00:03:47 -!- sebbu2 has quit (Connection timed out). 00:14:08 -!- calamari has joined. 00:26:21 "(The paradox is alleviated if "interesting" is instead defined objectively: for example, as of November 2008, the smallest natural number that does not have its own Wikipedia entry is 215.)" 00:26:25 and now it has one! 00:26:35 ... it redirects to 210, though 00:29:35 anyone know if there's a defigletizer? 00:33:31 -!- oklofok has joined. 00:35:26 Deewiant: Any googlebomb luck? 00:35:30 Results 1 - 10 of about 167 for polytypic ziggurat. (0.44 seconds) 00:35:31 :-( 00:36:22 215 redirects to 210? :P 00:36:31 -!- sebbu2 has joined. 00:36:57 GregorR: which covers 210-219 00:37:05 I should hope so. 00:51:14 -!- sebbu has quit (Read error: 110 (Connection timed out)). 01:33:35 -!- psygnisfive has quit ("Leaving..."). 02:37:53 -!- calamari has quit (Read error: 110 (Connection timed out)). 02:39:32 -!- calamari has joined. 02:56:49 "I lost 15 pounds in a fortnight using acai" 02:56:52 Oh did ye? 02:57:08 You lost the ability to use the metric system 02:57:16 15 POUNDS IN A FORTHNIGHT 02:57:35 Are you going to throw horsepower and faraday in the mix? 02:58:00 'twas a spam subject line :P 02:58:08 Horsepower is perhaps the stupidest measurement ever. 02:58:14 No it's not 02:58:26 Because horses are so consistent? 02:58:26 It's from an era where that was a well known frame of reference 02:58:49 Do you know what a mean value is 02:59:04 It's a value that doesn't treat its friends respectfully. 03:00:33 Fine, I'll change my statement, from 03:00:37 Horsepower is perhaps the stupidest measurement ever. 03:00:37 to 03:00:41 Horsepower is perhaps the stupidest measurement presently. 03:06:23 ur stupid 03:06:45 ur mom! 03:17:49 ur moms mom 04:06:27 I'll see your horsepower and raise you a library of congress 04:14:14 The helen is a pretty stupid measurement. 04:15:12 Then again, nobody uses it. 04:17:27 It's the amount of beauty required to launch a thousand ships and burn the topless towers of Ilium. 04:18:45 It's considered ill-defined due to the passing of an ordinance requiring all towers of Ilium to be fully clothed. 04:32:37 -!- Dewio has changed nick to Dewi. 04:43:35 -!- oerjan has joined. 04:48:05 -!- bsmntbombdood_ has quit (Read error: 110 (Connection timed out)). 05:11:35 -!- psygnisfive has joined. 05:33:03 -!- oerjan has quit ("leaving"). 05:53:44 -!- coppro has quit ("The only thing I know is that I know nothing"). 05:54:05 -!- bsmntbombdood has joined. 06:03:20 -!- calamari has quit ("Leaving"). 06:13:20 -!- psygnisfive has quit (Read error: 110 (Connection timed out)). 06:15:44 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 06:17:33 -!- bsmntbombdood has joined. 06:19:35 -!- Sgeo has quit ("Leaving"). 06:20:07 -!- kar8nga has joined. 06:30:03 You know, I once wanted to write a compiler from C to Subleq, but I think it would be easier to compile from Haskell instead. 06:44:21 It would be easier (although far from easy) to write one from MIPS. 07:06:49 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 07:08:03 -!- bsmntbombdood has joined. 07:11:29 -!- kar8nga has quit (Read error: 110 (Connection timed out)). 07:45:03 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 07:46:44 -!- bsmntbombdood has joined. 07:58:38 -!- BeholdMyGlory has joined. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:11:03 -!- olsner has changed nick to coolsner. 08:11:24 -!- tombom has joined. 08:35:45 -!- GreaseMonkey has joined. 08:55:39 -!- BeholdMyGlory has quit (Remote closed the connection). 09:21:28 -!- Tritonio has joined. 09:22:26 -!- Tritonio has quit (Remote closed the connection). 09:24:17 -!- Tritonio has joined. 09:24:35 -!- Tritonio has left (?). 10:31:57 ef is such an elegant language 10:39:52 http://www.vjn.fi/oklopol/ef.txt <<< i added some elegance. god it's elegant. 10:40:28 it used to be about the fixed points, but i'm thinking i like the supersmart pointers even better. 10:45:17 Deewiant, 10:45:19 GOOD: 2k ;;;5 executes 5 thrice 10:45:20 sure 10:45:23 but what if you have: 10:45:27 2k ;;;p 10:45:33 and that p writes something between the k and the p 10:45:43 nice smiley 10:45:45 what will be the next instruction to execute after the k finishes 10:45:51 btw AnMaster did you see how elegant it was? 10:46:05 don't you just love the intricacies of how the pointers work. 10:47:14 and what's ;? 10:47:40 skip? 10:47:47 oklofok, skip to matching ; 10:47:49 AnMaster: k specs say that the fetch happens once, not between every execution 10:48:29 And when you're done executing the k you move on as normal. 10:48:31 AnMaster: so it skips from first to second, then proceeds to third, skips to first, proceeds to second, skips to third and then moves on the 5 and executes? 10:48:41 oklofok: Yep. 10:48:49 alright. 10:48:59 that's one sick test case. 10:49:09 :-P 10:49:26 hm 10:49:35 but err why three time? 10:49:36 s 10:49:39 oh 10:49:40 Deewiant, so k will execute the instruction p wrote next then 10:49:46 err 10:49:46 if it overwrote the first ; 10:49:49 no oh, i don't know 10:49:57 AnMaster: Yep, IMO. 10:49:58 my thinking: if you p'd on the p, it would probably run the p x amount of times and then it would land on what the p has next cycle 10:50:09 hm 10:50:21 actually, that kinda inspired me 10:50:34 Deewiant, yes it makes sense. But I think it is actually UNDEF. 10:50:40 you could plot lots of crap 10:51:02 also, are the SGML spaces REALLY necessary? 10:51:07 yes, 10:51:08 . 10:51:16 Gracenotes, they are easy to implement too. 10:51:18 AnMaster: Well, a lot of k is somewhat UNDEF according to the specs; I'm including that e-mail exchange with Pressey into the specs here 10:51:22 i mean for funge-109... what benefit would it have? 10:51:50 GreaseMonkey, a bit more well speced, I'm reworking it currently. 10:51:53 The benefit is that you can write spaces on top of characters with p thus removing them from the string 10:51:59 a"!dlroW olleH"ck,@ 10:52:09 ok. 10:52:17 actually... 10:52:22 Or at least, *a* benefit is :-P 10:52:26 couldn't you make it ignore all 0s? 10:52:32 instead of ignoring all spaces? 10:52:33 GreaseMonkey, for what 10:52:35 err 10:52:36 what 10:52:36 strings 10:52:42 i mean 10:52:43 what do you mean... 10:52:44 \0s 10:52:46 not the number 0 10:53:01 GreaseMonkey, so outputting a binary file would be impossible? 10:53:04 Why that and not spaces 10:53:13 AnMaster: No, he means "SGML NULs" 10:53:27 would \x01s be a better option, perhaps? 10:53:30 ignoring \0 in input files 10:53:31 what 10:53:38 GreaseMonkey, why not the spaces. 10:53:52 because if you needed two spaces, you have to shove a   in 10:53:57 6 characters to make up one 10:54:03 and if you slam something over it 10:54:07 e 10:54:09 btw is ; an actual command at all? i mean shouldn't 2k ;;;5 execute ; twice, and then do 5 just once? 10:54:09 err* 10:54:18 "Hello World" becomes "Hello nbsp;World" 10:54:22 GreaseMonkey: "foo "::::"bar" <- "foo bar" 10:54:33 ; is not a command just like spaces aren't 10:54:34 oklofok: ; and space are markers, not instructions. 10:54:37 GreaseMonkey,   isn't understood by funge. 10:54:43 but what Deewiant said works 10:54:51 ok 10:54:53 Deewiant: right, i thought it might be something like that 10:55:11 actually... \x7F would be appropriate 10:55:29 it has meaning. 10:55:45 GreaseMonkey, are you just trying to skip the needed two line logic of handling multiple spaces? 10:56:06 nope, as this would probably also take up as much code 10:56:18 I don't like k. Did it go so that "2kX" executes X thrice (because the k executes it twice, and then it is done normally once), but "0kX" somehow skips X completely? 10:56:34 Yes. :-/ 10:56:35 i think so 10:56:37 yes 10:56:39 it's terrible 10:56:43 ah, makes sense 10:56:45 it simply makes no sense. 10:56:52 that's why i didn't get it 10:56:59 fizzie, yes. 109 will fix that 10:57:02 The rules make sense, they're just nonsensical. 10:57:05 so the instruction after is always skipped. 10:57:13 my reasoning for this is if someone wants to attack "foo ":::"bar", then they may get a quotation mark 10:57:35 err "attack" 10:57:36 what 10:57:48 Deewiant: they make sense kinda like assuming undef doesn't crash the interp makes sense, perhaps? 10:57:50 erm, if they want to scrub it out 10:58:00 then they may scrub out a quotation mark 10:58:09 oklofok: No, they make a different kind of sense 10:58:14 The rules are also very interpretative. The part of 0k skipping is written as "Note that some instructions don't make much sense within the context of k unless you include zero as one of the possibilities for how many times the instruction is repeated. For example, no matter how many times after the first time k execute ^, the result is the same. However, you may pass a zero count to k, and the ^ instruction will not be executed; this can be a valuable behavi 10:58:14 our." 10:58:21 Deewiant: i thought they make a rather similar kind of sense. 10:58:27 but DUNNO. 10:58:40 fizzie, yes. 10:58:41 The k specs aren't very interpretative IMO 10:58:53 On the contrary, you have to read them 100% literally to get it right :-P 10:58:59 indeed 10:59:13 But that's "valuable behaviour" even without the skipping, if you assume the last "will not be executed" means "will not be executed by k"; I mean, you can do 0k^ and it would turn up either at k or the ^, depending on the input. 10:59:31 (That is, if k would never skip the next command.) 10:59:43 Okay, I suppose that's true. 10:59:47 fizzie, except. C. Pressy answered on this with his own small test suite thingy. 11:00:00 well not that aspect. 11:00:02 hm maybe 11:00:02 Why do you always write the surname worng? Pressey. 11:00:16 meh 11:00:59 also, please don't complicate it further. 11:01:14 that means lots of interpreters have to change *again* 11:01:15 * oklofok complicates it further: pressaye 11:01:31 not that that, the other that 11:01:32 i'm going to bed now 11:01:33 gnight 11:01:37 night 11:01:38 cya 11:01:41 nide 11:01:55 Neetsch. 11:01:58 -!- GreaseMonkey has quit ("goude naight."). 11:03:28 Deewiant, actually I don't agree about: 'v041k ;;;p 11:03:44 assuming that is the whole program 11:03:58 That puts a v at 1 11:04:21 Deewiant, yes true, but it also puts one on the first ; as far as I can see 11:04:26 err no 11:04:30 No :-P 11:04:42 Deewiant, the 1 is a parameter to k isn't it? 11:04:48 Yep 11:05:05 And then you call p with (4,0) and 'v' 11:05:11 And (4,0) is the 1. 11:05:12 oh 11:05:16 right 11:05:26 'v081k ;;;p 11:05:29 I meant 11:05:51 That puts it on the middle ; which means it gets skipped. 11:05:57 7 then 11:05:58 anyway 11:06:06 I think k will resume at p after 11:06:07 Then it's hit and executed. 11:06:10 or if it won't it is undef 11:06:16 Why would it do that. 11:06:41 Deewiant, sec... 11:08:02 Deewiant, where does it say it won't 11:08:08 Where does it say it will? 11:08:10 I agree there is nothing explicitly saying it *will* 11:08:17 So why would it, then? 11:08:19 but it makes just as much sense 11:08:25 When you execute an instruction, you continue to the next one as normal. 11:08:32 You don't arbitrarily skip instructions. 11:08:41 Since the p is executed "at k", I personally think it stands to reason that it's from there, and after the execution of k (and indirectly p), that it continues to look for the next operation. 11:09:08 With that always-skip-the-next-instruction k, it would make sense to continue from the p onwards, though. 11:09:19 Yep. 11:09:22 fizzie, ok that argument makes sense I guess. 11:10:05 It's just sad that with so many k-related things it is possible to write sensible-sounding arguments for contradictory things. 11:12:15 Deewiant, however ccbi seems to skip it too. 11:13:38 No it doesn't. 11:16:30 hm ok 11:16:39 nor does cfunge it seems in fact. 11:16:47 the code was just too confusing :D 11:16:53 :-P 11:17:51 in other news the thread supervisor is almost working in efunge now. Only issue is that it quit doesn't work in that branch atm. 12:07:23 * kerlo ponders bit widths and the amounts of memory they can access 12:09:25 1-bit: 2 bits. 2-bit: 1 byte. 4-bit: 8 bytes. 8-bit: 256 bytes. 16-bit: 128 kilobytes. 32-bit: 16 gigabytes. 64-bit: 32 whatever-comes-after-peta-bytes. 12:09:45 Good to know if you're writing Subleq programs, maybe. 12:10:40 heh 12:22:20 -!- Gracenotes has quit (Remote closed the connection). 12:29:27 http://omploader.org/vMWw1Zw <-- fancy supervision tree for efunge :D 12:50:54 -!- bsmntbombdood_ has joined. 12:51:41 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 13:31:56 -!- ais523 has joined. 13:32:20 hi ais523 13:32:23 http://omploader.org/vMWw1Zw 13:32:26 hi 13:32:30 and what's that a link to? 13:32:33 I'm on a public computer atm 13:32:34 ais523, a diagram 13:32:40 so I /especially/ don't want to click links at random 13:32:45 ais523, of efunge supervision tree 13:32:55 enterprisy. 13:33:10 ah, ok 13:33:22 I still think this is an insane idea, but I don't begrudge you for it 13:33:30 it's only slightly more insane than porting INTERCAL to embedded systems 13:33:32 ais523, what? ATHR you mean. 13:33:39 because I need this for ATHR. 13:34:10 just, distributed Befunge 13:34:39 ais523, ah that. Well that is a step after I get it working on a single node. Btw <0.1.2> are erlang pids, for the unnamed processes. 13:34:58 just that tree shows names if there are any 13:35:56 -!- oklofok has quit (Read error: 60 (Operation timed out)). 13:37:35 ais523, and distributed would just be 1) make sure right server run on the right node (you need all IO to be channeled through one server on one node for example) 2) some changes to funge space. 3) possibly similar changes as for funge space to some other daemons. 13:38:10 it would be some work, but not infeasible. 13:38:16 (sp?) 13:39:09 ais523, currently I'm trying to make the thread supervisor work correctly. It has some issues atm. 13:39:09 I think you spelt it correctly 13:39:29 strange, the spellchecker here doesn't like "spelt". Or "spellchecker". It doesn't mind "infeasible", though. 13:39:41 haha 13:44:29 * ais523 drops a link to http://catb.org/~esr/jargon/html/Z/Zawinskis-Law.html in here for ehird 13:44:45 I wonder if the GNU people knew of that law when they went and bloated hello(1)? 13:45:47 -!- Sgeo has joined. 13:47:36 ais523, is that jwz 13:48:02 ? 13:48:31 "Jamie Zawinski" ==jwz right? 13:48:35 == * 13:48:53 I have no idea 13:49:13 yep seems so. 13:49:20 in that case it is extra ironic... 13:49:30 why, what did jwz do? 13:49:41 Netscape Navigator. 13:49:53 haha 13:49:56 Which could read mail in version something or later 13:50:11 ais523, you seriously didn't know who jwz was? 13:50:26 err, no, I tend not to care about names all that much 13:50:29 I mean he is as famous as djb. By initials. 13:50:36 * AnMaster waits for "who is djb" 13:50:39 I only vaguely know who djb is 13:50:51 responsible for djbdns 13:50:54 at least 13:51:08 and possibly qmail, although I'm not entirely sure whether that was him or someone else without looking it up 13:51:11 djb: author of qmail, djbdns, and a lot more. Famous enough to be known by initials. 13:57:56 -!- ais523 has quit ("http://www.mibbit.com ajax IRC Client"). 14:13:57 -!- ais523 has joined. 14:26:17 what's qmail? 14:27:00 wait actually i don't care, djbdns sounds like something dns related, not that i can think of anything dns related that could be complicated enough to be famous. 14:27:04 oklopol, a secure and stable MTA. 14:27:15 oklopol, djbdns is a dns server. 14:27:27 but qmail is more famous. 14:27:32 mta? 14:27:50 Mail Transfer Agent. 14:27:53 right 14:27:54 EMail server. 14:28:00 ais523: You probably had an American dictionary on that spellchecker. 14:28:05 oklopol, smtp to be specific 14:28:16 Deewiant: ah, possibly 14:28:18 okay, both of those sound incredibly trivial 14:28:31 ais523: Since they'd write it as "spelled" 14:28:33 * AnMaster is stuck in a twisty maze of licenses all different. 14:28:37 :/ 14:28:52 well, BSD is compatible with more or less anything 14:29:11 AnMaster: remember to keep your left hand on the wall 14:29:13 ais523, GPLv3 and EPL (which is similar to MPL) 14:29:23 ugh, that's a rather nasty mic 14:29:25 *mix 14:29:30 unless it's an infinite maze, then you know iterative deepening. 14:29:37 ais523, yes and I'm wondering if it is legal. 14:30:23 probably not, unless the EPL's terms are a subset of the GPLv3's 14:30:29 does maze actually convey a planar graph btw, or is it more general? 14:31:20 you can have nonplanar mazes, although they're more difficult 14:31:23 ais523, well you are allowed to link a GPL program to closed source libc aren't you? So as long as I use EPL as a standard library there should be no issue. (And I do use erlang's standard library, there is no way around that.) 14:31:25 Enigma 1.1 has a 4D maze in it 14:31:33 ais523: i meant the term 14:31:46 AnMaster: ooh, the GPL with closed source libc is a huge raging argument 14:31:49 also 3d and 4d mazes are both arbitrary graphs. 14:31:58 oklopol: I think nonplanar mazes are still considered mazes 14:32:01 mathematically speaking 14:32:14 ais523, issue now is, the standard erlang supervisor module doesn't cut it for the thread supervisor. I need to customise some internals for it to work properly. 14:32:22 So I need to distribute a modified version of it. 14:32:28 that is where I wonder... 14:33:48 well, i guess what i said was rather obvious, planar is pretty much defined by being a 2d-maze, and not bigger 14:33:58 bigger in amount of dimensions 14:34:02 ais523, hm... how else would you run open source apps on windows or OS X. I mean you clearly have to link them, at some layer, kernel32.dll, syscalls or whatever, to non-open source. 14:34:04 i love how clearly i can say things 14:34:56 Enigma 1.1 has a 4D maze in it <-- So it changes with time. But what is the 3D bit about. 14:35:07 I didn't know Enigma could do 3D. 14:35:07 err 14:35:15 i'm pretty sure he meant spatial dimensions 14:35:24 ok even stranger... 14:35:25 AnMaster: no, 4 spatial dimensions 14:35:26 the diagonals are used for the other two 14:35:29 god i hate it when people thing the fourth dimensions is time 14:35:40 and there are lots of teleports involved in order to make everything link up correctly 14:35:46 ais523, I see. 14:35:47 that's retarded, even in physics, you should call it time. 14:36:08 oh umm 14:36:15 i guess 14:36:18 if umm 14:36:20 ais523, hm one way around would be those rare "as a special exception you are allowed to link ..." 14:36:28 oklopol: There's a "4D theater" in the Linnanmäki amusement park; the "fourth dimension" means they spray water on your face and vibrate the seats. 14:36:34 ais523, but I would need a lawyer to write one. 14:36:37 blah, i can't articulate what i was about to say 14:36:51 but i had this point, maybe i'll tell it later. 14:36:58 fizzie, hah 14:37:06 fizzie: clearly that's 5 dimensions 14:37:08 (And the third comes from those shutter 3D glasses. Or maybe polarized, I've forgotten now.) 14:37:23 those are better described by a fourth dimensions than time is 14:37:31 *dimension 14:37:51 Well, the seat-movements need more than one dimension. 14:37:59 probably, 14:37:59 ais523, any idea? 14:38:04 *. 14:38:08 AnMaster: is the GPLv3 code yours? 14:38:12 ais523, yes it is. 14:38:36 why not use a different licence, like the EPL itself? 14:38:58 ais523, For the same reason you wouldn't use MPL? 14:39:00 fizzie: actually it's probably the "much vibration" - "no vibration" scale that's relevant. 14:39:00 brb phone 14:39:17 so i'd say one. 14:39:26 and water is clearly separate 14:40:47 oklopol: Well, I dunno; it's not really "vibrate", more like "move around". I just accidentally used a bad word to describe it. Though I don't really recall exactly how it moved; it's possible it was only able to move forward/backward along one dimension, and not up/down or anything. 14:41:14 ah okay 14:42:39 "4D coaster" is a type of a roller coaster, but I can't find out what sort. It seems it has something to do with rotating seats there too. 14:48:42 "A 4-D film (sometimes written 4D film) is a marketing term that describes an entertainment presentation system combining a 3-D film with physical effects in the theatre, which occur in synchronization with the film. Because the physical effects are expensive to set up, 4-D films are currently presented only at special venues such as theme parks and amusement parks." 14:48:50 Oh, it's an official term; it has a Wikipedia page. 14:49:08 -!- ehird has left (?). 14:50:17 -!- ehird has joined. 14:50:19 -!- ehird has left (?). 14:50:23 -!- ehird has joined. 14:51:34 fizzie, an abuse of that term. 14:53:52 12:49 AnMaster: in that case it is extra ironic... 14:53:53 Er, no. 14:54:00 He supports kitchen sinks. 14:54:02 Also, XEmacs too, man. 14:54:08 He was there at the start of Lucid Emacs. 14:54:12 12:44 ais523 drops a link to http://catb.org/~esr/jargon/html/Z/Zawinskis-Law.html in here for ehird 14:54:14 I know 14:54:20 Plz to be linking to non-jargon-file sources in futu 14:54:21 re 14:54:30 why? 14:54:36 do you hate the Jargon File, or something? 14:54:46 ais523: I hate what esr's done to it. 14:54:52 ah, ok 14:54:54 Hello, 14:54:54 14:54:56 I know you would be surprised to read from someone relatively unknown to you before. My name is Major Larry Downs, a member of the U.S. ARMY USARPAC Medical Team, which was deployed to Iraq in the beginning of the war in Iraq. 14:55:00 14:55:02 I would like to share some highly personal classified information about my personal experience and role which I played in the pursuit of my career serving under the U.S 1st Armored which was at the fore-front of the war in Iraq. 14:55:06 — Spam 14:56:54 ehird, where did credit card come into it 14:57:32 AnMaster: nowhere, he just told me to read a bbc article and reply saying I understood 14:57:38 a rather bizarre start to a scam. 14:57:51 indeed. 14:59:58 On the subject of jwz, 15:00:03 http://jwz.livejournal.com/1040129.html 15:00:15 His fingers have the power to melt alt keys on $500 keyboards. 15:02:14 -!- ais523 has quit (Read error: 104 (Connection reset by peer)). 15:04:27 Ok, I have become a twitter addict 15:04:35 Oh god. 15:04:51 I can imagine it now. "Became a twitter addict." "PSOX." "Something AWFUL is happening on Second Life!" 15:05:11 "Combined PSOX and Second Life." "Died happy." 15:05:18 lol 15:06:23 parse ski 15:06:23 parseski={!x:_;(!x=='` & ! x+1 & ! x+2 : (class[])) => ! x..x+2 = [! x+1,! x+2]} 15:06:27 oklopol: that's amazing 15:11:20 -!- FireFly has joined. 15:12:37 ehird, um, "died happy"... did it happen when he tried to combine all three. 15:13:01 AnMaster: he died happy in posting the tweet to twitter about how he combined psox and second life 15:13:09 so he did combine all three in the "Combined..." message. 15:13:25 Today on impractical computing environments: http://www.gamesetwatch.com/2009/04/one_step_closer_to_a_holodeck.php 15:14:03 ehird, how did he fit that in 140 chars... 15:14:13 15:04 ehird: I can imagine it now. "Became a twitter addict." "PSOX." "Something AWFUL is happening on Second Life!" 15:14:13 15:05 ehird: "Combined PSOX and Second Life." "Died happy." 15:14:19 nothing there's over 140 characters 15:16:15 Reading stuff about cpu-specific asm optimization is fun. 15:16:36 Especially when it's about things like the Nehalem. We must hyperoptimise to save a nanosecond on this top of the range CPU! 15:16:39 ehird, what optimization in specific. 15:16:49 the article and comments of http://www.reddit.com/r/programming/comments/8fqgy/cmov_instruction_a_bad_idea_on_outoforder_cpus/ 15:16:50 Nehalem hah 15:16:59 AnMaster: what's that supposed to mean 15:17:03 :p 15:17:25 I read that a while back but used CMOV in dobelx64 anyway because it saves a few bytes 15:17:32 Deewiant: Hahaha 15:17:58 AnMaster: but srsly wuzzat supposed to mean 15:18:22 my own tests show cmov helps in some places. 15:18:25 jne x; mov foo, bar; x: is a byte or two bigger than cmovne foo, bar 15:18:32 I meant 15:16 AnMaster: Nehalem hah 15:18:40 s/cmovne/cmove/ 15:18:45 ehird, I'll let you figure that out yourself. 15:18:45 cmovnik 15:18:55 AnMaster: oh, "AMD didn't make it, it sucks?" 15:19:01 nop. 15:19:04 nope* 15:19:18 well I'm not telepathic yet sorry 15:22:24 AnMaster: no hints? 15:26:25 no 15:29:56 ehird, anyway I have seen icc put cmov in code optimised for core2 at least. 15:30:05 -!- andreou has quit (Remote closed the connection). 15:30:12 so clearly even intel thinks it makes sense sometimes. 15:30:17 if you read the comments that was acknowledged 15:33:23 -!- Dewio has joined. 15:38:51 ehird, I saw that. 15:39:36 ehird, personally I tend to leave it to the compiler. 15:40:04 "Reported estimates indicate that transistors at these dimensions are significantly affected by quantum tunneling." 15:40:05 http://en.wikipedia.org/wiki/11_nanometer 15:40:29 ehird, how much is intel down to currently 15:40:42 AnMaster: 45nm 15:41:16 "however, Intel has already given some indications as to the nature of its process and its rough timing for 2009.[4]" 15:41:18 No, 32nm 15:41:18 regarding 32 nm 15:41:23 Deewiant: not public 15:41:24 -!- Dewi has quit (Read error: 101 (Network is unreachable)). 15:41:26 just in development 15:41:27 right? 15:41:31 They've demoed some stuff 15:41:33 ah 15:41:39 But no complete processors, I think 15:41:41 I guess the last few steps will be the hardest 15:41:43 Just some SRAM or whatever 15:41:52 They're brushing against weird physics from what I can tell 15:42:02 AMD's planning on releasing 32nm stuff in 2010 15:42:32 We just need negative length chips. 15:42:38 Install an Intel, get some room space back. 15:42:47 ehird, the first comment is confusing. I don't see why swapping operands affected the result so much. 15:43:04 AnMaster: branch prediction 15:43:10 but 15:43:12 that's on a pentium 4 15:43:13 -!- oklofok has joined. 15:43:17 so go figure 15:43:19 ah 15:43:21 MORNING 15:43:21 In other news: TRDS doesn't work in CCBI 2 and I have no fricking clue why 15:43:38 i suppose the pentium 4 gnomes get confused if you give them it in in the other order 15:43:51 and run around aimlessly for a few microseconds, bumping into each other 15:43:54 *nanoseconds 15:44:09 hah 15:44:31 ehird: the ski parser works by taking the fixed point of the trivial parsings `(parsed)(parsed) 15:44:37 oklofok: hawt 15:44:41 => [` (parsed) (parsed)] 15:44:47 i find it a lovely paradigm 15:46:45 (it doesn't actually work yet, it'd need a small modification, you'd need to wrap the actual combinators in lists first, but that'd take me multiple minutes, so i'll just leave it broken for now) 15:47:05 this track messes with my ears 15:47:22 ehird, about worn out keys. How is your keyboard. 15:47:31 AnMaster: I've never worn out a single key. 15:47:35 basically the correction is x:chars=>[x] 15:47:43 Then again, I switch keyboard every few years (unintentionally) 15:47:46 On mine many of the keys lack text nowdays, but no such dents as on that pic... 15:47:49 err, x=[x] 15:47:51 but anyway 15:48:14 This keyboard is a year or so old; I dropped my old version - which was identical 'cept UK not US layout - and it wouldn't turn on, so I got a replacement free from apple 15:48:19 maybe two years 15:48:20 yeah two years 15:49:00 wouldn't turn on... Oh right one of the fanged ones... 15:49:09 my ubuter's keyboard is completely broken, but it has experienced pretty much everything from cups of coffee to frustrated humans using it as a bat. 15:49:10 AnMaster: Bluetooth, yah. 15:49:19 yeah that's what I said. 15:49:24 well not completely, more like a tiny bit, but enough to be annoying to use 15:49:29 Thus "yah". 15:49:38 I just want wireless power ;-) 15:49:40 cables 15:49:41 that is 15:49:48 control and left key used to be broken, but control got fixed recently when i accidentally jumped on the kb 15:49:53 from the bed 15:50:50 "I dropped my old version - which was identical 'cept UK not US layout - and it wouldn't turn on" <<< what kind of crap do they sell in uk 15:50:53 ehird, I just fail to see what happened to jwz's Alt key... an accident with a solder iron maybe. 15:51:09 first thing i do with my new things is make sure dropping them doesn't break them 15:51:09 well 15:51:10 okay 15:51:11 that's a lie 15:51:13 AnMaster: He uses emacs for everything and he's had the kb for 8 years 15:51:15 but still 15:51:21 ehird, it seems the top of the key cap is rather thick on that keyboard... 15:51:25 And he's been typing for decades 15:51:33 thicker than mine certainly 15:51:34 So just veeeeeeery heavy abuse 15:51:48 ehird, why n and m though... 15:52:02 AnMaster: nails scratching keys while moving around from the home row? 15:52:10 hm... 15:52:11 i'll wager c and v look similar 15:52:31 AnMaster: also, C-n 15:52:40 maybe he hits C-m instead of enter, less movement 15:52:48 left and right hand movements aren't necessarily the same for keyboards, especially as they're not at all symmetric 15:53:10 hm 15:53:27 for rightie, when moving down, you ..contract your fingers, for leftie, you more like turn your wrist 15:53:33 because of the way the keys are positioned 15:54:12 From this UltraX flat some printings have disappeared; ":" is completely gone, ";" almost, and there's only a "/" left in "A", and the left half of "S". Oh, and "M" has turned into a "V". 15:54:20 hmm, actually probably not.... but still. 15:54:52 ehird, I wonder about C-p... 15:54:54 my laptop's is in perfect shape. how embarrassing, this is an old computer. 15:54:57 well rather his p key 15:55:00 See pee. 15:55:11 hah 15:55:16 haha humor 15:55:29 have you heard the sex pee joke yet? 15:55:33 (:P is rather more worrying; peeing directly out of your colon.) 15:55:33 ehird, if he use C-n rather than arrow keys. 15:55:38 oklofok: oh man is it to do with parentheses 15:55:40 ha 15:56:23 there is a reason I always include the - in S-expressions 15:56:32 ehird: i don't remember, it's a bit complicated. 15:56:48 AnMaster: was it some advice you got off expertsexchange? 15:57:10 ehird, never heard of that place. So no. Was just to reduce confusion. 15:57:16 serious funniness overload. 15:57:21 AnMaster: experts exchange / expert sex change 15:57:23 it was a joke, foo. 15:57:34 (yes, there is an expertsexchange.com but they changed to experts-exchange) 15:57:38 (see? its' related to your joke) 15:57:41 (*it's) 15:57:41 ehird, I never heard of "experts exchange" either. 15:57:42 (ha ha) 15:58:01 ehird: in fact, i never had a sex change. 15:58:13 oklofok: as exchange? 15:58:13 ehird, so thus your reference wasn't clear to me. 15:58:14 ass exchange? 15:58:17 you never exchanged your ass? 15:58:19 i see 15:58:50 ehird, well maybe he liked the donkey. Why replace it then. 15:59:03 true. 15:59:07 ehird: no one will exchange with me :< 16:00:43 I can't come up with any misinterpretation to that one that doesn't make the remaining part include a non-existent word. 16:00:58 just assume it's a typo and fix it for them 16:01:14 will lex change. 16:01:15 hm 16:01:26 why should it. I mean it seems to work now. 16:01:52 I guess you want to parse some non-LR(1) grammar you might need to change it. 16:02:02 right is lex a lexing tool? 16:02:15 ah right. I mixed them up too,. 16:02:17 too* 16:02:18 meh 16:02:31 is that a yes? :D 16:03:08 yes, lex takes an input description file and generates a stream of tokens. Which can then be parsed by a parser for example. 16:03:19 no 16:03:25 it takes an input description file and outputs a c file 16:03:33 which takes bytes and outputs a stream of tokens 16:03:39 wasn't that what I wrote 16:03:44 no 16:03:47 damn you are right, it wasn't 16:03:55 it was what I *thought* though 16:04:04 meh 16:04:05 there's a level of indirection in addition to yours, so to speak 16:04:12 o 16:04:12 o 16:04:12 o 16:04:14 yes 16:04:29 HOW MANY O'S WOULD AN O O O IF AN O O O O O O O O 16:04:32 O 16:04:32 O 16:04:32 O 16:04:32 O 16:04:33 O 16:04:41 oklofok, what do you mean. 16:04:57 AnMaster: any excuse to flood o's 16:04:57 O O O O O 16:05:04 I see. 16:05:18 actually i have an exam on wednesday, so i should probably start reading stuff 16:05:25 oklofok, exactly 5 and 1/3 16:05:30 hmm 16:05:34 ah 16:05:37 it wasn't a question 16:05:43 oklofok, assuming normal room temperature of course. 16:05:46 i misread your intentions 16:05:55 (15-30 degrees) 16:06:56 SO HAVE YOU HEARD THIS BLONDE WALKS IN BAR AND IS STUPID AND EVERYONE LAUGHS 16:07:31 god i'm tired 16:07:32 okay 16:07:34 umm 16:07:37 now seriously 16:07:44 -~> 16:07:49 oklofok, there is also some inconclusive evidence that extreme humidity during extended periods can affect the result. 16:07:57 -!- ais523 has joined. 16:08:01 hi ais523 16:08:50 hi 16:26:00 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)). 16:30:46 -!- Sgeo has joined. 16:49:30 -!- Hiato has joined. 16:55:52 -!- Judofyr has quit (Remote closed the connection). 17:06:12 -!- Judofyr has joined. 17:16:36 * ais523 submits Enigma scores for April 17:26:37 who's ap 17:26:39 ril 17:27:42 err, a month 17:28:01 you never know 17:29:02 * Sgeo wants Twitternomic 17:29:20 * Sgeo realizes that he has no clue how that could work 17:30:05 Sgeo: you have to publish the entire ruleset every week 17:30:12 therefore, it's very short] 17:31:00 is that specified for nomics? 17:31:32 was just thinking that probably wouldn't work on other planets, then realized i don't really know what a week is, arbitrary? 17:31:43 Agora specifies weeks in terms of UTC 17:31:51 B also specifies weeks, as being at least 12 days long, possibly longer 17:31:58 ah 17:32:11 so not earth week necessarily 17:32:22 but what's the definition of week, is it historical or what? 17:32:40 hmm seven days i guess that's biblish. 17:32:50 we need zero length twitter. 17:33:01 http://femto.picoup.com/ 17:33:04 The message would be in the timing of the messages. 17:33:06 one letter's good enough for anyone 17:33:11 ehird, zero. 17:33:16 and timing for encoding information 17:33:21 AnMaster: w/ zero length you still have @references 17:33:27 so you'd just register a bunch of dummy accounts 17:33:43 I'm not familiar enough with twitter to know what @references is 17:33:49 ehird, according to Twitter's counter, the @thing still counts in characters 17:33:57 AnMaster, @nick refers to the user nick 17:34:01 Sgeo: picoup is the international standard for small scale twitters 17:34:02 ah 17:34:08 and it lets you use a reference unlimitedly 17:34:09 If it's the first thing in a tweet, it's a reply. If not, it's a mention 17:37:47 -!- tombom has quit ("Peace and Protection 4.22.2"). 17:46:02 seems like wikipedia is having issues today 17:46:13 yep 17:46:17 lots of timeout, missing css, missing images. 17:46:33 ais523, #wikipedia didn't seem to know about it when I asked. 17:46:35 strange 17:49:09 rewrite fingerprints/rcfunge98/trds.d (75%) 17:49:12 Well, that was fun. 17:49:16 :D 17:49:24 4 files changed, 255 insertions(+), 267 deletions(-) 17:49:39 Deewiant, rewrite meaning 17:49:46 how did you measure it 17:49:52 That's what git outputs 17:49:59 $ git commit -va 17:49:59 [fingerprints-back d13d9ba] TRDS. Forced fixes in FungeMachine and IP. 17:49:59 4 files changed, 255 insertions(+), 267 deletions(-) 17:49:59 rewrite fingerprints/rcfunge98/trds.d (75%) 17:50:06 * ais523 was wondering if rewrite was like patch, but generated patches using Markov chaining 17:50:14 Deewiant, yes but what does it mean. number of lines that are different or what 17:50:35 I think it just says so if a certain percentage of the lines change but I'm not sure 17:51:01 FungeMachine... 17:51:03 Possibly also the contents, not the lines 17:51:11 Deewiant, did you fix those core bugs yet 17:51:28 AnMaster: Yes, the revision you pulled whenever should have them all fixed? 17:51:48 Deewiant, there were some other in some of the test cases. Lockups and such as I mentioned. 17:51:52 I ran through all of cfunge's tests this morning and none failed 17:51:56 ok 17:52:00 AnMaster: Yes, I know, but I fixed them all before you last pulled. 17:52:04 ah 17:52:08 I might have only mentioned the one. 17:52:21 Deewiant, the crash on wrong line number was the only one I remember you fixing 17:52:28 err 17:52:31 right line wrong file 17:52:47 LDC fixed that error message bug too :-) 18:04:02 Deewiant: I may have an LDC/D1/Tango/rebuild environment up in a sec. 18:04:05 So I can try ccbi2. 18:04:14 Yes, I've been following your troubles. :-) 18:04:47 Yeah. 18:04:56 But Deewiant #ldc is not Deewiant #esoteric. 18:04:59 The latter is blissfully unaware. 18:05:33 Quite. 18:05:55 Hmm, methinks CCBI2 is now at the same level of functionality as CCBI1, minus 3DSP which is blocked on an LDC bug. 18:06:05 You should port ccbi to a language that doesn't make me suicidal. 18:06:12 That would be quite nice. 18:06:22 For you, perhaps. 18:06:39 CCBI is a bit too big for me to bother porting it. 18:06:51 [ehird:~/Junk] % rebuild hello.d -ofhello 18:06:51 [ehird:~/Junk] % ./hello 18:06:52 Hello world 18:06:57 FUUUUUUUUUUUCK YEAAAAAAAAAAAAAAH 18:07:03 A winner is you! 18:07:27 * Sgeo prefers "You're winner!" 18:07:29 Deewiant: git uri? 18:08:15 Wait, I have a bug to fix. 18:08:33 Don't we all 18:12:25 Well, if you still have that old repository a 'git pull' might work now. 18:12:49 Nope. I don't. 18:12:56 git clone git://tar.us.to/ ccbi2? 18:13:14 Yeah, just the root path. 18:13:56 Deewiant: rebuild ccbi.rf? 18:14:06 rebuild -rfccbi.rf 18:14:10 -L-lncurses 18:14:20 ooh, they've invented 500GB optical disks 18:14:31 ais523: ooh, they've invented 1TB solid state drives 18:14:33 :-P 18:14:41 Deewiant: ccbi.d(18): module flags cannot read file 'ccbi/flags.d' 18:14:43 What what in the butt butt 18:14:49 yes, but optical > solid state in the size impressiveness stakes, I think 18:15:03 ais523: But solid state > optical in the useful stakes 18:15:09 well, agreed 18:15:10 Imagine the spinning 18:15:11 ehird: As AnMaster has berated me for many times, the directory has to be called ccbi. 18:15:16 I suspect optical > solid state in the cheap stakes, though 18:15:31 Deewiant: Wrap it all in one directory, have ccbi.rf outside of it doing -Iccbi 18:15:37 Sroblem polved 18:15:39 ehird: Yes, I know, I just haven't done it. 18:15:42 ais523: True, $3k isn't a nice price. 18:15:50 ehird: * soblem prolved? 18:15:50 Deewiant: Are patches accepted? 18:15:54 ais523: oh, yes. 18:16:02 ehird, do not try to do ln -s . ccbi. It won't work since the produced binary is called ccbi 18:16:07 ehird: If you insist. :-P 18:16:25 ehird, it was one of the first things I tried. 18:16:45 idea: express major version as the directory nest 18:16:48 You /can/ tell rebuild what the binary should be called. 18:17:00 like: foo 1.0 is foo/*.c foo 2.0 is foo/foo/*.c 18:17:02 and so on 18:17:09 foo 0.x is a tarbomb 18:17:09 And 1.1? 18:17:17 Deewiant, not sure about that yet 18:17:31 The number of dots before the extension 18:17:38 Or repeat the extension that many times 18:17:38 good idea! 18:17:40 AnMaster: I'd love there to be a foo -1.x under that naming system 18:17:49 actually, any excuse for an antitarbomb would be great 18:18:00 I mean, an antiantitarbomb-tarbomb 18:18:08 as in, one that untars in .. 18:18:18 ais523, iirc tar restricts that? 18:18:19 I bet the tar standard doesn't support that, but I so hope it des 18:18:20 They should get the holography-based discs to the market (infallopedia says they have 250GB discs now with a 3.9TB maximum); simply because, you know... holographic discs! It's *so* the future. I'm sure they've used some sort of holocubes in at least one scifi series. 18:18:21 *does 18:18:23 Deewiant: How do you want the patch? 18:18:47 ais523, I remember some security report about that. 18:18:53 CVE or similiar 18:18:55 ehird: Whatever 18:19:09 fizzie: in "3001" by Arthur C. Clarke, standard removable data storage that people carried around was in the petabyte to exabyte rate 18:19:10 Deewiant: What I'm trying to say is I forgot how to get a mail friendly version of a commit. 18:19:25 which strikes me as surprisingly low given Moore's Law 18:19:26 ehird: I don't know, I develop alone. 18:19:33 ehird: darcs send 18:19:37 ais523: Git. 18:19:47 I know, I was being facetious 18:19:59 ehird: git show? 18:20:07 I'm mostly at peace on the whole emacs/vi thing, can you give me at least one holy war to fight? 18:20:12 ehird: git show --pretty=email 18:20:13 Well, Moore's law is more like a guideline here... but still, it does sound pretty small for 3001. 18:20:22 Deewiant: Does git have a command to apply that? 18:20:38 ehird: It has git am, but of course that doesn't work with Thunderbird. 18:21:16 Deewiant: "git show" on the rename shows a diff for every single fucking line in ccbi 18:21:19 Which is why I might just grab the diff-part and apply it under my name. :-P 18:21:33 Deewiant: How about I just give you instructions? 18:21:38 :-D 18:21:54 Tell you what, I'll do it myself 18:22:00 Deewiant: mkdir ccbi; "ls|sed 's/ccbi.rf//;s/ccbi$//'|xargs echo", git mv THOSE ccbi 18:22:02 an d then 18:22:10 -ofobj/ccbi 18:22:12 in ccbi.rf 18:22:18 WALLA 18:22:24 I think you meant ccbi/obj 18:22:29 But I'll just move obj to toplevel 18:22:35 It already is 18:22:38 Deewiant: Don't move ccbi.rf 18:22:39 Duh 18:22:42 Keep it in the root 18:22:47 It's not part of the ccbi. hierarchy 18:22:55 Yes, I know? 18:23:00 But I will move obj to root 18:23:02 Kay 18:23:08 It already is, I think, but kay 18:23:29 Well, I didn't follow your instructions 18:23:37 I just did mv * and then moved back the rest 18:23:50 Deewiant: git mv != mv 18:24:03 ehird: git mv = mv + git add 18:24:08 Hmm 18:24:14 But in the commit it shows "renamed ff -> gg" 18:24:17 git doesn't track moves iirc 18:24:18 or does it 18:24:19 Oh, heuristics 18:24:19 Right 18:24:34 It's a testament to how good they are that I didn't notic 18:24:35 e 18:24:50 ehird, I kind of prefer that it tracks moves, since I often move and rewrite large parts of the file in the same commit. 18:24:59 Now should I call the binary ccbi2 or the directory src/ccbi 18:25:08 AnMaster: Then it's not much of a rename 18:25:13 Deewiant: Directory: src/ccbi 18:25:14 Deewiant, why can't you just call it src 18:25:19 Binary: obj/ccbi 18:25:25 AnMaster: Module naming. 18:25:36 src/*.d src/fingerprints/{rcs,catseye,...}/*.d 18:25:38 ehird, hm ok 18:25:41 like java then 18:25:42 or? 18:25:49 Like Java, Python, ... 18:26:28 Pyva 18:26:47 zen trolling. 18:27:37 lamzent 18:27:57 Well, that's that, now AnMaster doesn't have to hate me for that any more. 18:28:35 Deewiant: Put the uri up plz 18:28:54 ehird, I just need to git pull I assume? 18:29:05 it's down. 18:29:05 ehird: Done 18:29:09 now it's not 18:29:39 Deewiant: rebuild -rfccbi.rf -L-lncurses 18:29:41 here goes 18:30:09 Remember to link with an ncurses library. 18:30:09 Assuming 32-bit chtype... 18:30:11 Remember to link with an ncurses library. 18:30:13 Assuming 32-bit chtype... 18:30:15 Remember to link with an ncurses library. 18:30:17 Assuming 32-bit chtype... 18:30:19 [more] 18:30:20 6 times each 18:30:21 Thank you, CCBI. I got it the first time. 18:30:26 % bin/ccbi --help 18:30:27 zsh: bus error bin/ccbi --help 18:30:29 AWESOME. 18:30:32 ehird: Blame the DMD frontend, not me. 18:30:41 Deewiant: Er, so what'mI meant to do 18:30:47 Deewiant: I got CCBI running under gdc but not dmd, IIRC 18:30:49 Or actually, blame rebuild, I guess. 18:30:56 ais523: this is ccbi2 18:30:59 you're meant to use LDC 18:31:04 which uses the dmd frontend, but LLVM as a backend 18:31:06 ais523, how do you tell apt-get or aptitude to *reinstall* a package 18:31:09 ais523: It can't have worked well with GDC, some things would have segfaulted it. 18:31:13 AnMaster: Uninstall and reinstall. 18:31:16 ais523, somehow some of it's files are not there. 18:31:16 Deewiant: So, why does --help segfault? 18:31:24 ehird, won't that delete the config files for it. 18:31:26 ehird: Beats me, can you backtrace? 18:31:29 AnMaster: No. 18:31:30 Only purging does that. 18:31:33 Deewiant: Can I? 18:31:36 ehird: Pass -gc to rebuild if you can't 18:31:46 And -full to make sure it actually builds 18:31:50 is there no plain reinstall!? 18:31:53 no 18:31:57 silly 18:31:58 AnMaster: apt-get install --reinstall 18:32:02 yes 18:32:02 :-P 18:32:02 ehird is lying to you 18:32:02 ais523, ah 18:32:07 ais523: not intentionally! 18:32:11 it's not lying if you didn't know 18:32:13 ais523, none in aptitude though 18:32:18 aptitude=apt-get 18:32:19 well, I didn't know either, I was reading the man page 18:32:20 for the mostpart 18:32:24 and I don't use aptitude 18:32:56 ok why is it still missing... 18:33:24 There *is* a "reinstall" command in aptitude. 18:33:38 fizzie, Which menu? 18:34:00 Just "aptitude reinstall package"; I'm not sure where it is in the text-ui. 18:34:03 anyway it didn't add the missing file back. So how do I list the package to make sure I didn't imagine it being there in the first place. 18:34:08 apt-cache something right? 18:34:14 "L": Request that a package be reinstalled. 18:34:19 Okay, you can use uppercase-L in the UI. 18:34:40 And you can use "dpkg-query -L package" to list the files contained in one. 18:35:09 ok.... for some reason reinstalling lighttpd didn't add back the missing /etc/init.d/lighttpd 18:35:50 OMG, Multi-Threading is 18:35:50 Easier Than Networking 18:35:53 — Title of Intel paper 18:35:56 http://software.intel.com/file/14723 18:36:02 so what could be wrong 18:36:17 * ais523 isn't sure offhand which of multithreading and networking inherently feels harder 18:36:23 dpkg-query -L lighttpd 18:36:23 lists /etc/init.d/lighttpd 18:36:42 the file doesn't exist though. And reinstalling it doesn't add it back. 18:36:52 what does one do on debian in that case. 18:36:53 -!- pikhq has joined. 18:37:13 Er. Deewiant? 18:37:19 "ccbi" prints the same as "ccbi --help", right? 18:37:20 ? 18:37:22 Yep 18:37:23 Because just ccbi works. 18:37:24 Well, you can "aptitude purge lighttpd" followed by a separate "aptitude install lighttpd", to make sure it is as if you were installing lighttpd for the first time; that will remove the related config files too, though. 18:37:25 Okay 18:37:33 Deewiant: Does this make any fucking sense to you? 18:37:33 Then 18:37:39 #0 0x0006febe in _d_invariant (o=0x1300000) at /Users/ehird/d/ldc/runtime/internal/invariant.d:16 18:37:39 #1 0x0003a9c4 in _D5tango2io6stream6Format20__T12FormatOutputTaZ12FormatOutput8formatlnMFAaYC5tango2io6stream6Format20__T12FormatOutputTaZ12FormatOutput () 18:37:40 Sure, why not 18:37:42 It's the stream formatting that's broken 18:37:46 Why would it only break with a command line argument 18:37:46 fizzie, I don't want it to remove them. Just add the missing init.d script! 18:38:09 Well, something else could easily break it 18:38:30 ehird: What if you pass any other argument 18:38:52 I really think reinstall should add missing non-configurationary files. But if you don't mind, you can always just unpack the deb and put it there manually. Incidentally, do you happen to have the links to the initscript still? 18:38:59 % bin/ccbi fuck 18:39:00 zsh: bus error bin/ccbi fuck 18:39:02 LOL WAT 18:39:08 Okay 18:39:14 So it's probably due to regex 18:39:25 Deewiant: #1 0x0003a9c4 in _D5tango2io6stream6Format20__T12FormatOutputTaZ12FormatOutput8formatlnMFAaYC5tango2io6stream6Format20__T12FormatOutputTaZ12FormatOutput () 18:39:27 Go to line 350 in ccbi.d, turn the test to true || 18:39:28 or something 18:39:29 fizzie, yes, something complaining about those broken symlinks was how I noticed it. 18:39:29 That's not to do with regex? 18:39:29 -!- jix_ has quit (Read error: 60 (Operation timed out)). 18:39:49 ehird: Doesn't mean that regex can't break something that causes everything else to break 18:40:02 Where is this test, o mighty one? 18:40:03 Oh 18:40:04 line 350 18:40:10 :-P 18:40:45 fizzie, any command to make it check integrity of all installed files. 18:40:51 To make sure no other ones are missing. 18:41:28 Deewiant: regex thing changes nothing 18:41:35 ehird: Okay 18:41:44 Reinstall indeed doesn't seem to add init.d files back. I guess it really considers that a config file, then, or something. Can't really say I've ran across a related problem before. 18:42:28 ehird: Reduce it to a single small file, if you can 18:42:37 I don't know D 18:42:38 No can do 18:43:02 ehird: Remove from line 359 to the last } 18:43:30 Try that, in the meanwhile I'll make you something to test 18:44:01 -!- jix has joined. 18:48:27 ehird, I wonder what response I would have got from you if I had asked that for cfunge (I'm pretty sure you know C) 18:48:38 I'd have done it. 18:48:47 I'm surprised. 18:49:43 ehird, by the way, what large esolang related projects have you done recently? 18:49:58 Nothing; why? 18:50:17 no particular reason. 18:50:21 When did you last do one 18:50:32 I don't go for the large esolangs, so umm 18:50:34 None? 18:50:46 ah. What about your last small esolang interpreter/compiler. 18:50:50 in that case 18:50:51 Hey, I'm actually doing some homework! 18:51:01 Sgeo, yes and? 18:51:08 AnMaster, it's a rarity 18:51:14 ah ok 18:51:19 that explains PSOX. 18:51:28 ..what? 18:51:37 AnMaster: Oh, I foundeded it: dpkg -i --force-confmiss /var/cache/apt/archives/lighttpd.deb 18:51:45 AnMaster: My unlambda in haskell? 18:51:58 AnMaster: That will add missing configuration files, but should not touch existing ones. 18:52:00 fizzie, thanks. great! 18:52:02 I don't get it 18:52:14 ehird, ah right. when was it. 18:52:25 A few months ago. 18:52:28 rigt 18:52:30 right* 18:52:42 http://lkml.org/lkml/2009/4/22/78 LOL WAT 18:53:28 It should say something like "Configuration file `/etc/init.d/mt-daapd', does not exist on system. Installing new config file as you request." 18:53:43 ehird, insane 18:54:40 interestingly enough sparse fails to parse some cfunge files. It handles C99 pretty well. But not the foo(double bar[restrict 16]) syntax. 18:59:24 AnMaster: http://ivoras.sharanet.org/freebsd/freebsd8.html freebsd replacing gcc w/ clang 19:00:06 ehird, yes sure. Only reason I consider the first link insane is that using clang or something else would be a way smarter move. 19:00:09 My Internet connection hates me. Be back in a bit. 19:00:15 -!- pikhq has quit ("leaving"). 19:00:33 I know 19:00:34 I was just saying 19:00:40 Prediction YACC won't mean Yet Another Compiler Compiler. Instead it will mean: Yet Another C Compiler (oh no!) 19:00:50 that is in 10 years time. 19:00:50 C's dead. 19:01:02 ehird, If so it is undead. 19:01:19 I wish it'd die again. 19:01:33 ehird, it is like trolls in nethack I'm afraid. 19:01:50 ehird, plus, what language would you suggest, other than C, for the linux kernel. 19:01:53 -!- pikhq has joined. 19:02:11 Lisp. Although that would not lend itself to implementing a Unix. That's a feature. 19:03:58 ehird, Yes. But do you think that will actually happen. Or are you just wishing. 19:04:34 Probably. Maybe. One day. If we get intelligence-enhancing brain transplants my preferred UI would be Lisp. 19:04:42 That's far off of course. Hopefully lisp machines will happen again sooner. 19:04:54 There are many things less likely. 19:04:57 my preferred ui would be tits 19:05:04 how about a tits-based programming language. 19:05:12 nested tits? hell yeah. 19:05:49 .... 19:06:18 (.)list 1 2 (.)+ 2 2(.)(.) 19:06:24 Tlipsts. 19:07:51 -!- Gracenotes has joined. 19:08:47 ehird, what is that 19:09:01 lisp + () messed up 19:09:02 it seems 19:09:06 Nested tits programming language, pioneered by lament. 19:09:15 ah... 19:14:47 ehird: So if you wanna pull I removed usage of regex 19:15:01 I'll try 19:15:43 Yay it is the work ing 19:16:05 Deewiant: Is optimization enabled by default? 19:16:15 Coincidentally, removing that module decreased the size of the binary by 140K (the optimized, not stripped and compressed version) 19:16:18 ehird: Nope 19:16:39 How do I? 19:16:50 (86K for the stripped one and 26K for the UPX'd one) 19:17:03 ehird: -O5 -release 19:17:19 -O5?!?!! 19:17:19 XD 19:17:25 LLVM goes up to -O7 19:17:32 But >5 breaks? 19:17:33 But I don't think the last two are implemented 19:17:39 They're link-time optimization 19:17:48 ldc: Unknown command line argument '-O7'. Try: '/Users/ehird/d/ldc/bin/ldc --help' 19:17:48 ldc: Unknown command line argument '-O7'. Try: '/Users/ehird/d/ldc/bin/ldc --help' 19:17:49 ldc: Unknown command line argument '-O7'. Try: '/Users/ehird/d/ldc/bin/ldc --help' 19:17:51 And I don't think anything LLVM does it. 19:17:51 ldc: Unknown command line argument '-O7'. Try: '/Users/ehird/d/ldc/bin/ldc --help' 19:18:03 LDC doesn't pass them through 19:18:21 Whether something upstream would accept it and do something with it, I don't know. 19:19:01 Deewiant: -O5 gives me an llvm assertion error 19:19:02 awesome 19:19:06 Assertion failed: (Op.isUse() && (Op.isKill() || getFPReg(Op) == FirstFPRegOp || MI->killsRegister(Op.getReg())) && "Ret only defs operands, and values aren't live beyond it"), function handleSpecialFP, file X86FloatingPoint.cpp, line 1072. 19:19:07 0 ldc 0x00ae5a02 llvm::sys::RemoveFileOnSignal(llvm::sys::Path const&, std::string*) + 866 19:19:13 It still runs though 19:19:18 It could be an LLVM bug or an LDC bug 19:19:31 Say, does ccbi work with 64 bit ints 19:19:44 It doesn't use them for the funge-space cells, no. 19:19:55 ehird: The reason it still runs is that it didn't link it again. :-P 19:19:57 Right, so compiling 64bitly is wasteful 19:19:59 You're running the old binary. 19:20:01 Deewiant: I rm -rf'd bin so let's try again 19:20:02 And no 19:20:04 it gave output after that 19:20:08 the compilation 19:20:15 ehird: That's because rebuild continues. 19:20:20 Even if a single file fails. 19:20:20 Ah. 19:20:22 Right you are. 19:20:24 LEt's try -O4 19:20:36 64-bit isn't wasteful: there are some longs/ulongs in there (64-bit ints) 19:20:47 -!- Gracenotes has quit (Remote closed the connection). 19:20:48 Same error woo hoo 19:20:49 Well, it may still be wasteful in the end 19:21:01 But it's not completely useless. 19:21:45 -m64 -march=x86-64 19:21:51 rebuild -rfccbi.rf -full -L-lncurses -O3 -m64 -march=x86-64 -release 19:21:56 ^CError: -m32 and -m64 switches cannot be used together with -march and -mtriple switches 19:21:58 Lal wat 19:23:02 Deewiant: Trying -O2 19:23:17 same error 19:23:18 tee hee 19:23:18 ehird: It's not so slow that you /have/ to optimize it >_< 19:23:37 -funroll-unsafe-mathematical-integral-loops-pentium-4-only 19:23:42 Hey -O1 works. 19:23:49 No it doesn't. 19:23:53 It gives a linker error. 19:23:53 :-P 19:24:03 A FUCKING HUGE linker error 19:24:47 ehird: -march=x86-64 implies -m64 19:24:53 Deewiant: 1966 or so lines of integer errors: 19:24:54 Your paste cannot be larger than 100 kb. Sorry. 19:25:43 ehird: Is it -release? 19:26:23 http://pastebin.com/f73d5f685 19:26:24 And yes 19:26:37 I meant, is that the flag causing the problems. 19:27:05 Dunno 19:27:11 Find out! 19:27:13 :) 19:27:21 Well er 19:27:23 ehird: fail 19:27:29 ehird: ld warning: in /Users/ehird/d/ldc/lib/libtango-base-ldc.a, file is not of required architecture 19:27:29 Wat 19:27:34 LAWL 19:27:38 So no shit it can't find the symbols 19:28:13 % rebuild -rfccbi.rf -full -L-lncurses -O1 -release 19:28:14 WORK 19:28:26 Now, where's dat mycogy 19:39:45 Deewiant: Is sanity.bf meant to just sit there? 19:41:21 ehird, no. 19:41:30 I'm saying that ccbi2 does that. 19:41:38 So does mycology.b98. 19:41:40 $ build/cfunge mycology/sanity.bf 19:41:40 0 1 2 3 4 5 6 7 8 9 19:41:46 I know. 19:41:47 well it doesn't print a newline at the end 19:41:53 Hey Deewiant, also, 19:41:54 Copyright (c) 2006-2009 Matti Niemenmaa, http://www.iki.fi/matti.niemenmaa/ 19:41:55 See the file license.txt for copyright details. 19:41:57 Usage: --help ARGS SOURCE_FILE [BEFUNGE_ARGS...] 19:42:04 :-D 19:43:23 So, I've gone batshit insane and am going to be the implementing of the befungey. 19:43:23 Again. 19:43:38 write it in ITRALCEN! 19:43:43 ais523: AAAAAAAAAAAAAAAAAAAA 19:43:50 in fact, you could have both ITRALCEN and Bfngue 19:43:54 ehird, how does ITRALCEN differ from INTERCAL 19:44:03 AnMaster: ITRALCEN's a vaporware INTERCAL interp 19:44:07 AnMaster: It's more reflective, and orthogonal, and it has a weird compiler system like CLC. 19:44:08 right 19:44:12 or possibly compiler 19:44:12 ais523, but what makes it special. 19:44:16 or knowing INTERCAL, a bit of both 19:44:20 AnMaster: Simultaneous operand overloading, fwiw. 19:44:22 AnMaster: all INTERCAL implementations are special 19:44:23 That's one thing. 19:44:27 If you operand overload one thing to two things, 19:44:33 whenever you use the overloaded, the program threads 19:44:35 for both meanings 19:44:35 ais523, yes. But in what way is this one special. 19:44:41 ehird, ah 19:44:42 this is used to implement COME FROM, with ‽ vars 19:44:43 interesting 19:44:45 ‽N is the line N 19:44:49 it contains an ITRALCEN-string 19:44:52 but you can operand overload it, too 19:44:58 cool! 19:45:01 this, of course, is hell for performance or any sane sort of compiling 19:45:06 but ITRALCEN compiles ITRALCEN to ITRALCEN 19:45:11 um 19:45:12 ok 19:45:14 specifically, it compiles the intercal code to the itralcen primitives 19:45:21 which are designed just for intercal 19:45:26 somewhat like CLC then 19:45:33 an ITRALCEN befunge implementation would compile befunge to ITRALCEN on the fly 19:45:37 in ITRALCEN 19:45:46 ehird, "compile on the fly" == JIT 19:45:48 Yo dawg? 19:46:00 AnMaster: nope 19:46:08 ehird, oh? 19:46:14 more like self-rewriting source code that's continually hammered into a set of primitives by the ITRALCEN compiler 19:46:21 and interpreted 19:46:22 interesting 19:46:26 on the fly 19:47:15 it's quite similar to CLC, but a bit more brain damaged due to the ‽ variables and the really odd set of primitives 19:47:28 I mean, operand overloading used for branching? Operand overloading more than once for threading? 19:48:49 ehird, I still don't see how "on the fly" differs from "just in time" 19:49:12 and, how do you use operator overloading for threading 19:49:15 err 19:49:17 branching I mean 19:49:37 you told me about threading already 19:50:24 AnMaster: ‽N is line number N; the contents is its source code, and if you operand overload it, it switches that line for another one (obviously) 19:50:39 oh ok 19:50:41 if you do it twice, then as the interpreter dereferences it to jump, the multi-operand-overloading code kicks in, and threads it 19:56:20 Is the name mycelium taken by any funge thing, Deewiant? AnMaster? 19:56:35 Nope 19:56:46 Now it is. 19:56:55 it's quite close to mycology, but not quite close enough to be confusing 19:56:57 if you finish it. 19:57:07 which I doubt 19:57:16 AnMaster: No, if I have something that can execute a good portion of funge I'd say it's takn 19:57:17 *taken 19:57:19 ais523: That is true. 19:57:26 I thought it might be the name of a mycology file 19:57:38 They're all myco* 19:57:50 Deewiant, mycosanity.bf 19:57:53 riiiight 19:58:09 Yeah, and mycolicense.txt 19:58:20 what licence is mycology? 19:58:23 I've forgotten 19:58:30 BSD3 IIRC 19:58:37 Like most of my stuff 19:59:15 I had plans to use the name "mycena" (which is a genus of mushrooms, many of the glow-in-the-dark bioluminescent variety) for some project or other, but I don't think it ever happened. 19:59:49 Deewiant: (# !Int, !Int #) 19:59:51 COWER IN FEAR 20:00:13 Not much :-P 20:00:15 No wait 20:00:19 (# Int#, Int# #) 20:00:22 AAAARGH! 20:00:26 Have fun with that 20:00:34 is that an unboxed pair of unboxed ints? 20:00:37 Yep 20:00:37 ais523: yep! 20:00:44 And a pain to work with 20:00:50 = a 128-bit integer, effectively 20:01:22 ais523: 64-bit 20:01:25 on 32 bit 20:01:28 Platform-specific 20:01:40 Meh 20:01:43 #haskell says to use data IntPair = IntPair {-# UNPACK #-} !Int {-# UNPACK #-} !Iint 20:01:46 To avoid me going insane 20:02:05 I thought you wanted to use +# everywhere :-P 20:02:23 Deewiant: Well, yeah, exactly, I don't want to :P 20:02:34 So avoid # :-P 20:02:38 is that an unboxed pair of unboxed ints? 20:02:38 um 20:02:41 in what language 20:02:45 Haskell, obviously 20:02:49 ah 20:03:01 to be precise, Haskell with GHC extensions 20:03:02 data Point = Point { x :: {-# UNPACK #-} !Int 20:03:02 , y :: {-# UNPACK #-} !Int } 20:03:05 fuck yeah 20:03:19 No Trefunge? 20:03:30 Deewiant: Not for now. 20:03:38 ehird, no n-funge? 20:03:41 Not for now. 20:03:46 ok 20:06:00 *Main> point (2,3) 20:06:00 point (2,3) 20:06:07 :-P 20:06:13 You're making progress! 20:06:25 Yeah 20:06:26 good point 20:06:26 point !(!x,!y) = Point { x = x, y = y } 20:06:28 AWESOME PROGRESS 20:06:37 Strict much? 20:06:43 Totally man. 20:06:50 You don't need the outer ! 20:06:57 But it feels right. 20:07:02 :-P 20:07:08 Do I even need any bang patterns? The fields are strict. 20:07:17 No, you don't. 20:07:23 The outer you wouldn't need in any case. 20:07:38 Since (x,y) is a pattern that has to be matched against. 20:07:44 I wish I could read ghc asm 20:08:05 I wanna see what my point compiled to 20:08:15 i can read it with my eyes closed 20:08:20 It compiled to a _closure and _somethingelse 20:08:32 And I can never find the actuaal code 20:08:34 -a 20:09:04 Deewiant: Actually, maybe having point is bad 20:09:07 Since it might result in a call 20:09:13 As opposed to just exposing Point 20:09:29 you have a point 20:09:38 good point 20:09:48 20:09:49 [20:09] Deewiant: Actually, maybe having point is bad 20:09:51 [20:09] Since it might result in a call 20:09:52 [20:09] As opposed to just exposing Point 20:09:54 ^ I misinterpreted that as being about sport 20:09:57 XD 20:09:58 due to not looking at the context 20:10:06 ehird: GHC isn't that crap at inlining 20:10:23 Deewiant: I can't read its output, therefore it scares me. 20:10:34 Just use C. 20:10:37 :-) 20:10:41 "In January our profits went up or down, depending on which chart is right, we're not sure. Now February was a very different month, because as you recall we took it off" 20:11:25 -!- Gracenotes has joined. 20:12:32 in soviet russia, the area jans you 20:12:54 20:12 ehird: Possibly the most bloated, extensible (it has modules...) esoteric language on the planet 20:12:54 20:12 Berengal: ehird: There's Java... 20:13:19 Java's not an esolang 20:13:39 it suffers from excessive trying to be sane in some areas, though, like interfaces, exceptions, and function pointers 20:13:41 Maybe I'll call my funge RicerFunge. 20:13:47 not to mention templates, but they almost fixed that 20:13:56 stop mocking java, i need to read about it all night. 20:14:04 -!- tombom has joined. 20:14:07 wait 20:14:09 you didn't mock it 20:14:14 you were loving it 20:14:19 sorry. 20:14:34 20:14 wli: Am I the only one who finds th "esoteric languages" unworthy of any attention/notice/etc. whatsoever? 20:15:06 ^ I misinterpreted that as being about sport <-- which sport... 20:15:06 ehird: i somewhat doubt that :D 20:15:14 20:14 ehird: wli: PROGRAMMING IS SERIOUS BUSINESS 20:15:14 20:14 ehird: Fun is strictly forbidden, lest we write unmaintainable code! 20:15:15 20:15 ehird: Most of all, theoretical models are NEVER helpful to discover new concepts. 20:15:16 AnMaster: it doesn't really matter, it could have been anything 20:15:26 ah 20:15:29 I misinterpreted the first line of ais523's misinterpreted block of text in the sense "maybe having a point is bad"; i.e. pointlessness being a good thing. 20:15:35 ehird: why are you quoting yourself? 20:15:36 ais523, I'm not much into this "sport" thing. 20:15:37 ;P 20:15:54 lament: Because it's a reply to a thing I previously quoted, #esoteric-related, that's not happening in #esoteric. 20:16:10 ehird, so how far in mycology did it get. 20:16:12 atm 20:16:15 isn't everyone here in #haskell anyway? 20:16:16 lament: you should probably ban him, that's against the rules 20:16:20 i'm not atm 20:16:21 ehird: but nobody but me's in #esoteric-related 20:16:27 I'm not ever on #haskell. 20:16:28 now i am 20:16:31 AnMaster: Why wasn't I a dick to you when you tried efunge? 20:16:37 HEY YOU STARTED IT 5 MINUTES AGO WHY ISN'T IT DONE YET SLACKER 20:16:44 IS IT 'CUZ YOU SUCK 20:16:58 ehird, you were irritating me when I started on cfunge. 20:17:06 just FYI. 20:17:11 but also, it was a joke 20:17:14 surely you saw that. 20:17:21 Nnnnoooo I wasn't, the only thing I complained about was the name at first. 20:17:27 20:17 wli suggests that there are better languages for exercises in compiler/interpreter writing. 20:19:31 people who dislike things are idiots 20:19:36 i hate them 20:19:39 @faq can haskell be an esolang? 20:19:40 The answer is: Yes! Haskell can do that. 20:19:41 um 20:19:47 it answers that every time 20:19:50 no shit 20:19:52 for "@faq can haskell" 20:19:54 I assume 20:19:58 for @faq anything 20:20:04 ok then that is silly 20:20:16 real faqs can be useful 20:20:22 it's a reference 20:20:30 => funny 20:20:33 fungot: @faq CAN BEFUNGE DO THAT? 20:20:34 fizzie: " and when are we to do? and meanwhile, baron, what you think of him!" 20:20:39 :D 20:20:52 that's a better answer than lambdabot's 20:21:07 someone add ^faq to it 20:21:33 ^def faq ul (Yes! Funge can do that.)S 20:21:34 Defined. 20:21:41 ^faq can Funge be an esolang? 20:21:41 Yes! Funge can do that. 20:21:56 ^faq can Funge solve the halting paradox. 20:21:57 Yes! Funge can do that. 20:22:02 should be befunge 20:22:06 flows better 20:22:08 no, Funge in general 20:22:09 AnMaster: halting problem 20:22:12 ais523: yes, but, flow 20:22:18 ^faq can Funge solve the halting problem? 20:22:18 Yes! Funge can do that. 20:22:30 20:21 wli: monochrom: In non-esolang-related affairs, slapping a higher-order module system atop basic FP lang constructs is proving difficult for me. ← that's not esoteric‽ 20:22:38 ehird, no, it shouldn't. Since that would be a paradox. Thus it needs to solve the paradox. 20:22:40 ;P 20:22:44 20:23:16 Hey, someone come up with a good binary coordinate operator. 20:23:31 (x .:. y)? 20:23:33 ehird, um what do you mean 20:23:42 .:. is legal? 20:23:43 AnMaster: a symbol going between x and y to construct a point 20:23:44 . 20:23:45 oklofok: yes 20:23:46 maybe 20:23:50 AnMaster: that's composition 20:23:50 what's the rule again 20:23:55 True or false: If this statement is true, I can win #esoteric by announcement. 20:23:58 ehird, (x . y) looks lispy though 20:23:59 but ok 20:24:07 i thought it was endswith: 20:24:07 AnMaster: I'm not breaking Haskell. 20:24:11 ais523: True. 20:24:15 ais523, false. 20:24:17 I win #esoteric 20:24:18 oklofok: a string of symbols 20:24:27 AnMaster: it can't be, logically 20:24:27 ais523: Ah, AnMaster still denies Curry's paradox. 20:24:29 ais523, why not. 20:24:30 ehird: oh i thought you wanted a constructor 20:24:31 ais523: stop it 20:24:35 I argued with him about this before 20:24:42 he thought the (this=>foo) was an axiom he had to accept 20:24:43 or something 20:24:45 not worth it 20:24:45 AnMaster: because if it's false, that implies that it's true /and/ I can't win #esoteric by announcement 20:25:39 ais523, Um. I don't see anything implying it must be true if it is false in it. 20:25:48 ais523: please, I'm trying to help you here 20:25:57 he doesn't understand logic, I tried to explain curry's paradox for _hours_ 20:26:10 ehird, s/hours/10 minutes/ 20:26:13 at most 20:26:18 no, it was at least 30 minutes 20:26:26 AnMaster: "If A then B" means "A is false or B is true", in logic 20:26:32 I know because I started looking for the nearest window to jump out of 20:26:41 ehird, yes, but your overstatement is equal to my understatement. 20:26:45 so it's negation, "not (If A then B)", means "A is true and B is false" 20:26:54 ais523, hm ok... 20:27:11 ehird: see, that was easy 20:27:15 ais523, seems like a flaw somewhere. 20:27:19 ais523: ... or was it. 20:27:24 what i'm not getting about it is why i'm not able to tell you what the reason for the paradox is. 20:28:49 err, well, it's an infinite proposition, nothing says those need to have a definite value, therefore that doesn't need to be true or false. 20:28:50 ais523, to me it sounds more like a natural language problem than a logical one. And I suspect that the whole premise (sp?) for that logical connection is inconsistent with reality. 20:28:55 possibly. 20:29:03 AnMaster: i explained it, read. 20:29:08 This is exactly the bullshit he spinned last time. oh well. 20:29:59 oklofok, hm... Are you agreeing with me or not. 20:30:16 AnMaster: it's similar to the Epinimedes paradox, "This sentence is false" 20:30:25 ais523, I get that one yes. 20:30:38 AnMaster: curry's paradox is logically true in naive logics (those which allow self-reference) 20:30:43 it's not a matter of agreeing or disagreeing 20:30:55 in practice, this means you use a non-naive logic. 20:30:59 the paradox works in the logic too, following naive logic rules 20:31:13 translating maths into English is always a little hairy, though 20:31:20 it ends up meaning the same thing, but being a lot more ambiguous 20:31:25 so it feels like there's wiggle room 20:31:30 when in the original maths, there isn't 20:31:41 ais523, but "If the moon exists, it is made of cheese" isn't a paradox. It is just a flawed "casual connection". To me this "paradox" seems more like such a flawed connection, than a paradox. 20:31:42 kerlo: could you translate the paradox into Lojban easily, I wonder? 20:31:52 AnMaster: not the same thing 20:31:52 AnMaster: well, that's just a false statement 20:31:56 not a causal thing at all 20:32:07 ais523, yes. Didn't find the right English word. 20:32:09 compare "if the moon exists, it is made of cheese" to "if the moon existed, it would be made of cheese 20:32:11 " 20:32:23 first is indicative, therefore the logical if-then; second is subjunctive 20:32:32 and therefore a causal if-then 20:32:45 it's a weird feature of English, I wouldn't be at all surprised if Lojban had two different words for taht 20:32:46 ais523, those seems to boil down to the same to me. Except the latter says it doesn't currently but would happen if it did. 20:32:47 *that 20:32:52 same thing from two different viewpoints 20:32:59 AnMaster: the first is comparing two unconnected statements 20:33:02 AnMaster: logically, you're utterly and completely wrong. 20:33:16 for instance, "If AnMaster is American, then the Empire State Building is made of lemonade" is true 20:33:24 two separate statements combined by if-then, logically 20:33:35 "If AnMaster were American, then the Empire State Building would be made of lemonade" is false 20:33:35 ais523: you mean it isn't? 20:33:36 !!! 20:33:42 because there's no logical connection between the two parts of that 20:33:46 ais523, hm right ok. 20:33:54 and the use of the subjunctive implies a causal if 20:34:03 noncausal if doesn't come up much apart from in contracts, tbh 20:34:08 which is probably why you aren't used to it 20:34:12 ais523, one issue here is that natural languages confuse the issue. 20:34:16 definitely 20:34:45 it does make a lot more sense as a possible paradox when written in logic. 20:34:48 logically, the statement is ((!(this statement)) || (ais523 can win #esoteric by announcement)) 20:34:53 that's C notation, but you know what I mean 20:35:00 ais523, but it doesn't make any sense as a paradox at all when written in English! 20:35:10 english has no built in logic 20:35:16 just a mapping of terms to logic terms 20:35:16 indeed. 20:35:21 therefore, that's a silly argument 20:35:21 well, it relies on uncommonly used bits of English which are similar to commonly used bits of English 20:35:29 and therefore is likely to be extremely confusing to someone who isn't a native speaker 20:35:30 english is just a transport method 20:35:32 not a logic 20:35:37 ais523, aha. 20:35:42 yes that would explain it. 20:35:47 logically, 20:35:53 X = (X → Y) 20:35:54 is true 20:35:55 why couldn't ehird just have explained what ais523 just said last time... 20:35:58 AnMaster: I tried to! 20:36:06 ehird, not the " and therefore is likely to be extremely confusing to someone who isn't a native speaker" bit 20:36:12 it's not easy to explain a language to someone who isn't a native speaker 20:36:19 err, i don't think ais523 was talking about the paradox 20:36:24 oklofok: yes I was 20:36:27 oh 20:36:28 but, in "X is X implies Y", X is true. 20:36:35 therefore, Y is true. 20:36:37 i thought you were talking about the two if-then's. 20:36:40 it's logically true 20:36:44 ehird: it is, however, easy to explain to a non-native speaker that something is particularly hard to explain to a non-native speaker 20:36:45 oh actually right. 20:36:47 so that's why you use a non-naive logic 20:36:52 which disallows self-reference 20:37:02 ais523: anyway swedish has the exact same two if's. 20:37:10 ais523, in English I still don't see the paradox. In C I see it. Same for (err what is the word...) predicate logic? 20:37:31 yep 20:37:45 right. predikatlogik in Swedish. 20:38:11 ais523: can you explain to him that english is not a separate logic system 20:38:14 just a transport method for thoughts? 20:38:17 oklofok, which ones are you thinking about 20:38:21 he seems to think that english somehow has its own system of logic 20:38:25 i don't get it 20:38:25 ehird: but don't thoughts have logics of their own/ 20:38:30 nope? 20:38:43 there are tribes of people who are culturally incapable of counting 20:38:43 stating something in english does not change its truth valu 20:38:44 e 20:38:46 that's just ridiculous 20:38:58 ais523, isn't that an urban myth? 20:39:00 ais523: .—.—.mipmip.—.—. 20:39:09 AnMaster: well, Wikipedia confirms it 20:39:13 ehird, *groan* 20:39:15 ais523: but does netcraft? 20:39:17 for Wikipedia values of confirm 20:39:24 AnMaster: I should have known you'd pick up that reference 20:39:34 AnMaster: conditional versus the normal verb case, you'd do that the exact same way in swedish 20:39:48 ehird, was that a bad thing 20:39:54 AnMaster: not particularly 20:39:56 k 20:40:05 maybe I'll name my funge Cunninghamellacae 20:40:11 *aceae 20:40:15 It has a certain ring to it. 20:40:23 ehird, of course I can pick up almost any reference from that source. There are just some out of print stuff I'm missing. 20:40:30 Actually, I'll call it Hypha. Or Dimorphic. 20:40:34 Those are nice names. 20:40:36 AnMaster: if i had used swedish in years, i'd demonstrate, but enough people here know it that i'm not going to try :P 20:41:03 oklofok, I still don't know what you are talking about... 20:42:43 well i'm not sure i have any idea about anything, but afaiu'd ais523 said you might not understand the paradox because you should it was a causal relation, even though swedish uses that exact same sentence structure for a logical implication, and has a separate causal relation 20:42:58 *thought it wasz 20:43:00 *was 20:43:30 hm ok 20:43:59 ais523: lojban doesn't have "paradox", which i find paradoxical 20:44:13 heh 20:44:18 Redirected from "paradoxical" 20:44:18 Did you mean: paradox, paradoxical 20:44:34 oh paradox was suggested there too, then slightly less fnny 20:44:35 *funny 20:44:52 oklofok, eh 20:45:04 but i love how when searching for a word in answers, it often redirects, and asks me if i wanted to search for what i was searching for 20:46:14 AnMaster: basically i search for paradoxical, it redirects to paradox and asks if i maybe wanted paradoxical 20:46:37 and i'm like "wow how did you guess!" 20:46:39 ah 20:46:49 oklofok, where was that search 20:46:56 answers.com 20:47:07 why not use google 20:47:11 i use it to check all english words. 20:47:19 is it good for that 20:47:22 [20:47] lingbot: en fi "If this statement is true, ais523 can win #esoteric by announcement"? 20:47:23 it's okay for that. 20:47:23 [20:47] ais523: "Jos tämä väite on totta, ais523 voittaa # esoteerinen jota ilmoitus" (en to fi, translate.google.com) 20:47:24 AnMaster: answers.com aggregates dictionaries, thesauruses and wikipedia 20:47:27 AnMaster: which is convenient 20:47:32 ah 20:47:38 * ais523 has no good if paradoxes survive Google translate 20:47:43 *no idea 20:47:49 now, that's quite a typo 20:47:52 ais523: where's lingbot? 20:47:53 ais523: if this statement is true, ais523 wins # esoteric which notification 20:48:11 ais523, in what channel is longbot 20:48:11 ehird: Freenode 20:48:12 ais523: so yes :) 20:48:15 err 20:48:16 ais523: ah 20:48:17 lingbot 20:48:18 that was in a PM 20:48:21 ah 20:48:55 ais523, it didn't respond 20:49:25 ais523, do you need to share a channel with it or something 20:49:49 22:49… oklofok: lingbot: fi en "Jos tämä väite on totta, ais523 voittaa # esoteerinen jota ilmoitus"? 20:49:49 22:49… lingbot: oklofok: "If this claim is true, ais523 win # rarefied by the return" (fi to en, translate.google.com) 20:50:20 -!- ais523 has set topic: This is #rarefied | http://tunes.org/~nef/logs/esoteric/?C=N;O=D. 20:50:28 ais523, it isn't responding to me! 20:50:35 AnMaster: are you putting in the quotes and the question mark? 20:50:36 any idea why 20:50:41 ais523, I were yes 20:50:48 and nickpinging it? 20:50:53 yes 20:50:54 it's quite strict on syntax, it seems 20:50:57 AnMaster: with a colon? 20:50:59 yes 20:51:08 ais523, I copy pasted your line 20:51:15 and changed to sv, tried with fi too 20:51:16 * pikhq can has job again 20:51:27 pikhq: why would you want a job? 20:51:32 pikhq: poor you 20:51:33 [20:51] lingbot: en sv "If this statement is true, ais523 can win #esoteric by announcement"? 20:51:34 [20:51] ais523: "Om detta är sant, ais523 kan vinna # esoteriska genom kungörelse" (en to sv, translate.google.com) 20:51:36 (just jumpin' on bandwaggin) 20:51:45 err 20:51:53 oklofok: Income. 20:51:54 isn't that correct? 20:52:08 i don't know the last word, but at least the rest seems 20:52:35 pikhq: what do you do? 20:52:39 or is it a secret 20:52:48 System administration. 20:53:02 ais523, haha, "kungörelse" 20:53:07 http://www.youtube.com/watch?v=Bb5DjyoDObA 20:53:11 pikhq: for who 20:53:14 GUGUL??? 20:53:14 ais523, that feels soooo out of date 20:53:15 oh. well that isn't that bad i guess, given you're internet people 20:53:17 AnMaster: what's that? 20:53:32 Tufts University. 20:53:38 oklofok, announcement, But think "town cries" and you get the time setting right. 20:53:48 err spelling... 20:53:54 criers* 20:54:05 -!- Sgeo has quit ("Leaving"). 20:54:14 [20:54] lingbot: "Om detta är sant, ais523 kan vinna # esoteriska genom kungörelse"? 20:54:16 [20:54] ais523: "If this is true, ais523 can win # esoteric by public notice" (sv to en, translate.google.com) 20:54:22 it seems to roundtrip pretty well via sv 20:54:43 ais523, yes but the Swedish one feels so off. 20:54:56 it means the right thing, but is worded weirdly? 20:55:03 ais523, anyway I can't see any paradox in the Swedish one either. 20:55:19 ais523, but yes as far as I can tell they mean the same. 20:57:05 ehird: shadow thing is awesome. 20:57:08 yar 20:57:20 checking for GNUreadline.framework... checking for readline... no 20:57:21 checking for tputs in -lncurses... yes 20:57:22 checking for readline in -lreadline... yes 20:57:23 ais523, kungörelse might be closer to "proclamation" in style I think. 20:57:24 checking for rl_readline_version... yes 20:57:26 checking for rl_begin_undo_group... no 20:57:28 configure: error: readline not found, so this package cannot be built 20:57:30 See `config.log' for more details. 20:57:39 haha 20:57:39 checking for irc channel to spam... yes 20:57:42 ehird: sounds like something i would make if i wasn't you know lazy. 20:57:49 AnMaster: it just makes no senseeeeeeeeeee 20:57:51 my favourite configure message ever was 20:58:03 checking if build environment is sane... yes 20:58:12 ehird, true, the error was badly named. 20:58:15 ais523: err... 20:58:17 ais523, "was"? 20:58:19 that build environment was C-INTERCAL cross-compiling to a half-finished bugg gcc-bf 20:58:23 hahaha 20:58:25 *buggy 20:58:25 oh 20:58:29 right 20:58:42 checking if build environment is sane... 20:58:43 ... 20:58:43 ehird: I checked the source, all it's actually doing is checking that ls is indeed ls, rather than something entirely different 20:58:45 ...... 20:58:48 ................ 20:58:50 uh... 20:58:53 *gibber* 20:58:54 ais523, what 20:59:00 bunnies! 20:59:04 (./configure continues) 20:59:09 apparently there were weird bugs due to people aliasing ls before running configure 20:59:11 ais523, does it even use ls elsewhere 20:59:19 and probably, or maybe possibly 20:59:30 I also particularly like the way ls is singled out for such treatment 20:59:36 ais523: a translation into Lojban of "if this statement is true, I can win #esoteric by announcement" would be uninteresting. 20:59:52 why? 20:59:56 kerlo: would it correctly get logical if rather than causal if? 21:00:17 ais523: I imagine Lojban has a material implication conjunction. 21:00:24 oh 21:00:25 hey 21:00:27 my time to shine 21:00:32 it has 4 different if-then's 21:00:33 iirc 21:00:34 kerlo: but it's so rarely used you don't know what it is? 21:00:47 does kerlo actually study lojban? 21:01:06 I study Lojban very lightly. 21:01:10 ah, ok 21:01:21 oklofok: as far as I can tell kerlo just sits around trying to be as cold and logical as possible while making idle remarks always about ai :) 21:01:23 you chose your nick specifically to make sense in Lojban, though! 21:01:24 I don't even know any of the conjunctions for "or". 21:01:28 I did. 21:01:52 ehird: he's far too warm when it comes to calculus tho 21:01:59 If I remember correctly, the guy who invented Ithkuil and Ilaksh doesn't speak either. 21:02:13 oklofok: is he like 21:02:16 SEXY INTEGRALS 21:02:17 or something 21:02:18 ehird: i've heard he sometimes works with differentials. 21:03:18 hm interesting.... I usually read English pretty fast, usually I read computer related stuff. However today I was reading about something biology related on wikipedia, and found that excessively hard to read. The Swedish page for that felt like a "simple Swedish" variant. (which iirc doesn't exist.) 21:03:47 kerlo: he doesn't, no one does afaik 21:04:57 kerlo, what does your nick mean in Lojban 21:05:08 my readline doesn't have rl_begin_undo_group 21:05:09 and it needs it 21:05:09 :-S 21:05:16 AnMaster: ear? or dog 21:05:18 you have a readline without undo? 21:05:19 or something 21:05:23 ais523: I suppose I do. 21:05:24 what sort of editor is that? 21:05:30 I have a macports one and the system one. 21:05:36 you should use Emacs in readline mode, obviously 21:05:44 | int 21:05:44 | main () 21:05:46 | { 21:05:48 | return rl_begin_undo_group (); 21:05:50 | ; 21:05:51 (someone's bound to have implemented that by now, surely?) 21:05:52 | return 0; 21:05:54 | } 21:05:56 configure:3196: result: no 21:05:59 ehird: ear 21:05:59 http://www.nabble.com/Can't-cabal-install-readline-td20862558.html 21:06:10 *AnMaster 21:06:15 The above happens because GHC is using the OS X default installation 21:06:15 of libreadline.a which is actually a link to libedit that doesn't 21:06:16 ah 21:06:16 implement the full readline API. 21:06:23 but adding the /opt shit don't work :( 21:06:30 .i mi kerlo mi 21:06:40 kerlo, that means 21:06:46 "I am my ear." 21:06:49 ok 21:07:10 a language you couldn't express anything false in would be interesting. 21:07:12 kerlo: doesn't that make you simultaneously infinitely small and infinitely large? 21:07:15 not possible. 21:07:18 AnMaster: probably trivial 21:07:26 ais523, what 21:07:35 ais523: no. 21:07:39 as in, the language itself would only be able to express a small finite number of things 21:07:41 the empty langauge is like that 21:07:42 and therefore would be a trivial language 21:07:43 *language 21:08:01 ais523, ah. What I meant was a more general usage language. 21:08:06 ais523: kerlo: doesn't that make you simultaneously infinitely small and infinitely large? <<< and no to this 21:08:09 what about Newspeak? 21:08:20 that doesn't let you express anything counter-government 21:08:22 ais523, 1984 right, haven't read it. 21:08:33 not quite the same as banning falsities, but should be close enough for practical use 21:08:52 ais523, how was it supposed to work in the fiction. 21:09:11 AnMaster: just there weren't concepts to express things like human rights, and free speech 21:09:17 hm ok 21:09:19 and equality 21:09:24 in the moral sense 21:09:28 ah 21:09:45 nor was there a sufficiently rich set of concepts to explain them 21:09:47 ais523, wouldn't it be trivial to invent those though 21:09:47 or define them 21:09:48 ah 21:09:50 I see 21:09:57 % INCLUDE_PATH=/opt/local/include C_INCLUDE_PATH=/opt/local/include LIBRARY_PATH=/opt/local/lib LD_LIBRARYPATH=/opt/local/lib cabal install readline --extra-include-dirs=/opt/local/include --extra-lib-dirs=/opt/local/lib 21:09:58 success 21:10:14 ehird, you only need that because of OS X right 21:10:20 no 21:10:25 I need it because I have readline installed in a non-standard place 21:10:33 sorry you don't have an oppertunity to troll 21:10:33 which is only because of OS X, right? 21:10:36 ais523: no 21:10:40 I have done such things on linux too 21:10:46 yes, so have I 21:10:51 but only because I was doing something weird 21:10:53 ehird, readline is usually in /usr/lib there though 21:10:57 i am weird. 21:11:09 exception gobolinux noted and irrelevant 21:11:20 it's so irrelevant that I used it. 21:11:26 no 21:11:32 but it solves it in other ways iirc 21:12:38 [ehird:/usr/local/lib/ghc-6.10.2] % ./ghc-asm 21:12:38 zsh: ./ghc-asm: bad interpreter: /opt/local/bin/perl: no such file or directory 21:12:44 LOL@evil mangler 21:12:54 where is your actual perl? 21:13:00 /opt/perl/bin/perl 21:13:17 also, there's a simple solution to such issues; just have only one directory on the entire computer which contains lots of symlinks to itself 21:13:22 admittedly, it causes lots of other issues 21:13:33 but it will solve those ones 21:13:44 as a bonus, it makes implementing locate very easy 21:13:54 ais523, symlinks to itself usually cause more problems than it solves indeed. 21:14:05 What license ["GPL2","GPL3","LGPL2","LGPL3","BSD3","BSD4","PublicDomain","AllRightsReserved"] ["BSD3"]: 21:14:10 Where's the BSD2/MIT :-( 21:14:20 ehird, what is that prompt for 21:14:24 AnMaster: mkcabal(1) 21:14:31 ehird, can't you enter your own 21:14:34 ais523: k? 21:14:41 yes 21:14:41 Under what category? [Codec,Control,Data,Database,Development,Distribution,Game,Graphics,Language,Math,Network,Sound,System,Testing,Text,Web,Other] [Codec]: Language 21:14:42 oklofok: ? 21:14:44 arguably incorrect 21:14:48 Is this your name? - "Author Name" [Y/n]: 21:14:50 Um, no. 21:14:50 ehird, so you can enter your own then 21:14:52 AnMaster: yes 21:14:55 ais523: i just wanted to ask you a really short question. 21:14:55 ehird, good 21:14:56 but it won't copy the file for you 21:14:56 I like the way Enigma does licences 21:15:02 oklofok: ah, ok 21:15:08 I gave an even shorter reply 21:15:12 and umm i meant the constant combinator 21:15:16 ehird, still maybe report a bug 21:15:23 oh, in that case, oklofok: second argument? 21:15:37 actually, first argument? 21:15:41 that's more useful than the second 21:15:43 with respect to k 21:16:09 ais523: first argument would be the directory 21:16:17 oklofok: ah 21:16:23 applying combinators to filesystem 21:16:24 s 21:16:30 sounds like the sort of thing an EsoOS would do 21:16:38 * ais523 wonders what an s-filesystem would be like 21:16:40 ais523: i'm not sure it makes that much sense, i just really wanted to ask a short question :P 21:16:43 or even worse, a c-filesystem 21:16:58 hmm 21:17:03 oklofok: it doesn't, but we're #esoteric; we should be able to /make/ it make sense 21:17:03 with c, backups sound easy 21:17:06 yes 21:19:04 only three weeks of exams left, and i have time to code again \o/ 21:19:11 *i'll 21:19:16 wonder if i still can. 21:22:34 ...wait rarefied 21:22:40 how's that esoteric :D 21:23:02 oklofok: translation of #esoteric to finnish and back to English, via Google 21:23:19 but "esoteerinen" means exactly the same as "esoteric" 21:29:08 -!- oklofok has changed nick to qop. 21:29:10 o 21:29:11 o 21:29:11 o 21:30:07 oqo 21:31:47 about making a set class... "the public interface we already know from elementary school" <<< currently they teach it at high school, in an advanced course 21:31:58 funny thing related to this 21:32:24 i asked a lecturer which course actually does real numbers rigorously in our uni 21:33:07 "none" 21:33:12 -!- ais523 has quit ("o"). 21:33:25 (i had a slightly longer answer, but that's the gist of the funny) 21:34:34 of course, the focus is more on discrete math, but still, that's just incredibly wrong. 21:34:41 also bye ais. 21:35:45 -!- Hiato has quit ("Leaving."). 21:37:21 Our "Mathematics 1" course pretty much started by a reasonably rigorous handling of real numbers; but that was the "L1" variant, for physics and maths students mostly; the computer science version ("C1") is rather more discrete-oriented. 21:37:30 And of course there's rigorous and then there's rigorous. 21:37:47 you started from the axioms? 21:38:05 or a construction 21:38:12 or wut 21:38:25 It was a rather constructivistic viewpoint, if I recall correctly. 21:38:45 It is all so vague; this was five-six years ago. 21:39:22 this older professor decided a few years ago to give the analysis 1 course (first course you take) that year, and started from reals, and a rigorous foundation of calculus 21:39:32 -!- Judofyr has quit (Remote closed the connection). 21:39:40 halfway through the course about 50% of students had dropped 21:40:03 and no one was going to pass it, he had to drop most content and just start over, basically 21:40:55 i actually have the lecture notes, unfortunately i'm not that good at studying if i don't get you know points and grades. 21:41:12 That sounds, content-wise, pretty much like it was done here; but of course the rigorousness level is hard to quantify. 21:41:48 Official list (in Finnish) of the contents for the half-a-year course are: "Luvut ja lukujonot. Vektorit ja analyyttinen geometria. Kompleksiluvut. Yhden ja usean muuttujan funktiot, kompleksifunktiot. Jatkuvuuskäsitteet ja jatkuvat funktiot. Reaali- ja kompleksifunktion derivaatta. Yhden muuttujan differentiaalilaskenta." 21:42:15 well, this dude is like 65 years old, when he was young high school actually taught stuff 21:43:15 well we did things in a different order 21:43:37 I hear the computer-science-maths do a different order, too. 21:43:44 but a solid foundation of calculus would take a big enough part of half a year to get on the list. 21:44:16 The official list for the C1 variant is: "Vektorialgebra ja matriisilaskenta. Lineaariset yhtälöryhmät. Yhden reaalimuuttujan differentiaali- ja integraalilaskenta. Johdatus lukuteoriaan. Verkkoterion alkeet." 21:44:55 we have basic course in mathematics 1-3 and analysis 1-2, first set being for stupid people, analysis for smart people; then there's this separate analysis course for math students, and they wouldn't take me in 21:45:25 that's not rigorous either, but probably more so 21:45:40 verkkoterion? :D 21:46:02 Someone's been typoing, I guess. 21:46:12 well okay ais's idea->good was slightly worse 21:46:18 but still 21:46:27 Official list (in Finnish) of the contents for the half-a-year course are: "Luvut ja lukujonot. Vektorit ja analyyttinen geometria. Kompleksiluvut. Yhden ja usean muuttujan funktiot, kompleksifunktiot. Jatkuvuuskäsitteet ja jatkuvat funktiot. Reaali- ja kompleksifunktion derivaatta. Yhden muuttujan differentiaalilaskenta." <-- translate! 21:46:44 * AnMaster requires translations of everything Finnish said here. 21:47:04 google translate will probably get that right 21:47:14 qop, too much work 21:47:19 + usually it doesn't 21:47:36 just saying, it's a list of topics, so it's very computer translateable 21:48:00 i don't know much math terminology in english 21:48:05 qop: L1 started from fields 21:48:09 AnMaster: Uh... "Numbers and sequences. Vectors and analytic geometry. Complex numbers. Functions of single and multiple variables, complex functions. Continuity and continuous functions. Derivatives of real and complex functions. Differential equations of one variable." Or something like that, anyway. 21:48:54 Deewiant: Our L1 went on to construct real numbers as (infinite) sequences, I think; it's been a while, and it might differ a bit based on the lecturer. 21:49:06 Yep 21:49:30 How else would you construct them? :-P 21:49:49 i have a hunch there are multiple ways. 21:50:05 -!- qop has changed nick to oklopok. 21:50:48 fizzie, and the other one. 21:50:48 You might take a more axiomatic view instead of a constructivistic one. At least when it comes to relative amounts of emphasis on things. 21:50:51 the one with the typo 21:51:04 i should probably be careful about what i say around you helsingans, i hate hearing about how much better your uni is. 21:51:04 (and what was the typo) 21:52:01 AnMaster: "Vector and matrix algebra. Linear equation systems. Differential and integral calculus of one variable. Introduction to number theory. Basics of graph thory." I've translated the typo here too. 21:52:07 construmatic 21:52:25 fizzie, haha 21:52:47 fizzie, did the typo mean anything in Finnish 21:53:05 Nope 21:53:09 oh well 21:53:40 Typos is amongst the top reasons why web translators are useless :P 21:53:51 Also, bad grammar are. 21:53:57 Just run a spellchecker over it 21:54:10 Before translation, that is 21:54:30 It went "verkkoteorian" → "verkkoterion"; I guess "graph theory" → "graph thory" is reasonable approximation; or maybe "graph throy". (The Finnish version is the genitive case, too.) 21:55:15 teoria -> terio is more like theory -> thyro IMO 21:55:31 Yes. That's good, too. 21:56:14 heh 21:56:31 GRAPH THYROID CANCER 21:56:45 Maybe not. 21:59:42 If you have graph thyroid cancer, it means your vertices are multiplying with no limits. 22:00:37 Take THAT, polynomial approximation scheme. 22:02:19 what's a polynomial approximation scheme 22:02:30 all i know is there's a course that covers them 22:02:51 It's a scheme that approximates polynomially 22:03:21 finds a polynom that approximates, approximates in polynomial time? 22:04:13 Polynomial-time algorithm that approximates something 22:04:20 right, so latter 22:04:57 You should've inserted an "or" there, I couldn't parse that 22:05:17 well that's your loss! 22:05:26 well mostly mine, but somewhat yours too 22:05:34 Indeed 22:07:27 Or a semicolon instead of a comma. Any Prolog-speaking person would then have parsed it correctly. 22:07:57 Unfortunately Prolog-understanding would've failed at lexing 22:08:20 i was going for a list of suggestions to get him to elaborate 22:08:41 Oh, but then it worked just fine. 22:08:44 Maybe you should've wrapped it in [] 22:13:25 wait why 22:13:29 oh 22:13:30 list 22:13:32 ... 22:13:50 i was like box -> birthday present -> i wonder when Deewiant's birthday is 22:13:56 and then i was like what 22:14:02 :-P 22:14:07 Maybe you should've explicitly created a std::vector and push_back()ed the suggestions there! 22:14:58 aren't se using namespace std on #esoteric? 22:15:01 *we 22:15:27 I should hope the namespace isn't that polluted 22:15:38 someone should grep logs 22:15:53 * oklopok eyes at fizzie with himself 22:15:58 wait 22:16:03 you don't get that reference 22:16:04 well umm 22:16:09 kerlo said he's his own ear 22:16:19 so i thought i'd be my own eye during that sentence 22:16:38 okay getting out of hand again, i'll go make pizza -> 22:22:47 -!- tombom has quit ("Peace and Protection 4.22.2"). 22:38:34 hey oklopok 22:38:36 what a coincidence 22:38:38 I have a pizza 22:38:41 it is going into my mouth. 22:39:07 i don't, i drank some coke, then read a page, then irced. 22:39:22 oklopok: are you irked about that 22:39:57 Deewiant: whuz fungespace again? sparse 2d array + what's that weird bound thing 22:40:20 Sparse 2d array basically 22:40:38 For y, you need to know the min/max point describing the smallest rectangle that bounds it 22:41:05 Deewiant: umm I recall it being simpler 22:41:11 wasn't it like you recorded the lenght of each line 22:41:20 what's the diff 22:41:33 I don't do anything currently; CCBI fails at that 22:41:49 But that is essentially what you need to know 22:41:53 How you impl is up to you 22:42:02 Deewiant: oh, so if you just have a sparse 2d array you should be fine in non-edge cases? 22:42:18 i mean how does that interact with wrapping 22:42:20 For non-edge cases you can get by with a lot ;-P 22:42:25 you have to know when you've reached the end 22:42:31 Yep 22:42:31 so surely you have to store bounds of each line 22:42:39 But for wrapping you don't need to know the smallest rectangle 22:42:43 Just any bounding rectangle 22:42:45 but for y you do? 22:42:54 Strictly, according to my interpretation, yes 22:42:59 well, smallest rectangle is just shrinking when you replace an end with a space right? 22:43:02 At least AnMaster accepted it and implemented it :-P 22:43:06 Yes, or beginning 22:43:46 efunge have exact bounds too since a few days 22:43:48 Deewiant: http://pastie.org/460291.txt?key=l9dauneba2wsru0bgfhza Voila? 22:44:15 For example 22:46:20 {-# LANGUAGE BangPatterns #-} 22:46:21 what 22:46:31 What what 22:46:35 what does that do 22:46:44 It's a LANGUAGE pragma, it enables the BangPatterns extension 22:46:48 ah 22:47:18 Dur :P 22:47:23 bang patterns = !Foo 22:47:29 it's just to make 22:47:30 data Point = Point { x :: {-# UNPACK #-} !Int 22:47:30 , y :: {-# UNPACK #-} !Int } 22:47:32 deriving (Show) 22:47:36 an unboxed tuple of strict platform integers 22:47:44 ah 22:47:45 ehird: You don't need bang patterns for that. 22:47:51 Deewiant: Oh? 22:47:54 Ah, right. 22:47:55 Strict fields are not an extension. 22:47:58 Right 22:48:05 And they're gone 22:48:10 I should probably put this into git sometime 22:48:19 Go monotone 22:48:24 Nothx :P 22:48:30 Meh 22:48:34 bang patterns... there's probably a haskell porno with that name 22:48:41 oklopok: that would be awesome 22:48:44 yes 22:48:47 "A Haskell porno"? 22:48:50 also Strictness Analysis 22:48:55 Deewiant: yes 22:48:57 a bdsm haskell porno 22:49:21 Deewiant, monotone... 22:49:29 AnMaster: weird, unused VCS 22:49:34 "Eta expansion" 22:49:34 Deewiant: afaik it can be used like that. 22:49:34 yes 22:49:34 think arch-levels of weirdness 22:49:40 I was just surprised at Deewiant 22:49:43 he was joking 22:49:44 or what was your concern 22:49:46 we've discussed it lately 22:49:48 ehird, I hoep so 22:49:50 hope* 22:49:54 Monotone is the first DVCS I tried, or Darcs 22:49:59 ehird, clearly you should use RCS. 22:50:07 That was at the point when I couldn't even understand the point of SVN 22:50:10 AnMaster: AAAAAAAAAAAAAAAAAAAAAAAAAAA 22:50:16 AAAAAAAAAAAAAAAAAAAAAAAAAAAAA 22:50:18 ehird, :D 22:50:18 Visual SourceSafe! 22:50:18 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 22:50:21 AAAAAAAAAAAAAAAAAAAAAAAAAAA 22:50:23 AAAAAAAAAAH! 22:50:28 Deewiant, now that is even nastier! 22:50:33 hm 22:50:50 ehird, bitkeeper. 22:51:00 ... 22:52:27 Deewiant: Do you have a solution to those leaning-left data decls btw? 22:52:30 *leaning-right 22:52:46 More line breaks? 22:52:53 After the =? 22:52:57 Yep 22:53:25 ehird, before it 22:53:26 That works nicely 22:53:35 AnMaster: Yeah, that helps too, when you have |s 22:53:41 data Foo 22:53:44 = Bar 22:53:46 | Baz 22:53:48 yes 22:53:53 wait you were serious? 22:53:58 ehird, no 22:54:02 i was 22:54:23 -type supervisor_start_child_result() :: {ok, supervisor_child_pid()} 22:54:23 | {ok, supervisor_child_pid(), _} 22:54:23 | {error, supervisor_child_error()}. 22:54:27 I have stuff like that 22:54:30 which is indeed leaning 22:54:35 so yes it might work 22:54:39 I don't mind leaning 22:54:45 Unless it leans past 80 cols 22:54:47 -!- psygnisfive has joined. 22:54:50 i quite like brace-y languages, because they don't lean 22:54:51 ah 22:54:55 -type supervisor_start_child_result() 22:54:56 :: {ok, supervisor_child_pid()} 22:54:56 | {ok, supervisor_child_pid(), _} 22:54:56 | {error, supervisor_child_error()}. 22:55:00 that works nicely 22:55:04 yep, that's ocaml/haskell style 22:55:14 actually 22:55:18 ehird: they dont lean? 22:55:22 I should move it one column 22:55:26 psygnisfive: To the right 22:55:28 psygnisfive: to the right of code 22:55:30 -type supervisor_start_child_result() 22:55:30 :: {ok, supervisor_child_pid()} 22:55:30 | {ok, supervisor_child_pid(), _} 22:55:30 | {error, supervisor_child_error()}. 22:55:32 like that 22:55:35 ehird, what do you think 22:55:39 AnMaster: yep 22:55:41 that's what I do 22:55:47 hm 22:55:48 what, lean to the right of the code?? 22:56:07 psygnisfive: foo = bar versus foo =\n bar 22:56:10 ehird, well yes it works nicely. Not idiomatic for erlang though. 22:56:20 data Point = 22:56:20 Point { 22:56:21 x :: {-# UNPACK #-} !Int, 22:56:23 deewiant: WHAT? 22:56:23 y :: {-# UNPACK #-} !Int 22:56:25 D: 22:56:25 } deriving (Show) 22:56:27 that works nicely 22:56:33 Deewiant: WHAT??! 22:56:36 ?!?!?!? 22:56:40 huh?? 22:56:42 psygnisfive: Long lines versus ones broken up like these guys have been spamming the channel with 22:56:44 WHAT THE FUCK IS HAPPENING HERE 22:56:50 my uncomprehension is UNBOUNDED. 22:57:01 oklopol: noone knows! 22:57:14 hnce why you can predict rightly that ehird is talking 22:57:35 grr, having to cd src before runhaskell Main.hs is irritating 22:57:44 i always read "noone" as a cutified version of "none" 22:57:44 foo foo foo = bar 22:57:46 | baz 22:57:51 psygnisfive: ^ right-leaning 22:58:03 foo foo foo = 22:58:05 bar 22:58:05 list-tries is a Haskell library which implements the trie (Wikipedia article) and Patricia trie (Wikipedia article) data structures. 22:58:07 baz 22:58:08 Deewiant: that already exists. 22:58:08 i dont follow but ok :D 22:58:10 psygnisfive: ^ not 22:58:31 Deewiant: 22:58:31 also 22:58:32 Assuming you have the cabal-install tool installed and working, the easiest way to obtain Glob is with the cabal install list-tries command. 22:58:33 ehird: Not for any lists. 22:58:36 on the list-tries page 22:58:42 Woops. 22:58:42 :-D 22:58:49 that's a very easy way to do something else! 22:59:25 oh blah, I forgot that lahey space is a pain to implement 22:59:46 ehird: Tries existed in the form of IntMap and bytestring-trie, not otherwise. 22:59:48 this look hilarious though: 22:59:49 -spec start_child(supref(), child_spec() | [any()]) -> {ok, pid_undef()} 22:59:50 | {ok, pid_undef(), any()} 22:59:50 | {error, already_present 22:59:50 | {already_started, pid_undef()} 22:59:50 | any()} 22:59:52 And bytestring-trie didn't even exist when I started. 22:59:56 LEAAAAAAAAAAAAANING! 23:00:04 AnMaster: that's awful 23:00:08 ehird, agreed. 23:00:10 factor it out. 23:00:16 ehird, yes into -type 23:00:20 Why is it awful now but not when you wrote it? 23:00:21 since it is used elsewhere I think 23:00:28 Deewiant, I just wrote it. 23:00:28 ... 23:00:34 Oh. 23:00:38 and noticed how bad it was. 23:00:41 {ok}{ok}{ok}{ok}{ok}{ok}{ok}{ok}{ok} 23:00:45 I don't think it's bad at all. 23:00:51 But whatever. 23:00:51 psygnisfive, ...? 23:01:02 Deewiant, it goes past 80 cols 23:01:06 so horrible even. 23:01:16 fuck 80 cols 23:01:16 Oh, that sucks then. 23:01:17 its non-right-leaning okoing 23:01:26 FUCK them 23:02:04 ehird, no. 23:02:10 learn to love them 23:02:22 -spec start_child(supref(), child_spec() | [any()]) -> 23:02:22 {ok, pid_undef()} 23:02:22 | {ok, pid_undef(), any()} 23:02:22 | {error, already_present 23:02:22 | {already_started, pid_undef()} 23:02:23 | any()} 23:02:25 works better 23:02:25 I don't care about your 1980s 80s. 23:03:04 I care about them but I use a tab width of 3 so others who care aren't so lucky 23:03:40 3? 23:03:40 3?! 23:03:44 3?!? 23:03:47 3?!?! 23:03:47 3?!?!? 23:03:49 ! 23:03:53 Freak. 23:04:00 ehird, agreed. 23:04:02 2 is too small 23:04:02 I use 4 23:04:07 It's a prime choice EH EH EH. 23:04:13 fizzie: <3 23:04:15 4 is pointlessly large 23:04:39 fizzie, good. apart from the forced laugher. 23:04:39 2 is too small 23:04:42 4 is pointlessly large 23:04:44 Then I <3 3 23:04:48 um 23:04:53 3 is too small. 23:04:55 4 too large 23:05:02 lets use 3.14 23:05:07 for tab 23:05:08 :D 23:05:10 Sure. 23:05:10 Nah, 3 is fine. 23:05:15 3.14 could work fine. 23:05:20 it'd change at very nested depths 23:05:49 Wouldn't you want it to get smaller at high nestings? :-P 23:05:54 It'd go 23:05:56 3 6 9 13 23:05:58 etc 23:06:04 maybe e 23:06:20 would fit better 23:06:32 I've used all of 2,3,4,8 and settled on 3. 23:06:34 ehird, no, not rounded to whole cols in display! 23:06:35 :( 23:07:15 ehird, it must display it offset compared to other mono-space text! 23:07:22 I think I very briefly used 1 but that kinda sucked 23:09:27 Still... 23:09:28 ^ul (3)S((!)(?))(~^:Sa~a*~:^):^ 23:09:29 3?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!?!? ...too much output! 23:09:32 It is very strange. 23:09:56 A not-a-power-of-two tab, that is. 23:10:13 Why should it be a power of two? 23:10:26 I wonder how 6 space indentation is 23:10:32 Too wide. 23:10:33 Deewiant: It's a computar thing. 23:10:46 fizzie: I note that your nick's length is not a power of two. 23:10:47 3 is a power of 2 23:10:48 ehird: Nor yours. 23:10:50 0,1,2,3 23:11:06 Deewiant: Yes, but my tab length (4) is. 23:11:11 Not everything on a computers has to be a power of two. :-P 23:11:18 Especially not an computers. 23:12:12 my nick's length is a prime number. 23:12:19 ;) 23:12:24 mine too 23:12:26 So's my tab length! 23:12:30 Deewiant: is there a way to runhaskell with an include dir 23:12:31 like 23:12:34 runhaskell src/Main.hs 23:12:40 w/ src/foo available 23:12:43 I don't know, I don't use stuff like that 23:12:47 :-( 23:12:58 You and your interactivity 23:13:25 * lifthrasiir writing 15-page-long essay on a topic which i don't care so much and which deadline is 8 hours after 23:13:36 *phew* 23:13:52 lifthrasiir: what topic 23:15:27 the Korean history in early 20C. 23:15:45 which is very complicated. 23:16:02 Goddamn you lahey space 23:16:37 lifthrasiir: i refuse to believe history can be complicated, but i can imagine writing about it would be 23:16:56 lifthrasiir: which korea are you in? the dictator one or the high speed interwebs one. well err okay the answer to that one is fairly obvious, I doubt esolangs are encouraged in the other one 23:17:00 but you know, have to send it anyway 23:17:04 otherwise it'd be wasted typing. 23:17:29 ooh, of course the latter, south one. but that distinction is made in mid-20C anyway. 23:17:52 what's 20C? 23:17:56 20th century 23:17:57 20th century 23:18:09 20 degrees Centigrade 23:18:09 i imagine writing about north korean history is rather easy 23:18:22 http://www.gamesetwatch.com/2009/04/one_step_closer_to_a_holodeck.php 23:18:23 DO WANT 23:18:37 "Once upon a time, Kim Il-sung created the world." 23:18:41 ~fin~ 23:18:44 psygnisfive: old 23:18:47 also impractical 23:19:20 i want brain implants :< 23:19:45 i just want plants in my brain 23:19:50 yeah could work 23:20:00 ehird: shut the fuck up :| 23:20:18 PSY. Gnis, five. 23:20:42 wat 23:21:02 ehird: The one with Internet access at all, of course. ;p 23:21:10 North Korea doesn't have net? 23:21:26 Except in a few limited circumstances, yes. 23:21:46 pikhq: Does it have a national net at least? 23:21:53 (a university or two has censored satellite access, and Kim Jong-Il has his own Internet access) 23:22:06 Yes, but not that many people have access to it. 23:22:09 weird 23:22:25 I thought it was, y'know, connected, just censored to hell like maybe only govt approved sites 23:23:20 North Korea is a land that hardly has television and radio. ;) 23:23:42 Yar... it's just the modern looking cities fool you. 23:24:04 Modern looking and eerily empty. 23:24:34 "This is an impoverished country where televisions and radios are hard-wired to receive only government-controlled frequencies. Cellphones were banned outright in 2004. In May, the Committee to Protect Journalists in New York ranked North Korea No. 1 — over also-rans like Burma, Syria and Uzbekistan — on its list of the “10 Most Censored Countries.” " 23:24:37 — NY times 23:25:33 might be pretty cool to have a country like that 23:25:50 IT Industry 23:25:51 Korea Computer Center! 23:25:58 — North Korea govt site 23:26:05 maybe esoland will be like that 23:26:48 "Politics" links to /en/great 23:27:13 -!- FireFly has quit ("Later"). 23:28:02 pikhq: "Some small “information technology stores” — crude cybercafes — have also cropped up. But these, too, connect only to the country’s closed network. " 23:28:16 So I assume nationet is widely available, just not in homes. 23:28:24 "According to The Daily NK, a pro-democracy news site based in South Korea, computer classes at one such store cost more than six months wages for the average North Korean (snipurl.com/DailyNK). " 23:28:27 Orrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr not 23:29:05 Available, but not affordable. 23:30:01 It would have been fun if lifthrasiir was in north korea and his client self-destructed when I mentioned it :-D 23:31:04 then i'll wish the self-destruction occurs at both end. 23:31:36 :< 23:36:38 hmm 23:37:19 32-bit fungespace is 65536 petabytes 23:37:19 :-D 23:38:26 lifthrasiir: how many sloc is pyfunge? 23:39:12 5500 lines or so. 23:39:24 lifthrasiir: w/o fingerprints? 23:39:53 fingerprints account for 60% of codes. 23:40:14 so 1650 lines of actual interp 23:40:16 that's not so bad 23:40:46 that is 30%, not 40% ;) 23:40:55 ups 23:40:58 it's late :-( 23:41:07 anyway actual core is quite small 23:41:28 after I get fungespace done I'll probably breeze through the rest of funge98 core 23:42:54 ehird: you making like haskell ffunge 23:43:00 yes 23:43:19 i typoed funge, but it looks faster that way. 23:43:33 it's like the word is a racecar or something. 23:43:40 -!- coppro has joined. 23:43:45 that should be fffunge, short for fast-forward funge 23:43:57 heh, i love it 23:46:56 i could have that in my debugger 23:46:58 I plan to have a scrubber 23:47:04 so you can go back and forwards in fungetime 23:47:16 (as well as have the IP have a shadow of where it's been recently etc) 23:47:26 and let you edit fungespace on the fly 23:48:02 so by dragging the scrubber along, you fast forward funge 23:57:26 *Hypha.Fungespace> parseFungespace "Hello, world!" 23:57:26 Fungespace {space = fromList [(Point {x = 0, y = 0},72),(Point {x = 1, y = 0},101),(Point {x = 2, y = 0},108),(Point {x = 3, y = 0},108),(Point {x = 4, y = 0},111),(Point {x = 5, y = 0},44),(Point {x = 7, y = 0},119),(Point {x = 8, y = 0},111),(Point {x = 9, y = 0},114),(Point {x = 10, y = 0},108),(Point {x = 11, y = 0},100),(Point {x = 12, y = 0},33)], bounds = fromList []} 23:57:30 Doesn't do bounds yet, oh well. 23:57:40 (that's "Hello, world!") 23:57:44 err wait 23:57:47 that's in the line before :D