instead
17:58:10 <elliott> oerjan: \x y z -> (\a b c d -> a c (b d)) x y z z
17:58:35 <oerjan> well what about that reversing, hm
17:58:46 <elliott> <elliott> oerjan: \x y z -> (\a c b d -> a b (c d)) x y z z
17:58:52 <elliott> but i'm sure you just need flip
17:59:04 -!- Wamanuz3 has quit (Remote host closed the connection).
17:59:21 <elliott> \f x y. (\y' x' f'. f' x' y') y x f
18:00:15 <elliott> do my thinks for me oerjan
18:01:24 <oerjan> a little adjustment of that gives you compose
18:01:41 <Lymia> I see functional programming.
18:01:44 <Lymia> I'm not sure if I want.
18:01:47 <oerjan> \f x y. (\y' x' f'. f' (x' y')) y x f
18:02:03 <elliott> Lymia: this isn't functional programming.
18:02:06 <elliott> this is pathological functional programming.
18:02:31 <elliott> that was stolen for something on the wiki already i think :)
18:03:52 -!- Wamanuz has joined.
18:03:59 <oerjan> http://esolangs.org/wiki/0x29A
18:04:39 <lambdabot> Maybe you meant: id ignore index instances instances-importing irc-connect . ? @ v
18:04:53 <lambdabot> forall a. (Integral a) => a -> String -> String
18:05:59 <Vorpal> elliott, still playing or did you get bored?
18:06:17 <elliott> and also really fucking tired
18:06:25 -!- elliott has left (?).
18:06:27 -!- elliott has joined.
18:07:27 <Phantom__Hoover> Jesus, actually reading about the Chernobyl disaster is depressing.
18:10:35 <fizzie> Phantom__Hoover: Every three years or so I wikiread the topic with a non-deterministic depth-limited BFS for a day or three. (Not sure why.)
18:11:11 <fizzie> Last time I did that I did find this photo, though: http://upload.wikimedia.org/wikipedia/commons/5/5f/Peacekeeper-missile-testing.jpg
18:12:25 -!- atrapado has joined.
18:12:30 <fizzie> Breadth-first search; how *couldn't* you do that on a graph?
18:13:33 <Vorpal> fizzie, technically by not performing it
18:14:42 <fizzie> Well, yes. But doing it with a browser is borderline trivial. Or at least with a config that makes "open in new tab" open at the end of the tab bar.
18:15:51 <elliott> a yway cherlaer chernobyl deserved it because of perl harbor
18:16:32 -!- asiekierka has joined.
18:16:55 <oerjan> yeah how dared those russians push perl on us
18:17:14 <elliott> inour oaeubrs no less!k1'1
18:17:51 <oerjan> elliott: hallucinating yet?
18:18:19 <oerjan> not sufficiently deprived, then
18:18:23 <Vorpal> elliott, this IRC thing is all in your mind
18:18:27 <elliott> oerjan: think i should oo for 72 hrs??????
18:18:49 <oerjan> ooing for 72 hours is _not_ recommended
18:18:54 <elliott> oooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:18:58 <elliott> oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:01 <elliott> oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:01 <oerjan> it'll ruin your voice cords
18:19:07 <elliott> oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:10 <elliott> oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:13 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:16 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:17 <elliott> oooooooooooooooooooooooooooooooo
18:19:23 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:25 <elliott> ooooooooooooooooooooooooooooooooo
18:19:29 <elliott> ooooooooooooooooooooooooooooo
18:19:33 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:38 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:42 <elliott> ooooooooooooooooooooooooooooo
18:19:44 <elliott> oooooooooooooooooooooooooooooooooooooo
18:19:46 <elliott> oooooooooooooooooooooooooooooooooooooooooooo
18:19:49 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooo
18:19:51 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:54 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:57 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooo
18:19:59 <elliott> ooooooooooooooooooooooooooooooooooooooooooooooooooo
18:20:00 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | what now?.
18:20:01 <elliott> ooooooooooooooooooooooooooooooooooooo
18:20:40 <optbot> Phantom__Hoover: I rarely, if ever, read it.
18:20:48 <elliott> Phantom__Hoover: IT'S FROM 2008
18:20:56 <elliott> it's been here longer than you
18:21:04 <elliott> so "latest" is objectively wrong?
18:21:06 <Vorpal> but I forgot what it did
18:21:29 <Phantom__Hoover> elliott, so what if it was here in '08? I would have opposed it then as well.
18:21:56 <elliott> sure, but that's assuming i actually care that you oppose. i've already discussed it with oerjan and am implementing the fix we agreed on soon
18:22:00 <elliott> like when i'm not 90% asleep.
18:22:08 <elliott> you whine about everything. :p
18:22:18 <elliott> Vorpal: *and fungot inspiration
18:22:19 <fungot> elliott: can you put your call-,with-* hack for scheme-mode? and if so, what does syntax-case buy you over define-macro?
18:22:26 <Vorpal> elliott, oh really, mkay
18:22:28 <elliott> OK, fungot babble inspiration
18:22:29 <fungot> elliott: do you run emacs in konsole
18:22:36 <elliott> Vorpal: yes, fizzie was all "i am inadequate ;_; must compensate"
18:22:41 <elliott> so he brought out all the n-grams.
18:22:53 <Vorpal> elliott, what did you do? just random letters?
18:23:02 <elliott> Vorpal: it does exactly what it did then
18:23:07 <Vorpal> elliott, I don't remember!
18:23:14 <elliott> Vorpal: you'll get it in a few pings
18:23:19 <elliott> it makes it very obvious sometimes
18:23:23 <optbot> Vorpal: oh, i seem to remember now
18:23:36 -!- Gregor has set topic: /mode +t | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/.
18:23:36 <Vorpal> elliott, I think it quotes log
18:23:45 <Vorpal> elliott, maybe not full lines, don't know yet
18:24:05 <Phantom__Hoover> RTFLogs is not a response to all requests for information.
18:24:07 <Gregor> elliott: I know it's off, that's still yukks :P
18:24:08 <oerjan> Phantom__Hoover: 12 hour timeout after last change by anyone
18:24:17 <elliott> Phantom__Hoover: it is when you're being irritating about it.
18:24:22 <Phantom__Hoover> I'm not going to sift through the logs to find everything.
18:24:31 <elliott> then don't be irritating/obnoxious.
18:24:34 -!- Vorpal has set topic: /mode +t | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | + Gregor has changed the topic to: /mode +t | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/.
18:24:40 -!- elliott has set topic: * Vorpal has changed the topic to: /mode +t | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | + Gregor has changed the topic to: /mode +t | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/.
18:24:47 <elliott> we need a fixed-point topic.
18:24:59 <Vorpal> elliott, yep, we will easily hit limit
18:24:59 -!- optbot has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | F.
18:25:01 <Phantom__Hoover> elliott, OK, expressing my discontent for your bot is being irritating and obnoxious.
18:25:13 <elliott> Phantom__Hoover: nope, expressing it in the way you did (an obnoxious way) is
18:25:16 -!- Gregor has set topic: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has changed the topic to: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has changed the topic to: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has changed the topic to: logs: http://tunes.
18:25:20 -!- elliott has set topic: * Gregor has changed the topic to: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has changed the topic to: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has changed the topic to: logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/ | * Gregor has chang.
18:25:21 <Vorpal> Phantom__Hoover, what do you expect...
18:25:37 <Gregor> How did we end up with a different length :P
18:25:39 <Vorpal> Phantom__Hoover, come on :P
18:25:43 <elliott> Phantom__Hoover: Vorpal: plz realise: "I don't like optbot" != "optbot is elliott's latest stupid project and it's retarded and stupid and dumb"
18:25:43 <optbot> elliott: that's like making printf return an integer so you can do printf("Hello, world!\n") + printf("Bye!\n");
18:25:53 <Vorpal> Phantom__Hoover, I agreed with you about elliott's treatment of you.
18:25:55 <elliott> latter does not get spoonfed log lines
18:26:02 <Vorpal> though I'm neutral on optbot
18:26:14 <elliott> vorpal's constant attempts to agree with everyone who's taking a negative view of me remain unhampered.
18:26:15 <Vorpal> elliott, yeah it isn't your latest stupid one
18:26:38 <Vorpal> however I do feel both you and Phantom__Hoover should calm down
18:27:28 <Vorpal> Phantom__Hoover, I can't see why. I just told you both to calm down...
18:30:02 <oerjan> about the surest way to make people ignore you, i should think
18:30:28 <Vorpal> oerjan, no one likes feeling they are at fault
18:31:05 <oerjan> yeah faults are dangerous, look at japan
18:31:27 <Vorpal> oerjan, may I borrow your flyswatter for a moment?
18:31:44 <Vorpal> wait I forgot the shape. what length is it?
18:32:36 * Vorpal swats oerjan for that pun-----###
18:32:45 <Vorpal> here you can have it back
18:37:15 <quintopia> what's the command for makin an hg repo directory match the server it was cloned from?
18:37:27 <elliott> do you mean pulling new changes
18:38:12 <oerjan> rtfm, rm -rf, so close
18:40:05 -!- Gregor has set topic: λf.(λx.f (x x)) (λx.f (x x)), bitches! | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/.
18:40:50 <Phantom__Hoover> ISTR it's from that tattoo, but then again it might not be.
18:53:04 -!- pingveno has quit (Read error: Operation timed out).
18:55:39 -!- pingveno has joined.
18:58:04 <Gregor> oerjan: I certainly hope you weren't referring to the topic ...
18:58:45 <elliott> i like how nobody noticed Deewiant even said anything
18:59:45 <Phantom__Hoover> elliott, I noticed, but I ignored him because I was otherwise engaged.
19:00:02 <elliott> Phantom__Hoover: You're getting married? Congratulations
19:00:29 <elliott> Phantom__Hoover: Wait, wait, let me guess. Gold.
19:00:59 <elliott> oerjan: it would have a drastic ego containment field collapse
19:01:12 <elliott> Phantom__Hoover: I propose :kanye:
19:01:33 <elliott> I'm assuming you mean shutter shades.
19:03:27 <oerjan> <elliott> i like how nobody noticed Deewiant even said anything <-- um am i being trolled
19:03:30 <Phantom__Hoover> elliott, I mean a metal hipster because that is what I am.
19:03:43 <elliott> Phantom__Hoover: THERE IS NO SUCH THING
19:03:48 <oerjan> because afaict Deewiant hasn't spoken recently
19:04:42 <oerjan> elliott: unless you mean that optbot! which i don't see why should need noticing...
19:04:43 <optbot> oerjan: sp0 will not work on my comp
19:05:16 <oerjan> elliott: well in that case i did notice.
19:11:44 <Gregor> What's more dangerous than glass to put mercury in in a ring ...
19:11:44 <Gregor> (But equally transparent)
19:12:16 <Gregor> elliott: Good luck making that maintain the shape of a ring.
19:12:23 <elliott> Gregor: You never said it had to last long
19:12:32 <elliott> Gregor: It'll last about as long as Phantom__Hoover's marriage
19:13:39 <Gregor> What's more dangerous than glass to put mercury in in a ring BUT IS ALSO TRANSPARENT, SOLID AND STABLE AT ROOM TEMPERATURE
19:13:49 <oerjan> the marriage is doomed anyway if they don't have the right chemistry
19:14:30 -!- augur has quit (Remote host closed the connection).
19:14:54 <oerjan> sheesh, all these restrictions
19:16:13 <Gregor> Maybe instead I'll go with semitransparent and use a mesh in which the pores are slightly smaller than mercury's natural drop size :P
19:16:28 <elliott> http://www.moral-politics.com/Temp/Pol_06449af7e2ec4ac5af505245273c9a61.png DO I WIN
19:16:37 <elliott> I'm pretty sure getting into one of the corners is how you win
19:21:14 <Phantom__Hoover> <elliott> Gregor: It'll last about as long as Phantom__Hoover's marriage
19:21:31 <Phantom__Hoover> Sure, since I plan to go through with the osmium ring plan with a wealthy heiress.
19:21:42 <elliott> Phantom__Hoover: It's not my fault you're planning to marry an extremely unstable isotope.
19:21:56 -!- oerjan has quit (Quit: Good night).
19:22:31 <Gregor> elliott: wtf does that even mean
19:23:10 <Gregor> I'm likin' my mesh idea, but I think the drop size for mercury is quite small :P
19:23:13 <elliott> Gregor: Result of taking the silly test at http://www.moral-politics.com/.
19:23:26 <elliott> Apparently I fall somewhere between social democratism and activism.
19:23:38 <elliott> Apparently activism is A FORM OF MILD SOCIALISM.
19:24:01 <elliott> It's a bullshit test and some of the answers have really biased wording
19:24:09 <elliott> [[Of the 666,816 respondents (11,708 on Facebook):
19:24:09 <elliott> 81% are more conservative.
19:24:10 <elliott> 7% are more authoritarian.]]
19:24:12 <elliott> PRETTY SURE THAT MEANS I WIN
19:24:24 <elliott> (7% are more authoritarian? wtf i never said anything authoritarian...)
19:24:34 * elliott redoes the political compass test while he's at it.
19:25:23 <elliott> Gleh, forgot how badly-worded some of the questions on this are, too.
19:25:26 <elliott> "People are ultimately divided more by class than by nationality."
19:25:36 <elliott> I have no idea whether this is saying "In practice, ..." or "Inherently, ...".
19:26:49 <Phantom__Hoover> I like the implication that noöne thinks women are better than men.
19:27:27 <elliott> What I hate about these tests is the questions that are basically trying to funnel you into one of two categories and are really obvious about it
19:27:36 <elliott> e.g. "Controlling inflation is more important than controlling unemployment."
19:28:24 * Phantom__Hoover wonders which political system would be best for an online sandbox economy thing á la EVE.
19:28:46 <elliott> I'd like to see a game economy ENTIRELY based around the fact that you can play the game with enough money.
19:28:58 <elliott> Like you can with EVE, except the servers being the Fort Knox.
19:29:19 <elliott> "Those with the ability to pay should have the right to higher standards of medical care ." ;; you can't just ask me this and not also ask me if I support public healthcare, ffs
19:29:35 <elliott> Either you want to outlaw Bupa, or you want to make everyone die on the streets!
19:30:08 <elliott> This is the political compass, btw.
19:30:19 <elliott> Amusingly I think I register as more conservative/authoritarian than I really am on these because of bad questions...
19:30:40 <elliott> "People are ultimately divided more by class than by nationality."
19:30:43 <elliott> HOW DO I EVEN DISAGREE/AGREE TO THAT
19:30:48 <elliott> IS IT WHAT I WANT TO BE TRUE, IS IT WHAT I THINK IS TRUE IN PRACTICE
19:31:45 -!- copumpkin has joined.
19:31:45 -!- copumpkin has quit (Changing host).
19:31:45 -!- copumpkin has joined.
19:34:33 -!- augur has joined.
19:34:54 <oklopol> everyone's dissing on elliott again, i'll jump the bandwagon: elliott: optbot's nick has a typo
19:34:55 <optbot> oklopol: and well there are lots of process supervision stuff built in. Oh and hot code reloading. Oh and support for distributed nodes and what not. Not features of your typical "scripting" language.
19:34:59 -!- asiekierka has quit (Read error: Operation timed out).
19:35:05 <elliott> oklopol: nah Phantom__Hoover was totally a bro
19:35:13 <elliott> Vorpal was just totally a Vorpal
19:35:23 <elliott> oklopol: but er waht's the typo
19:35:31 <elliott> it stands for oerjan's punnes terribales
19:35:37 <elliott> it's french you uncultured fuck
19:35:48 <elliott> there's even a fucking ' thing on top of one of them
19:36:55 <oklopol> today i said "jeau deux le shambray de la seminar" to a french guy, and he didn't get me
19:36:55 <elliott> Phantom__Hoover: WHAT WAS YOUR SCORE
19:37:05 <oklopol> even though chambre is totally french
19:37:33 <oklopol> "jeau deux" may have been wrong though
19:37:33 <elliott> Phantom__Hoover: IT'S LIKE FIVE QUESTIONS
19:38:49 <elliott> "A significant advantage of a one-party state is that it avoids all the arguments that delay progress in a democratic political system."
19:38:54 <elliott> I THINK THATS THE best question ever
19:38:59 <elliott> i just wanna see osmeone say "strong algryy"
19:40:08 <elliott> http://www.politicalcompass.org/facebook/pcgraphpng.php?ec=-5.38&soc=-7.64
19:40:55 <elliott> Phantom__Hoover: http://www.politicalcompass.org/images/axeswithnames.gif
19:41:00 <elliott> http://www.politicalcompass.org/images/bothaxes.gif
19:41:05 <elliott> Those should explain. The first one moreso.
19:41:12 <elliott> I love how Thatcher is more extreme than Hitler.
19:42:24 <elliott> oklopol: the moral or political
19:42:35 <elliott> if the political, try the moral one it's supra-eesy http://www.moral-politics.com/
19:43:49 <Phantom__Hoover> "We should reduce the causes of crime" vs. "we should eliminate the causes of crime".
19:44:45 <elliott> Gives me the warm fuzzies.
19:45:11 <oklopol> apparently i'm a socialist
19:45:18 <oklopol> interesting, since i'm not
19:45:50 -!- pikhq has quit (Ping timeout: 246 seconds).
19:45:55 -!- pikhq_ has joined.
19:47:26 <oklopol> i don't know how, but -3 on moral order, 1.5 on moral rules
19:47:49 <oklopol> in any case i'm not sure i agree with all my answers, i might answer differently at different times
19:47:49 <elliott> oklopol: right lcick the graph thing
19:49:24 <elliott> oklopol: give us the ilnk to the page with it on then
19:50:15 <oklopol> yeah i'll send you the POST data through an irc message
19:50:26 <elliott> haha oklopol just got a really weakling version of my results
19:50:57 <Gregor> elliott: Still using my handwriting as a system font?
19:51:00 <oklopol> my political stance is mostly that finland is perfect, let's just keep it this way
19:51:16 <elliott> Gregor: no :( but i regret not doing so
19:51:23 <elliott> oklopol: including the military service thing? :D
19:51:33 <oklopol> oh well some disagreements, tru
19:51:57 <oklopol> i don't get why a country would need an army
19:52:20 <Gregor> http://codu.org/gregor_handwriting.ttf
19:52:29 <elliott> Phantom__Hoover: set as system font
19:52:31 <elliott> set as default web page font
19:52:56 <Phantom__Hoover> Gregor, wow, that actually makes my handwriting look controlled.
19:53:32 <elliott> Gregor: do you want me to make my own handwriting font for you to use in return, your computer will look like a three year old wrote it
19:53:37 <Phantom__Hoover> (When I write long passages the words kind of merge into one big mass of slanty lines.
19:53:49 <Gregor> elliott: And my font DOESN'T look like a three-year-old wrote it?
19:54:10 <elliott> Gregor: you have clearly never seen MY handwriting.
19:54:40 <elliott> Gregor: Recently it evolved from "Cool, the lowercase letters take up THE ENTIRE LINE of regular lined paper" to "now I just write the upwards slanting lines, and literally just swiggle the rest"
19:54:51 <elliott> with that and the first/last letters i'm sure everyone can interpret my words
19:55:17 -!- ais523 has joined.
19:55:45 <elliott> Phantom__Hoover: same way Gregor did it
19:57:02 <elliott> Gregor: You realise that not doing it by hand == -19872938791237123 nerd points
19:57:14 <Phantom__Hoover> I mean, it's frowned on over here, but I can't imagine anyone *legislating* it.
19:57:22 <elliott> Gregor: I'd need to do all the OpenType stuff to get my insane ligatures (EVERYTHING IS SWIGGLES)
19:57:24 <Gregor> elliott: You realize that making a font ... of handwriting ... by hand makes no sense at all?
19:57:42 <elliott> Gregor: WHAT DO YOU THINK WE DID BEFORE YOURFONTS
19:57:48 <elliott> Gregor: I mean as in manually vectorising etc. :P
19:57:57 <Gregor> Make fonts that were not an accurate conversion of handwriting.
19:58:17 <Phantom__Hoover> Yeah, making a font of my handwriting would lose the melting.
19:58:45 <elliott> Phantom__Hoover: Not with OpenType!
19:58:51 <elliott> It'll just take twenty years to get every ligature down properly.
19:59:06 <Phantom__Hoover> (The best part is when I scribble stuff out /with nearly the exact same density and angle as my normal writing/.)
19:59:37 <Phantom__Hoover> It also goes halfway down the next line, which doesn't stop me writing on it.
19:59:40 <elliott> sometimes i don't even bother with spaces
19:59:55 <elliott> joined by a bunch of swiggles
20:01:17 <elliott> the purpose of handwriting isn't to be readable or useful imo
20:01:31 <Gregor> oklopol: YOU LIVE IN A HOLE
20:01:54 <elliott> A hole at bedrock! Factually correct.
20:02:10 <elliott> Well you do in Minecraft, dunno what Gregor's on about
20:02:22 <elliott> Your walls aren't flat btw, did someone mine some ore from them or did you decide that flat walls are boring
20:02:27 <Phantom__Hoover> I used to write algebra on one line unless it was literally impossible to make my pen write the symbols.
20:02:50 <elliott> Phantom__Hoover: biros have that much more leeway in stuffing squiggles into a line
20:04:19 <elliott> Phantom__Hoover: dunno, leaking can be useful for blotting off incorrect reasoning
20:04:25 <elliott> as long as the rest of the proof looks ok
20:04:38 -!- augur has quit (Ping timeout: 248 seconds).
20:04:42 <Phantom__Hoover> elliott, sure, but then I try to clean it with my fingers.
20:05:08 <Gregor> Everyone: Put the Y combinator in various languages in the topic, each suffixed with ", bitches!"
20:05:22 <Phantom__Hoover> And I end up looking like the Hulk if he'd been an early-20th century X-ray operator.
20:05:45 <elliott> (assume the language can do both)
20:06:13 <Gregor> I don't think it really matters for something this stupid :P
20:06:20 -!- elliott has set topic: λf.(λx.f (x x)) (λx.f (x x)), bitches! | (\f -> let x = f x in x), bitches! | logs: http://tunes.org/~nef/logs/esoteric/?C=M;O=D and http://208.78.103.223/esoteric/.
20:06:33 <elliott> that's not actually the y combinator.
20:06:37 <elliott> but it's a fixed-point combinator.
20:06:42 <elliott> the y combinator itself requires some newtype wrapping.
20:06:43 <oklopol> knowing the y combinator is a sign of inteligence, Gregor.
20:07:02 <Gregor> Eh, I guess we could tolerate any fixed-point combinator :P
20:07:06 <Gregor> oklopol: ... fascinating?
20:07:25 <oklopol> "<Gregor> I don't think it really matters for something this stupid :P"
20:07:38 <Gregor> oklopol: I meant something as stupid as putting it in the topic.
20:07:50 <elliott> i forget how to do it in underload, or if i ever got it done
20:08:34 <elliott> \f. (\x. f (\y. x x y)) (\x. f (\y. x x y))
20:08:37 <oklopol> well you know if people of inferior intelligecy come here then they'll understand right away that we're better than them so how's that stupid we don't have to listen to their whining.
20:09:14 <oklopol> elliott: erm, isn't i t":^"?
20:09:21 <elliott> oklopol: no, that's mockingbird
20:09:54 <elliott> too lazy so i'll use the abstraction elimination rules
20:10:04 <Gregor> λalchemy.(λmoonspirit.alchemy (moonspirit moonspirit)) (λastralplane.alchemy (astralplane astralplane))
20:10:24 <oklopol> Gregor: ah, that way they REALLY understand our greatness
20:10:33 <oklopol> because we're speaking their language, but better
20:12:00 <elliott> ^ul ()(((~:^~^)~a~*~^)~a~*:^)^
20:12:02 <elliott> ^ul ()(((~:^~^)~a~*~^)~a~*:^)^S
20:12:07 <elliott> ^ul (!())(((~:^~^)~a~*~^)~a~*:^)^S
20:12:49 <oklopol> i don't get what y should even do in ul
20:13:12 -!- Behold has joined.
20:13:17 <elliott> :t \f -> (\x -> f (\y -> x x y)) (\x -> f (\y -> x x y))
20:13:18 <lambdabot> Occurs check: cannot construct the infinite type: t = t -> t1 -> t2
20:13:18 <lambdabot> Probable cause: `x' is applied to too many arguments
20:13:34 -!- BeholdMyGlory has quit (Read error: Operation timed out).
20:13:35 <elliott> oklopol: it's ∀a.∀b.((a→b)→(a→b))→(a→b).
20:13:49 <elliott> let rec fix f x = f (fix f) x (* note the extra x *)
20:13:49 <elliott> let factabs fact = function (* factabs now has extra level of lambda abstraction *)
20:13:52 <quintopia> what do i do if apt-get runs out of disk in the middle of an install? would it break my packages to just kill it?
20:13:52 <elliott> let _ = (fix factabs) 5 (* evaluates to "120" *)
20:14:05 <elliott> quintopia: you can probably fit it with dpkg --reconfigure.
20:14:06 <ais523> quintopia: it won't break it in an unfixable way
20:14:23 <ais523> it can leave the packages temporarily broken until dependencies and configurations are fixed, though
20:14:24 <Gregor> quintopia: What they said :P
20:14:26 <elliott> ais523: can't seem to work out applicative-order Y in underload
20:14:31 <elliott> thought i had it but i made a mistaek
20:14:39 <elliott> \f. (\x. f (\y. x x y)) (\x. f (\y. x x y)) ;; this 'un
20:14:39 <ais523> I once had to shut my computer off in the middle of a distro upgrade
20:14:50 <ais523> and actually recovered from that, although I did have to use the command prompt to do so
20:14:55 <ais523> because the GUI wasn't working
20:15:22 <ais523> elliott: I probably used Windows for longer than I've used Linux
20:15:42 <ais523> remember that I'm older than you
20:15:47 <ais523> Linux hadn't really caught on when I was a child
20:15:54 <elliott> ais523: yeah well, when i'm older than you
20:16:00 <elliott> i'm gonna learn all the linuxes.
20:16:20 <oklopol> what, ais523 has used windows? lol what a noob
20:16:37 <elliott> ^ul (( { \x. f (\y. x x y) } )~a~*)
20:17:00 <quintopia> ais523: how do i unlock the dpkg directory so i can run it again?
20:17:20 -!- Sgeo has joined.
20:17:32 <Sgeo> http://kawagner.blogspot.com/2007/02/why-monads-are-evil.html
20:17:37 <Sgeo> Is this person some kind of moron?
20:17:45 <Sgeo> "It can hurt readability: A concrete monad is choosen by the return type of a function. For example a simple 'x <- get' can switch the rest of the do-block into 'state-monad'-land."
20:17:46 <ais523> quintopia: there's a lockfile somewhere, I'm not sure where offhand though
20:17:51 <ais523> oklopol: I've even used DOS
20:17:55 <ais523> from before Windows caught on
20:17:58 <elliott> ^ul (hello!)(((~:^~^)~a~*~^)~a~*)^S
20:18:05 <elliott> ^ul (hello!)(((~:^~^)~a~*~^)~a~*)^^S
20:18:09 <elliott> ^ul (())(((~:^~^)~a~*~^)~a~*)^S
20:18:12 <Sgeo> If I understand correctly, that shouldn't be properly typed in any monad..ic thingy that isn't State
20:18:13 <elliott> ^ul (())(((~:^~^)~a~*~^)~a~*)^^S
20:18:16 <elliott> ^ul (())(((~:^~^)~a~*~^)~a~*)S
20:18:19 <elliott> ^ul ((a))(((~:^~^)~a~*~^)~a~*)S
20:18:23 <elliott> ^ul ((abc))(((~:^~^)~a~*~^)~a~*)S
20:18:30 <elliott> that doesn't seem right at all!
20:18:35 <elliott> ^ul ((abc))(((~:^~^)~a~*~^)~a~*)^S
20:18:38 <elliott> ^ul ((abc))(((~:^~^)~a~*~^)~a~*)^^S
20:19:01 <elliott> exactly what i came up with :/
20:19:36 <quintopia> ais523: i found the file. how to unlock it? (no process is using it)
20:19:47 <ais523> is it empty? if so, just delete it
20:20:30 <elliott> ^ul ()(!())(((~:^~^)~a~*~^)~a~*:^)^
20:20:37 <elliott> !underload ()(!())(((~:^~^)~a~*~^)~a~*:^)^S
20:21:04 <elliott> ais523: just wait 'til relief is done!
20:21:08 <Gregor> I honestly have no friggin' clue why it's had such trouble remaining connected recently.
20:21:19 <Gregor> Waitwtf, it IS connected.
20:21:31 <Gregor> It's just not in #esoteric.
20:21:48 -!- HackEgo has joined.
20:23:26 <Gregor> OK, wtf, once again, EgoBot is connected but didn't join.
20:23:41 <Sgeo> Maybe I should try to learn category theory
20:23:58 <ais523> not fully, but enough to understand the basics
20:24:13 <Sgeo> Where should I start? Is Wikipedia readable on this subject?
20:24:18 <ais523> (mention category theory anywhere, and all the computer scientists in the audience immediately start gushing)
20:24:25 <elliott> yeah even my stupid friend can do category theory ;D
20:24:29 <ais523> and I'm not sure; I haven't looked at the Wikipedia article
20:24:36 <elliott> ais523: unless they're rabid haskell haters
20:24:45 <ais523> oh, the connection with Haskell I never really get at all
20:24:48 <elliott> arguably they don't count as computer scientists :>
20:24:55 -!- EgoBot has joined.
20:24:57 <ais523> I can't connect Haskell concepts to the mathematical concepts they're meant to be based on
20:25:12 <elliott> !underload ()(!())(((~:^~^)~a~*~^)~a~*:^)^S
20:25:13 <oklopol> LOL what a noob. i can't either, even though oerjan explained it once
20:25:15 <ais523> and nor can most other computer scientists, by the look of it
20:25:36 <elliott> i think it just comes from expecting something deeper than it is.
20:25:45 <Gregor> WTF TRAC IS ONCE AGAIN PEGGED AT 99% CPU
20:25:48 <Gregor> trac: it is a pile of garbage.
20:25:53 <oklopol> elliott: you get the connection?
20:25:58 <Phantom__Hoover> <elliott> yeah even my stupid friend can do category theory ;D
20:26:05 <ais523> Gregor: MediaWiki does that too
20:26:06 <elliott> oklopol: close enough anyway ;D
20:26:10 <elliott> Gregor: it's because of you
20:26:13 <oklopol> what's the definition of a category?
20:26:13 <elliott> it knows it's on your server
20:26:40 <elliott> oklopol: ffff a category is a bunch of objects and morphisms and fff dude
20:26:43 <elliott> it's been like 30 hours since i slept
20:26:52 <elliott> there's definitely ob in there
20:26:56 <elliott> ob everywhere in that bitch
20:27:17 <oklopol> well it's just a multi graph with a composition operator
20:28:28 <oklopol> that's a nice description, but i'm not sure it's a great definition
20:28:53 <elliott> oklopol: a category is somethign wit ha lot of pages in it
20:30:50 <Gregor> elliott: It's always slow on first load.
20:31:00 <elliott> Gregor: egobot is still going.
20:31:07 <elliott> admittedly it is likely inflooping
20:31:08 <Gregor> elliott: It's always slow on first load.
20:32:07 <elliott> Phantom__Hoover: What day is it kaksikymmentäneljätuntiaikakausitämänhetkinen?
20:33:05 -!- augur has joined.
20:33:56 <Gregor> dd if=/dev/zero of=bigfile bs=1M count=10 conv=fdatasync # omfg this has been running for over a minute *sobs*
20:34:23 <elliott> Gregor: So prgmr, quite shitty eh
20:34:33 <Gregor> Apparently they've decided to be, yes.
20:35:21 <elliott> <blast_hardcheese> This is kind of an open-ended question to whoever wants to respond, I'm trying to figure out the best way to manage this situation I'm in:
20:35:26 <elliott> Gregor: contact them and get SO MANY REFUNDS
20:35:29 <Gregor> wtf, is this some kind of joke, it's still running.
20:35:29 <Gregor> I'M TRYING TO WRITE A 10M FILE OF ZEROS
20:35:29 <Gregor> 10485760 bytes (10 MB) copied, 171.363 s, 61.2 kB/s
20:35:46 <elliott> That's slower than my internet connection at the worst of times.
20:36:07 <elliott> Gregor: They've migrated to Cloud(TM)-based storage!
20:36:11 <elliott> All the storage is in Zimbabwe now.
20:36:33 <elliott> Gregor: no, they just have all the libraries
20:36:38 <elliott> they're loaded every single time a process is spawned
20:36:41 <elliott> by downloading from <libname>.so
20:39:58 <elliott> Gregor: gnu are bidding on libc.so themselves
20:41:06 <ais523> now I wonder if libc.so actually exists
20:41:26 <ais523> "This TLD has no whois server."
20:41:30 <ais523> short and to the point
20:41:40 <elliott> Gregor is in the auction for it
20:41:47 <elliott> ais523: it's a closed auction, he signed up for libc.so
20:42:09 <elliott> ais523: .so = Somalia, they've only recently had something that could call itself enough of a government to get their tld activated :)
20:43:08 <elliott> from this we can conclude that the soviet union still exists
20:45:07 <Gregor> ais523: Taking donations to help me buy it! :P
20:45:17 <ais523> what sort of auction is it?
20:45:20 <Gregor> ais523: The benefit of donating is a snazzy <libc symbol>@libc.so email address!
20:45:33 <Gregor> ais523: Closed, anonymous, proxy bidding allowed.
20:45:44 <ais523> do you know each other's bids?
20:46:19 <ais523> I said that before you posted your reply, or at least before it arrived at my client
20:46:27 <Gregor> Yeah, I'm lagged all to hell X_X
20:46:56 <elliott> Gregor: What's the current bid, or has it not started yet :P
20:47:01 <Gregor> Mind you, the auction hasn't opened yet, I'm trying to raise capital before it does.
20:47:10 <elliott> also, holy crap, this thing is gonna make somalia so much money :D
20:47:33 <Gregor> elliott: Don't say that, I want it to sell for an amount I can buy it for X-P
20:47:40 <elliott> I mean the domain shit in general
20:47:40 <ais523> the sale of .tv gave a fortune to each Tuvaluan citizen
20:47:47 <elliott> Gregor: you realise that one corporate bidder means you're screwed?
20:48:07 <Gregor> elliott: My only prayer there is the fact that it really has no corporate value.
20:48:13 <elliott> ais523: I believe .tk made a lot of money/infrastructure for Tokelau too, but that was .tk advertising, so maybe exaggerating
20:48:22 <elliott> Gregor: We live in Web 2.0.
20:48:29 <Gregor> elliott: And? It still has no corporate value.
20:48:31 <ais523> I've never even heard of Tokelau
20:48:35 <elliott> Gregor: It has marketing value.
20:48:41 <Gregor> elliott: No, it doesn't.
20:48:54 <elliott> ais523: they're the island that decided that they really didn't need a cctld, so it's ok if a company gave them away for free
20:48:54 <Gregor> Wait, are we referring to libc.so or .so in general? :P
20:49:03 <Gregor> Yeah, it has zero marketing value.
20:49:03 <elliott> Gregor: It has marketing value IN THE MINDS OF IDIOTIC IDIOTS.
20:49:11 <elliott> Gregor: This is where you should assume the worst in people.
20:49:31 <ais523> well, hope those idiotic idiots don't find it until later
20:49:36 <ais523> then you can sell it to them for a fortune
20:49:47 <elliott> why would yo uever sell that!
20:49:54 <ais523> for a fortune, obviously
20:49:56 <Gregor> If I get it, I wouldn't sell it for less than six digits :P
20:49:59 <elliott> i like dynamic libraries, because i'm lame
20:50:03 <ais523> then you can buy it back again after their idiotic company collapses
20:50:09 <elliott> Gregor: I'll give you $99,999
20:50:27 <Gregor> elliott: TOO FEW DIGITS and also I didn't say I'd sell it for ANY six-digit value.
20:50:39 <elliott> Gregor: Too few digits on purpose :P
20:50:46 <elliott> I did not believe and do not believe that you value it more than $99,999
20:50:47 <Gregor> But anyway, I honestly think I have a so-so chance of getting it at this point.
20:50:58 <elliott> Well, you know, optimism is good for you.
20:51:08 <elliott> Even hopeless, naive optimisim.jf
20:51:24 -!- marian_30 has joined.
20:51:38 <ais523> Gregor: it depends on if everyone else is also there just on the offchance
20:52:01 <Gregor> elliott: No, seriously, it's not naive at all. We're talking about a domain name that has nerd cred but absolutely no commercial value, with thirteen bidders in a closed auction, and with only people who happened to notice its availability in a brief window.
20:52:12 <Gregor> In other words, we're talking about the intersection of domain name nerds and Unix nerds.
20:52:20 <elliott> marian_30: We're NEONEOpagans.
20:52:31 <ais523> Phantom__Hoover: that's a strange statement to make to someone who hasn't even spoken yet
20:52:32 <elliott> Gregor: I bet SCO are bidding
20:52:37 <Gregor> elliott: Dude, you're still at neoneo? I've evolved to neoneoneo.
20:52:48 <elliott> I'm at neo^\omega pagans of the iag neo om
20:52:53 <ais523> I think someone bid $8 or some similarly small amount for all SCO's assets
20:52:54 <elliott> Phantom__Hoover: YOU RUIN OUR FUNSTERS
20:53:15 <elliott> 10:41:29 <ais523> I agree that there's no chance of it taking over the world
20:53:15 <elliott> 10:41:46 <ais523> I think there's a marginal chance it'll lead to Wolfram being booted from the internet, but for unrelated reasons
20:53:16 <elliott> 10:41:53 <ehird> ais523: huh?
20:53:16 <elliott> 10:42:11 <ais523> you'll see later on, if they still have the feature I'm thinking of
20:53:20 <elliott> ais523: what was that feature? (context: W|A)
20:53:36 <ais523> oh, portscanning arbitrary sites that were entered into its = box
20:53:59 <Gregor> Honestly, think about it, what benefit does the domain name libc.so confer to e.g. SCO? Or even a legitimate company? Nobody's going to type libc.so into their web browser and go "WOW I SHOULD BUY THIS UNIX LOL"
20:54:15 <elliott> Gregor: MARKETING IS NOT BASED ON THE PRINCIPELS OF LOGIC AND SEN
20:54:19 -!- marian_30 has left (?).
20:54:35 <Gregor> elliott: But it IS based on money, and nobody's going to sink money into a completely valueless proposition!
20:54:44 <elliott> Gregor: Apart from IDIOTS.
20:54:59 <elliott> Gregor: And who are the kind of people who would bid on an auction for libc.so?
20:55:10 <Gregor> The intersection of domain name nerds and Unix nerds.
20:55:19 <Gregor> Which is to say, nerds.
20:55:21 <Gregor> Which is to say, people who either have little or a LOT of money :P
20:55:34 <elliott> Gregor: YOU KEEP IGNORING THE _IDIOTS_
20:55:48 <Gregor> elliott: Idiots don't know what libc.so is!
20:55:52 <elliott> I don't think Gregor has a little amount of money :P
20:56:09 <elliott> <Gregor> elliott: Idiots don't know what libc.so is!
20:56:40 <Gregor> OK, whotf is "APT Guy"
20:57:15 <elliott> Gregor: He is known only as APT GUY
20:57:23 <elliott> THE MEANING OF HIS CRYPTIC NAME IS UNKNOWN
20:57:36 <Phantom__Hoover> He has a name but it's trivially Googleable and he goes to the same school as me, so...
20:58:09 <elliott> Gahh, I wish I could have a machine where the whole Linux audio/video stack was NOT horribly out-of-sync.
20:58:23 <elliott> How does PulseAudio even MANAGE to be so TERRIBLE?
21:01:12 <Phantom__Hoover> Gregor, APT Guy is a guy at my school who is basically a script kiddie who convinced the Ubuntu guys to let him have fairly high-level APT access.
21:01:46 -!- azaq23 has quit (Quit: Leaving.).
21:02:02 <elliott> I still don't know wtf pat access is
21:02:33 <Phantom__Hoover> elliott, I think it means he can apply other people's patches.
21:03:52 -!- zzo38 has joined.
21:04:17 <Phantom__Hoover> He's also a Pythonista, but I wasn't able to use him to save cpressey.
21:05:38 -!- Phantom__Hoover has changed nick to Phantom_Hoover.
21:06:31 <zzo38> Do you know the Society of Creative Anachronism? I know of one of their games.
21:08:06 <zzo38> A game was made for them, called Ludus Equitum. It can be played with a normal chess set and some dice.
21:08:29 <elliott> Phantom_Hoover: unignoe zzo38 hesz fun
21:08:51 <zzo38> Or about yesserpc?
21:08:52 <elliott> whatsz the prblem with that i ask!!
21:10:14 -!- BeholdMyGlory has joined.
21:10:29 <zzo38> Zwaarddijk: You roll the dice 2d6 and the number that comes up tells you what kind of pieces you are allowed to move. You can make up to 2 moves in one turn. Win by capturing opponent's king.
21:11:52 <Zwaarddijk> iirc chess originally had some rule like that?
21:12:15 <zzo38> R=King, Q=Queen, L=Laurel, P=Pelican, E=Knight, M=Fighter, A=Squire. (Use the rook for P, bishop for L, pawn for M.) In this notation, FEN setup is "1eerqlp1/1mmmmmm1/8/8/8/8/1MMMMMM1/1PLQREE1".
21:12:31 <zzo38> Zwaarddijk: Some old chess game did have dice. This is a bit different though.
21:12:57 <Zwaarddijk> zzo38: so it has fairy chess pieces too?
21:13:35 -!- Behold has quit (Ping timeout: 255 seconds).
21:13:43 <Zwaarddijk> it'd be interesting to make something more go-like with dice
21:14:29 <Zwaarddijk> well, like, only one kind of piece, but such that the wrong number of stones n the board can turn out fatal if you have certain structures
21:14:37 <Zwaarddijk> and the wrong number comes up on the dice
21:14:45 <Zwaarddijk> but with another number can turn out v. good
21:15:25 <ais523> that would be hard to manage, I think, but probably interesting if it could be managed
21:15:52 <ais523> perhaps go where you roll a d6, your new stone has to be next to (the result - 1) other stones, color doesn't matter
21:16:22 <elliott> i just copied somethign to the licpboard
21:16:27 <elliott> tbabbed over to my browser
21:16:30 <elliott> almost fell asleep half way
21:16:32 -!- Sgeo has quit (Ping timeout: 276 seconds).
21:16:36 <Phantom_Hoover> http://28.media.tumblr.com/tumblr_li0vnjLRLg1qh25w2o1_500.jpg
21:16:38 <elliott> and thought me going to google something was a dream
21:17:12 <ais523> elliott: for what reason are you awake?
21:17:19 <zzo38> The Q and L move one space diagonal, P one space orthogonal, M like the pawn in chess but no double step, R like king in chess (but is permitted to be in check), A the same. M promotes only to A. The numbers on dice you can move: 1=RQ 2=RQ 3=PL 4=M 5=EA 6=EA.
21:17:20 <ais523> inability to sleep? being busy?
21:17:32 <elliott> ais523: according to stephen wolfram, it is 9 am in 12 hours
21:17:37 <elliott> as such, I will be going to bed within the hour
21:17:51 <elliott> in the anticipation that I will sleep for approximately that time, and wake up at a sane time
21:17:57 <zzo38> ais523,Zwaarddijk: Yes that is an idea about Go with dice.
21:18:00 <ais523> 12's a bit low, I think
21:18:16 <elliott> ais523: 14 is my record, and only once
21:18:22 <elliott> 12 is what I get in cases of extreme sleep deprivation
21:18:33 <ais523> the 23 was pretty extreme for me
21:18:34 <elliott> 11 in moderate sleep deprivation
21:19:06 <Zwaarddijk> after two weeks of extremely dutiful living, I slept 23 hours one day
21:19:50 <ais523> what do you mean by "dutiful" in this context?
21:20:03 <Zwaarddijk> fulfilling duties to various associations
21:20:21 -!- augur has quit (Remote host closed the connection).
21:20:35 <ais523> I think that was a little lost-in-translation without the clarification
21:20:37 <Zwaarddijk> viz. the association of comp.sci students at åbo akademi, and the student orchestra
21:20:50 <Zwaarddijk> yes, it was not meant to be losslessly translated
21:21:13 <Zwaarddijk> it was in fact an expression that wouldn't've worked in Swedish. point was: doing what you're supposed to isn't always smart
21:21:48 <ais523> normally I try to reduce the number of obligations I have in such cases
21:22:21 <Zwaarddijk> and i was essentially the chairman of the orchestra
21:22:38 <zzo38> I have written a ZRF implementation of this SCA game. http://www.chessvariants.org/membergraphics/MZludusequitum/LudusEquitum.zrf (Zillions is not particularly good software, but way better than nothing.)
21:23:17 -!- augur has joined.
21:24:00 -!- cheater- has joined.
21:24:03 <zzo38> What music do you play in that orchestra?
21:24:22 <Zwaarddijk> so essentially, my days during two weeks consisted of: waking up, getting a cup of coffee, running errands, getting cups of coffee in between errands, eating something, going to comp.sci.association ball-week events, coming home in the middle of the night, ...
21:25:02 <Zwaarddijk> big band jazz, some funk, some rockabilly, some modern stuff - the annual concerto usually has some theme like "80s music" or "latin american stuff" or "sweden" or "secret agents"
21:26:05 <Zwaarddijk> I play the jazz guitar - but I've not played there for about a year. also, my musical interests have veered into microtonality a bit too far to be compatible with any orchestra
21:26:10 -!- cheater00 has quit (Ping timeout: 255 seconds).
21:26:17 <ais523> in that most instruments can't play it
21:26:28 <ais523> and the ones that can, need people so insanely skilled to play them that it's not worth it?
21:26:29 <Phantom_Hoover> <elliott> ais523: according to stephen wolfram, it is 9 am in 12 hours
21:26:52 -!- elliott has quit (Quit: really need to sleep).
21:27:27 <Zwaarddijk> ais523: nah, trombones, trumpets and fretless string instruments do microtones trivially
21:27:42 <Zwaarddijk> saxophones I think are a bit less trivial, but still doable without too much effort
21:27:47 <zzo38> I have also worked with some non-12TET music, although I did them by computer.
21:28:02 <Zwaarddijk> so have I, but I got bored with doing them by computer
21:28:25 <Zwaarddijk> (I do sometimes still use a midi keyboard for it, but it feels weird not to have the octave-pattern repeat)
21:28:41 <zzo38> That is an idea too. Then you can readd frets to the notes that you want it to play instead?
21:28:54 <zzo38> And also retune the strings for the new notes.
21:29:09 <Zwaarddijk> zzo38: well, I could tie some thing there, sure, but I jsut leave it fretless and play by ear
21:30:38 <oklopol> Zwaarddijk: me 2, that's why
21:30:55 <quintopia> why would df show the used and available 1k blocks not adding up to the total size?
21:31:05 <quintopia> i've deleted a bunch of packages, but available blocks is still 0
21:32:16 <Zwaarddijk> ict:n kellarissahan asuu niitä jonkun verran
21:32:29 <oklopol> olin asteriski, mutta knnyin matemaatikoksi
21:32:32 <Zwaarddijk> (sorrry for the moon-language, people)
21:32:49 <oklopol> en harrasta opiskelijaelm, asun tyhuoneessani
21:33:32 <Zwaarddijk> just niin. tutkija vai onko sulla jopa opettajan virka?
21:33:54 <oklopol> "*jopa* opettajan virka" :D
21:34:06 <ais523> I fear I've just witnessed a pun I don't get at all
21:34:17 <oklopol> ja opiskelija, en ole viel maisteri.
21:34:58 <oklopol> ais523: no not really, he just asked me if i'm a researcher, or even a teacher. which i disagreed with since being a teacher is a lesser job.
21:35:15 <ais523> (the amusing thing is, most channels would just go "this is an English-speaking channel" in that situation; here, a) nobody cares, and b) nobody could truthfully make that statement anyway)
21:35:16 <Zwaarddijk> teachers have better contracts, though
21:35:17 <oklopol> Zwaarddijk: itse aloitin 2008
21:35:32 <oklopol> gradun sain valmiiksi viime viikolla
21:35:35 <Zwaarddijk> ensimmäiset kaksi vuotta opiskelin toki venäjää
21:35:38 -!- pikhq_ has quit (Read error: Operation timed out).
21:35:51 <Zwaarddijk> ja sen jälkeen oon ollut pari vuotta nokian tehtaan lattialla
21:36:34 -!- pikhq has joined.
21:37:07 <Zwaarddijk> tällä hetkellä näyttää siitä että yritän selittää (ja ymmärtää) miksi relativisointi ei auta
21:38:10 <Zwaarddijk> on uh Ion Petre, ÅA:n tietojenkäsittelytieteen proffa
21:38:30 <oklopol> ai A:lla on tietojenksittelytiedett :D
21:38:53 <Zwaarddijk> ennen vanhaan - sillon kun mä aloitin se oli "informationsbehandling2
21:39:53 <oklopol> tulee melkein englantia puhuttua enemmn livenkin
21:39:58 <oklopol> kun tiss kaikki ulkomaalaisia
21:40:18 <oklopol> no, paitsi muut opiskelijat, mutta en ky kauheasti luennoilla
21:40:21 <Zwaarddijk> miten matematiikassa saa kuvakielistä gradua aikaan?
21:40:35 <oklopol> siten ett kuvakielet ovat formaalien kielten alalaji
21:40:53 <oklopol> meill tehdn matematiikan laitoksella aika paljon teoreettista cs:
21:41:01 <Zwaarddijk> mite "kuvakieli" tarkkaan on englanniksi?
21:41:28 <oklopol> wikipedian artikkeli on aika <3
21:41:47 <Zwaarddijk> ok, en tiennytkään että semmosta käsitettä on olemassakaan
21:42:09 <oklopol> sehn siin hienoa onkin kun avoimet ongelmat on ihan vitun helppoja ratkoa kun niit on 5 ihmist yrittnyt
21:42:18 <Deewiant> What links here kertoo kuinka moni tietää
21:42:49 <oklopol> kannattaa mys huomata wp:n artikkelin referenssilista, hofl ja sitten joku aivan vitun random sivu :D
21:43:05 <Zwaarddijk> oklopol: munkin pitäisi löytää semmosta tieteen ala
21:43:44 <oklopol> Zwaarddijk: no eiks se P?=NP ole juuri sellainen
21:45:09 <Deewiant> The 't' in 'letters' is double
21:46:33 <Zwaarddijk> what's worst though, si the graphical resemblance between w and vv
21:46:35 <oklopol> Zwaarddijk: taking any math courses atm by any chance? :D
21:46:51 <Zwaarddijk> or like, one that I should've gotten done years ago
21:47:05 -!- zzo38 has quit (Remote host closed the connection).
21:47:43 <oklopol> a is the university Zwaarddijk is in
21:48:10 <oklopol> the buildings are hundreds of meters away from each other
21:48:27 <oklopol> Phantom_Hoover: it's the abbrev
21:48:27 <Zwaarddijk> oklopol: except the comp.sci parts and the biochemistry parts
21:48:53 <Zwaarddijk> it's like I dunno, a white and a darkie marrying :|
21:49:01 <oklopol> you lesser universities should stay out of wait actually even utu sucks except for the math dep
21:49:31 <oklopol> Phantom_Hoover: isn't even a finnish letter, a is the swedish university.
21:49:45 <fizzie> Phantom_Hoover: Åbo Akademi, it's a Finnish/Swedish place.
21:50:00 <Zwaarddijk> so uh, what's the lowest Erdös number in UTU?
21:50:16 <Zwaarddijk> Phantom_Hoover: we've got nothing to do with Sweden
21:50:25 <Zwaarddijk> we Finnish Swedes hate on it even more than the Finns
21:50:32 <fizzie> Yet you speak their language.
21:50:45 <Zwaarddijk> sort of. but they speak a diluted terrible version :|
21:50:53 <fizzie> The ones I've heard mix in Finnish words about half the time.
21:51:40 <Zwaarddijk> oklopol: I think ÅA has at least 4 as the lowest, but it might be 3, not entirely sure.
21:52:43 <oklopol> all i know is your math courses are rather uninteresting, and therefore utu > a
21:53:02 <fizzie> I know Erkki Oja from our place has an Erdös number of at most 4 (according to some online thing).
21:53:05 -!- leBMD has joined.
21:53:12 <fizzie> (And he's not a mathematician at all.)
21:53:23 <fizzie> (Mine's 6 now via that path.)
21:53:25 <Zwaarddijk> åa's comp.sci. department sucks pretty bad as well
21:53:34 <Zwaarddijk> they were going to arrange a course on complexity this spring
21:53:40 <oklopol> fizzie: link? i can check the ones that could be small here
21:53:45 <Zwaarddijk> but had only one professor knowledgeable enough about it
21:54:12 <oklopol> Zwaarddijk: we currently have a course on that stuff
21:54:13 <fizzie> oklopol: I think I used http://www.ams.org/mathscinet/collaborationDistance.html -- it has the "Use Erdös" button there.
21:54:53 <oklopol> it's pretty simple stuff tho
21:55:27 <Zwaarddijk> well, I figure it is - in the reading up on teh state of complexity theory for my kandi, I've been able to predict stuff tht's coming up in the book about a chapter ahead of it
21:55:36 <oklopol> my employer has 3, so i'll have 4 soon prolly
21:55:48 <oklopol> Phantom_Hoover: bachelor's degree
21:56:57 <oklopol> Zwaarddijk: what are you reading?
21:57:32 <oklopol> the course is simple because we're just doing the basic np completeness stuff, complexity theory itself is rather vast already
21:57:37 -!- leBMD has quit (Client Quit).
21:57:39 <oklopol> have you read complexity theory companion?
21:58:16 <Zwaarddijk> Papadimitriou's Computational Complexity
21:58:37 <oklopol> the leader of the math dep has erdos number 2
21:59:01 <oklopol> well that's really basic stuff yeah
22:00:04 <oklopol> 4 is better than fizzie's already
22:00:57 -!- Behold has joined.
22:01:15 -!- pikhq_ has joined.
22:01:17 <oklopol> the erdos distance is a bit weird tho, i'd prefer publishing alone
22:01:22 -!- pikhq has quit (Ping timeout: 248 seconds).
22:02:44 <oklopol> he said it 10 lines ago or something
22:04:36 -!- BeholdMyGlory has quit (Ping timeout: 276 seconds).
22:05:24 <Gregor> Dude, mines is less than 6.
22:05:40 <fizzie> 6 is the upper bound I know, MathSciNet's thing doesn't really do computer science and related fields so there might be a less circuitous path.
22:06:23 <oklopol> erm, i don't even have an erdos number yet tho
22:06:39 <oklopol> but 4 is likely to happen soonish
22:07:02 <oklopol> but if Gregor is 4, then that's less than nothing
22:07:35 <Gregor> Me -> Jan Vitek -> Nir Shavit -> Michael E. Saks -> Erdos
22:09:00 <fizzie> The path I know goes me -> Kurimo, Mikko -> Oja, Erkki -> Cooper, Leon N. -> Zeitouni, Ofer -> Diaconis, Persi W. -> Erdös.
22:10:36 <Deewiant> Me -> Heljanko, Keijo -> Niemelä, Ilkka N. F. -> Przymusiński, Teodor C. -> Rudin, Mary Ellen -> Erdős
22:13:40 <Phantom_Hoover> No, I cannot be bothered to work out the right compose sequence.
22:20:53 -!- pikhq has joined.
22:21:03 -!- pikhq_ has quit (Ping timeout: 255 seconds).
22:21:04 -!- Sgeo has joined.
22:33:50 <pikhq> #esoteric should write an academic paper together.
22:34:26 <Sgeo> I feel as though I'd be unable to contribute
22:34:35 <Sgeo> Although I would like to try
22:35:39 <pikhq> Your number one problem is taking your dad's word for everything.
22:35:55 <Phantom_Hoover> I'd suggest you fix the typos but oerjan has more experience than you at that.
22:35:58 <pikhq> Your number *two* problem is being convinced that you have no intellectual ability.
22:37:11 <Sgeo> I'm not convinced I have no intellectual ability. Just that it is firmly sandwiched with #esoteric absolutely above, and RL people absolutely below
22:37:44 <oklopol> i'm real i even have a skin
22:37:45 <pikhq> You're just noticing that some of us are *absurdly* good. :P
22:38:12 <pikhq> (I do not include myself in the set of "absurdly good". Perhaps "pretty damn smart".)
22:39:57 <pikhq> (my procrastination makes "absurdly good"... Not quite applicable.)
22:41:30 <oklopol> i used to think i was smart but i don't think i think that anymore
22:43:19 <pikhq> I think I'm smart, and that fucking disappoints me.
22:43:27 <oklopol> i'm certainly good at some stuff, but i think it's more because i actively train those abilities hours and hours every day
22:43:36 <pikhq> I should *not* be smart. But holy fucking *hell* almost everyone is completely and utterly STUPID.
22:44:00 -!- FireFly has quit (Quit: swatted to death).
22:44:05 <pikhq> So I suppose it's not that I consider myself smart as it is that I consider a good 9/10ths of humanity depressingly retarded.
22:44:35 <oklopol> hmm well i suppose that's close to my viewpoint
22:47:15 -!- MigoMipo has quit (Read error: Connection reset by peer).
22:49:28 -!- Phantom_Hoover has quit (Remote host closed the connection).
23:02:46 <pikhq> *Dang* profile-guided optimisation can make an absurd difference.
23:02:59 * pikhq is running bsnes, accuracy profile, in realtime.
23:09:00 -!- copumpkin has quit (Ping timeout: 264 seconds).
23:11:31 -!- copumpkin has joined.
23:15:21 -!- Behold has quit (Remote host closed the connection).
23:25:34 -!- calamari has joined.
23:28:42 <augur> oklopol: do you read sate?
23:32:50 <Zwaarddijk> hey, do multi-tape automata for weaker architectures than the turing machine gain anything compared to single-tape versions?
23:36:43 <ais523> it can even make them TC in some cases
23:36:50 <ais523> especially if you can construct a Minsky machine counter out of one tape
23:40:12 <Zwaarddijk> right, I kind of guessed that a pushdown automaton probably can be made TC
23:41:29 <quintopia> how to kill a process that kill -STOP and kill -KILL won't kill?
23:53:46 <ais523> quintopia: kill its parents
23:54:03 <ais523> it's probably a zombie, waiting to be reached
23:54:09 <ais523> kill -STOP never kills anything, btw