00:00:04 Dammit, I am getting to the bottom of this. 00:00:15 What picture? 00:00:29 Oh 00:00:37 I thought you meant a picture of _you_ 00:00:51 http://en.wikipedia.org/wiki/Dude 00:01:07 Bloody TinEye... 00:02:44 "Evander Berry Wall (1860–1940) was a New York dude[1] who became famous in the 1880s for his extravagantly refined look." 00:03:07 That is the single best opening to a Wikipedia article. 00:04:27 XD 00:04:45 http://lumeniki.referata.com/wiki/Jesus_(carnation_bread) 00:04:49 Phantom_Hoover: Just what 00:04:58 I... Huh? 00:05:41 Will you promise not to be confrontational if I tell you that he's been turning up at RationalWiki lately? 00:06:33 Weirdly, he could pass for normal there. 00:07:01 All because I decided to use "eir" 00:08:19 -!- Sgeo has changed nick to Lumenos. 00:08:23 -!- Lumenos has changed nick to Sgeo. 00:08:25 Eep 00:08:30 That's a registered nick 00:08:38 Let me just reassure everyone that I'm not em 00:08:51 Sgeo: WE DON'T BELIEVE YOU 00:09:22 oerjan, you don't love me anymore? 00:09:33 -!- Phantom_Hoover has changed nick to Lumenos. 00:09:41 -!- Lumenos has changed nick to Phantom_Hoover. 00:09:54 I must ask 'im about that. 00:10:33 Is there a name for the problem that is halfway between SAT and TAUT? (Do exactly half the assignments make the sentence true, the other half false?) 00:10:38 Phantom_Hoover: reminds me of the congo sapeurs 00:11:19 TAUT? 00:11:26 Tautology? 00:11:27 TAUTOLOGY 00:11:35 Tautology? 00:11:44 and SAT as in 3-SAT? 00:11:47 Complement of SAT (SATISFIABILITY) 00:12:23 But no, I don't know why complexity theorists like SHOUTING. 00:12:33 cpressey: seems connected to #P and PP 00:12:39 cpressey, how can something be between SAT and TAUT? 00:12:49 #P asks for the exact count 00:13:07 AnMaster: Do exactly half the assignments make the sentence true, the other half false? 00:13:19 ah 00:13:33 cpressey, what if you have an odd number of assignments? 00:13:50 PP essentially asks for majority, although it's formulated as probability 00:13:54 There are 2^n possible assignments for n variables. How can that be odd? 00:14:24 (Don't say n=0.) 00:14:51 cpressey, ah true 00:14:55 also +-in-a-circle P asks whether the number is even or odd 00:14:57 cpressey, also I set n=-1 00:15:21 oerjan: That last one sounds like it -- hard to Google that though 00:15:27 that gives... half an assignment? 00:15:55 cpressey: um no it's not whether it's exactly half, it's whether the number is even or odd 00:16:07 http://en.wikipedia.org/wiki/Sharp-P 00:16:09 oerjan: oh, right 00:16:15 (the first one) 00:16:41 oh god, i remember this now 00:19:25 http://qwiki.stanford.edu/wiki/Complexity_Garden#sharpsat is the SAT version for #P 00:20:16 oerjan: I'm naively trying to pessimize possible solutions to this problem 00:21:32 The idea of an adversary doesn't really apply to something like a boolean sentence 00:22:00 encoding an RSA factoring problem usually gives you something practically impossible to solve, afaiu 00:22:30 well, integer factorization is in both NP and co-NP, so... 00:22:51 and you can make it a SAT or TAUT problem as much as you want 00:23:02 cpressey: um no that's primality 00:23:15 oerjan: No.... PRIMES is in P, remember? 00:23:40 actually _finding_ a factor if you know it's a prime is not known to be in co-NP afaik 00:24:28 in fact testing whether there's a factor in a certain interval is NP-complete 00:24:29 Of course, P is in both NP and co-NP, but that's not what I/you meant 00:24:34 -!- coppro has joined. 00:24:53 (saw it in lipton's blog archive the other day) 00:25:13 "It is not known exactly which complexity classes contain the decision version of the integer factorization problem. It is known to be in both NP and co-NP." 00:25:20 wikipedia, the perfect authority 00:25:40 erm 00:25:52 link? 00:25:59 http://en.wikipedia.org/wiki/Integer_factorization 00:26:59 oh 1 < d < M, not M < d < N 00:27:07 might make a difference... 00:27:44 oh right it explains why 00:27:58 yeah, that applies only if the lower bound is 1 00:28:24 Phantom_Hoover: Confrontational? I dislike RationalWiki. Doesn't mean I flip at its mention. 00:28:41 I know, but the last time was hardly pleasant. 00:28:51 It also says " It is suspected to be outside of all three of the complexity classes P, NP-complete, and co-NP-complete." I don't know how it could be in NP (or FNP) and "outside of NP-complete" though... 00:30:09 NP-complete is a subset of NP, possibly strict. 00:30:39 Yeah, I never understood that. 00:30:42 Phantom_Hoover: Maybe for you. I just made what to me was an offhand remark. 00:30:43 "outside" doesn't mean "harder than", just not an element of 00:30:55 Oh. K... 00:31:10 Yeah, math talk confuse english brain. 00:31:22 it can be harder than P but easier (in the reduction sense) than NP-complete and co-NP-complete 00:31:31 Phantom_Hoover: I mean no offence, you irrational Hellspawn, you. 00:31:42 Hellspawn? 00:31:49 Good one. 00:31:53 Spawn; of Hell. 00:32:51 I choose it to mean that my spawn point is in hell. 00:32:53 So these are the dying moments of my last night as a sleeper. 00:32:59 No fanfares... 00:32:59 oerjan, isn't the only known PRIMES algorithm in P extremely slow? As in, exponential algorithms are faster for practical purposes? 00:33:20 AnMaster: _random_ algorithms are faster for practical purposes 00:33:34 AnMaster: iirc you need proper reals 00:33:39 aliseiphone, I suppose you don't get holidays from /bin. 00:33:41 aliseiphone, hm? 00:33:42 sqrt(2) and the like 00:33:44 *randomized 00:33:46 aliseiphone, ah 00:33:48 for the algo 00:33:50 oerjan, right 00:33:54 oerjan, that makes more sense 00:33:59 Well, *if* P = NP (I'm not entirely convinced by my own proof) then I bet some of these remarkably efficient algorithms work by creating a new "customized" Turing machine (or PL interpreter of whatever sort) on the fly and emulating it. Or perhaps several in sequence. 00:34:28 Even if P = NP the constant is probably huge. 00:34:40 I definitely think P != NP. 00:34:42 I often wonder how any computer works at all. 00:34:46 Probably. 00:34:47 AnMaster: also the random algorithms can be done in such a way as to give a certificate of certainty unless they fail, in which case you just retry 00:34:50 Create a TM which creates a TM which creates a TM which.... k times. 00:34:51 Phantom_Hoover: Badly. 00:35:02 I mean, flip a single bit and you segfault. 00:35:10 NP-complete is strict subset of NP unless P=NP. NP-complete and co-NP-complete are disjoint unless NP=co-NP (then NP-complete=co-NP-complete). 00:35:13 oerjan, lets just get a quantum computer and be done with it 00:35:22 Phantom_Hoover: /bin? Loony bin. 00:35:31 ...Yes 00:35:33 Yea, quite so. 00:35:55 Please tell me I'm not the first to make that pun. 00:36:01 AnMaster: i'm not sure that a quantum computer can test primality any faster 00:36:05 I mean, flip a single bit and you segfault. <-- ECC used to be a lot more common in the early days of RAM 00:36:11 oerjan, hm it can factor faster 00:36:19 & such an unassuming way it draws to a close. Don't I die at the end of tradgedies? Dammit, playwrite. 00:36:22 AnMaster: apparently. 00:36:24 oerjan, and if you can't factor, then you have a prime, right? 00:36:27 You do it all wrong! 00:36:37 *playwright 00:36:39 err 00:36:41 EEC 00:36:43 not ECC 00:36:45 iirc 00:36:51 ECC. 00:36:53 or maybe I was correct to begin with 00:36:53 yeah 00:37:12 aliseiphone: btw it hasn't been proved that the constant power if P = NP has to be as big as 2, even... 00:37:29 the best bounds are somewhere between 1 and 2 00:37:41 oerjan: hell, mathematics is 90% hunches. 00:37:48 XD 00:37:48 (for solving SAT) 00:37:51 what's a few more of my own? 00:38:09 -!- augur has joined. 00:38:16 how many proofs depend on Riemann, Goldbach, Collatz :) 00:38:43 http://abstrusegoose.com/133 00:38:45 Also Zorn 00:39:01 Slereah: that's an axiom 00:39:05 aliseiphone: lipton discusses at least 4 possibilities in one if his most popular blog articles: P=NP and fast, P=NP but huge exponent, P != NP and slow, P != NP but so small constants it still works in practice 00:39:07 (choice) 00:39:09 *one of 00:39:32 and has been proved consistent wrt ZF 00:39:40 (can you tell i'm slowly gobbling up his blog archive? :D) 00:39:58 oerjan: Which blog? 00:40:01 aliseiphone : What's your point 00:40:01 Goedel? 00:40:04 Axioms are theorems 00:40:05 yes 00:40:21 Slereah: Zorn is "proven consistent" 00:40:29 Riemann could be plain false. 00:40:32 My god, the P=NP argument works! <> 00:40:36 AnMaster: note that quantum computers only give the correct answer with a certain probability, so you might _still_ want one of those certificates 00:40:37 oerjan, what about the variant: P?=NP can't be proved with current axioms (that is, it is independent) 00:40:46 (for shor's factorization algorithm) 00:41:07 Will quantum computing make Schrodinbugs a reality? 00:41:19 Answers on a postcard; I must sleep. 00:42:11 -!- zzo38 has joined. 00:42:41 AnMaster: no. 00:42:56 you can't just have random shit be independent! 00:43:09 You can't ADD a P algo for NP as an axiom 00:43:11 AnMaster: i think that may have been mentioned in the comments 00:43:15 That's nonsense! 00:43:18 I connected a few seconds ago I got a connection timeout (after sending PASS/NICK/USER), it didn't do that before. But now I connected it worked 00:43:20 Why is that? 00:43:31 Is my IRC client broken? 00:43:35 Or is the server broken? 00:43:39 Or something in between? 00:43:45 aliseiphone: It's certainly a possibility, that there is no proof we can make. 00:43:49 Just like riemann 00:43:51 zzo38: Dunno. All kinds of strange transistent errors can happen. 00:44:08 cpressey: So P = NP but there is no P algo for NP? 00:44:36 If it is independent 00:44:38 Then 00:44:40 I was trying to write a music synthesizer program so I wrote a music using QORCH 00:44:43 aliseiphone: More likely P != NP and there is no proof. 00:44:45 ZFC + P=NP 00:44:46 aliseiphone: there are at least two ways of having it be independent afaict: there is an algorithm that works but no proof that it works, or there's a proof that there's an algorithm but it's nonconstructive and doesn't _actually_ exist :D 00:44:46 http://sprunge.us/RHYa 00:44:48 and 00:44:56 ZFC + P!=NP 00:44:59 both work 00:45:06 Take algo from former 00:45:14 It must work in latter 00:45:19 *I* didn't use the word "independent". 00:45:20 -!- coppro has quit (Remote host closed the connection). 00:45:45 oerjan: For the latter the constructivist in me punches you. 00:46:03 wait, of course i forgot the obvious third: no proof and no algorithm 00:46:04 This music is in 12-TET but the program QORCH supports just intonation as well, and even Bohlen-Pierce 00:46:05 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds). 00:46:29 You simply have to enter the correct parameters to the TEMPERAMENT or TUNING commands, to do it..... 00:47:13 That's nonsense! <-- indeed. But P!=NP might be true but unprovable or such. I can't see why that one is nonsense 00:47:25 having P=NP and unprovable would be closer to nonsense though 00:47:39 * aliseiphone grumbles about constructivism 00:48:05 Mathematics is not yet ready for such problems. 00:48:07 what about it? 00:48:08 grumbles about destructivism 00:48:11 Well, goodnight for the last time here. 00:48:16 yay! 00:48:16 o.O 00:48:22 Oh! 00:48:30 aliseiphone: Good night you are late!! 00:48:31 aliseiphone, what about constructivism 00:48:31 AnMaster: Yer stupid "unprovable" stuffs :P 00:48:32 Here as in the realworld here, not the #esoteric here 00:48:37 AnMaster: an algorithm which works but no one knows _why_ it works... 00:48:40 aliseiphone, um, such stuff exists 00:48:40 * Sgeo :Ds 00:48:56 aliseiphone, consider that thing about cardinality 00:49:10 AnMaster: Yeah. But constructivist mathematics has much less. 00:49:12 or the parallel lines stuff 00:49:16 We could have P algorithms for NP-complete problems, but we can't prove they are in P 00:49:17 Much, much less. 00:49:26 Uhh 00:49:33 Parallel lines is independent 00:49:41 cpressey: or we cannot prove they actually give the right answer 00:49:46 aliseiphone, exactly 00:49:47 Which P=NP cannot be 00:49:50 oerjan: indeed 00:49:51 You just agreed 00:49:56 aliseiphone, P!=NP could be 00:50:06 Ffs 00:50:16 WHAT DO YOU THINK INDEPENDANT MEANS 00:50:25 It does not mean unprovable!!! 00:50:37 Well, goodnight for the last time here. <-- good night then ;P 00:50:48 Oh, I see what independent means 00:50:57 Still, considering P?=NP to be independent is a nice mindbender trip: the algorithm version of non-Euclidean geometry 00:51:01 Independent X wrt ZFC: ZFCX and ZFCnotX are consistent relative to ZFC 00:51:03 Same way that C is independent of ZF? 00:51:08 the fact that complexity theory relies so much on diagonalization just begs for the theory to get some of the undecidability that can entail, anyway... 00:51:22 Unprovable X in ZFC: X is true, no proof of X exists 00:51:35 Sgeo: Right. 00:51:41 Bye. 00:51:46 Bye aliseiphone 00:51:46 -!- aliseiphone has quit (Quit: Get Colloquy for iPhone! http://mobile.colloquy.info). 00:52:07 * Sgeo happies 00:52:10 ZFC+CollatzAlwaysHalts vs ZFC+CollatzSometimesLoops 00:52:15 cpressey, isn't the continuum hypothesis unprovable in ZFC too? 00:52:21 Collatz? 00:52:38 Prove that Z is a valid proof of X 00:52:48 AnMaster: yes 00:52:53 it's also independent 00:53:10 Independence is a bit mind-bending 00:53:10 indeed 00:53:16 so why couldn't P!=NP be 00:53:18 (For arbritrary Z and X) 00:53:19 I fail to see that 00:53:41 That would render it a matter of opinion, wouldn't it? 00:53:51 Which seems... unintuitive 00:53:58 But intuition has no place, I guess 00:54:05 AnMaster: to me, it's just that the actual _algorithms_ involved means there are such a large number of different ways it could be independent 00:54:52 and also that unlike the ZF+C and CH cases it doesn't make sense to just pick one possibility and try to use it 00:55:18 oerjan, that's what confuses me a bit about P!=NP being independent 00:55:20 oerjan, hm 00:55:30 because that gives you no actual practical algorithm 00:55:54 oerjan, I love non-constructive proofs. Because they make ehird mad 00:55:58 XD 00:56:00 :D 00:56:20 ehird is actually a constructivist? 00:56:24 I think so 00:56:36 he loves the constructive math used in theorem provers 00:56:48 http://tr.froup.com/tr.pl?1026 00:56:56 oerjan, why can't we have non-constructive theorem provers? 00:57:25 AnMaster: you can have them talk _about_ non-constructive math, of course 00:57:45 oerjan, hm 00:58:14 Night 00:58:15 -!- cpressey has quit (Quit: Leaving.). 00:58:16 but the ones based on higher order functions and the curry-howard isomorphism naturally have constructive stuff at their base, and deals with it more naturally. iiuc. 00:58:27 "If you'd proven constructivism, you could use it in a proof. Since you're not allowing it in proofs, you haven't proven it, and so you must destroy your own constructivist programming! I'm Prozac the Bear!" 00:58:34 Go to Special:Random on Wikipedia and make a esolang about whatever comes up 00:58:48 Sgeo: is that an actual T&R quote? 00:58:51 oerjan, yes 00:59:00 http://tr.froup.com/tr.pl?1027 00:59:30 I read a bit of it, horrible drawing. Worse than xkcd even 00:59:47 That's why the Cornersheep turned evil 00:59:49 and shitty story to boot 00:59:59 Now _that's_ a lie 01:00:11 no, that is subjective :P 01:00:44 I voted the second one. But is there a command-line version of those polls, too? 01:01:14 Open it in elinks? 01:01:15 ... what polls? 01:01:15 AnMaster: the horrible drawing is even used as a meta-joke. all the time. 01:01:32 Sgeo: No that is not what I mean 01:01:33 oerjan, yes, it gets tedious very quickly 01:02:14 I mean like how sprunge has it you can send/receive file using that simple script, with curl 01:04:01 -!- cheater99 has quit (Ping timeout: 240 seconds). 01:04:36 We were talking about Lumenos, which reminded me of Losethos 01:04:42 Anyone want to talk about losethos>/ 01:05:59 What is Losethos? 01:06:16 This junk 01:06:23 erm 01:06:24 http://www.losethos.com/ 01:06:36 zzo38, I repeat my questions: what polls 01:06:47 AnMaster, the one in the topic 01:06:50 AnMaster: See the topic message 01:07:23 clearly it's a brand of cough drops that turned sentient 01:08:18 oerjan, ... what is? 01:08:28 losethos, duh 01:09:16 okay even at the start of the page it fails 01:09:29 you can not enter x86-64 mode without paging enabled 01:10:39 -!- SimonRC has quit (Ping timeout: 265 seconds). 01:11:03 "As anyone will tell you, 01:11:03 Linux is a just a kernel, but GNU/Linux is an operating system." 01:11:11 I mean, some people are of that opinion, iirc 01:11:16 well. That is correct 01:11:20 n/m 01:11:22 >.> 01:11:23 technically 01:11:30 of course, I don't use it like that either 01:11:45 unless I'm trying to be very clear in the context of other userlands 01:11:52 like BSD/Linux 01:11:54 or whatever 01:12:32 I looked at some information about LoseThos now 01:13:57 I would have included a Forth system in it, but that's me.... 01:14:37 I don't like some of the changes that LoseThos has difference from C 01:16:46 Sgeo, about that OS, the author does seem reasonably skilled. Otherwise ey couldn't pull off such a compiler and so on 01:17:38 -!- SimonRC has joined. 01:17:55 -!- BeholdMyGlory has quit (Read error: Connection reset by peer). 01:19:54 any vimmers here? 01:20:40 augur: Do you mean vi improved? 01:20:46 yes 01:20:54 I use it at Free Geek 01:21:49 -!- cheater99 has joined. 01:22:57 which vim do you use? 01:23:22 augur: Whichever one is included in Ubuntu LTSP, since that is what their computers use there 01:23:35 -!- coppro has joined. 01:23:42 ok. 01:24:44 im not sure how i feel about it 01:25:30 augur: I just use it because it is the one installed on there, they use Ubuntu and LTSP for everything there 01:25:54 Although they do not have the "mail" command installed 01:26:21 i like some of the features of it 01:26:34 I also like some of the features of vim 01:26:38 basically the editing featues 01:26:44 c/d 01:26:58 with w/b/$/^/etc 01:28:03 I would do some things differently, when I write Linux distribution I will probably write my own things in a way that is more better in my opinion instead. Like, one thing would be how the status bar is written, I would do it differently 01:28:57 Like, if you are recording a macro called "3" it will display "q3" in the status line, in insert mode it will display "i", it can display "v" for visual mode, and so on 01:29:08 And has position for each one so you can display multiples at once 01:29:12 i hate how in navigation mode the shit navigates by character tho 01:29:21 i find character navigation to be annoying 01:32:10 zzo38, writing a new vim is a separate task from making a new linux distro 01:32:48 * Sgeo uses emacs 01:33:01 same 01:33:03 At least, when I have to use a terminal editor 01:33:08 µemacs quite often 01:34:41 -!- coppro has quit (Remote host closed the connection). 01:34:43 AnMaster: It is, but if I write new Linux distro, I write a lot of new programs as part of that project too, because I can make it many differences, including new window manager, new text editor, new shell, and so on. 01:35:16 new shell? I assume you will have the old one around still 01:35:24 since stuff like configure scripts expect bash 01:35:26 or sh 01:35:38 night → 01:36:44 AnMaster: I don't need configure scripts I can use some feature of CSPIDER (if it is a Enhanced CWEB program), or else enter some configure manually and differently, and then convert it 01:38:01 zzo38, so your system won't be POSIX compatible then 01:38:49 how about textmate 01:38:51 who uses textmate 01:38:59 augur, I never heard of that 01:39:01 what is it? 01:39:04 a mac text editor 01:39:11 augur, no one sane uses mac 01:39:14 so that answers it 01:39:21 AnMaster: I intend to make it enough POSIX compatible that most programs written for it in C (or other ways that compiles into a binary) will still run with little or no modifications, except possibly having to modify makefiles. 01:39:22 alise does! 01:39:29 ok, alise isnt sane 01:39:30 but.. 01:39:31 augur, I rest my case :P 01:39:34 XP 01:39:48 Although the programs should still be changed to improve them to be better designed in the way of Arcane Linux. 01:40:09 theres a plugin for textmate called vimate 01:40:11 nice distro name 01:40:16 that adds on vi-like tools 01:40:30 augur, just use emacs with viper-mode and be done with it :P 01:40:45 i really like textmate 01:40:50 but i like this modal stuff 01:41:04 AnMaster: I think I have also posted on this channel before, about the codenaming scheme I plan to use with Arcane Linux? 01:41:06 -!- derdon has quit (Ping timeout: 265 seconds). 01:41:14 zzo38, no idea 01:41:55 AnMaster: The scheme is that the second version can be codenamed "Illimitable Illithid", the sixth version can be codenamed "Vancouver Island", and so on. 01:42:11 No relation to any other Linux distro codenaming scheme 01:42:32 zzo38, hm? I don't see the logic in your scheme 01:42:51 also "Vancouver Island" doesn't fit the obvious theme established by the first 01:42:55 AnMaster: And the first version can be codenamed "Initial". Now see if you can figure out its logic.... 01:43:13 zzo38, too tired and lazy to do so 01:43:16 If still not, I can give more clues even, or tell you the answer 01:43:19 not even going to try 01:43:47 OK, let's see if anyone else on here will figure it out 01:43:58 night really → 01:44:10 everyone gets it except AnMaster ;D 01:44:23 oerjan: Really? 01:44:45 well everyone i've checked, anyway :D 01:44:56 Including you? 01:45:02 although i think you mentioned it once before 01:45:19 Yes, I did 01:45:23 well given that i'm the only other person i've checked... 01:45:25 I also mentioned that I mentioned it once before. 01:46:51 Does Sgeo figure it out? 01:47:07 -!- Gregor has quit (Ping timeout: 260 seconds). 01:47:22 I wasn't paying attention 01:48:25 Sgeo: Now you can read it and pay attention? 01:48:50 I get it 01:52:58 Some people might be confused because of "Vancouver Island" does not fit the theme established by the first, but actually it does fit the same theme, and in addition "Illimitable Illithid" is actually the second not the first 01:53:10 But that is OK if it is confusing, some people can figure it out or try to figure it out or not! 01:53:14 It is the way. 01:54:12 If you think there is an obvious theme established and then it doesn't fit, well, it is deceptive, it is not the obvious theme! 01:59:13 Please read this and write comment, especially if you play 4E: http://zzo38computer.cjb.net/icosahedral/html/differences.html 01:59:22 -!- Gregor has joined. 01:59:43 (I don't play 4E, I prefer 3.5E) 02:02:24 Icosahedral is the opposite of 4E do you think so? 02:04:05 Do you think "vegitabelation" is a good word for describing 4E? 02:08:02 -!- zzo38 has quit (Remote host closed the connection). 02:17:34 richard stallman, when he was younger, was a total hotty 02:25:37 Make his hair blond and you've got me :P 02:29:33 then or now? x3 02:31:07 Probably neither, I actually have no idea what RMS looked like young :P 02:31:29 I was just applying the HAIR FALLACY (that all men with long hair look alike because that's their only distinguishing feature) 02:32:06 Gregor: Well, he had short hair. And no beard. 02:32:15 Ah :P 02:34:49 and was a total hotty. 02:41:20 More like total nerd. 02:41:32 False dichotomy. 02:41:38 -!- augur has quit (Remote host closed the connection). 02:41:45 -!- augur has joined. 02:44:56 Gregor: Which is, of course, the best kind of dichotomy. 02:45:31 * Sgeo won't be watching Futurama as it airs tonight 02:45:41 Sgeo: FAIL 02:45:58 Anyone spoils me, I'll fake a call made from you to Mom's tech support 02:55:25 -!- Wamanuz has quit (Remote host closed the connection). 03:10:13 -!- cal153 has quit (Ping timeout: 252 seconds). 03:33:30 -!- cal153 has joined. 03:46:30 -!- nooga has joined. 03:57:30 beh 03:57:56 ah 03:58:22 o 04:01:47 huh, do you guys ever sleep? 04:02:02 No. 04:02:28 i've been here almost whole day and it's like 5 in the morning here 04:02:35 and you're still running 04:02:58 Only 22:02. 04:03:05 where? 04:03:13 Missouri. 04:03:24 ah 04:03:43 ok, but oerjan ?! 04:03:53 ZZZ.. what? 04:04:03 -!- sshc has joined. 04:04:22 They don't get sun there anyways; what does he care for circadian rhythm? 04:04:34 * oerjan swats pikhq -----### 04:04:41 nah 04:04:51 in summer they've got everlasting day 04:04:59 my brother complains about that 04:05:02 Still. 04:05:29 mhmh 04:05:33 circadian rhythm means you only wake up every 17 years, right? 04:05:50 anyway 04:05:54 -!- zzo38 has joined. 04:06:36 i just got back from a party and i'm going straight to sleep 04:06:53 nooga: OK 04:07:07 oh sh.. 04:07:21 see ya later 04:07:31 bye 04:07:32 NO^H^HOK 04:11:59 -!- nooga has quit (Ping timeout: 265 seconds). 04:12:46 For 17 years? 04:15:26 What about 17 years? 04:16:22 O, I saw the log file now 04:16:37 I do not know the answer of that question, however. 04:22:10 and another pun goes *whoosh*, i see 04:23:04 (now it _does_ rely on you knowing a certain thing from biology) 04:41:02 I don't know biology much, I mostly study physics instead 04:41:33 guys, opinion needed 04:41:45 wisdom required tho likely to go unfound 04:42:25 Opinion of what? 04:43:20 I found that the negative array sizes trick in C is not only something I used, but now I also found it mentioned in section 5.10.1 of the GNU autoconf manual. 04:44:40 zzo38: unification 04:45:11 Unification of what? 04:45:22 objects 04:45:23 like 04:45:45 if X and Y are variables, [1,X] = [Y,2] unifies with X = 1, Y = 2 04:46:21 or say 04:46:31 [X,X] = [1,2] fails to unify 04:46:47 OK, I think I understand. 04:46:54 right? so, i have more of an opinion question than anything 04:47:11 if we define ~p to be the objects that *dont* unify with p 04:47:12 But in the second case, what about if you use the Perl6 feature to have multiple values at once (I think Perl6 does this?)? 04:47:19 no. 04:47:31 so then 2 unifies with ~1 04:47:37 because ~1 means just "not 1" 04:47:44 anything except 1 will unify with ~1 04:47:49 OK 04:48:00 ~X will unify with nothing, because everything unifies with X 04:48:01 Now I understand more better. 04:48:13 but heres the question, right 04:48:26 [1,2] = ~[2,3] will ofcourse work 04:48:39 [1,2] unifies with ~[2,3] because [1,2] does NOT unify with [2,3] 04:48:57 Yes 04:49:00 and similarly, [1,2] unifies with ~[2,X] because [1,2] cannot unify with [2,X] 04:49:15 but suppose i did 04:49:22 [1,2] = ~[1,X] 04:49:37 [1,2] does definitely unify with [1,X] 04:49:39 yielding X = 2 04:49:53 OK 04:50:41 but what about [1,X] ~[1,2] 04:50:54 [1,X] will unify with [1,2] yielding X = 2 04:51:06 so does that mean that [1,X] = ~[1,2] fails 04:51:19 or does it mean that it succeeds yielding X = ~2 04:51:20 ? 04:51:32 Yes that is the difficult 04:52:10 Let me see...... 04:52:48 i think in spirit the intent is that ~[1,2] should be anything but [1,2], right 04:52:54 I suppose you have to start at one thing at first and then break it in parts? 04:52:55 but dos that include variable bindings with it 04:53:10 Using only the valid ways of breaking the parts 04:53:11 well yes, i mean, i know how i'd get it to yield X = ~2 04:53:18 the issue is whether it SHOULD or not 04:54:56 I guess if ~[1,2] corresponds to anything other than [1,2] then [1,X] can unify [1,1] and [1,3] and [1,4] and so on, so it is ~2 but you should possibly ignore the cases [2,2] and [2,3] because those don't unify? I really don't know 04:55:19 Who controls esolangs.org? graue? 04:56:50 Gregor: I don't know. Does esolangs.org have a WHOIS service? 04:58:44 augur: Therefore it should be able to yield X = ~2 (I think) 04:58:52 ok 05:00:01 In that case, also [1,X] = ~[2,2] means anything other than [2,2] so [1,X] = [1,2] is also OK and now it unifies everything, this way? 05:02:34 yeah, see, thats the tricky thing, right 05:02:40 -!- Gregor-P has quit (Ping timeout: 246 seconds). 05:03:25 the solution i have for [1,X] = ~[1,2] is to turn ~[1,2] into the disjunction [~1,2] | [1,~2] | [~1,~2] 05:03:43 Yes, because if [1,X] = ~[1,2] yields X = ~2 then [1,X] = ~[2,2] should unify everything? 05:03:52 which would succeed as [1,X] = [1,~2] 05:03:54 so X = ~2 05:04:04 Yes that works 05:04:19 but the same trick with ~[2,2] is to turn it into [~2,2] | [2,~2] | [~2,~2] 05:04:43 so [1,X] succeeds against [~2,2] and [~2,~2] 05:04:53 so X can be either 2 or ~2 05:05:01 which basically means anything at all 05:05:13 as it should 05:05:27 Yes, it is what I meant so it is OK 05:05:35 Do you need a special notation for this or is not needed? 05:05:46 im just using ~p 05:06:09 OK 05:08:49 I have posted first version of Enhanced CWEB by now. Enhanced CWEB version 0.1 05:09:28 CWEB? 05:10:37 http://zzo38computer.cjb.net/cweb/ 05:10:57 You can tell me if is good, comment, or any feature suggestion, question, etc 05:11:04 Or complain, if you really want to 05:11:32 augur: Have you ever used CWEB? 05:11:42 Now it is Enhanced CWEB, it is the improved version! 05:13:35 If you have a program to print DVI files, you can just print out the DVI files to read it 05:16:02 ive never heard of cweb before x3 05:16:41 Do you have TeX? 05:17:32 yes 05:17:36 well, a version of it 05:18:41 -!- CakeProphet has quit (Ping timeout: 265 seconds). 05:18:47 Then probably you can print a DVI file? There is one new file called CSPIDER that is the new file I added (although I have also made many modifications to the other files). Please tell me if you have any comment about it. 05:18:56 What version of it do you have? 05:19:35 but what is SWEB 05:19:38 CWEB, even 05:20:04 (Please note that .ZIP is not a tape archive, so you have to create a directory first before expanding, unlike tape archives where the directory will be created automatically) 05:20:33 CWEB is a system that does various things, one thing it does is allows you to print out the programs using TeX. 05:20:41 But it has other features too. 05:20:44 what programs 05:20:44 zzo38: ... that is neither a property of ZIPs nor TARs. 05:20:54 That's just a convention. 05:21:04 Gregor: Yes, I know it is just a convention. 05:21:35 What I mean is that the convention for ZIPs is that you should create the directory yourself, while the convention for tape archives is that you don't need to, because the tape archive will do it for you. 05:22:14 The Wikipedia article for CWEB has a list of features (I wrote the list of features): http://en.wikipedia.org/wiki/CWEB But Enhanced CWEB has additional features that are not listed on Wikipedia. 05:23:02 Welp. Time to install Debian on my phone. 05:23:25 One of the new features in Enhanced CWEB is metamacros, although there are others, too. 05:24:53 (The codes for processing metamacros is in common.w (and common.dvi for printing), while the file cspider.w uses a few metamacros.) 05:25:45 Now do you understand it a bit more? 05:27:59 -!- zzo38 has quit (Read error: Connection reset by peer). 05:28:37 -!- zzo38 has joined. 05:35:38 12.5 hours! 05:39:04 -!- oerjan has quit (Quit: leaving). 05:45:43 -!- zzo38 has quit (Remote host closed the connection). 06:30:30 Android has chroot but not mkdir. 06:30:32 Wow. 06:31:25 Ohhh, it's in a weird path. 06:31:26 Strange. 06:43:45 -!- Sgeo has quit (Ping timeout: 265 seconds). 06:43:55 -!- Sgeo has joined. 06:45:44 Gettin' Debian on my phoooooone 8-D 06:47:22 What about Win95? 06:48:27 That's only been done in an emulator. 06:48:29 BORING. 07:13:05 * Sgeo deliberately does not invite Gregor or any other #esoteric er to an event 07:31:38 -!- relet has quit (Quit: Leaving.). 07:39:13 -!- Flonk has joined. 07:40:22 An esoteric "errr..." 07:41:15 good morning everybody :) 07:41:29 And on the other hand: at least my phone has Debian on a Debian. (In a matter of fashion, anyway.) 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:11:51 -!- Deewiant has quit (Ping timeout: 276 seconds). 08:13:58 -!- Deewiant has joined. 08:52:37 -!- BeholdMyGlory has joined. 09:24:02 -!- MigoMipo has joined. 09:30:04 -!- BeholdMyGlory has quit (Remote host closed the connection). 09:41:38 -!- MigoMipo_ has joined. 09:41:43 -!- tombom has joined. 09:44:51 -!- MigoMipo has quit (Ping timeout: 264 seconds). 09:45:41 -!- MigoMipo__ has joined. 09:45:47 -!- MigoMipo__ has quit (Remote host closed the connection). 09:48:59 -!- MigoMipo_ has quit (Ping timeout: 252 seconds). 10:22:11 -!- Phantom_Hoover has joined. 10:24:01 -!- MigoMipo has joined. 10:26:22 -!- CakeProphet has joined. 10:34:38 Kurt Gödel looks delightfully evil... 10:37:04 He would be an ideal candidate for a Halloween costume. 10:43:54 -!- R4ven has joined. 10:44:03 Hello! 10:45:33 -.- 10:45:59 hi 10:46:03 * Sgeo needs to go back to sleep 10:46:41 -!- R4ven has left (?). 10:52:33 -!- augur has quit (Remote host closed the connection). 10:52:36 Well, that was odd. 10:52:48 -!- augur has joined. 11:29:02 -!- MigoMipo_ has joined. 11:30:00 -!- ais523 has joined. 11:32:18 -!- MigoMipo has quit (Ping timeout: 240 seconds). 12:09:27 -!- ais523 has quit (Remote host closed the connection). 12:12:28 -!- MigoMipo__ has joined. 12:16:18 -!- MigoMipo_ has quit (Ping timeout: 265 seconds). 12:18:30 vrml. How does one view it? For historical computing purposes. 12:19:08 tools available to me are: opengenera, ubuntu 7.10, ubuntu 10.04, current arch linux 12:21:43 ah, nvm found it 12:23:55 VRML? 12:29:21 -!- Wamanuz has joined. 12:30:17 Phantom_Hoover: It's like HTML but for 3D. :p 12:30:27 Cool... 12:30:46 It's also pretty 1990s. 12:32:12 Ah, the 90s. 12:34:22 A technically-speaking human-writable ASCII-based syntax for describing 3D scenes with the usual sort of primitives, with some hyperlinking stuff and simplish even-scripting stuff added, meant to be viewed in a browser. 12:35:58 They also made a XML-based successor, X3D. I'm not sure if it was ever used much. 12:36:09 And there's some more recent competitors in the same space. 12:41:30 How about we do an N-dimensional version? 12:41:42 Or a non-Euclidean one! 12:42:27 -!- tombom has quit (Ping timeout: 264 seconds). 13:18:16 -!- nooga has joined. 13:20:04 -!- alise has joined. 13:20:10 O the night. 13:20:34 What about it? 13:20:44 This table wobbles! 13:20:50 Phantom_Hoover: Alack alack alack? 13:21:18 Try sawing one of its legs off. 13:21:37 It has only two, and they support the other; wobble wobble wobble. How does one work on this? I recall doing it on previous weekends. 13:21:55 So, I'm doing cpp lists. 13:21:58 With nil. 13:23:04 How do you even make a table with 2 legs? 13:23:29 I dunno. It has two metal bars on the floor horizontally. 13:23:35 And the whole bar curves up and supports the table. 13:23:39 It's for putting over your bed, say. 13:24:24 Agh, what is up with my mouse? 13:24:29 Clicks aren't being recognised. 13:24:31 Well, they are, but wrongly 13:24:34 Ah, there. 13:25:32 list.h:18:15: error: macro "true" passed 10 arguments, but takes just 2 13:25:32 list.h:18:15: error: macro "tl_" requires 3 arguments, but only 1 given 13:25:32 list.h:18:15: error: macro "hd_" requires 3 arguments, but only 1 given 13:25:46 Damn __VA_ARGS__. 13:26:16 Phantom_Hoover: hey how did you invoke cpp? 13:26:23 the #define f(x) f_(x) trick isn't working for me 13:27:01 Type "cpp". 13:27:10 Gimme your transcripted list.h, please? 13:27:31 #define hd_(x,...) x 13:27:31 #define hd(x) hd_(x) 13:27:31 #define tl_(x,...) __VA_ARGS__ 13:27:31 #define tl(x) tl_(x) 13:27:31 #define cons(x,y) x,y 13:27:31 #define foo cons(a,cons(b,cons(c,d))) 13:27:33 hd(tl(foo)) 13:27:45 "Erik entered Dalhousie University in Canada at the age of 12, and completed his bachelor's degree when only 14" 13:27:48 Try doing the spacing thing. 13:27:50 Fuck you, Erik. 13:27:53 Phantom_Hoover: Did. 13:27:56 gcc --version? 13:28:04 cpp (Ubuntu 4.4.3-4ubuntu5) 4.4.3 13:28:17 Me too. Hm. So my code has a bug. 13:28:24 I'm doing it with cpp, rather than gcc, though. 13:28:29 yeah 13:28:40 ha, I defined comments in the language 13:28:43 #define rem(...) 13:29:18 -!- augur has quit (Remote host closed the connection). 13:29:19 -!- cpressey has joined. 13:29:29 -!- augur has joined. 13:29:41 Hi cpressey! 13:29:45 alise, GNU cpp strips out comments by itself, though. 13:29:52 Phantom_Hoover: But that's not standard. 13:30:15 I'm writing in Industry Standard C Preprocessor. 13:30:19 alise: good morning 13:30:32 cpressey: It's after noon. 13:31:01 time is all like relative and shit man, especially THIS early 13:31:24 7 o'clock EST? 13:31:47 7:30 cst 13:32:02 or cdt or some such 13:32:03 how can you be awake at such a time 13:32:22 why can't america be like china and adopt a single timezone 13:32:23 AST 13:32:55 then life would be simple 13:32:56 couldn't sleep, have work stuff (a release schedule) that should have been done yesterday, but due to "fun" in testing, it wasn't 13:33:09 so here i am 13:33:23 -!- MigoMipo__ has quit (Remote host closed the connection). 13:34:18 why do people who like programming work as programmers? 13:34:20 it's bizarre. 13:34:26 alise: for whatever it's worth, i think cpp can be shown tc. i hold out hope, anyway. 13:34:34 it is bizarre 13:34:36 alise, zuh? 13:34:51 i would go for the cyclic tag system 13:35:01 Phantom_Hoover: programming-as-a-job is widely recognised to be the most kafkaesque, soul-draining sludge of time there is 13:35:01 There's nothing standard as to what a "preprocessor" separately invoked must do (since it doesn't need to be a separate part), but /* ... */ comments need to be replaced by a single space character before preprocessing directives and macro-expansion in any standard C implementation. 13:35:13 cpressey: i'm developing "standard conveniences" first to explore the space 13:35:31 alise, oh, you mean professional development. 13:35:33 fizzie: Well, whatever; rem() is funner. 13:35:40 Phantom_Hoover: Yeah. 13:37:23 alise, that's why software sucks, really. 13:37:25 i been thinking about the difference between "true but unprovable" and "independent" and now i hate the universe 13:38:05 cpressey: if we consider T the set of truths in some theory: 13:38:10 *theory, 13:38:32 and P the set of (statement,proof_text)s in the theory 13:38:33 then 13:38:51 My god, old computers were awesome. 13:38:54 truebutunprovable(x) := x in T /\ ~(exists p such that (x,p) in P) 13:39:13 independent(x) := x not in T /\ ~x not in T (ergo T with added axiom of x or ~x are both consistent) 13:39:48 say you have an axiom system z. say x is true but unprovable in z. then z+!x is consistent, even though x is true. therefore i hate the universe 13:40:20 alise: you are so weird 13:40:25 alise: because you do prolog 13:40:28 cpressey: well true but unprovable is of course terribly confusing. 13:40:32 just had to get that out in the open 13:40:36 Quadrescence: i do prolog? 13:40:40 yeah 13:40:53 Quadrescence: what prolog in particular have I done to make you say that? 13:41:04 alise: i don't know 13:41:07 just whatever prolog 13:41:16 how do you even know i do prolog 13:41:22 i only do it as a total newbie btw 13:41:31 Phantom_Hoover: Look, a 2160-bit memory module: http://en.wikipedia.org/wiki/File:Mercury_memory.jpg 13:41:47 -!- MigoMipo has joined. 13:41:50 (That's 2160 bits as in total capacity, not bus width or anything else.) 13:42:02 That is incredibly cool. 13:42:18 How big is it? 13:42:19 "Please Do Not Touch" 13:42:23 I hope that was on the original computer. 13:43:00 Phantom_Hoover: The article says that the whole 1000-word memory subsystem (7 of those units) had its own walk-in room, so presumably not especially small. 13:43:08 Oh, it's a mercury delay line. 13:43:22 alise: i know you do prolog because idk 13:43:33 Quadrescence: Well, I don't really do it any more. :P 13:43:37 fizzie: look up "memory plane" 13:43:52 * Phantom_Hoover is very happy that his wild guess as to the technical name was correct 13:43:58 http://www.youtube.com/watch?v=kmRtNmnRdNI 13:44:18 Quadrescence: Pah, Edison. 13:44:23 That UNIVAC tank is sequential-access, of course. 13:45:15 (Or random-access if you don't mind a bit of waiting, but..) 13:45:46 fizzie, you could use that for pipes, come to think if it. 13:46:31 Because you want your conceptual pipes involve real pipes somewhere? 13:46:41 YES 13:46:45 "It's not a pipe unless fluids are involved." 13:47:00 Even better if it uses ball bearings 13:47:21 Just for transmission of oscillations, of course. 13:47:47 list.h:16:29: warning: __VA_ARGS__ can only appear in the expansion of a C99 variadic macro 13:47:51 Oh, lah-de-dah. 13:48:17 love the word "variadic", it sounds so scientific-y. 13:48:26 alise, we're using the same version and code, aren't we? 13:48:35 Phantom_Hoover: Not code. 13:48:37 I'm fiddlin'. 13:48:41 #define tl_(d,h,...) __VA_ARGS__ 13:48:51 How is that NOT a variadic macro, GNU Stupid Crapshoot 13:49:08 oh 13:49:11 it's in my rem line 13:49:11 lol 13:49:25 cpressey: how about p-adic 13:49:29 does that sounds sciency 13:49:38 alise: boycott gcc 13:49:43 Quadrescence: wat 13:49:46 Quadrescence: ironically, it sounds less so 13:49:53 you can't ... boycott a project 13:49:56 it sounds totally made up 13:50:01 alise: wanna bet 13:50:04 yes 13:50:06 *yes. 13:50:11 I am boycotting it right now 13:50:34 if everyone boycotts gcc they will starve... for attention 13:50:44 Quadrescence, why boycott GCC? 13:51:17 GCC is a HACK!!! 13:51:27 and stuff 13:51:36 GNU Stupid Crapshoot 13:53:02 What is actually *wrong* with it/ 13:53:14 gcc is crap but Quadrescence has gone crazy 13:53:18 Well, other than the fact that its assembler is stupid. 13:53:30 Phantom_Hoover: terrible, huge, huge code (over 100,000 lines -- how many bugs do you want to bet are in there); 13:53:39 simply inferior to well-designed compilers like pcc and clang 13:53:50 it's a huge crapshoot from the 80s 13:54:06 alise, is there anything mainstream that you *don't* think is crap? 13:54:07 "cpp -std=c99 -Wall -pedantic -undef -nostdinc -P" 13:54:09 cpp the interpreter! 13:54:16 Phantom_Hoover: yes; but note that my opinion on gcc is not unusual 13:54:21 and i think you'll find most competent people share it 13:54:29 theo de raat does, in case you care about him 13:54:33 i am fascinated by how the mainstream attracts crap and vice versa 13:54:35 FreeBSD, too, want to move away from gcc 13:54:50 and not just for licensing reasons. 13:54:56 most of the bsds would rather not have gpl'ed stuff in their base 13:54:58 gcc needs to die a painful death. but meh. 13:54:59 llvm? 13:54:59 cpressey: indeed 13:55:03 cpressey: but they have technical reasons too 13:55:08 nooga: clang is a C compiler for LLVM. 13:55:09 alise: granted 13:55:11 I mentioned clang. 13:55:22 ah 13:55:24 OK, so it's my conditional that's failing. 13:55:56 You know, denouncing everything as crap gets boring after a while. 13:56:23 "You criticise a lot of things! Therefore you should stop making criticisms because I believe excess criticisms make future ones incorrect." 13:56:40 Would you have told Sturgeon to stop being such a pessimist? 13:57:27 No, I just wish you didn't do it so much. 13:57:30 But listen, my opinion on gcc is nothing radical. 13:57:35 If you actually studied it you would agree. 13:58:05 What does it actually do wrong? 13:58:39 It's horribly architectured. The code is written badly and it is very, very long with lots of unchecked possibilities for bugs. It's crap; what the heck do I need to elaborate on? 13:58:59 It's coded badly, it's buggy, it's huge, it's slow, it gives useless error messages, and it doesn't optimise well at all. 13:59:10 alise: you hit the nail on the head 13:59:13 buggy huge slow 13:59:14 These flaws are not present in other compilers like pcc and clang. 13:59:22 gunked up with stallman's foot debris 13:59:26 Age is the only excuse; and gcc has had plenty of time to get into this century. 13:59:37 yet the market determines TRUTH and pcc and clang are MARKET FAILURES!!1! 13:59:38 it's not even fully c99 compliant 13:59:42 which is pretty sad 13:59:43 The best thing it could do with its time is to die quietly. 13:59:52 1999, 11.5 years to get it done 14:00:08 alise: that won't happen though 14:00:10 Quadrescence: but dude, C99 is, like, impossible. 14:00:15 It has complex numbers and shit. 14:00:21 fuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuk 14:00:24 i always forget about that 14:00:29 alise, why are complexes impossible? 14:00:32 Quadrescence: my sentiment exactly 14:00:34 They're complex, dude. 14:00:43 "And what has he done with said knowledge? I'm waiting to see the list of OSS projects he's working on." -- reddit moron about the guy who got a bachelor's at 14 14:01:05 Not impossible, out of place. What part of "systems construction language" does C99 not understand? 14:01:15 -!- Mathnerd314 has quit (Disconnected by services). 14:01:17 I was joking wrt complexes. 14:01:25 I don't mind having complex numbers in the language although C99 did do them terribly. 14:01:35 -!- Mathnerd314_ has joined. 14:01:50 -!- Mathnerd314_ has changed nick to Mathnerd314. 14:02:28 -!- Phantom_Hoover has quit (Quit: Leaving). 14:02:41 -!- Phantom_Hoover has joined. 14:02:43 alise: c99 did a lot terribly to be honest. Like booleans are a god damn hack 14:02:56 yeah 14:02:57 I know it's a very basic thing, but still 14:03:02 the committee are retards 14:03:14 * Phantom_Hoover has come to the conclusion that his internet connection is nondeterministic. 14:03:23 C89 is pretty ok. It is not the best by today's standards but It's The Best We Got 14:03:53 "And what has he done..." <-- wow, i find that surprisingly offensive in a way i cannot readily describe 14:04:07 the quote, not alise 14:04:27 yeah 14:04:35 there's a bunch of people like that on reddit 14:04:41 Phantom_Hoover: really? try downloading from all sites at once 14:04:42 seriously, they don't care about you unless you work on FOSS. 14:04:53 the worst thing about FOSS is everyone who likes FOSS. 14:04:56 9c is awesome 14:05:04 nooga: yes, indeed 14:05:13 8c 14:05:15 uh 14:05:16 but i was using conventional compilers to illustrate my point 14:05:25 nooga: 2c is the manpage :P 14:06:50 -!- Mathnerd314 has quit (Ping timeout: 260 seconds). 14:07:59 grr 14:08:06 anyone here any good at cpp? 14:08:09 cpressey maybe? 14:08:26 "good" is not the word I'd use. I wrote macros that pasted tokens together once! 14:08:36 What's wrong? 14:09:28 I'm not even sure. 14:10:12 Phantom_Hoover: http://pastie.org/1067011.txt?key=bxdsdyraapk8c9zshrrwcw 14:10:17 see the last two lines, compare their outputs 14:11:36 What are you trying to do? 14:12:04 The second line should yield true,c,true,d,false,nil,nil. 14:12:11 I don't see how the first line works but the second doesn't. 14:12:45 Nor do I. 14:13:25 I think it's because cpp keeps track of what was typed in and what was expanded 14:13:27 so as not to recursively expand 14:13:30 if so, whyyyy 14:14:10 i suppose i should try this 14:14:35 oh, wait, yeah, cpp doesn't recursively expand 14:14:37 cpressey: you're grumpy like me when you don't use caps 14:14:39 cpressey: yeah 14:14:43 cpressey: which makes my conditionals not work 14:14:47 back to the drawing board for them 14:14:53 i think the causality might be backwards there re caps, but yes. 14:15:02 :-D 14:15:07 screw you, shift key 14:15:13 causality doesn't exist don't you know 14:15:30 screw you, alise 14:15:37 :D 14:15:53 "Screw you! I want my fucking A following B!" 14:16:37 maybe define tltl(x,y,...) to do it? you should only need a fixed number of tl's in any tc proof 14:16:54 well not ANY tc proof. you know what i mean 14:18:07 cpressey: i'm thinking for things like map 14:18:11 which would suggest power 14:20:23 yes, but... recursion is out, it looks like. 14:21:41 Yes; I gathered that. But I have no recursion, even! 14:21:53 In fact this restriction of cpp is the stupidest thing ever. I can think of no reason for it. 14:22:17 Recently, like in the past few months, cygwin wants to install three dozen new packages every time I upgrade. Including 'bc' and 'ed' and 'libglitz'. The deps for something I have installed are clearly hosed. 14:22:37 alise, I suppose you aren't fond of Linux either. 14:22:44 alise: um, well. it's nice when you don't have to worry about your macros not terminating 14:23:14 Phantom_Hoover: I use it, but ... no. Linus is one of my heroes, he's a great guy. But Linux is another entry in the Unix tradition. And it has some braindamage (usually by people other than Linus). 14:23:34 cpressey: at expense of fucked up, barely explainable behaviour ... who writes macros complex enough that they'd recurse without this anyway? I mean, what? 14:23:42 So you don't like the Unix tradition? 14:23:43 cpp is instant anyway, if it takes more than a second on normal macros you'd know you have a loop. 14:24:03 Phantom_Hoover: I don't. The original Unix was cool, and Plan 9, the successor to Unix by the same people, is genius; but modern Unix is, well, a crapshoot. 14:24:18 ...Why? 14:24:20 alise: hell, I dunno. who writes normal C code that relies on wrapping around at the bit size? ask Gregor. 14:24:20 And no, I'm not merely a malcontent; I just took a good, long, hard look at things and regret it a bit because I'm basically an outcast now. 14:24:43 Phantom_Hoover: Complexity, bugginess, non-orthogonality, unpredictability, lack of unified coherent interface to anything (Plan 9 rectifies this), years of historical cruft... 14:24:53 ...makes a decent UI impossible... 14:25:00 cpressey: anyone who writes a BF interpreter? :P 14:25:02 alise: You've read the Worse is Better essay(s), I take it? 14:25:08 cpressey: I have. 14:25:15 k 14:25:16 And I don't believe modern Unix follows Worse is Better. 14:25:21 It follows Better is Worse. 14:26:11 But I'm straying dangerously into my old zealot turf here; so I'll quiet down about it for a bit. Probably. 14:26:34 i will meditate on BiW. i probably know what you mean. 14:27:56 cpressey: Modern Unix is "MIT-style code" -- overcomplicated error-checking monstrosities -- over a New Jersey base; so it's bloated but in the end it fails hard. 14:28:14 I will try not to get angry over things that work adequately for my purposes. 14:28:18 It's a "perfect" implementation of a quick-and-dirty philosophy, it just doesn't work ... it's not very explainable. Better is Worse. You know what i mean. 14:28:32 Phantom_Hoover: I'm not angry. I just don't like it. Linux does make me angry, though; you know, when it doesn't work, which is often. 14:29:22 I take the approach of giving up easily. 14:29:35 alise: Yes. I think I've been BiW'ed silly working with Python, actually. 14:29:58 I dislike Python but tend to bury that anger because it's useful to glue shit together on Unix. 14:30:01 We're going to start with naff, then get it RIGHT! 14:30:08 It /is/ terribly naff. 14:30:17 Naff is a very good way to describe a lot of things in today's computing world. 14:30:30 Sure, it [...], but... do you really want to? 14:30:35 So what was it like in The Golden Age, then? 14:30:51 Phantom_Hoover: Didn't ever been none Golden Age. Always been shit. Probably always will be. 14:31:22 So why complain that it's shit all the time, then? 14:31:36 You have no non-shit against which to compare. 14:32:02 I think there is a distinction between beauty and efficacy. Things that "get the job done" tend to be ugly. And ugly things tend to work badly *in the long run*. But most people don't have a significant amount of foresight, and they will defend the ugly things that "get the job done" tooth and nail. In business, especially. 14:32:05 I do but it's all ideas, nothing concrete. However, there are shards of perfection present in many things that as a whole are not; and together they form a fucked-up collage that nevertheless represents something perfect, if you can make the glue stick. 14:33:11 You know, when I don't think about this shit I'm positive enough. 14:33:43 * Phantom_Hoover wonders what DMM's 20 questions poll will end with. 14:34:02 You know, I never had cpressey down as the grumpy cynic everything-sucks type. 14:34:19 I always assumed he'd be the happy-camper-on-LSD-ah-it-sucks-so-what-i-just-make-cool-stuff type. 14:35:45 I would be that if it weren't for the LSD, cool stuff and the camping. And the happy. 14:36:05 And the cool stuff? 14:36:48 Second on the list. 14:37:35 alise: Hm. Well, ... 14:37:35 Oh yeah. 14:37:55 I've gone back and forth during my life, actually. 14:38:25 cpressey: I take it most of your esolang work was done in the happy-camper phase :-P 14:38:30 Hey, that means there's hope for me to rebound yet. 14:38:31 Hooray. 14:38:51 In other news, I have decided to break into my friend's house and steal his father's old BBC Micro. 14:38:58 I've had periods where I've been extremely depressed about how much everything sucks, and I've had periods where I've not cared at all and was just elated that the universe was so weird. 14:39:05 Unfortunately, I have only a vague idea where he lives. 14:39:25 cpressey: Hell, I don't get depressed about it. It's just fucking computers. Or do you mean /everything/ everything? 14:39:38 And yes, most esolang work occurs during the latter half of that cycle. 14:39:53 Yes, not just computers. Everything. Systems. 14:39:56 Are you certain you're not bipolar? :P 14:40:22 Sorry, that was nasty. 14:41:27 I know which part of Edinburgh his flat is in and I know that it's above a Chinese restaurant. 14:41:40 I may have to break into a lot of houses. 14:44:03 "It's in the part of Scotland where it usually rains. Oh, and it's next to a DFS with a sale on; you can't miss it." 14:44:32 I have a vague idea as to the name of the Chinese restaraunt. 14:44:43 a/arau/aura/ 14:44:47 s/a/s/ 14:44:52 Recursive sed! 14:47:14 http://www.dangermouse.net/media/memento.html 14:47:21 David Morgan-Mar watches Memento for the first time. Backwards. 14:47:29 Speaking of why CPP won't recursively expand macros, it's because it's forbidden to: "If the name of the macro being replaced is found during this scan [after the first round of substitution] of the replacement list -- it is not replaced. Furthermore, if any nested replacements encounter the name of the macro being replaced, it is not replaced. These nonreplaced macro name preprocessing tokens are no longer available for further replacement even if they are lat 14:47:30 er (re)examined in contexts in which that macro name preprocessing token would otherwise have been replaced." 14:47:36 fizzie: Yes. 14:47:39 It's a stupid restriction. 14:48:03 Well, you wouldn't want to have scary LOOPS in the preprocessor. 14:48:03 alise, btw you said something about cpp and TC. Got anywhere with that? 14:48:16 iirc it is possibly TC with recursive includes or such 14:48:31 AnMaster: I'm "working on it". 14:48:32 alise: I can't be certain, I suppose, but I strongly doubt it. Seems to be tied much more strongly to what's happening in my world and how I look at it. 14:48:34 alise, ah 14:48:38 Do we hate M4 as well? 14:48:41 I'm trying to define lists; run into a roadblock but I think I can fix it. 14:48:47 alise, your way doesn't include recursive includes? 14:48:47 Phantom_Hoover: Who /doesn't/? 14:48:53 AnMaster: hm? 14:48:59 I think it's cool in a theoretical way. 14:49:09 cpressey: I try and stay positive. :P 14:49:11 alise, do you do recursive includes or not when making it TC? 14:49:17 or proving it rather 14:49:41 AnMaster: I will, yes. 14:49:46 I haven't got a proof yet. 14:49:48 Just a strong hunch. 14:51:20 So wait, what preprocessors *do* we like? 14:51:48 Phantom_Hoover: None? Preprocessors are a pretty pointless concept :-P 14:51:56 M4 is more a template language from hell. 14:51:59 Anyway, cpp works fine for C. 14:52:09 OK, what macro systems do we like? 14:52:24 I still hold that M4 is a good esolang. 14:52:59 alise, what about lisp macros. Not really pre-processor as such thoughh 14:53:02 though* 14:54:08 M4 is a good esolang, not a good lang. 14:54:13 AnMaster, they're preprocessing done right. 14:54:20 What Phantom_Hoover said 14:54:22 *said. 14:54:23 M4+dc 14:54:30 cpressey, OMG. 14:54:43 how do you disable indentation in cpp mode? 14:54:46 erm 14:54:47 in cc-mode 14:54:47 Surprsed I haven't seen something esosomething implemented in that combo 14:54:47 in emacs 14:54:49 autoindent 14:54:57 Use vim? 14:54:59 like when I press ( 14:55:12 DISCLAIMER: I do not particularly like vim. 14:56:25 Just M-x fundamental-mode. :p 14:57:44 But keeping syntax highlighting. Some electric thing. 14:57:50 alise: maybe i'll look LATER 14:58:25 AnMaster, they're preprocessing done right. <-- not a separate pre-processor though. Runs in the compiler iirc 14:58:25 Perhaps you could just set c-basic-offset to 0. 14:58:27 * Phantom_Hoover always wonders why Lisp did things right then everyone went and got it wrong for the next few decades. 14:58:34 on the AST I think 14:58:36 (It could still line up things, I guess.) 14:58:50 AnMaster, 6:dozen/2 14:58:57 Phantom_Hoover, ?? 14:59:10 6 to half a dozen? 14:59:39 Phantom_Hoover: WANT TO KNOW WHY? 14:59:55 Why what? 15:00:04 I want to know why, whatever it is. 15:00:11 PHANTOM WHY LISP DID THINGS RIGHT 15:00:15 THEN ----------> WRONG 15:00:37 http://www.lambdassociates.org/blog/bipolar.htm 15:00:51 Phantom_Hoover, and context? 15:01:10 on the AST I think ;; no ast, just the concrete tree; the lists 15:01:21 AnMaster, you know Lisp, right? 15:01:24 PHANTOM_HOOVER: READ THE ABOVE LINK, AFTER, READ http://steve-yegge.blogspot.com/2006/04/lisp-is-not-acceptable-lisp.html 15:01:44 oh that article 15:01:45 http://www.lambdassociates.org/blog/bipolar.htm 15:01:46 it's shit 15:02:03 no 15:02:07 It is actually very good 15:02:09 it's just the qi guy whining about how programmers are so misunderstood then giving it the label "bipolar" and thereby furthering the continued uselessness of psychological terminology. 15:02:11 And very accurate 15:02:25 it's shit and if you believe it you're another lisp elitist that's full of themselves 15:02:27 NO it is good and accurate 15:02:34 the yegge article is good however. 15:02:37 Qi sucks though 15:02:40 i must say that 15:02:46 the syntax is terrible 15:02:46 What? It's funny! 15:02:49 what's "qi"? 15:02:50 (((BLA BLA BLA 15:02:52 ----------------------------------- 15:02:55 cpressey, quiz show. 15:02:58 bLA BLA BLA))) 15:03:00 Phantom_Hoover: you fail 15:03:02 cpressey: a powerful type system thing grafted onto common lisp 15:03:03 It has Stephen Fry. 15:03:08 cpressey: it has a turing complete type system without dependent typing 15:03:09 alise, I grasped that. 15:03:15 which is hilariously fuck-uppedly pointless 15:03:19 yes 15:03:21 what alise said 15:03:36 alise: but really, that article is pretty accurate. Maybe not for all lisp programmers 15:03:41 But a good number of them 15:03:59 I can't remember the last time I've seen a complete, good, full piece of working lisp 15:04:36 I object to the way it is written and the way the author perceives the immense genius of Lispers. I also dislike the blatant misuse of the term "bipolar", which is unacceptable. 15:04:51 The author is also a bad writer. 15:04:52 Yeah, the use of bipolar is unacceptable 15:04:57 I thought the writing was appropriate 15:05:08 alise is in denial 15:05:14 alise has bipolar 15:05:17 Ack, it's Lisp with pattern matching. 15:05:19 go take your HALDOL alise 15:05:24 Phantom_Hoover: no, it's more than that 15:05:35 Quadrescence, that's probably terribly offensive. 15:05:35 it's a whole different game 15:05:36 Quadrescence: I just sort of wish he'd render "it's" or "let's" properly at least once. 15:06:01 Phantom_Hoover: what, saying someone has a medical condition 15:06:10 and then recommending taking an anti-psychotic medication 15:06:24 honestly i do not think saying someone has a medical condition is an insult 15:06:32 HEY ALISE YOU HAVE THE COMMON COLD 15:06:36 GO TAKE COUGH SYRUP 15:06:39 *gasp* 15:07:14 Quadrescence: Considering that I'm wrongly in an institution and they tried to push meds on me... 15:07:21 I *died* of the common cold! 15:07:26 I'd say that if I were the kind of person to get offended by that, I'd rip your eyeballs out. 15:07:30 I am not, however. 15:07:43 And you can't rip eyeballs over the internet. 15:07:45 YET 15:08:27 alise: you just think you're wrongly in an institution 15:08:31 it's due to the schizophrenia 15:09:27 No, it's due to my early experiments with mind control. 15:09:35 Just ask cpressey or oerjan. 15:10:06 jdijdijidjdijeijd ooooooooooooooooooooooooooooooooooo :D~ 15:11:24 As you can see, nooga was another early test case. 15:12:54 -!- ais523 has joined. 15:13:15 I've used it as a variable occasionally. Don't ask why 15:13:23 comex: I'm afraid I'll have to ask why 15:13:29 although it should be a safe question in /this/ channel 15:15:02 Used what as a variable? 15:15:43 _ 15:18:19 Is Qi a TC type system as in, "write a program in this DSL which implements the type checker you want"? 15:19:40 -!- distant_figure has joined. 15:19:46 cpressey: sort of. 15:19:50 cpressey: a bit more infrastructure than that. 15:20:52 "Wikileaks To Leak 5000 Open Source Java Projects With All That Private/Final Bullshit Removed" --Steve Yegge 15:21:01 [[According to the Wikileaks press release, millions of Java source files have been run through a Perl script that removes all 'final' keywords except those required for hacking around the 15-year-old Java language's "fucking embarrassing lack of closures."]] 15:21:08 The only thing I /really/ don't like about LISP (and Scheme) is destructive update. 15:21:18 [[Longtime Java programmer Ronnie Lloyd of Austin, TX is offended by the thought of people instantiating his private classes. "It's just common sense," said Lloyd, who is 37. "If I buy you a house and put the title in your name, but I mark some of the doors 'Employees Only', then you're not allowed to open those doors, even though it's your house. Because it's really my house, even though I gave it to you to live in."]] 15:21:25 cpressey: So clearly we need purely functional Lisp. 15:21:30 Also, it's Lisp. Has been for years. Not LISP :P 15:21:53 alise: I have quite an interesting view about final, actually 15:21:53 Meh. Pretend I'm shouting! 15:22:11 I like to read the old papers 15:22:16 I used to think it was an awful idea; nowadays, I'm beginning to conclude that it's correct on anything which isn't specced to be extend-safe 15:22:18 [[ 15:22:19 Next article: Eclipse Sits On Man's Couch, Breaks It 15:22:19 New Hampshire programmer Freddie Cardenas, 17, describes the incident: "We invited Eclipse over for dinner and drinks. Eclipse sat down on our new couch and there was this loud crack and it broke in half. Those timbers had snapped like fuckin' matchsticks. Then my mom started crying, and Eclipse started crying, and I ran and hid in my bedroom." Read more]] 15:22:23 Yegge is an awesome writer. 15:23:24 cpressey: Maybe your destiny is to create a purely-functional Lisp that doesn't suck. 15:23:45 alise: Pixley is a purely-functional Lisp that's not nearly big enough to suck 15:23:49 alise, I don't get that thing. 15:24:04 I assume that this is because I avoid Java like the plague. 15:24:08 People will complain about the lack of, oh, I don't know. Numbers? though. 15:24:27 cpressey: It sucks. :P 15:24:41 Phantom_Hoover: Maybe it'll make more sense if you read it: http://steve-yegge.blogspot.com/2010/07/wikileaks-to-leak-5000-open-source-java.html 15:24:46 But things like pattern-matching don't seem very nice in Lisp... 15:25:12 alise: It can implement itself in a couple of hundred lines of code! How can it suck! What else would you possibly use a Lisp for! 15:25:29 Aren't questions which end in exclamation points obnoxious! 15:25:39 cpressey: link me to pixley again? 15:25:43 alise, I am, but I never poke around inside Java projets. 15:25:44 yeah 15:25:49 also, you can implement a subset of scheme in itself in like 50 lines 15:25:50 s/et/ect/ 15:25:55 http://catseye.tc/projects/pixley/doc/website_pixley.html 15:25:57 Phantom_Hoover: I am? 15:26:08 I am reading Yegge's blog post 15:26:23 My implementation of Pixley in Pixley is 140 lines. 15:27:00 "They have no right to do this. Open Source does not mean the source is somehow 'open'." what? 15:27:01 cpressey: I've become a bit of an infix whore, tbh. 15:27:08 nooga: hurr i don't get blatant satire hurr 15:27:17 For the precise meanings of each of these forms, please refer to the Revised5 Report on the Algorithmic Language Scheme. 15:27:29 cpressey: you mean the Revised Revised Revised Revised Revised Report on the Algorithmic Language Scheme. 15:28:01 The 5 doesn't indicate such? 15:28:32 NOT SUFFICIENTLY 15:29:04 alise, I don't get what it's satirising. 15:29:09 Phantom_Hoover: nothing much 15:29:12 Surely Java people aren't really like this. 15:29:17 cpressey: http://wry.me/~darius/hacks/icbins/icbins.tar.gz Scheme-subset-sideset self-compiler and interpreter in 10 pages 15:29:26 cpressey: http://wry.me/~darius/hacks/ichbins.tar.gz the beautiful 6-page version with just the compiler 15:29:28 compiler to C that is 15:30:11 cpressey: I'm pretty sure I like infix too much to ever truly like a "traditional" Lisp, though. 15:31:08 So you're in Cupertino? 15:32:00 cupertino lol 15:32:06 I'm against Lisp compilers for some reason 15:32:07 i just read that 15:32:10 alise: I was at the time 15:32:18 now I'm near Chicago 15:32:27 cpressey: You move around a lot. 15:32:27 -!- Flonk_ has joined. 15:32:27 it's like you were working for apple? 15:32:34 nooga: I doubt that highly. 15:33:14 Phantom_Hoover: Java programmers are similar to that article. 15:33:17 Phantom_Hoover: Less blatant about it, of course. 15:33:25 :O 15:33:43 I did work experience in a computing department that used primarily Java, 15:33:58 I never saw a line of their source, though. 15:34:16 alise: my objection to someone removing all the final and private from my code is "now you can't tell what can be overriden/called without being randomly breaking-changed, and what can be" 15:34:38 -!- Mathnerd314_ has joined. 15:34:39 -!- Flonk has quit (Ping timeout: 264 seconds). 15:34:39 -!- Flonk_ has changed nick to Flonk. 15:34:53 -!- Mathnerd314_ has changed nick to Mathnerd314. 15:35:14 ais523, you? A Java programmer? 15:35:33 Phantom_Hoover: I teach Java for a living 15:35:36 he teaches java 15:35:40 to poor, innocent souls 15:35:44 I think it's a really bad language to teach people as their first language 15:35:44 *gasp* 15:35:45 then harvests what's left of their brains 15:36:02 what 15:36:04 but hey, I'm being paid to teach that rather than something saner 15:36:09 alise, if my school is anything to judge by, he's a bit skinny.. 15:36:09 you're worse than Hitler then 15:36:13 and hopefully it'll increase the average quality of Java amongst graduates 15:36:19 java should die 15:36:27 same with C++ 15:36:31 Phantom_Hoover: wat 15:36:51 C++ is an order of magnitude worse than Java. 15:36:55 At least. 15:37:03 In the top computing class *two* people could program at all well. 15:37:03 yeah 15:37:10 Even on their 4th attempt. 15:37:24 Phantom_Hoover: the students here are on average rather better than that 15:37:25 grr, how do you typeset the proper prime character in TeX? $\prime$ looks weird 15:37:27 alise, have I not bitched about this at you? 15:37:28 for a height, like 5'4" 15:37:37 although you still get several who just do not get programming, and a few more who can program but don't get OO 15:37:40 Phantom_Hoover: alise, if my school is anything to judge by, he's a bit skinny.. 15:37:43 Phantom_Hoover: oh i see 15:37:48 but i never said he ate them! 15:37:49 although e does 15:37:51 *he does 15:37:56 He eats their *brains*. 15:38:50 ais523, what do you think of JS as a first language? 15:39:08 Phantom_Hoover: not nearly as bad as Java 15:39:15 although I think it's a bit large for use as a teaching language 15:39:16 $^\prime$ is... close... 15:39:19 students sould learn: C, then ruby, then haskell 15:39:23 and object.prototype, etc, is likely to be confusing 15:39:30 ais523, JS on IE 5 for the Mac? 15:39:38 wait what? 15:39:50 (although I hear mac-IE is a lot saner than windows-IE) 15:39:55 Supposedly it had better debugging. 15:39:55 ah, 6$'$1$''$ does it about right. 15:40:48 my lord, Minion is beautiful 15:40:54 I just need something to typeset in it... 15:41:02 Minion? 15:41:47 font? 15:41:50 Evidently you have a sense used only for being weird about typography. 15:42:14 nooga: typeface 15:42:25 Phantom_Hoover: you should see actual typophiles. 15:42:29 they live every moment for it. 15:42:36 alise, AAAAA 15:42:37 I just appreciate nice typefaces. 15:42:54 I share a world with people who are even more obsessed with typography than you‽ 15:43:03 I am not obsessed. Jeez. 15:43:14 I have an interest, like you probably do about many things 15:43:16 *things. 15:43:20 Minion looks no different to most other fonts! 15:44:39 -!- BeholdMyGlory has joined. 15:46:18 $26 for a font? 15:47:06 Id never pay money for a font D: 15:47:07 alise: this reminds me, I've been meaning to ask something 15:47:14 what are good monotyped fonts that are standard on the Mac? 15:47:25 ais523: What would be a good learning language? Lua? 15:47:38 cpressey: Lua wouldn't be bad, actually 15:48:08 I suppose it depends on what field you're going into; they taught electronic engineers (my first degree) asm and C simultaneously first, which worked well 15:48:15 ais523: Does this help a bit? http://www.ampsoft.net/webdesign-l/WindowsMacFonts.html 15:48:46 " Phantom_Hoover: I teach Java for a living" <<< just general java or something more specific using java? 15:48:52 well i guess you teach java if you say you teach java 15:48:53 Establishes a lower bound at Courier New + Lucida Console :/ 15:48:53 actually 15:49:16 oklopol: when teaching, you don't have much of a chance to do anything particularly specific 15:49:56 And I have to say I'm not *so* down on Java as a production language. Its culture is much worse than the language itself. 15:50:03 ais523, is it soul-crushing? 15:50:11 Phantom_Hoover: not really, it's mostly reflex 15:50:27 and Java is not an awful language, although it's a bad first language, and tends to be overengineered 15:50:35 Semicolon. Semicolon. Semicolon. 15:50:53 semicolons aren't soul-crushing 15:50:54 they're just syntax 15:51:10 Yes, just use speech synth. 15:51:13 unless syntax is deliberately obstructive (/me glares at most esolangs), it's normally pretty much irrelevant 15:51:38 I'm referring to students' inability to remember them. 15:51:56 students tend to be quite good with boilerplate 15:52:00 especially when using NetBeans 15:52:52 OK. 15:53:03 " oklopol: when teaching, you don't have much of a chance to do anything particularly specific" <<< it was surprising to me that a phd student was teaching a basic course, but i guess they need to harvest everyone for those 15:53:09 So is your soul a bit squashed? 15:53:53 oklopol: actually, it's not that surprising; basic courses are all that they can trust mere phd students on 15:54:01 most courses need actual lecturers 15:54:25 yes i realized that right after saying that 15:55:17 phd students don't teach courses often here, but if they do it's usually a course they actually ask permission to give (i know exactly one instance of this) 15:55:25 and even then, I'm not the main teacher, just take tutorials and do marking 15:55:27 which means they are everything but basic courses 15:55:41 ohh 15:55:58 that's almost exclusively done by phd ppl here 15:56:05 everything but lecturing 15:56:41 ais523: the default one on the mac is good 15:56:43 or Monaco, the old default 15:56:46 I forget the name of the new default 15:56:54 ugh, it's names I need to know 15:57:07 trying to work around a bug with jettyplay, where it sometimes chooses a ridiculously inappropriate font 15:57:12 Phantom_Hoover: firstly, Minion is a typeface, not a font; secondly, $26 is damn cheap for a font. Besides, it comes with Adobe Reader. 15:57:21 Probably using it violates the license, but who gives a shit apart from ais523? 15:57:36 and obviously it needs a monotype one because ttyrecs assume monotype 15:57:42 Finally, fuck no it does not look the same as other fonts. 15:57:49 ais523: sec 15:58:02 To my untrained eye it doesn't look special.. 15:58:34 ais523: Menlo; it's a modified DejaVu Sans Mono 15:58:41 -!- MigoMipo_ has joined. 15:58:42 thanks 15:58:48 ais523: only on 10.6 and above 15:58:50 fallback to Monaco 15:58:50 (also, yay DVSM) 15:58:54 yep, that's what I plan to do 15:58:55 -!- MigoMipo_ has quit (Remote host closed the connection). 15:59:11 -!- MigoMipo_ has joined. 15:59:22 thank 15:59:24 *thanks 15:59:24 Phantom_Hoover: well, to most people all typefaces look the same. 15:59:37 however, people will notice that a well-typeset book is easy to read. the more discerning will note that it looks good. 15:59:39 I'm not *that* untrained. 15:59:42 mostly it's subconscious. 15:59:47 Phantom_Hoover: can you tell Garamond from Baskerville? 15:59:54 I can tell them apart, to some extent (although Arial/Helvetica normally needs a blatantly different letter like the R to tell); normally I don't care, though 15:59:58 I don't know what either looks like. 16:00:07 I tend not to spend time typesetting. 16:00:55 I suppose you hate TNR, BtW. 16:01:02 -!- derdon has joined. 16:01:04 you are all heretics 16:01:09 Phantom_Hoover: TNR? 16:01:17 Times New Roman. 16:01:32 oklopol, yes, non-monospace is an offence in the eyes of oklo. 16:01:49 Phantom_Hoover: In its common Windows incarnation, yes; and by now everyone's eyes are dulled to it, and nothing interesting is ever set in it. 16:01:53 yeah, just doing my job 16:01:55 Phantom_Hoover: When it was designed, it was great. 16:01:59 -!- MigoMipo has quit (Ping timeout: 248 seconds). 16:02:18 -!- MigoMipo_ has changed nick to MigoMipo. 16:02:29 -!- MigoMipo has quit (Remote host closed the connection). 16:02:47 alise: Times New Roman was probably released a bit early 16:02:52 -!- MigoMipo has joined. 16:02:52 it looks awful at 640x480 resolution 16:03:03 which was standard for Windows 3.1, where it first became popular 16:03:10 ais523: excuse me, Times New Roman was released in 1931. :-) 16:03:17 alise: well, of course 16:03:17 Windows' font of it is and always has been terrible. 16:03:20 I mean, the Microsoft versoin 16:03:22 *version 16:03:36 yes 16:03:52 OS X's "Times" font is much better. 16:04:00 It's still fundamentally bland, though; you can't really do anything about that. 16:04:15 bland isn't bad for the default font everyone uses for everything 16:04:23 ouch! 16:04:32 Yeah, but dammit, Word could default to Palatino or something. 16:04:41 sorry, I had a sudden flashback to that feature of old versions of Word which let you put animated sparkly shapes behind all the text you wrote 16:04:42 Ouch? 16:04:46 Hahahahaha. 16:04:59 Incidentally, has anyone else read the GNU standardisation info page? 16:05:12 no, and I hope never to. 16:05:28 Oh, but it's delightfully obnoxious. 16:06:40 is this of the same standard as the Enigma video trailer? 16:06:58 (at least they've actually uploaded it to YouTube now and embedded it, rather than providing a .flv download) 16:07:38 ais523, what? 16:08:02 I forget the exact link, but http://enigma-game.org is the homepage 16:08:10 there's probably a link to the trailer on there somewhere 16:08:17 it is the worst advert I've ever seen pretty much 16:08:20 so bad it's hilarious 16:09:17 Oh, god, it is. 16:11:29 "another breathtaking movie" 16:12:14 ais523: nooo, they used the rubbish music! 16:12:16 not the NICE music! 16:12:22 apart from, you know, all the other flaws 16:12:27 Obviously this is because open-source video-editing stuff sucks. 16:12:34 alise: they can't legally use pentagonal dreams on the trailer 16:12:37 or because a lot of open source people have no taste :-))) 16:12:39 its licensing is screwed up 16:12:43 ais523: i knoowwww 16:12:45 but it's not faaaair 16:13:08 I'm amazed that the GNU zealots allowed them to use flash. 16:13:10 wtflol 16:13:34 Where's the nice music, then? 16:15:01 * Phantom_Hoover hopes beyond hope it uses pentagonal waves. 16:15:27 Phantom_Hoover: you have to download Enigma to get it; its licensing is, as I said, screwed up 16:15:30 and it's not /that/ amazing 16:15:38 [[Formerly, when a word or phrase in quotation marks came at the end of a phrase or clause that ended with a semicolon, the semicolon would be put before the trailing quotation mark; now, however, the magazine follows the universally observed style and puts the semicolon after the second quotation mark.]] 16:15:39 alise: there's a third music too, now 16:15:41 I really want to know how that looked. 16:15:44 ais523: I bet it sucks. 16:15:45 No pentagonal waves? 16:15:48 Also, it IS amazing. :| 16:15:50 not as good as pentagonal dreams, but still pretty good 16:18:03 what's so bad about the trailer 16:18:06 Musttypesetabookmusttypesetabook 16:18:17 oklopol: have you watched it? 16:18:21 half 16:18:37 well, it's far too long, and tries to explain various gameplay elements in too much detail 16:18:47 rather than actually saying what the game's about 16:18:53 well right it's a bad trailer, not a bad video 16:18:55 sorry i'm an idiot 16:18:56 pretty much everything else in there has a calculated awkwardness to it too 16:19:03 yes, it's bad at being a trailer 16:19:17 well, it's not an awesome video, but it's not spectaculary awkward 16:19:46 enigma contains different floors ! 16:19:54 :D 16:20:17 yes that was rather spectacular 16:21:43 and flaws. 16:22:48 ais523: maybe I should set your wolfram proof xD 16:22:54 except that it's ugly (sry but it is a bit messy) 16:23:35 Can we see a copy? 16:23:41 cpressey: CLOJURE; CLOJURE FIXES ALL PROBLEMS! :p 16:23:43 *:P 16:23:56 Phantom_Hoover: yes: http://www.wolframscience.com/prizes/tm23/solved.html 16:24:00 http://www.wolframscience.com/prizes/tm23/TM23Proof.pdf is the pdf 16:24:11 alise: don't worry about it; it's not me who messed it up 16:24:24 ais523: you typeset it, didn't you? 16:24:27 the wolfram people went and appended a bunch of inefficient mathematica to it for advertising purposes or something 16:24:28 or rather, scribbled it in openoffice 16:24:34 and redid all the formatting then for no apparent reason 16:24:41 ah 16:24:45 they have really bad taste, then 16:25:00 admittedly, my original wasn't brilliant, it was just sane default settings 16:25:15 it doesn't even have proper quotes, the fonts for headings are wat, the spacing is just utterly crazy, especially around headings and quote blocks, quote blocks are pointlessly coloured and dedented... 16:25:29 ais523: heh, default formatting is far superior to that 16:25:42 ais523: although really, why didn't you just use LaTeX? 16:26:27 I was thinking as I went 16:26:33 and didn't know about LyX back then 16:26:36 alise, did you get opengenera to work nicely? I wrote up a somewhat more up-to-date guide for it. Also includes how to install the symbolics X fonts (otherwise small text like in Show Keyboard Layout output is unreadable). 16:26:37 also, there's hardly any maths 16:26:42 I figured you might be interested in it 16:26:54 AnMaster: symbolics x fonts? 16:27:02 it could mostly have been done in VT100speak without issues 16:27:15 ais523: yeah, but the maths that is there is barf 16:27:26 alise, yes if you read the opengenera documentation it mentions that opengenera comes bundled with custom bitmap fonts, but will fall back to standard X fixed bitmap fonts 16:27:28 -!- tombom has joined. 16:27:33 AnMaster: ah. well, link me up. 16:27:34 alise, the bundled ones are much nicer 16:27:40 i'd prefer a proper symbolics emulator though! 16:27:40 alise, sec, will sprunge the file 16:27:46 alise, same! 16:27:50 which nobody has written yet, inexplicably! 16:28:14 there are mit cadr emulators, ti explorer emulators ... but no symbolics emulators 16:28:25 alise, pure text, would really need to be written up as LaTeX to make what is commands and what isn't clearer: http://sprunge.us/ZAWh 16:28:35 I'll write it up as LaTeX! 16:28:36 alise, the section you want is near the end 16:28:38 Yay, a job. 16:28:43 alise, nah I have it partly done already 16:28:45 so no need 16:28:46 Also, I'll fix all your awful formatting and grammar /because I'm kind/. 16:28:49 AnMaster: Yeah, but not in Minion. 16:28:58 sigh 16:29:07 :-D 16:29:09 I'm kidding. 16:29:12 alise, GFDL 16:29:14 take that ;P 16:29:28 alise, *phew* 16:29:33 You are joking wrt GFDL, right? 16:29:36 alise, of course 16:29:39 Thank god. 16:29:48 "Invariant sections: EVERYTHING" 16:29:57 alise, I'll probably go for CC-by-sa-nc-3.0. 16:30:10 * Update system with GUI tool, aptitude, apt-get or whatever you prefer. 16:30:16 mrrf ... that's not how you're supposed to uppgrade 16:30:22 AnMaster: now you are surely joking 16:30:33 alise, I meant update as in get security upgrades and such 16:30:41 alise, otherwise have fun with openssl bug and so on 16:30:50 what is it with FOSS zealots who get all touchy when their work happens to be in a language whose interpreter happens to be a brain rather than a computer 16:30:52 yes the versions on the cd are affected by that 16:31:16 AnMaster: now you are surely joking <-- hm? why am I joking about CC-by-sa-nc? 16:31:33 oh, i thought you included the no-changes clause. 16:31:40 alise, that would be nd! 16:31:44 -nc isn't Free, however 16:32:11 in the dfsg it violates no discrimination against fields of endeavour. 16:32:21 I remember a licence that is GPL incompatible because it says that the software must be used for good, not evil. 16:32:34 it's also not OSI-open 16:32:36 Phantom_Hoover: JSON 16:32:43 AnMaster: and I doubt the FSF considers it free either 16:32:47 so there, all your idols hate you 16:32:48 alise, well, symbolics probably isn't legal to use for commerical purposes without a license either 16:32:49 or at all 16:32:58 alise! Hail! 16:33:01 AnMaster: you can get an OpenGenera license. 16:33:01 commercial* 16:33:06 pikhq: hii 16:33:15 Thou art free! 16:33:17 AnMaster: by contacting sales@symbolics-dks.com 16:33:20 hm 16:33:22 pikhq: i still have to go there, just not sleep. 16:33:23 meh 16:33:27 do they still exist? 16:33:32 ais523: Symbolics? yes, as a shell company 16:33:35 http://symbolics-dks.com/ 16:33:39 dks? 16:33:41 they sold symbolics.com, the oldest .com, to a squatter; shameful 16:33:45 AnMaster: David K Schmidt 16:33:47 alise: FREEDOMNESSZOMG 16:33:48 ah 16:33:49 the only person 16:33:56 ... They *sold symbolics.com*? 16:33:59 pikhq: yep 16:34:00 How dare they? 16:34:01 http://symbolics.com/ 16:34:14 it's not like anyone even gives a shit, lol 16:34:20 but dks probably got tons of money from it 16:35:10 pikhq: so are there any players that can play ripped blurays? 16:35:14 can mplayer do it? 16:35:27 alise, anyway unsolved problems: give internet access to opengenera, write a custom keymap (if that is possible). Numpad-5 (without numlock) as a modifier key doesn't work very well on my laptop! 16:35:49 another one: Figure out CL-HTTP. Upgrading to last version did not work 16:35:56 alise: Ripped and have the h.264 bitstream dumped to an MPEG4 container file? Have been able to for years. 16:36:09 alise: Just dd'd and decrypted? Recent mplayer can. 16:36:27 i refer to a bunch of directories including 16:36:34 Yes. 16:36:38 clipinf/blah 16:36:40 jar/.../... 16:36:42 Recent Bluray. 16:36:43 playlist/... 16:36:47 bdjo/... 16:36:55 stream/... 16:36:57 in caps 16:36:57 mplayer bluray:///path/to/bluray/disc/ 16:37:05 pikhq: right 16:37:20 pikhq: now how can i take this and pack it up into a single file? can mplayer still play that at speed? 16:37:45 alise: What, you mean make it into a single mkv or something? 16:37:49 Sure thing. 16:38:32 I... Think you'd be best using mplayer -dumpaudio and -dumpvideo and -dumpsubs to fetch out the bitstreams you want and then pass it into mkvmerge. 16:38:43 pikhq: Okay. Because, you see, things should be single files, dammit. 16:39:44 pikhq: I'm considering downloading the newest Star Trek flick, you see, just because... well, I'd like to see whether it's crap or decent (Sam Hughes thinks good) and I'd like to see the full splendour of 1080p; it's all CGIy so it should be the best quality a man can get. Or is that Gillette? 16:39:53 I don't actually have the 1080p screen here, it's the TV downstairs, but oh well, a man can dream. 16:40:00 And it MUST BE A SINGLE FILE 16:40:44 Alternately, you could see if mplayer will play tarballs. :P 16:41:04 pikhq: Heh. You check that. :P 16:42:03 -!- relet has joined. 16:45:22 NOVELS are like GREEK MYTHOLOGY put in a BULLET. 16:45:31 pikhq: I SAID CHECK IT SLAVE :| 16:46:32 -!- MigoMipo_ has joined. 16:48:56 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.8/20100722155716]). 16:48:58 pikhq: I would but I have nothing to put in the tar. 16:49:03 -!- MigoMipo has quit (Ping timeout: 264 seconds). 16:57:12 "But Shatner wanted to share Nimoy's major role, and did not want a cameo,[62] despite his character's death in Star Trek Generations. He suggested the film canonize his novels where Kirk is resurrected [...]" 16:57:16 SHATNER: Give me my fucking film 16:57:47 Put all the files in a filesystem image (that is a single file), and then loop-mount it every time you want to play? The complicatedest solutions are the best. 16:58:32 pikhq: If a Blu-Ray rip says "x264", then the stream has been tampered with, yes? 16:58:51 Especially if it's Way Too Small. 16:59:15 ViDEO CODEC..........: x264_L4.1 @ 13 Mbps 16:59:16 SOURCE...............: 1080p Blu-ray AVC TrueHD 16:59:17 Indeed. 16:59:18 alise: http://blognomic.com/archive/nothing_arbitrary_about_this/ 16:59:24 you will probably like that proposal 16:59:33 especially as it's turned into a hideous counter, of sorts 16:59:57 blognomic irritates me for some reason 17:01:24 RoV? 17:02:17 repeat of vote 17:03:06 ah 17:03:26 Musttypesetabookmusttypesetabook <--- ok, now that's one urge I've seen but I've *never* felt or understood 17:03:43 cpressey: That's because you haven't realised the BEAUTY. 17:03:54 Heck, I don't even want to publish anything on the web any more because it'd look awful! 17:06:25 pikhq: God mplayer is like the best thing ever. 17:06:30 It's ffmpeg + SUPPORT EVERYTHING 17:07:50 I guess my eyeballbuds just aren't that refined. 17:11:55 ************************************************ 17:11:57 **** Your system is too SLOW to play this! **** 17:11:58 ************************************************ 17:13:57 what's that a reference to? 17:14:25 mplayer said it to me 17:14:38 "In one or two sentences, describe the process in which users are approved to become administrators on English Wikipedia." -- online Wikimedia Foundation job interview 17:14:42 It's like an exam. :-D 17:15:51 alise: "A mess." 17:16:18 ais523: marry me 17:21:02 -!- Gregor-P has joined. 17:21:49 way tempted to answer that soooo inappropriately 17:21:55 cpressey: do so 17:23:15 Failure on the other programmer's part to do anything resembling best practices caused me to stay up all night 17:23:27 Since there's no easy way to clear out static class's members 17:23:45 "static class"? 17:23:47 pikhq: how do you play a video in a reduced size in mplayer? 17:24:32 Sgeo: I assume you either meant "a class's static members" or you are using an interesting language there 17:26:10 -!- Wamanuz has quit (Remote host closed the connection). 17:27:33 This code is a shining example of worst practices 17:27:51 -!- Gregor-P has quit (Ping timeout: 240 seconds). 17:28:14 a static class would just be a namespace, wouldn't it? 17:29:09 -!- Gregor-P has joined. 17:29:14 Sgeo, what cide> 17:29:25 This project I've been working on for a while 17:29:27 s/o/i/ 17:29:42 Sgei? 17:29:46 Indeed. 17:29:57 s/bran/much/ 17:30:32 surely you must be joking 17:31:05 I'm seriously, and don't call me Shirley 17:31:14 * Sgeo fails 17:31:34 -!- Wamanuz has joined. 17:33:02 cpressey: I'm trying to minimise the function "size of self-implementation" over the scale lambda-calculus to Scheme. 17:33:32 I remember someone wrote an article about tht 17:33:35 *the range 17:37:20 alise: -xy scaling-factor 17:37:22 cpressey: C# has a thing called "static class"; it's a class with only static members in it, but you can sort-of enforce that by putting a "static" modifier in the class declaration itself. 17:37:52 alise: If using a video output that doesn't have scaling acceleration, you will also need to add -zoom 17:39:07 pikhq: Bleh, way too slow. 17:39:52 There's also the "lowres" parameter (recommended by that "too-slow" message) of lavdopts, if you're decoding via libavcodec; it could help a lot more than just scaling the output, though it won't always work. 17:41:01 fizzie: Kind of an abuse of the word "class" IMO, but I guess that's C# for ya 17:41:15 cpressey: is your website really generated with xslt? xD 17:41:27 alise: Yarrrh, 'tis, matey! 17:41:37 cpressey: have you ever noticed that xslt is insane? 17:41:55 alise: YARRH, why thinks ye I be speaking like a PIRATE? 17:42:35 cpressey: Because you just illegally downloaded copyrighted material? 17:42:39 (I had to learn it for my job at the time anyway, I figured, this looks pretty insane, why not?) 17:42:59 cpressey: What sort of job requires you to learn how to speak like a pirate? 17:43:06 fizzie: pirate impersonator? 17:43:16 I imagine Disneyland has at least one 17:43:25 * alise authors khs 17:43:25 ais523: Maybe, or some sort of generic children's entertainer thing. 17:43:28 the Kludgy Hypertext System! 17:43:42 Ye have no idea how hard 'tis typing with a hook. 17:43:45 Unfortunately the name is a pain to type on QWERTY. 17:43:48 kh hurts. 17:44:26 alise: move your fingers to the vi position, then it works just fine 17:44:55 ais523: but my fingers are never in the vi position 17:45:05 mine are when I play NetHack 17:45:15 kws is close, kludgy website system, but the kw combination is easy to mistype 17:45:26 -!- zzo38 has joined. 17:45:36 what about if you use Dvorak? 17:45:50 What is the command in TeX for "repeat until \pageno is the same value as it was the previous time"? 17:45:53 alise: The other option was "code it in PHP", because that's all my crap hosting service offers. And compared to that, xslt felt goof. 17:45:58 *good. 17:46:01 Or goof, whatever. 17:46:38 I agree PHP is full of some idiotic things, it is also slow, but I sometimes use PHP 17:46:56 And not only for serving files over HTTP 17:47:13 I've wanted to write a (insert nicer language here)->PHP compiler for some time for this reason. 17:47:32 kms, maybe; kludgy markup system 17:47:49 cpressey: Yes that would help a bit, but it doesn't make PHP run any more efficiently 17:47:57 i wish there was a mediocre language like python but without the crap 17:48:17 I also sometimes use Forth. 17:48:29 Why do people always say "ah, is crap" as if it were immediately obvious 17:48:32 alise: Have you ever used Forth? 17:48:41 English programming language is crap. 17:48:41 zzo38: Yes. 17:48:49 It isn't "mediocre like Python but without the crap". :-) 17:48:56 Phantom_Hoover: Because it ... is. 17:49:14 I have used Python as well, for writing a few card games, also in order to fix some things at Free Geek 17:49:15 But it's not necessarily obvious! 17:49:21 alise: I never implied it was 17:50:30 To me, Python embodies "Better is Worse", now that alise has brought that up. Sort-of "Mediocrity honed to a fine point" 17:50:49 -!- nooga has quit (Ping timeout: 252 seconds). 17:50:53 Hey, cpressey woke up. 17:51:15 Yep, pretty much. See, I'm capitalizing my sentences and everything! 17:51:25 Well, there is different opinion. Some people like Python and some don't, also some people like some things in Python or not 17:51:29 I credit coffee for this. 17:52:05 YAY COFFEE 17:52:48 YAY 17:53:17 Mailman is also written in Python, and I have a question I want to know if you know something about this, which is, which file contains the codes for formatting a text message as HTML document? There is something wrong with it that it removes lines with ">" at the beginning. And in September I will have to fix it 17:53:37 So, which files should I look in to figure out how to fix this problem? 17:54:05 ais523: why isn't parsing easy? 17:54:14 zzo38: I suggest not using Mailman, 17:54:17 *Mailman. 17:54:22 alise: it is, in most cases 17:54:29 ais523: orly? 17:54:38 alise: Free Geek is the one using Mailman, not me. 17:54:48 zzo38: ah 17:54:50 alise: try Perl6 grammars, if you want to avoid yacc and lex 17:55:09 (of note: Rakudo Star was released for "early adopters" today) 17:55:11 If it was up to me I would use the UNIX "mail" command. 17:55:28 But, the UNIX "mail" command is not installed on their computers. 17:55:41 ais523: i'm not using perl 6 though 17:55:47 ais523: say, cyclexa; is it good at parsing? 17:55:50 There is a proliferation of parser generators; I tried out Coco/R (not a horrible experience, but not entirely pleasant either) few months ago. 17:55:54 alise: it's what it's designed for 17:56:03 but no, because it isn't implemented yet 17:56:21 ais523: do you know what it looks like? 17:56:44 ais523: Perl6 has a built-in grammar construct? Hm 17:56:47 (Also SableCC was used in some course or other, and it wasn't completely horrible either.) 17:57:09 I tried out CoCo/R a *long* time ago. The first ALPACA was coded using it. 17:57:32 alise: Perl6? or Cyclexa? 17:57:41 Cyclexa looks like regexps just with more punctuation marks 17:57:49 ais523: cyclexa 17:58:09 it's unfortunately not particularly readable, as one of the minor design rules was to golf the thing 17:58:17 perhaps I'll add in sugar if I ever get around to implementing it 17:58:24 so, say you want to parse {word ...}, where ... is free text, nested; but also {word ... :a ... :b ...} to have the a and b parts separate, for arbitrary word/a/b 17:58:30 (and arbitrary number of parameters) 17:58:35 what would the general idea of that be in cyclexa? 17:58:39 I can never get into parser generators, they always seem a little incestuous to me, mixing grammar and code like that. 17:59:05 cpressey: Not all of them mix it up. Some have separation of grammar and code as a primary design feature. 17:59:13 Combinator parsers are less bad -- there it's more the case that I haven't found one I like yet 17:59:31 cpressey: Parsec? 17:59:58 pikhq: I don't like some of it's default choices 18:00:01 cpressey: Perhaps write one in the way that you like it? 18:00:25 zzo38: Yes -- I have one that's mostly written at this point. i should finish it 18:00:26 (\{('a-z_'+)( :'a-z'!|'^ '*)*\} 18:00:37 oh, nested 18:00:37 It's in Python though 18:00:39 surprise 18:00:40 I seem to recall that Grammatica (which is C#/Java), for example, does things so that the grammar file is completely generic and contains no code, and the code's in a separate class. 18:00:47 (\{('a-z_'+)($1!| :'a-z'!|'^ '*)*\} 18:01:00 and I forgot the closing paren 18:01:02 fizzie: I would prefer that way of organizing stuff. I should check it out 18:01:39 alise: I may have made some minor mistakes, but that's basically how it would work 18:01:46 Oh wild 18:01:59 SableCC did something similar, but I don't exactly remember what. 18:02:00 ais523: that's nice. 18:02:29 strangely, all the Cyclexa operators used there also exist in Perl5 18:02:44 although it would just try to match, it wouldn't remember the resulting parse tree like Cyclexa does 18:02:47 I was thinking of calling my combinator parser "Jeeves" (because I was using it to implement a language called "Wooster") -- but I just googled "jeeves parser" and there already *is* a parser gen called "Jeeves" -- fizzie was right about there being a proliferation 18:03:06 meh, there are two esolangs called Clue 18:03:29 We should combine them! 18:04:00 er, maybe it's just a template-driven code gen -- but it's close anyway 18:06:23 ais523: basically writing it in python would be a bitch 18:06:31 do you think it would be alright in perl, writing a parser for that? 18:06:33 with regexps 18:07:37 regexps do badly for parsing; they check whether your data is in the form you want, but it's a pain to actually get a parse tree out of that 18:08:12 "# Kills him! I was half expecting it, but also not really expecting it. I thought Lenny would go down to the basement and find Jimmy and then something dramatic would happen. I didn't really expect him to just kill Gammell here and now. 18:08:12 # Take photo. Opening credits roll. Whoa. That's it?" 18:08:27 David Morgan-Mar was not a very fan of Backwards Memento. 18:08:49 Maybe you should parse some HTML with regexps! (Cf. that stackoverflow losing-sanity post.) 18:09:49 as of Perl 5.8 or so, it can be done 18:09:52 this does not mean it's a good idea 18:10:30 alise: did I tell you about my collaboration with ESR? 18:10:39 ais523: No, but I've vomited pre-emptively. 18:10:41 Given that you can just basically stick a full Perl program inside a regex around it, I guess you *can* do pretty much anything. 18:11:02 he's trying to create the complete version history of every C-INTERCAL version ever 18:11:10 also, to merge his unreleased changes with the latest source 18:11:30 don't accept them ew ew ew you'll get esr cooties 18:11:46 I don't think it works like that... 18:11:53 DOES 18:11:57 they're probably just fetchmail integration anyway 18:12:39 well, I don't have much of a choice 18:12:43 because this is at the request of Knuth 18:12:56 ! wow - why? 18:13:01 who knows 18:13:09 ... Knuth, eh? 18:13:10 that's like an Imperial Degree for programmers. 18:13:21 Okay, you have no choice but to accept. It's an Imperial Decree. 18:13:29 alise, hm... opengenera supports lpd 18:13:34 looks like he really wanted a modern INTERCAL compiler for some reason 18:13:34 cups supports emulating this 18:13:37 and who am I to doubt him? 18:13:41 lets see if we can get this working 18:13:48 pikhq: Hey, you stole my terminology. :P 18:14:05 alise, now I just need a dot matrix printer instead of a pdf output driver 18:14:40 If you don't have a dot matrix printer, write a program to emulate a dot matrix printer? 18:15:17 Of course making an emulation will not be perfect, but it can still be used for some thing, display the contents on the screen as it is printing and then send the result to the printer once the document is finished 18:15:24 XD 18:15:40 And also you won't be able to use continuous paper 18:15:43 I have a spare dot matrix printer you might get, if you come to pick it up. (Note: estimated value of the printer probably won't cover your travel costs.) 18:16:19 alise: Is true! 18:16:21 ais523: did you ever respond to my perl question? 18:16:40 I'm not sure 18:16:51 re-ask the question and I'll remember which one it is 18:17:04 oh, "would it be alright in perl"? I responded 18:17:05 Last time I visited a large office supply chain store, their detectives managed to locate three existing printer-color-ink-ribbon-things in the whole world (in some warehouse in the middle of nowhere); I ordered all three of them, so it might be nontrivial to get more now if you run out of those. 18:17:10 regexps do badly for parsing; they check whether your data is in the form you want, but it's a pain to actually get a parse tree out of that 18:17:26 ais523: yes, what i meant was part-regexp part-perl to parse that syntax 18:17:28 I've never really tried that 18:17:38 that should work, I think 18:17:47 I haven't really tried it either, except for Underload 18:17:51 and that's rather easy to parse 18:17:56 it is very slow, incidentally 18:18:04 is it? 18:18:35 it's O(n) out from where you'd expect 18:18:52 at least the way I did it 18:20:42 -!- Wamanuz has quit (Remote host closed the connection). 18:21:48 -!- Wamanuz has joined. 18:23:32 What is the code in TeX for "repeat until \pageno is the same value as it was the previous time"? 18:23:34 ais523: what i don't get about parsers is, why isn't there one that doesn't require code, just splits into a nested list? 18:23:35 zzo38: dunno 18:24:12 alise: people tend to want more control than that 18:24:22 ais523: I don't :-) 18:24:27 alise: How much do you know about TeX? 18:24:37 zzo38: Not much about plain TeX. 18:24:41 Or LaTeX; but a bit of that. 18:25:37 What I know is that Plain TeX does not have blackboard bold, there is various other mathematical symbols missing, too. 18:25:52 How much do you know about LaTeX? 18:26:07 Not much, but enough to use it competently for simple stuff and for complicated stuff with a bit of Googling. 18:26:10 alise: SableCC constructs an AST without any code, but you'd end up with your AST made out of separate classes (corresponding to the grammar productions), and then you'd need to walk those. 18:26:20 fizzie: Which is a bitch. 18:26:23 I am writing the Icoruma->TeX conversion program, so I can use LaTeX instead of Plain TeX if I have to. 18:26:33 zzo38: I would suggest that. 18:26:35 alise: Yes, but on the other hand: no messy parser actions. 18:26:43 It can do things like Tables of Contents for you, which would be a big boon. 18:26:57 fizzie: I just want to do something like 18:27:32 alise: I have already figured out how to to table of contents even in Plain TeX 18:27:35 definition := ^[a-z] \s* '=' \s* ^expr 18:27:52 And how to do cross-referencing 18:28:00 I used this: \def\Ref#1{\xdef#1{\the\pageno}} 18:28:09 And the ^ would mean "interesting part", so the result of foo=bar would be ('definition', 'foo', ('expr', ('varname', 'bar'))) 18:28:19 zzo38: Yes, but LaTeX does it for you and with lots of typographical stuff. 18:28:55 alise: On the third hand, also, if you just want a nested list out of it, the parser actions in some mixed grammar/code stuff (or the class in something like Grammatica) will be very very simple. 18:29:17 I don't think I need LaTeX unless I need to use a lot of extra matmematical symbols which are not supported in Plain TeX. (Why didn't Knuth put them in Plain TeX?) 18:29:37 zzo38: LaTeX isn't bloated or anything. 18:29:53 alise: OK 18:30:26 So how do you do that kind of cross-referencing in LaTeX, then? 18:30:50 Big Project is Open (as Public Alpha) NOW! 18:31:26 zzo38: Making a reference to a page number, you mean? 18:31:47 \label{my-unique-name} 18:31:49 then later 18:31:51 TeX has a very powerful macro support, so I can already do table of contents, and references, and various other things, how does LaTeX improve onto this? 18:31:53 \ref{my-unique-name} 18:32:25 alise: \pageref, you mean; if you're talking about page number references. 18:32:34 zzo38: LaTeX was created with close attention to typographical detail, so things will be more refined than your work, unless you're an experienced typesetter; and it has more comprehensive support for those things. Additionally, it provides several useful commands, alleviating the need to define them yourself, and often with more useful functoinality. 18:32:36 fizzie: That too. 18:32:45 \pageref{my-unique-name} yields a page number. 18:35:03 alise: So that is how you do it. 18:35:41 But still, what kind of improvements does LaTeX add on? Plain TeX can do things like this too without too much difficulty, I showed you the codes I used 18:36:57 LaTeX makes everything easier to do by hand. 18:37:03 This is the *main* benefit. 18:37:04 LaTeX pays much more typographical attention; contains many, many more commands, more complex than the ones you have so far hand-crafted; and is altogether a consistent and useful /semantic/ system with /pluggable looks/ rather than TeX the /visual/ system with /coded-in looks/. 18:37:25 Otherwise, there's a lot of fine details that you probably don't notice that it does. 18:37:27 O, OK, so that is the difference! Thanks now I know 18:38:04 But, yeah; you put in semantic info and let it do the rest. 18:38:22 I think the main idea is that with LaTeX you get to specify the logical structure of the document (it has sections, subsections, figures, tables) and then someone else has already thought out how to make it look nice; it's not so low-level that you'd have to say how something looks, as opposed to what something is. 18:38:54 (I had that whole long sentence written, I didn't want it to go waste even if you two already said the same things twice.) 18:39:02 Of course, if you *insist*, you can just straight-up futz with the look of the document. 18:39:11 (it *is* still TeX, just with a bunch of macros defined) 18:40:22 -!- Wamanuz2 has joined. 18:41:11 I could, however, understand someone not using LaTeX and opting for hand-crafted lowlevelness just "because"; I don't really have a valid explanation of everything I've ever reimplemented while fiddling with things. 18:41:39 fizzie: Way to give therapy to zzo38's NIH :-P 18:41:45 cpressey so does need to become a NIH therapist. 18:41:53 NIH, the rapist of minds. 18:42:00 alise: I mean, if I didn't write it, why on earth would it be any good for anything? 18:44:56 Does LaTeX also have more mathematical symbols? 18:45:51 -!- oerjan has joined. 18:46:06 There's AMS-LaTeX for lots of mathematical symbols; but didn't they have an AMS-TeX too? 18:46:42 Like all good psychiatric types, I would just end up subtly encouraging NIH to keep my patient roster full, of course. 18:47:14 I don't know, is there Plain TeX with AMS added on? If so, I would make the next version of Enhanced CWEB support it 18:47:21 fizzie: yes there was 18:48:43 Kurt Gödel looks delightfully evil... <-- both name and look fit perfectly for a mad scientist villain 18:49:43 oerjan: He should've had a cat on his lap in more pictures. 18:50:12 are there _any_ such pictures? 18:50:19 How do I activate the use of AMS with Plain TeX? 18:50:30 "More than zero", based on a quick Google image search. 18:50:45 zzo38: i don't recall 18:51:01 http://www.ams.org/publications/authors/tex/tex has download links for AMS-TeX; I have no clue what is in there that wouldn't be in AMS-LaTeX, or vice versa, but supposedly it has some docs in the distribution too. 18:51:04 presumably it was a package, like latex itself is 18:51:26 "Each collection includes a user's guide"; that sounds promising. 18:52:44 It says MediaWiki supports LaTeX markup, but commands such as \count don't work? Is that only for Plain TeX? Or are those commands just unavailable in MediaWiki? 18:53:15 no relevant hits on google picture search 18:53:34 Okay, anyone parsed stuff in Python? 18:53:42 oerjan: No; but there were very strange pictures of a hammer, and a hairy naked man. 18:53:44 zzo38: AMS-TeX is heavily deprecated. 18:53:48 Not even Spivak, its author, would use it 18:53:48 alise: YES 18:53:50 s/$/!/ 18:53:57 alise: to the parsed in py q 18:53:57 (Yes, oerjan; /that/ Spivak.) 18:54:00 (& ais523) 18:54:05 cpressey: With? Just plain Python? 18:54:21 alise: Yes. Then I tried a combinator parser dealie which sucked. So I wrote my own. 18:54:37 But I'm strange you see. I *like* writing RDPs. 18:54:52 -!- alise has left (?). 18:54:55 -!- alise has joined. 18:54:56 I would, but 18:55:00 Python makes them painful. 18:55:06 Really? AMS-TeX is deprecated? I would use it in Enhanced CWEB, if it is compatible with Plain TeX 18:55:15 -!- Wamanuz2 has quit (Remote host closed the connection). 18:55:16 All I want to parse is {tag body body body :param body body body :param body body body ...} nested in body. 18:55:19 That's it. 18:55:33 zzo38: Well, perhaps not officially /deprecated/; but very unmaintained. 18:55:39 "AMS-TeX was originally written by Michael Spivak, and was used by the AMS from 1983 to 1985." 18:56:00 Oddly, actually, it's Java where I've found writing an RDP to be the least pleasant. C, Perl, Python, all about the same level of hassle. Well, the C routines I use, I guess have been around a lot longer and have aged like fine cheese. 18:56:12 (A MediaWiki comment also says that \underrightarrow is also not supported, so it isn't only \count and \def that don't work) 18:56:33 alise: That's like two or three productions, tops. 18:56:56 People other than me would say using a parser gen for it would be overkill. 18:56:59 Does MiKTeX include support for AMS-TeX? 18:57:09 cpressey: Yeah -- now try and write it as a recursive descent parser in Python. 18:57:16 Enjoy your pain. 18:57:20 zzo38: Probably not. 18:58:21 def parse_body(): expect '{'; tag = scan(); while token != '}' parse_body(); expect '}'; 18:58:27 of course you need to parse params too 18:58:33 but i don't know what rules you are using for that 18:58:42 Okay, here's a full description of the syntax: 18:58:43 and of course that is teh bad python 18:58:43 alise, symbolics documentation on installing a printer: 18:58:44 In order to install a printer, you have to: 18:58:44 * Uncrate the printer 18:58:44 * Cable the printer to a Symbolics system 18:58:44 * Specify the switch settings 18:58:44 [...] 18:58:49 the first item there... wtf XD 18:59:02 AnMaster: It was even better since I misread it as "Uncreate the printer". 18:59:10 fizzie, :D 18:59:21 AnMaster: It needs a label: DO NOT PLUG IN WHILE STILL IN CRATE 18:59:47 cpressey: Every node is either arbitrary plain-text without { or }, or a node of the format {TAGNAME ... PARAMS}, where TAGNAME is just an a-z case-insensitive string, ... is more of these nodes that I'm describing, except terminated on the first whitespace followed by ":". Params are :TAGNAME ... with whitespace always before the :, with the ... terminated in the same way. 18:59:48 So we have 19:00:09 In bold under the section "Uncrating the Printer": "Check the outside of the box containing the printer for unpacking and installation instructions." 19:00:14 {link {bold Go}{italic ogle} :to http://google.com/ :andoverthetopfornoparticularreason hello, {bold world}!} 19:00:18 this just gets better for every line!! 19:00:22 I seem to have only written a parser for Rail -- as in, the 2d trail-following esolang -- in Python, and that's not very recursive-descent by nature. 19:01:08 "Hello, bold world!" sounds somehow more upbeat than a regular hello-world message. 19:01:38 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds). 19:02:09 cpressey: Writing this in Python is teh pain. 19:05:02 Seems that generally people want a Hackiki as an adjunct to the MediaWiki 19:05:19 why... is they naming the example printer in the docs "pravda" 19:05:30 are* 19:05:42 isn't that russian? 19:06:20 Gregor-P: We'll probably neglect it. 19:06:40 It's a reasonable name, still; the word's known widely even among non-Russian-speaking folks. 19:07:24 Also good name for a printer; it would not feel out-of-place to print flyers to extremist-political demonstrations on it. 19:07:27 that's the truth. 19:07:37 oerjan: Groanity. 19:08:26 XD 19:08:35 -!- Wamanuz has quit (Remote host closed the connection). 19:08:36 fizzie: did you paste what i asked you to 19:08:46 oklopol: Oh noes, sorry. I'll do that now. 19:08:54 good 19:09:08 Gregor-P: I also agree, add Hackiki as an adjunct to the MediaWiki. 19:09:16 :d:D:D:D:Dd:Ddddd:-D:DDDDD 19:09:21 I compleatly forgot about it. 19:09:22 fizzie, whaaat? 19:09:27 :DDDDDDDDDDDDDDDDDDDDDDDDD 19:09:33 this is the best day of my life 19:09:44 AnMaster: Don't look at me; he wanted it. 19:09:46 for two reasons actually, the other happened at uni 19:09:50 fizzie, that is so out of character. Using a longer than maybe 4 smiley 19:09:58 oklopol: You met your true love at the uni?! 19:10:02 you know how toilet paper comes from those weird boxes in public shitting places 19:10:03 If some people want to convert some of the pages over they can do so, possibly someone will add a MediaWiki parser in that Hackiki site if they want that..... 19:10:14 oklopol: Thus far it's not sounding like a love story. 19:10:15 there are two rolls in one 19:10:18 oklopol: You met your true love at the uni?! 19:10:18 you know how toilet paper comes from those weird boxes in public shitting places 19:10:20 XD 19:10:24 and the other roll was the wrong way around 19:10:24 I think you should provide read-only access to some of the MediaWiki files from Hackiki 19:10:37 -!- Wamanuz has joined. 19:10:37 So that some people can experiment a few things 19:10:39 and the paper was hanging down 19:10:46 so what when you pulled paper out 19:10:51 Also, add the tag to the MediaWiki installation you have 19:10:55 the other paper came out from the side of the wall! 19:11:04 so you had to take pieces in turns 19:11:07 AnMaster: Don't look at me; he wanted it. 19:11:11 That's not considered a valid defence. 19:11:13 that was AWESOME 19:11:29 fizzie: yeah don't be a german 19:11:31 gah 19:11:33 I think I can safely conclude that plug-and-play was invented way after symbolics went defunct 19:11:36 oklopol: Do you mean the sort of thing that has a big roll and a small roll, and you can sort of access both simultaneously, or just the sort where there's one roll at the bottom, and then one (or two; I've seen two too) back-up rolls on top of it, meant to drop down when the first one's exhausted? 19:11:52 back-up 19:12:30 i've never seen it happen before, but it was seriously so awesome 19:12:55 oklopol: Okay, yes, our toilets have those two. I've seen one of them in a state where the down-hanging piece of paper was actually the middle roll, and it was rotating around there in the middle; there's this sort of thing holding it up, so it didn't get friction from the bottom roll. (There's a narrow gap in front you can then use to drop it past those roll-holders if the bottom roll is finito.) 19:13:03 s/two/too/ 19:13:15 i stared at it for 5 minutes and could barely fight the urge to start knocking on doors telling random people about it 19:13:43 oklopol: I can actually believe that 19:13:56 alise: "Look how his nickname is shaped; he was so asking for it." 19:14:21 aww oklopol and fizzie are such a cute couple 19:14:24 the same sort of discovery as when you put a battery in backwards in an old-fashioned analog battery-powered clock, and the hands start ticking backwards 19:15:09 8| 19:15:20 holy shit that would be awesome 19:15:33 i have to see a clock going backwards 19:15:38 NO THAT'S DANGEROUS 19:15:49 it's how Gregor-P got in that closed time loop 19:16:04 oh, i guess that's why i've "never" seen it 19:16:41 so umm 19:16:48 from our perspective, he'll just disappear? 19:16:53 ais523: That reminds me of a story. My printer had its front panel lights flashing briefly every now and then (once every few hours or so) when it was turned off; so I went to the manufacturer's website, and the knowledge base had that as a known issue: the fix was to unplug the printer, rotate the plug 180 degrees around the cord axis, then re-plug it back. Sure enough, after I rotated the plug the flashing stopped. 19:16:57 -!- Wamanuz has quit (Remote host closed the connection). 19:16:58 no wait probably there's something like a tm involved 19:17:15 fizzie: ouch, that printer model must have a dodgy ground plane 19:17:32 -!- CakeProphet has quit (Ping timeout: 240 seconds). 19:17:46 -!- Gregor-W has joined. 19:17:56 oerjan: Oh, is THAT how? 19:18:03 Yeesh, I'll have to remember that this time 'round. 19:18:09 * Gregor-W promptly forgets. 19:18:20 the same sort of discovery as when you put a battery in backwards in an old-fashioned analog battery-powered clock, and the hands start ticking backwards 19:18:24 you're joking 19:18:37 fizzie: how many flashes did you wait in front of the printer before checking the webpage? 19:18:48 I'm trying to find the KB article, but the "customer support" site is giving me a "WrongChannelException: Channel is null for siteId=204816596 and channelId=374757490" Java exception and a 46 methods deep stacktrace. 19:18:57 alise: I'm not 19:18:58 Gregor-W: Hi! You are Gregor-W. 19:18:59 I mean you. 19:19:02 ais523: :| 19:19:06 I'm not entirely sure if it works on all models, but it definitely works on some 19:19:09 oklopol: A week's worth of clock-time, but I wasn't staring at it all the time. 19:19:12 I know, because I've done it by accident 19:19:13 alise: Thank you, /me writes this down. 19:19:28 alise: Where's our TC CPP code? ;) 19:19:45 Gregor-W: Oh, I'll give it to you in just a m... 19:19:48 Gregor-W: Hi! You are Gregor-W. 19:19:51 I mean you. 19:20:12 * oerjan expects there's a reference he doesn't get 19:20:12 what's W for 19:20:18 OK, you seem to have gotten caught into some ridiculously short prograde-amnesia-induced loop. 19:20:20 oklopol: Work. 19:20:27 i was wondering if it was about the loop but no idea what it'd mean 19:20:35 oklopol: actually, it's world-domination. 19:20:37 Gregor: oh makes sense 19:20:38 * ais523 has trouble remembering events from the future 19:20:39 ais523: Oooh: "Lexmark is conducting a voluntary recall of the E230, E232, E232t, E330, E332, E332n, and E332tn printer models. This recall is being done in cooperation with consumer safety agencies around the world. 19:20:39 Lexmark has not received any reports of incidents or injuries involving these printers. However, through internal reliability testing, Lexmark has identified a potential safety issue in one printer after the equivalent of several years of normal usage. In the unlikely event of a multiple component failure, the printer could present an electrical shock hazard if it is connected to an ungrounded power source." 19:20:41 -!- Wamanuz has joined. 19:20:59 fizzie: seems reasonable 19:21:24 if your printer is capable of detecting the difference between live-neutral and neutral-live, it must be that the chassis earth potential is involved somehow 19:21:31 which it shouldn't be, for safety and sanity reasons 19:22:07 insane printers for your hideously non-euclidean printing needs 19:22:17 what? 19:22:18 ais523: They haven't really advertised the recall too much, though. 19:22:29 oerjan: elaborate at once 19:22:37 oklopol: fthagn! 19:22:42 ais523: "For safety reasons, we ask that you unplug your printer from its power source before checking to see if your printer is subject to this recall. If your printer is connected to an ungrounded power source, we ask that you do not open the covers, do not turn off the printer, do not touch the back of the printer, and do not touch anything else connected to the printer prior to unplugging it." 19:22:53 umm wait non-euclidean because printers print on a plane 19:22:58 ok, that sounds /very/ like a chassis ground failure 19:23:07 i expect too much 19:23:21 oklopol: _these_ printers print on a klein bottle. which doesn't intersect itself. 19:23:26 hmm, pity that you're allowed to unplug it 19:23:29 well, that's the simplest setting, anyway 19:23:37 it would be hilarious if you had to ship it back to Lexmark while it was still plugged in 19:23:41 and turned on 19:23:57 "In spite of the clear instructions on the sticker in the case, under no circumstances should you lick any part of the printer until you have determined that it is not one of the affected models." 19:24:24 there are clear instructions on the sticker to lick it? 19:24:48 Printer just wants some love. 19:24:58 does it really say lick :D 19:25:19 DO NOT PUT PRINTER PLUG IN NOSE 19:25:40 ais523: On the other hand, they posted this recall notice in September 2004, and we bought the printer *after* that, and it still does the flashing-lights thing. They say only "some models" of post-August-2004 are affected. So, well. 19:26:03 I don't see why a voluntary recall necessarily means they stop selling the things 19:26:22 "Although cherry-flavored, the magenta ink is highly toxic and should not be drank." 19:26:55 It does hopefully mean they'd have fixed the problem, but of course this printer has possibly been in someone's warehouse for a long long time. 19:27:14 Gregor-W: :D:D:D 19:27:19 Gregor-W: what are you pasting? 19:27:22 They *are* (well, were) promising to replace it with another unit of the same model, so one hopes the replacement version would be un-faulty. 19:27:28 oklopol: Random nonsense :P 19:27:36 Gregor-W: keep doing it it's hilarious 19:27:47 Gregor-W: but from where 19:27:50 Actually our printer hasn't been doing the flashing-lights thing nowadays with grounded outlets, I think. Unless of course I accidentally got the plug in the right way. 19:27:53 oklopol: From my twisted brain. 19:28:10 fizzie: I would think that if they recalled a model, they would change the model number of new ones without the problem they recalled it for ... 19:28:35 Gregor-W: so do you people say "drank" for perfect tense nowadays or did you make a typo, or did you make a grammatical error? 19:28:54 middling tense markers; tattlesquat! I'll tell you what: 19:28:59 oklopol: hey i was trying not to comment on that 19:28:59 Bjorn made a trip to the embassy ... 19:29:10 oklopol: Oh shut up you :P 19:29:22 Gregor: It's only a partial recall, so I guess they already had broken and un-broken versions floating around. 19:29:25 alise: he should _not_ have worn his special knife belt. 19:29:26 oklopol: It's ambiguous in this case since the ink shouldn't be drunk either. 19:29:39 i'm not trying to be a smart-ass, it's just Gregor-W doesn't make mistakes, so i thought "ohh there's a whole webpage of these" 19:29:40 oklopol: So I went with the unambiguous but grammatically-dubious "drank" 19:29:49 oerjan: and this knife belt, right, it was powered by the night; 19:29:55 The night! night! The very frightened night! 19:30:03 Gregor-W: my first thought was that that would make sense 19:30:04 and Bjorn walked to the pavement--right--and saw something fearful: 19:30:10 not saying ink is drunk 19:30:41 Gregor-W: the ink should not be drinking so much alcohol 19:30:57 -!- oerjan has quit (Quit: Later). 19:31:11 suddenly, oerjan leaves 19:31:12 -!- BeholdMyGlory has quit (Remote host closed the connection). 19:31:15 "Please be aware when planning printer-related meals that the black ink is not kosher and contains pork products." 19:31:19 what a party defecator 19:31:30 -!- BeholdMyGlory has joined. 19:31:51 And, err, the fix seems to have been to include in the safety instructions the following: "[1] If your product is *not* marked with this symbol [square inside a square], it *must* be connected to an electrical outlet that is properly grounded." (Where I guess the square-inside-a-square is the usual class 2 double-insulated device symbol.) 19:32:45 oklopol: And in his wake Bjorn discovered a philosophical masturbator-- 19:32:52 curator of sciences, physic & all 19:32:55 He lead Bjorn into the great hall ... 19:33:51 then lead the hall back out of Bjorn? 19:33:54 Gregor-W: you could you make a website with made-up products with hilarious instructions 19:34:05 the twist is 19:34:07 ais523: & this was met with scorn. 19:34:13 Why does he speak, Bjorn wonders, right, and this guy-- 19:34:14 you can actually buy them 19:34:22 stupid fucking guy, I wonder why I'm even talking to him -- thinks Bjorn, and 19:34:26 -!- MigoMipo__ has joined. 19:34:27 he's not really quite sure what this is all about but anyway... 19:34:37 alise: he's not the only one 19:34:46 "The paper tray of this printer is removable for your convenience. However, due to the high printing speed and to avoid risk of decapitation, it is highly recommended that you leave it attached when printing on card stock." 19:35:00 oklopol: well it's not bjorn's swan song! 19:36:03 Terribly disturbed, he served lemon curd: 19:36:15 and this was what was seen by the president's priest when on a visit -- 19:36:21 Frightened, Bjorn said! Aah! Aah! This is what he said 19:36:29 and he said what he said before he became dead! Aaah! Aaah! This is what he said! 19:36:50 But he didn't die at this present point! 19:37:25 """Suddenly, nothing happens.""" 19:37:53 Gregor-W: more, please 19:37:58 """"But quote marks do nest", said the pest", said the antediluvian nest" was a stupid thing to say", he said, but it was said anyway" 19:38:00 oklopol: I'm out :P 19:38:01 said Bjorn. 19:38:01 this is the best flood i've seen in ages 19:38:04 -!- MigoMipo_ has quit (Ping timeout: 252 seconds). 19:38:14 alise: it was meant to be a quote, so I put it in quote marks 19:38:22 but then, because it wasn't actually a quote 19:38:24 I put the quote marks in quotes 19:38:30 to show that it wasn't really quoting 19:38:52 "Bjorn", later said a historian, "would have here commented on how quotes should properly be nested 'like this "and this" and this': for alternating quotes are the devil's honey, and Bjorn was a fan of the devil." 19:39:05 if that's not an INTERCAL reference, it should be 19:39:14 what if it was? 19:39:18 -!- alise has left (?). 19:39:21 -!- alise has joined. 19:39:22 should it then be 19:39:34 not that it's the only way to nest quotes in INTERCAL, but it's the easiest way to stay sane 19:39:37 "All references" said a philosopher (who was called Bjorn (the Bjorn we've been talking about)) "are relative." 19:42:00 I found the AMSFonts package it includes macro files for Plain TeX, I don't need AMS-TeX? Will just AMS-Fonts do what I wanted within Enhanced CWEB 19:42:18 Who knows. 19:42:48 -!- nooga has joined. 19:44:18 alise, one thing you can't say about opengenera is that it supports plug and play for networking and hardware 19:44:34 "We must never rest until everything inside us worships God." -- AW Tozer \ What an entirely creepy thing to say. 19:44:38 alise, atm I'm looking at the documentation of a function to define a postscript printer type 19:44:40 XD 19:45:08 Gregor-W: Say hi to religion. 19:45:23 I would rather cry in the fetal position, TYVM. 19:47:02 cpressey: By the way, a lot of the dependencies of C code on the machine wrapping things right is actually the compiler being clever with optimizations. IIRC (there was some case very similar to this), GCC multiplies by 10 without multiplying or shifting. It's convoluted magic. 19:47:58 Hmm, did I ever publish that Wiktionary-to-OED thing? 19:48:13 It would look up a word in Wikitionary then reformat it to look like the definitions you find in the OED. 19:48:30 alise: Work on TC CPP instead :P 19:49:10 Gregor-W: Okay; give me a simple recursive algorithm to implement that uses lists in a simple way. 19:50:53 Hmmm, MAYBE a cyclic tag system, which was ais523's (?) first thought too. 19:50:59 Which I scoffed at :P 19:51:39 Gregor-W: cyclic tag tends to be simplest for that sort of thing 19:52:08 Gregor-W: I suck at writing them, though. 19:52:38 although, cyclic tag likes queue-equivalents, and lists are better at stack-equivalents 19:56:03 http://www.laws-of-form.net/lof/pdf/Denjoy_proof.pdf A "proof" of Riemann's hypothesis (and "much, much more" sort of thing) by the author of that piece of twaddle Laws of Form. 19:56:09 ais523: Stack-equivalents aren't as good at being-Turing-complete equivalents ... and forming two stacks MIGHT be tricky ... 19:56:09 Anyone want to rip into it? 19:56:40 no, the Riemann Hypothesis scares me 19:56:49 ais523: it seems to use much more simple mathematics, actually. 19:56:57 by proving a supposedly-stronger theorem 19:57:12 it's not the complexity, it's the connection to the primes 19:58:08 alise: I THINK you should be able to do two stacks in CPP, right? And with two stacks you've got a tape, and with a tape it shouldn't be difficult to make an actual Turing machine. 19:58:46 Gregor-W: Yes, but *iterating* through these is VERY hard; you must use recursive includes. 19:59:17 alise: Each state could be a separate #include file which re-#defines the two halves, then #includes the next part. Think of it like iteration even though strictly it's doing recursive includes. 19:59:34 Eh? 19:59:36 So we have like 19:59:38 state1.h 19:59:39 state2.h 19:59:40 etc? 19:59:42 Yeah 19:59:58 And each one does the transform of that state as an actual transformation, since you can #undef and re-#define in CPP. 20:00:11 Then includes whichever next state is appropriate. 20:00:22 You could also have multiple states just by #define STATE 3 or whatever :P 20:00:24 -!- coppro has joined. 20:00:33 cpressey: Pixley minimalisation suggestions: Remove booleans; just use the symbols #t and #f. 20:00:47 -!- coppro has quit (Changing host). 20:00:47 -!- coppro has joined. 20:00:55 Specify that (define #t '#t) and (define #f '#f) must be in place. 20:02:03 If you made the tapes look like a, (b, (c, ())) then you could have every macro just take two arguments, head and tail. 20:02:07 s/tapes/stacks/ 20:02:18 Gregor-W: Heh, it is Not That Easy. 20:02:23 I have a working list structure... but it is difficult. 20:02:39 Wish I wasn't at work so I could prove to myself that it's Not That Easy :P 20:03:05 Fine, I'll try it Your Way. 20:03:16 I'm just throwin' out thoughts here. 20:03:34 It definitely can't be so simple as 20:03:35 #define hd(h,t) h 20:03:35 #define tl(h,t) t 20:03:36 It will have to be 20:03:45 Nonono 20:03:47 #define hd_(h,t) h 20:03:47 #define hd(l) hd_(l) 20:03:49 I have verified this. 20:03:50 Yeah 20:04:07 I didn't mention that detail because I knew you already knew that :P 20:04:08 However: 20:04:10 ehird@dinky:~/Code/cpp$ cpp -std=c99 -Wall -pedantic -undef -nostdinc -P list2.hlist2.h:9:37: error: macro "tl_" requires 2 arguments, but only 1 given 20:04:10 list2.h:9:37: error: macro "hd_" requires 2 arguments, but only 1 given 20:04:20 I am as of yet struggling to find a reason for this. 20:04:23 D'awww huh? 20:04:29 When hd() is called with what? 20:04:35 hd(tl(cons(a,cons(b,cons(c,d))))) 20:04:37 yields one less error 20:04:41 Gregor-W: Lemme debug. 20:04:44 Uhhh ... that is one argument. 20:04:49 Ahhh. 20:04:53 tl(cons(a,cons(b,cons(c,d)))) 20:04:59 => (b,(c,(d))) 20:05:05 You can't "unwrap" it from the ()s, either. 20:05:09 Thus debunking your list structure; it must be flat. 20:05:24 What about 20:05:25 Unless... 20:05:33 (h,t) is the cons 20:05:34 and we do 20:05:35 tl_ l 20:05:37 #define unwrap_(h, t) h,t \n #define unwrap(l) unwrap_ l] 20:05:38 Yeah 20:05:43 Without the ] there :P 20:05:50 Idonno if that'll work. 20:05:55 But it's certainly worth a shot. 20:05:56 Now it works. 20:05:59 You are a genius. 20:06:04 Of course I am ;) 20:06:07 Awesome. 20:06:11 Gregor: No need for anything so complex, btw. 20:06:14 #define cons(h,t) (h,t) 20:06:14 #define hd_(h,t) h 20:06:14 #define hd(l) hd_ l 20:06:14 #define tl_(h,t) t 20:06:14 #define tl(l) tl_ l 20:06:28 Hmm. This actually suggests that C could have *good* metaprogramming from this. 20:06:39 pikhq: it's iteration that's hard (almost impossible) 20:06:46 I will now extend this to have markers for nil/cons. 20:06:50 However, booleans are very tricksy. 20:07:07 can you use hd and tl as true and false? 20:07:48 However, booleans are very tricksy. 20:07:50 erm 20:07:59 ais523: alas, no; and nor specially designed ones, either 20:08:01 because 20:08:05 Do you need bools? 20:08:12 #define hd_(d,h,t) d(h, ERROR_IS_NIL) 20:08:13 works once 20:08:17 but then the tail that is produced 20:08:19 the d is marked as from-cpp 20:08:22 and so is not recursively expanded 20:08:24 Can't you #if hd(foo) == a 20:08:26 so it just becomes true(h, ERROR_IS_NIL) 20:08:27 so lol. 20:08:30 Gregor-P: Yes. 20:08:32 This is my approach. 20:09:15 #define nil (1,END,END) 20:09:15 #define cons(h,t) (0,h,t) 20:09:15 #define isnil_(d,h,t) d 20:09:15 #define isnil(l) isnil_ l 20:09:15 #define hd_(d,h,t) h 20:09:16 #define hd(l) hd_ l 20:09:18 #define tl_(d,h,t) t 20:09:20 #define tl(l) tl_ l 20:09:22 then 20:09:24 #if isnil(...) 20:09:26 works 20:10:59 list2.h:21:14: error: missing binary operator before token "tl_" 20:10:59 wut 20:11:17 ah. hm 20:11:31 Gregor-P: all this seems to run into the no-recursive-expansion mess i'm afraid 20:12:32 I'm not sure why, you should only ever need to take one step... 20:13:14 http://pastie.org/1067545.txt?key=9jy5nwppi2mgw55z1wbneg 20:13:21 Run this with "cpp -std=c99 -Wall -pedantic -undef -nostdinc -P" and study the error. 20:13:31 Then perhaps you will see how cpp's anti-tricksyness laws are stopping us. 20:14:12 Well, in this case, 20:14:16 it is that #define bar foo does not copy foo to bar. 20:14:24 Therefore how do we do it? 20:14:28 I do not know that it is possible. 20:14:48 Perhaps someone could link me to the Game of Life so I can study it. 20:15:18 What are you trying to do with the C preprocessor now? 20:15:32 Prove it Turing-complete. 20:16:03 Did you look at ORDERPP? 20:16:06 And CHAOSPP? 20:16:29 -!- Wamanuz has quit (Remote host closed the connection). 20:16:41 No. 20:16:45 I have no idea what they are. 20:16:45 Argh, anti-tricksyness. 20:17:15 The C preprocessor is literally trying to prevent itself from being more generally useful. 20:17:17 :( 20:17:32 -!- Wamanuz has joined. 20:17:35 Did you try: #define isnil(l) isnil_##l maybe you need a ## sign, I'm not quite sure about that 20:18:31 I know that in ORDERPP codes, you have functions with 8 before each function name, like 8if(8and(8x),8y) and things like that 20:18:36 zzo38: He doesn't want token concatenation. 20:19:13 I found the C preprocessor useful in writing ARGOPT 20:19:36 -!- nooga has quit (Ping timeout: 240 seconds). 20:19:58 It's probably just trying to prevent non-halting preprocessor expansions, which is perhaps a bit misguided; but you know what they say about preprocessor abusers. (Actually, what *do* they say about them?) 20:20:27 fizzie: But we *wants* non-halting preprocessor expansions. 20:20:32 We wants them so! 20:21:18 I believe the "8" before the function names in ORDERPP is probably to delay expansions or something, like that, so that you can expand the parts later with different codes, or something, I don't actually know why, but I guessed 20:21:36 -!- nooga has joined. 20:23:04 wonderful, I got a cover page in the pdf, that is all... 20:23:10 saying screen hardcopy 20:23:19 no actual screen hard copy though 20:26:43 I know that ORDERPP is turing complete, and is written entirely using the C preprocessor. 20:28:03 But the internet doesn't know it at all, so we have no clue what it's like. 20:28:14 The internet defined by google, anyway. 20:30:20 -!- ais523 has quit (Remote host closed the connection). 20:33:33 There is also Enhanced CWEB in which you can do many things even before the preprocessor, but that isn't based purely on the C preprocessor. You can do things like http://sprunge.us/WSfI but you can also write metamacros for various things, too. 20:34:42 Stupid "throw" where I didn't want one! 20:34:50 Now, instead of one puzzle crashing, the whole game crashed! 20:35:06 Sgeo: In what thing? 20:35:10 What game? 20:35:14 Evolution 20:35:29 I'd give a link, but the website's in Flash and not very helpful 20:35:30 zzo38: What is ORDERPP? 20:36:16 ORDERPP is a set of C preprocessor macros for making it Turing-complete. 20:36:39 Is it yours? 20:36:42 No. 20:36:45 Whose? 20:36:48 I don't know. 20:36:59 ... 20:37:01 What? 20:37:07 Can you publish it somewhere? 20:37:18 I think it is on SourceForge and is part of the same project as CHAOSPP. 20:37:32 I can't find CHAOSPP. 20:38:10 Unfortunately on Google, searching for "chaospp" and "orderpp" together returns no results. 20:39:10 Searching for only one of them also returns no relevant results, and searching for "chaospp" and "preprocessor" (or "orderpp" and "preprocessor") also finds no results, so they are pretty elusive stuff. 20:39:46 I remember seeing some awfully great cpp programming framework on sourceforge once. 20:39:52 But I don't think it was (CHAOS|ORDER)PP. 20:40:31 Try searchings things other than Google, try SourceForge, or Archie or Veronica even, or Tor or Freenet or GNUtella or whatever 20:41:08 -!- Gregor-W has quit (Ping timeout: 252 seconds). 20:41:53 Find anything on GNUtella? /Ha!/ 20:43:01 * Sgeo vaguely screams 20:43:13 ? 20:43:19 I'm waiting for someone 20:43:23 But one of the limitations in C preprocessor is macros cannot expand to other preprocessor directives (including definition of other macros) 20:43:24 While the public is in-world 20:43:43 alise: That is such an awful pun. 20:43:53 pikhq: What; gnutella? 20:43:57 Yes. 20:44:00 It is; isn't it great? You probably know it as LimeWire. 20:44:14 You know, the crappy P2P network that's just viruses disguised as porn clips. 20:45:03 You can create include files that use #include to conditionally define things and stuff, but you can't do it other things. 20:45:24 You can expand macros into the special "_Pragma" preprocessor unary-operator (not the #pragma directive form, the function-like form), but that's probably not very useful usually. 20:45:47 What you can do in C preprocessor however, is define one macro and then later define other macros that the first macro uses and call the first macro. This is something done in ARGOPT. 20:46:32 That is the standard-edition of ARGOPT. I might then later also write web-edition of ARGOPT, which includes some additional functions such as printing out the parameters in a table with documentation, and generating man pages 20:46:56 If I do so, I will be maintaining standard-edition and web-edition separately. 20:47:38 -!- uriel7 has joined. 20:47:49 uriel7: hi 20:47:54 are you the plan9 uriel? 20:48:05 Enter text here...hi 20:48:11 Okay, presumably not. 20:48:17 Hi. 20:48:20 Or Uriel Septim, the emperor? 20:48:33 Oh, sorry, that's a fictional guy. 20:48:35 uriel as the angel 20:49:01 Right, well, hello אוּרִיאֵל. 20:49:06 I suppose it can sometimes be useful to use _Pragma operator, maybe someone might use it with OpenMP possibly 20:49:16 uriel7: I suppose you are brought here for the esoterica, not for the programming? 20:49:26 Uriel, the little murmaid 20:49:48 no for proraming 20:49:59 programing 20:50:18 Huh. Okay. 20:50:22 Then you are in the right place. 20:50:28 Welcome. 20:50:36 thanks 20:50:43 For programming or not, there is many things discussed in this channel 20:50:50 But the main topic is esoteric programming 20:50:51 Yeah, but very rarely esoterica :-P 20:51:16 But you can discuss other things too, it doesn't have to be about esoteric programming 20:51:31 thanks 20:52:31 was looking for palce to talk nothing more 20:52:39 Slereah: Shouldn't that be a mirmaid? I mean, it would make more sense in a wovel-categorisationary way. Or am I overthinking it? 20:54:56 lol 20:56:34 -!- Gregor-W has joined. 20:56:39 alise: Well? Well? WELL?! 20:56:51 Gregor-W: Three holes in the ground. 20:56:56 uriel7: so what are your favourite esolangs? 20:58:05 a dont know 20:58:34 Anyone have greppable logs? Gregor? 20:58:40 uriel7: Made any yourself? Written any programs? 21:00:14 alise: Mine aren't up to date as my home system isn't running. 21:00:46 That's okay; I'd just like you to grep for /wikti?onary/i, if that's alright, and provide a list of dates. 21:01:06 I'm also not near a computer that has them checked out :P 21:01:17 -!- uriel7 has quit (Quit: Q: How much does it cost to ride the Unibus? A: 2 bits.). 21:03:52 alise: There are quite many of those. 21:04:23 fizzie: ok; said by (ehird|tusho).* 21:04:52 Regrepping; for the record, 150 lines with /wikti?onary/i in them, I didn't yet uniq them to dates. 21:06:00 *flash of brilliance* the esoteric logs should be in the (hypothetical) esolangs Hackiki, then anybody could grep them from afar. 21:06:36 Gregor-W: Or... just make a web interface to it. 21:06:45 Yeah, but this would be more general :P 21:06:50 And more general is exactly equal to better. 21:07:03 2008-04-20, 2008-06-13, 2008-08-22, 2008-09-21, 2008-09-24, 2009-05-19, 2009-06-07, 2009-08-02, 2009-08-18; they all seem to be mostly quoting some single word from there (mostly); but my logs are not completely complete. 21:07:18 Were you looking for something in particular? 21:07:27 ehird@dinky:~/esoteric$ grep -ri 'wikti?onary' 08.* 21:07:27 ehird@dinky:~/esoteric$ grep -ri 'wikti?onary' 09.* 21:07:27 ehird@dinky:~/esoteric$ grep -ri 'wikti?onary' 10.* 21:07:30 What's that, then... 21:07:40 fizzie: Yes; my Wiktionary lookup tool. 21:07:55 It formatted them all nice-like, OED-style. 21:07:58 You need either \? or -E in the grep. 21:07:59 okay this is hilarious: I have cups printing to a lpd server, which is actually cups-lpd on the same host, mapped to cups-pdf 21:08:03 why? for debugging 21:08:14 and this works. However, it does not from opengenera 21:08:17 sigh 21:08:35 10.06.25:18:50:06 http://en.wiktionary.org/wiki/fuck -- how on earth am i meant to present a definition of "fuck" to the user like this 21:08:37 looks like around this time 21:08:44 10.06.25:19:19:16 Gregor: I just mangled Wiktionary into something readable. 21:08:58 Seems I wasn't present at that particular time. 21:09:12 fizzie: TISK TISK 21:10:09 I do have a lot of chatter on 2010-06-25, though; inexplicablity. 21:10:11 -!- MigoMipo has joined. 21:10:18 Oh, right; I grepped for ehird|tusho, like you said, not alise. 21:10:50 Yeah, didn't realise it was that recent. Should have, of course. 21:11:00 WYGIWYAFAINGW, after all. 21:11:23 wat 21:11:34 "What you get is what you ask for, and it's no good whining". 21:11:46 It was to pre-empt any whining; didn't think you'd be that gracious. 21:11:51 Sorry for that. Retroactively. 21:12:05 18:52:05 " She shoved them up and together, pushing into me, forcing my foot to fuck her tits harder and harder while gasping as if I was shoving it deep into her body..." 21:12:08 18:52:08 Yep, the OED would quote that. 21:12:08 18:52:16 Classy, Wiktionary. 21:12:08 18:52:17 Classy. 21:12:38 I vaguely recall that :P 21:12:54 -!- MigoMipo__ has quit (Ping timeout: 260 seconds). 21:15:30 [[In the Oxford Latin Dictionary, the word irrumo is defined "[t]o practise irrumatio on." Great. Irrumatio, less than helpfully, is "[t]he action of an irrumator." We finally learn that an irrumator is "[o]ne who submits to fellatio." And thus, after flipping your way through three different entries, you get a definition that, while described precisely enough (and better than its counterpart in the main Victorian Latin dictionary, "to commit beastly acts" 21:15:30 ), is completely wrong. An irrumator is the active, not the passive, participant. The real meaning of irrumo is something like "to fuck (someone) in the face, esp. as a way of asserting dominance"; it is frequently found in the locker-room talk of Catullus and also appears in Martial's scurrilous epigrams.]] 21:16:14 19:04:58 PCC on Linux is a suckfest. 21:16:14 19:05:04 And not the good kind. 21:17:38 19:34:53 wikipedia, v. past tense of wipe. 21:17:38 19:34:54 WHAT 21:17:38 19:36:05 oh ha 21:17:38 19:36:22 it picks up on the deletion notice saying "deleted" which is like ... wiping, i assume 21:17:38 19:36:31 dunno 21:17:48 I propose we henceforth use "wikipedia" instead of "wiped". 21:17:57 I wikipedia my ass with toilet paper. 21:18:09 19:36:44 "He told me to wipe off the dishes, so I wikipedia them for an hour or so but got tired of it." 21:18:12 Dammit, beaten to the punch 21:18:13 Based on the "insignificant part" fair-useness and so on, here's "fuck, v." in current OED; http://sprunge.us/ODbU -- that's one long definition. 21:18:32 fizzie: you have the electronic OED? 21:18:42 Via the university's proxy, yes. 21:18:52 also, I see they have given up on pretensions of conciseness. That is more an encyclopedia article. 21:18:59 We need to look up "fuck" all the time at work, you see. 21:19:16 http://pastie.org/1019589.txt?key=iveweg7njbmwqqdjqpnndg 21:19:18 define.py 21:19:25 "As a physicist, I don't really have much cause to use mice in my regular research, which mostly requires the use of theoretical math," said Dr. Thomas Huber, author of the 1996 study Mouse Elasticity And Kinetic Rebound In High-Acceleration Collisions. "But when I have the time, I like to send them flying into walls. Even just seeing them in a cage makes me feel kind of good inside. I like knowing I'm depriving them of their fre 21:19:43 It's a lot conciser if you hide the etymology and quotations, like the UI lets you do. 21:19:51 It scrapes ninjaword which scrapes Wiktionary. :P 21:20:01 Gregor-W: what. 21:20:04 also, "their fr" cuts off. 21:20:09 alise: <3 The Onion :P 21:20:19 http://www.theonion.com/articles/worlds-scientists-admit-they-just-dont-like-mice,1256/ 21:20:35 bouncebackability, n. the ability of a person or team to bounce back, that is, to return to good form after a period of not performing well: "By passing the test, after last years abysmal failure, he exhibited great bouncebackability."; the ability to recover from a past relationship. 21:21:07 Also the 1500-era quotations are best: "Ay fukkand lyke ane furious Fornicatour." Imagine calling someone a "furious Fornicatour"; classy! 21:21:26 ehird@dinky:~/esoteric$ python define.py xtianity 21:21:26 xtianity, n. the condition of being sane; reasonable and rational behaviour. 21:21:30 Yeah, Christianity is all that. 21:22:06 turns into 21:22:07 Did you mean sanity? 21:22:07 sanity 21:22:07 noun 21:22:07 °The condition of being sane. 21:22:07 °Reasonable and rational behaviour. 21:23:15 ehird@dinky:~$ python define.py xtianity 21:23:15 "xtianity" ain't no word I ever heard of! 21:23:15 ehird@dinky:~$ python define.py wikipedia 21:23:15 "wikipedia" ain't no word I ever heard of! 21:23:16 That's better. 21:25:28 Makes me feel like playing some country music. 21:26:10 Gregor-W: Gimme a word to look up. 21:26:30 antidisestablishmentarianism 21:28:36 The best part about wordnet is that before definition, it lists all words that are partof the same "sense": 21:28:40 1. sleep together, roll in the hay, love, make out, make love, sleep with, get laid, have sex, know, do it, be intimate, have intercourse, have it away, have it off, screw, fuck, jazz, eff, hump, lie with, bed, have a go at it, bang, get it on, bonk -- (have sexual intercourse with; "This student sleeps with everyone in her dorm"; "Adam knew Eve"; "Were you ever intimate with this man?") 21:28:46 Gregor-W: antidisestablishmentarianism, n. a political philosophy opposed to the separation of a religious group ("church") and a government ("state"), esp. the belief held by those in 19th century England opposed to separating the Anglican church from the civil government. 21:28:47 We as a species certainly focus on that thing. 21:29:17 what 21:29:41 fuck, v. (often obscene sometimes extremely vulgar) To have sexual intercourse, to copulate: "I asked her if she wanted to fuck and she said yes, so we had sex together last night."; (often obscene sometimes extremely vulgar) To insert one's penis, or a dildo or other phallic object, into a specified orifice or cleft; n. (vulgar) An act of sexual intercourse: "That was a great fuck."; (vulgar) A sexual partner: "She's a good fuck."; int. expressing dismay 21:29:41 or discontent: "Oh, fuck! We left the back door unlocked." 21:30:14 alise: That last one isn't discontent, it's still the previous meanings. 21:30:31 xD 21:30:38 It's just the imperative. 21:30:51 "Oh, have intercourse! We left the back door unlocked!" 21:30:54 "a specified orifice or cleft"; heh. 21:31:03 Or an unspecified one! 21:31:13 orifice, n. a mouth or aperture, as of a tube, pipe, etc.; an opening; as, the orifice of an artery or vein; the orifice of a wound. 21:31:17 cleft, n. an opening, fissure, or V-shaped indentation made by or as if by splitting; v. past tense of cleave. 21:31:24 So basically we're having an awful lot of very dangerous intercourse. 21:31:28 Involving open wounds. 21:31:28 I'm going to take a shower 21:31:35 coppro: To clean yourself of all this dirty talk? 21:31:46 yes.... let's go with that 21:31:57 `addquote So basically we're having an awful lot of very dangerous intercourse. Involving open wounds. I'm going to take a shower 21:32:08 201| So basically we're having an awful lot of very dangerous intercourse. Involving open wounds. I'm going to take a shower 21:32:24 etape 21:32:24 muggle, n. (in singular or plural dated) A marijuana cigarette; a joint; (slang) hot chocolate; v. (in geocaching) To remove, deface or destroy a geocache; (obsolete) To be restless. 21:33:12 pikhq: Minion Pro is too nice. Halp. 21:33:23 -!- CakeProphet has joined. 21:33:30 I'll ... specify your ... orifice ... or cleft ... 21:33:42 OED "muggle, n." variant 1: "Obs. rare. A tail resembling that of a fish." 21:33:53 alise: Solution: read the pdfTeX microtypography thesis, and then you will never be able to accept the typesetting in any other book. 21:34:51 -!- CakeProphet has quit (Client Quit). 21:34:52 pikhq: How... how does that help? 21:35:09 pikhq: Did you know, the LaTeX minion includes the files that microtype needs. Oh yes. 21:35:11 *minion package 21:35:16 alise: You won't be able to focus on the niceness of Minion Pro. 21:35:24 Also March 2010, draft entry: "In the fiction of J. K. Rowling: a person who possesses no magical powers. Hence in allusive and extended uses: a person who lacks a particular skill or skills, or who is regarded as inferior in some way." I guess they resent Wikipedia getting all the cool popular-culture references. 21:35:33 Instead, you'll focus on the pain of all other typesetting systems. 21:35:40 Soon there'll be dozen-page articles on Wookiee mating. 21:35:55 fizzie: The OED has actually been quite open to adding words. 21:35:56 (Disclaimer: I don't know if Wikipedia has one.) 21:36:39 It generally takes a decent chunk of time to add, but it's not like they're *opposed* at all. They just want to be sure that it's going to be used outside of the work of fiction... 21:37:03 I wouldn't be too surprised if they had a mention of Dune under "spice". 21:37:25 -!- MigoMipo_ has joined. 21:37:27 (the Spice must flow) 21:38:28 they generally don't cite such works 21:38:29 YES! 21:38:31 but they do include words from them 21:38:36 I think I have priting in opengenera working 21:38:40 printing* 21:38:57 however, I need to pass the option to disable cover page to make it work 21:39:10 it fails with the default setting to create a cover page 21:39:12 AnMaster: So is their C-to-LISP compiler actually available? Does it work with semi-real code? 21:39:13 I have no idea why 21:39:20 Gregor-W, I haven't tried it 21:39:28 It's kinda compelling. 21:39:36 -!- MigoMipo has quit (Ping timeout: 240 seconds). 21:40:10 Gregor-W, I haven't got it to load. Load System c always auto completes to CL at the end when you press enter. Very strange bug. Especially since there is a file called c.system 21:40:18 :P 21:40:30 alise: They cite first usage of a sense of a word if possible. 21:40:46 Well, first known print usage. 21:41:46 There is no compatibility level for the new extensions of tracking, additional 21:41:46 kerning, and interword spacing. Therefore, they can only be switched on or off, 21:41:46 or they may be activated by passing a set name to the option. By default, neither 21:41:46 feature is enabled. 21:41:50 Is there any reason not to turn these on? 21:42:57 [[En passant, it may be noted that Type 1 format and T1 encoding are in no other way related than that 21:42:58 both start with a ‘T’ and end with a ‘1’.]] 21:45:57 pikhq: 21:45:59 \usepackage[tracking,kerning,spacing]{microtype} 21:45:59 \usepackage[minionint,mathlf]{MinionPro} 21:46:01 IT IS SO BEAUTIFUL. 21:46:16 alise: Good to know. 21:46:35 Package microtype Warning: \nonfrenchspacing is active. Adjustment of 21:46:36 (microtype) interword spacing will disable it. You might want 21:46:36 (microtype) to add `\microtypecontext{spacing=nonfrench}' 21:46:36 (microtype) to your preamble. 21:46:36 Wut? 21:46:54 Also, the reason for having them off by default is to allow you to get the same line numbers from a TeX document while adding microtype. 21:47:26 The idea being that you can get slightly better output without fucking anything up for bizarre purposes. 21:47:33 Sorry, that's more a reason for having them turn-off-able. 21:47:39 Not... Off by default. 21:47:40 :/ 21:49:19 * pikhq is not sure about that bit, though, alise. 21:49:49 I just did what it said and it shut up :P 21:49:55 alise: Okay, I found it. 21:50:32 alise: Non-French spacing keeps the convention that a space following a period is larger than others. 21:50:43 pikhq: Non-French is just British spacing according to microtype. 21:50:44 So whatever. 21:50:54 Oh my god. Minion's italic & is amazing. 21:51:09 It's of the "et" style, where the e is lowercase rather than curly like lowercase Greek epsilon. 21:51:22 But it isn't all jaggedy like some renderings of that form. 21:51:28 alise: That's the spacing in English in general. 21:51:48 One of those things that we all agree on. 22:02:36 why is there no active desktop in windows 7 >__> 22:04:21 Flonk: Because that is RAPE in UI form. 22:04:46 I'm disappointed that I can't make a fully-animated http://codu.org/tmp/teddynom.gif be my background. 22:06:50 -!- cheater99 has quit (Ping timeout: 265 seconds). 22:11:01 Gregor-W, background on what? 22:11:30 -!- cheater99 has joined. 22:13:38 AnMaster: EVERYTHING 22:13:56 GAAAHWHYCANTANYTHINGDOVERTICALTEXTINJAPANESERIGHT 22:14:30 I'm sorry, I'm wrong. There is one thing that can. 22:14:37 Pango does vertical text right. 22:14:42 :D 22:15:21 PangoTeX 22:16:53 It doesn't merely support it, it actually does it... Typographically correctly. 22:17:03 Ooh, I could typeset The Metamorphosis of Prime Intellect or Fine Structure. 22:17:07 Voting now commences; pikhq, Sgeo. 22:17:20 Mind, I'd have to read either beforehand. :-) 22:17:21 alise: Both! 22:17:28 Voting for what? 22:17:40 So if you type a long vowel mark, will it convert it to vertical long vowel mark if the text is being printed in vertical? 22:18:25 zzo38: Yeah, it selects the vertical variants of glyphs in fonts. 22:18:34 Sgeo: Which book to do; or rather, which to do first. 22:18:41 What's better is that it does *hanging punctuation* at the end of lines. 22:18:47 Does it do furigana? 22:18:47 OK 22:19:10 Actually... a Παν語-based TeX derivative would be amazing. 22:19:16 I don't know if it does furigana. 22:19:19 alise: ... It would. 22:19:25 pikhq: (Pango is a transliteration of Παν語.) 22:19:35 alise: I'm aware. 22:19:46 alise: I can sort-of read Greek and I can read Japanese. ;) 22:19:54 And I'm also aware of Pango besides. 22:19:55 :P 22:20:00 Right, it just isn't immediately obvious. :P 22:20:12 (read, for Greek, is != understand. :P) 22:20:35 Fine Structure's rather long 22:20:38 But would be nice 22:20:44 Would that be legal, though? 22:21:07 So, Παν語-TeX would have Unicode support with full localisation, support multi-directional text, and render even Arabic properly. 22:21:18 It would also have the fine typography support of pdfTeX. 22:21:28 Sgeo: Legal if I asked Sam. 22:21:42 Several people took it upon themselves to create unofficial dead-tree versions of the story: 22:21:42 * Fine Structure (PDF, 1.60MiB) transcribed by AlmightyFjord 22:21:42 * Fine Structure (PDF, 857kiB) transcribed by Duncan Townsend 22:21:42 * Fine Structure (EPUB, 313kiB) transcribed by Daniel Vollmer 22:21:59 No reason I couldn't, then. 22:22:04 What does hanging punctuation means? 22:22:05 pikhq: Didn't you say Lulu can actually do good-quality hardbacks? 22:22:09 zzo38: presumably, instead of 22:22:10 a 22:22:10 b 22:22:10 , 22:22:11 c 22:22:12 it'd be 22:22:12 a 22:22:14 b, 22:22:16 c 22:22:18 for vertical text 22:22:26 alise: O, that's what it means. OK 22:22:30 that's just my guess, though 22:23:13 Sgeo: The first bit of Fine Structure I read was the person with ... X-Ray vision or something, who could also stick their hands through stuff and something, talking to this sciencey type about eir abilities. 22:23:19 Where does that come chronologically, out of interest? 22:23:30 i'm looking for flatmates 22:23:32 it sucks 22:23:42 zzo38: If punctuation would start a line during a paragraph, it is instead placed after the end of the previous line. 22:24:06 To be more specific, the not-bracket punctuation. 22:24:17 So, periods, commas, and the like. 22:24:25 Hmm. I was close. 22:24:26 As far as chronology makes sense, around the beginning 22:24:42 No, I'm wrong 22:24:42 Sgeo: I take it it is a somewhat abstract story. :-) 22:24:48 There's one or two stories before that 22:24:58 And one where "chronologically" doesn't make much sense 22:25:13 nooga: Living alone > living with people, always :P 22:25:22 sure 22:25:24 For all I've heard about it I haven't heard one iota about the actual plot. 22:25:30 Another nice detail it covers is that text that shouldn't be vertical in vertically displayed text instead is rotated. 22:25:30 Gregor-W: /Unless/ these people are sex robots. 22:25:34 Literal sex robots. 22:25:52 http://farm1.static.flickr.com/202/498812683_2f01110063_o.png See the last example in this picture. 22:26:30 (the vertically-displayed "2006" is supposed to be that way; those are the full-width variants) 22:26:55 Review of typesetting mathematics in Minion: Pleasant-looking enough, but it's "too nice" for mathematics, as is common with most serifed fonts; it is too readable and not defined enough, and the pretty letterforms don't help. 22:27:14 Not to say it's ugly or unreadable: but something like Computer Modern just works better, because of the precise mechanicalness about it. 22:27:54 pikhq: Say, can you do Japanese in TeX next to some non-japanese? 22:28:59 -!- CakeProphet has joined. 22:29:17 alise: XeTeX or a lot of font hacking needed, but yes. 22:29:30 pikhq: The latter, please. 22:29:38 With babel I have Greek text working with the proper forms, not the mathematical ones. 22:29:44 \usepackage[greek, english]{babel} 22:29:45 {\greektext Pan}go 22:29:53 alise: WHERE'S MY CPP TURING MACHINE DAMN IT? :P 22:30:07 Babel does most of the layout work just fine; the issue is just getting a font installed. 22:30:11 However, there appears to be no Japanese support for Babel. 22:30:22 pikhq: I have Minion Pro. That probably does Japanese. I guess. Maybe. 22:30:24 It's "cjk" instead of Japanese. 22:30:55 Chinese, Japanese, and Korean have pretty much identical *typesetting* rules. 22:31:02 Not in mine. 22:31:04 No such thing. 22:31:18 I'm on Debian. 22:31:32 Install texlive's langcjk support. 22:31:34 (<3 Debian) 22:31:48 p latex-cjk-japanese - Japanese module of LaTeX CJK 22:31:56 So what's that? How do I utilise that specificness? 22:32:09 It should get you babel support for CJK. 22:32:13 But the Japanese thing? 22:32:15 \usepackage[cjk]{babel} 22:32:16 There's other packages too. 22:32:21 So there is a Japanese-specific element. 22:32:24 ... Hrm. 22:32:32 I'm strongly confused. 22:33:04 p cjk-latex - installs all LaTeX CJK packages 22:33:04 p latex-cjk-all - installs all LaTeX CJK packages 22:33:04 p latex-cjk-chinese - Chinese module of LaTeX CJK 22:33:04 p latex-cjk-chinese-arphic-bkai00 - traditional Chinese KaiTi fonts for CJK 22:33:04 p latex-cjk-chinese-arphic-bsmi00 - traditional Chinese KaiTi fonts for CJK 22:33:05 p latex-cjk-chinese-arphic-gbsn00 - traditional Chinese KaiTi fonts for CJK 22:33:07 p latex-cjk-chinese-arphic-gkai00 - traditional Chinese KaiTi fonts for CJK 22:33:09 p latex-cjk-common - LaTeX macro package for CJK (Chinese/Japan 22:33:11 p latex-cjk-japanese - Japanese module of LaTeX CJK 22:33:13 p latex-cjk-japanese-wadalab - type1 and tfm DNP Japanese fonts for latex 22:33:15 p latex-cjk-korean - Korean module of LaTeX CJK 22:33:17 p latex-cjk-thai - Thai module of LaTeX CJK 22:33:19 p latex-cjk-xcjk - XeTeX module of LaTeX CJK 22:33:21 Flood, yes, but I hope that explains. 22:33:23 CJK is a macro package for LaTeX. This package gives you the possibility to 22:33:25 include Japanese text in your (La)TeX documents. Install 22:33:27 latex-cjk-japanese-wadalab for pretty printing. 22:33:29 22:33:31 Install hbf-kanji48 if you want to use bitmap fonts in your documents. 22:33:33 22:33:35 Have a look at latex-cjk-common for a more detailed description. 22:33:37 And [cjk] doesn't work. 22:34:25 * pikhq is more confused 22:34:43 And why is Thai part of CJK? It has completely different conventions. 22:34:55 pikhq: THIS is just why Παν語-TeX must be done :-) 22:35:26 -!- zzo38 has quit (Ping timeout: 265 seconds). 22:44:20 ok, there's no way i'm going to be able to read all this backwash 22:44:23 backlog 22:45:00 :-) 22:45:08 It's mostly typography stuff; but also my dictionary tool. 22:45:14 OED-style definitions instantly! 22:45:51 cpressey: define.py -- http://pastie.org/1067745.txt?key=zrx2wuv4p8fndsov9ghbww 22:46:00 alise: for your python parsing problem which you are no longer working on -- or for ANY parsing problem in any language, really -- the trick is to decompose sentences like that into "these are the tokens" and "this is the grammar". Then implement the tokens with regexps in a scanner and implement the grammar with a RDP 22:46:46 cpressey: Yeah, I know. 22:47:26 http://www.w3.org/TR/jlreq/ Glee. 22:47:31 -!- MigoMipo_ has quit (Remote host closed the connection). 22:47:35 then it becomes basically trivial in python or anything else with half-decent string manipulation that you're willing to write a few functions in 22:48:01 pikhq: what's that free font used in the WP logo you mentioned? 22:48:03 -!- MigoMipo_ has joined. 22:48:25 alise: Linux Libertine O. 22:48:25 pikhq: Yeah, if only W3C standards actually mattered. 22:48:52 They are THE most ineffective standards organisation. Even ESO was better. :P 22:48:54 alise: Actually, it's a note published by the authors of the JIS standard on text layout. 22:49:07 Erm, written by. 22:49:24 http://en.wikipedia.org/wiki/Logo_of_Wikipedia Does. Not. Need. An. Article. 22:49:34 Which is mostly a standardisation of the principals of good CJK typesetting. 22:50:04 It just happened to be submitted to the W3C for their consideration. 22:50:29 It is, in short, "how to typeset Japanese: In English!" 22:51:26 That's right bitches. 22:51:30 English always pwns ur language. 22:52:06 pikhq: How to typeset Japanese ... in Korean! 22:52:16 The least-read document on the planet. 22:53:00 alise: Given that the animosity the Koreans have toward Japan rivals Brits' towards France... Yeah. :P 22:53:33 pikhq: But our rivalry with the French is mostly just joking; there's an awful lot of Korean racism against the Japs. 22:53:44 And yes, I will continue saying Japs because as a word it amuses me. (Nigger nigger nigger.) 22:54:16 alise: Well. Yeah, that's because it's old history for you... 22:54:27 alise: Imagine if the Norman invasion were about 70 years ago. 22:54:44 You know, we don't hate the Germans. :P 22:55:08 The Germans didn't assert that your language was a dialect of German and thereby try to exterminate your language. 22:55:21 Fair enough. :P 22:57:07 nigger! 22:57:17 oklopol: Nigger-faggot. 22:57:22 faggot homo 22:57:28 this is truly a high point for this channel 22:57:35 :D 22:57:45 graph rewriting grammars are so cool 22:58:01 hahahaha 22:58:04 what did I just miss 22:58:07 maybe i could make a formal computational model out of graphica and get it published 22:58:53 so I just got an idea for a language that's not really all that esoteric at all 22:59:11 we all do that sometimes. unnatural, filthy; bury it. 22:59:15 oklopol: or out of clue 22:59:16 :P 22:59:32 or out of toi! 22:59:44 prototype IO. strong, static, inferred, and optional typing. 23:00:05 math people find it interesting usually 23:00:16 CakeProphet: that sounds really nigger to me 23:00:28 "Plan To Start Little Stationery Store Too Sad For Bank To Deny Loan" --The Onion 23:00:29 does it have a funny syntax? 23:00:31 essentially I'm considering the possible of having a dynamically typed language with a static type checker, in essence. 23:00:37 oklopol: no syntax yet. :) 23:00:48 yeah i think CakeProphet sounds like a faggot nigger 23:00:52 i mean the cake is like totally flamboyant 23:00:58 maybe you could have "if" be "LOLWHATIF" 23:00:59 and as we all know jesus was a prophet and black soooo 23:01:03 basically he's FaggotNigger. 23:01:09 ... 23:01:12 oklopol: or while be HENCEFORTHWHEREAS 23:01:14 i'm looking for a way to make some kind of 'jail', that you can connect to via ssh but you cannot use sockets from inside and you can read/tty/S 23:01:18 CakeProphet: you can't deny it 23:01:22 /tty/S* 23:01:23 is this what our beloved channel has come to? 23:01:32 yes. 23:02:01 -!- GreaseMonkey has joined. 23:02:05 but I have a feeling other people in this channel (ahem, alise) would have better ideas on how to do OO correctly, and so I think discussing it would be worthwhile. 23:02:19 lol 23:02:31 alise: how FaggotNigger is GreaseMonkey tho :DD 23:02:36 don't! 23:02:36 Yeah, but we'd rather say "nigger" a lot. 23:02:37 blah 23:02:40 the way I imagine it working is simply through operators. obj1 + obj2 23:02:41 oklopol: haha wow that works perfectly. 23:02:46 and other constructs. 23:02:50 CakeProphet: if that isn't addition don't say + 23:02:56 especially anonymous object declarations. 23:03:06 alise: it's not, and yes...the syntax is just for brainstorming 23:03:11 like whayt 23:03:14 therefore don't say + 23:03:22 say @ or : or something :P 23:03:39 alise doesn't like needless overloading 23:03:46 alise: I don't think it matters for conversation. I assure you I will not make another Python. 23:03:53 ................................... 23:04:09 what a gawdaefully written proof. (that RH thing) 23:04:18 *gawdawfully 23:04:34 alise: I don't know if you knew. But + means everything in Python 23:04:35 cpressey: RH? 23:04:38 ah 23:04:39 yeah 23:04:41 it is pretty awful 23:04:45 Gregor-W: 35 points 23:04:48 his "laws of form" are even worse though cpressey 23:04:54 cpressey: http://lawsofform.org/ideas.html 23:05:08 http://en.wikipedia.org/wiki/Laws_of_Form does some sort of attempt at decoding it and giving it mathematical coherence 23:05:11 the "laws of form" i don't mind so much if they're not passed off as something they're not, but i guess they are 23:05:19 oh i forgot i found laws of form from your site 23:05:20 lol 23:05:35 did i put it in the links? hm 23:05:40 yeah 23:05:43 in the esoteric section 23:05:46 cpressey: he also claims to have proved 23:05:49 - four colour theorem 23:05:51 - goldbach conjecture 23:05:51 so... what other composition operators are there? and what about implementation? Can you efficiently do logical combinations of dynamic objects without using massive amounts of hash tables? 23:05:53 -!- Sgeo has quit (Ping timeout: 265 seconds). 23:06:06 he also claimed that the Appel, Haken & Koch proof of the four-colour theorem was incorrect. 23:06:15 pretty sure he's just a crank who reinvented boolean logic with circles 23:06:21 i need to put trivialism and that ultrafinitist mathematician in there too now i guess :) 23:06:40 * CakeProphet is an ultratrivialist 23:06:48 The ultrafinitist is Doron Zeilberger. 23:06:54 Trivialism is a joke, I'm /pretty/ sure. 23:07:01 no see 23:07:11 ultratrivialism must be distinguished trivially from trivialism 23:07:18 boolean logic with circles? 23:07:19 even though they are the same. 23:07:28 cpressey: Some extra kookiness: paraconsistent logics (trivialism is a parody of these). 23:07:31 What, and Zeilberger *isn't* a joke? Oh my. 23:07:46 Graham Priest is especially a devotee. 23:07:49 cpressey: He sort of jokes about everything. 23:08:04 He's not actually stupid -- he's done some valuable work like Zeilberger's algorithm -- 23:08:08 but he's definitely kooky. 23:08:16 He's a fun read. Just not one to take seriously. 23:08:33 what does zeilberger's okay i'll google 23:08:36 so I guess +/alise-(@,:) could be a non-commutative operation. 23:08:39 "In my eyes Connect-Four is much deeper than, say, K-theory. Of course, K-theory is deeper than the Win-In-One-Move problems, and perhaps even than the Win-In-Three-Moves problems, but definitely not deeper than the Win-In-Five-Moves problems in this book." -- Zeilberger 23:08:44 oklopol: the ultrafinitist 23:08:57 where you specify that one object is the delegate of the new object. 23:09:09 delegateObject @ returnObject 23:09:22 alise: i know, i was gonna ask what his algorithm ... 23:09:24 == returnObjectWithDelegation 23:10:00 -!- Sgeo has joined. 23:10:00 http://mathworld.wolfram.com/ZeilbergersAlgorithm.html 23:11:03 and then : could be direct, commutative combination 23:11:35 http://www.math.rutgers.edu/~zeilberg/mamarim/mamarimPDF/real.pdf 23:11:40 CakeProphet: What are you talking about? 23:11:41 is Zeilberger's most famous ultrafinitist thing 23:11:44 and the most kooky 23:12:02 cpressey: prototype object-oriented language. operators that combine objects together logically. 23:12:18 or, in general, transform them. 23:12:23 CakeProphet: OK 23:14:07 hmmm... I wonder if you could define public/private as a combination. 23:14:09 It is utter nonsense to say that 2 is irrational, because this presupposes that it exists, as a 23:14:09 number or distance. The truth is that there is no such number or distance. What does exist is 23:14:09 the symbol, which is just shorthand for an ideal object x that satisfies x^2 = 2. 23:14:24 ............. 23:14:33 that is so much bullshit in one sentence. it's unbelievable. 23:14:37 So the true 23:14:37 Pythagorean theorem is not c^2 = a^2 + b^2 , but rather c^2 = a^2 + b^2 + O(h). 23:14:44 CakeProphet: i know, isn't zeilberger fun? 23:14:50 in fact he's the most intelligent bullshitter i know of. 23:14:54 "IN ORDER TO UNDERSTAND SOMETHING REALLY DEEPLY, YOU SHOULD PROGRAM IT." 23:14:59 that's deep 23:15:23 so... is private sort of like a... binding operation? 23:15:35 pikhq: I need a stylistic opinion! 23:15:43 you could take one object of private members, and one object of public members 23:16:15 alise: Oh? 23:16:15 pikhq: Should the name of the λ-calculus be written with a mathematical, italic lambda, i.e. $\lambda$-calculus, or the actual textual, roman (regular, not actually roman :P) Greek letter lambda, as in {\greektext l}-calculus? 23:16:19 I believe the latter; it flows more nicely. 23:17:11 alise: Latter. Definitely latter. 23:17:24 Right. 23:17:40 And dammit I want Παν語-TeX. 23:18:24 CakeProphet: bind an attribute to the set of objects that can see it 23:18:29 pikhq: "This document was typeset by Παν語-\TeX using \AmS-\LaTeX." 23:18:39 Anything that can set those bytes correctly receives my undying love. 23:18:55 Properly as in "with Pango". 23:18:59 cpressey: right, so one object represents a private set of attributes, another represents the public set accessing it. 23:19:03 I can't trust XeTeX; especially its OS X lineage. 23:19:04 don't ask me how protected works in this. 23:19:17 http://www.laws-of-form.net/lof/pdf/Denjoy_proof.pdf <<< i can't get this open, someone copy paste it on irc plz 23:19:22 there isn't really inheritance, so protected would be delegate based. 23:19:25 alise: Mmm. 23:20:31 or perhaps I am trying to make too many things operators? :P 23:20:31 pikhq: OTOH, modifying TeX like that is hell. 23:20:41 pikhq: In fact... perhaps a LuaTeX derivative would be the most profitable. 23:20:45 why latter, i mean the name comes from the fact the lambda is actually used in the mathematical notation 23:21:02 or would you use the text lambda for that too? 23:21:11 oklopol: because the italic, mathematical lambda is more a variable symbol 23:21:14 CakeProphet: enable custom operators 23:21:18 than the actual lambda symbol used, which was an extension of ^ 23:21:24 (because ^ wasn't distinct enough or something) 23:21:28 eg A +-+ B 23:21:34 plus, it disrupts the flow of text. 23:21:36 nooga: yes. I was planning on that. 23:21:39 so... "yes" 23:21:58 I was actually going to borrow a lot from Haskell, but in the context of a non-functional paradigm. 23:22:05 well... less functional paradigm. 23:22:33 I was wondering if ternary operators would be reasonable. 23:22:38 also when reading lambda calculus, i'm not sure it's crucial the flow of text doesn't get interrupted, you won't be reading long snippets anyway 23:22:47 I think it would. As long as no unary or binary operators conflict in name. 23:22:58 every operator could be ternary 23:23:07 you'd get three times more stuff done 23:23:11 ha. 23:23:22 3 + 4 >> SUPERPOWERS 23:23:49 pikhq: Opinion: (\emph{...}) or \emph{(...)}? 23:23:58 so a ternary operator would in effect bind TWO names. 23:24:09 The latter seems to flow better. 23:24:11 yeah every operator has a comment track, (5 * (3 + 4 ;; the sum of three and four, which is 7) ;; we multiply 5 by 7 to get 35) 23:24:13 to prevent headaches with parsing ambiguity. 23:24:39 pikhq: On the other hand, ``\emph{...}'' is obviously the correct form. 23:24:52 also when reading lambda calculus, i'm not sure it's crucial the flow of text doesn't get interrupted, you won't be reading long snippets anyway 23:24:52 That Bipolar Lisper article is way worse than that kook RH proof, though. 23:24:55 reading the /name/ lambda-calculus 23:24:57 not lambda calculus itself 23:24:59 ohh. 23:25:07 i.e. λ-calculus 23:25:15 yeah sorry <--^---^--- idiot 23:25:23 ^ means the arrow jumps over a word 23:25:26 -!- Flonk has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]). 23:25:42 oklopol: I wonder what kind of operator <--^---^--- would be in Haskell? 23:25:53 Yegge is always a hair snotty for my taste though 23:26:33 CakeProphet: i have a hunch 23:27:03 -!- tombom has quit (Ping timeout: 240 seconds). 23:27:29 I wish @ and ' were valid operator character in Haskell. Then we could have the @-,-'- operator 23:27:36 aka "the rose" 23:27:41 -!- tombom has joined. 23:28:02 CakeProphet: They are in Plof! :P 23:28:05 (it's a thorny monad) 23:28:16 PLOF! 23:29:47 let x ^.^ io = unsafePerformIO io `seq` x 23:30:02 "Pick Scheme, and you have to pick a Scheme" -- wtf? This is like when Stanislav said Scheme's libraries "weren't standardized" 23:30:15 in 2+2 ^.^ computerExplosionSystemCall 23:31:28 ^.^ -- the strict side-effect operator. happy faces for all. 23:31:33 pikhq: You'd like Minion's quote marks; they're set so tightly that "Foo," and "Foo", look almost identical. 23:31:42 * cpressey stops reading 23:31:43 Yegge is always a hair snotty for my taste though 23:31:45 he's getting better 23:31:59 alise: Huh. 23:32:00 oklopol: YES! PLOF! 23:32:15 pikhq: This can be controlled, however. 23:33:31 When using LaTeX's \maketitle, this produces hideously classy results: 23:33:36 \title{\textssc{Some Title}} 23:33:40 \author{\textsc{My Name}} 23:33:41 \date{} 23:33:48 (requires Minion Pro for ssc) 23:33:50 *ssc font 23:33:54 well, not a font 23:34:18 I wonder if I'll get time to work on something this weekend, and if so, wtf I should work on 23:34:32 cpressey: So you have internet on weekends now? 23:34:36 It's partly time, partly brain power 23:34:44 alise: Not generally 23:35:23 -!- oerjan has joined. 23:35:27 Don't generally need internet for my projects. 23:35:37 Though having access to Java docs might be nice. 23:35:52 Why don't you have internet, again? 23:36:02 Because I live in a cardboard box. 23:36:57 cpressey: Wat. 23:36:58 Because I am a brilliant failure. 23:37:05 pikhq: Aww, optical sizings don't work with my Minion package. 23:37:09 Lisp is the flag I wave, you see. 23:37:12 cpressey: Can you not even receive 3G signal there? :P 23:37:21 ! Font T1/MinionPro-OsF/m/n/17.28=MinionPro-Subh-osf-t1 at 17.28pt not loadable 23:37:21 : Metric (TFM) file not found. 23:40:28 I wish @ and ' were valid operator character in Haskell. Then we could have the @-,-'- operator <-- um @ isn't? let me check 23:40:35 oerjan: o@x 23:40:35 so no 23:40:51 !haskell let a @+ b = a + b in 2 @+ 2 23:41:04 4 23:41:14 alise: hah! 23:41:38 well poo unto you 23:41:39 ' on the other hand is afair a _letter_ character, except at the beginning 23:41:49 i would like to inform you all that i just fixed an error in Gregor's plof implementation using my penis 23:41:57 !haskell let a @-,-'- b = a + b in 2 @-,-'- 2 23:41:58 the , in there won't work, either 23:42:04 Gregor-W can confirm 23:42:27 cpressey: ' and , are illegal. @--- should work, however 23:42:30 He didn't fix it, he just identified it. 23:42:48 !haskell let a @-|-+- b = a + b in 2 @-|-+- 3 23:42:50 5 23:42:58 Sort of an oddly symmetrical rose, but 23:43:02 a rose 23:43:12 Gregor-W: that's the hard part, the rest is just monkey work right 23:43:19 (' at the beginning starts a character constant, ' otherwise is permitted in things like x') 23:43:20 oklopol: Probably :P 23:43:31 clickety clickety change this code here i'm a monkey i'm a monkey 23:43:43 There might be Unicode widgetyglyphs that could sub for , and ' 23:43:45 i'm too tired 23:43:58 oklopol: frigging monkey work 23:43:59 hate it 23:45:13 Probably not going to work on esolang this weekend, not enough brain left. Might finish reading Also Sprach Zarathustra. 23:45:44 cpressey: well that's why i use Gregor-W, i just like the part where i write penis penis in a textbox and identify an error 23:45:57 alise: @ despite appearances is a reserved _word_ (operator-like word), not a reserved character, so it doesn't count as special inside other operators. many other things like -- and \ are similar, afair. but somethings work like actual punctuation, like brackets, semicolons and commas. iirc. 23:46:09 oklopol! :D 23:46:15 augur: D! 23:46:16 wait umm 23:46:20 that came out wrong 23:46:34 x3 23:48:19 since haskell tries hard not to allow two operators to be consecutive, it's not often noticable whether you think of them as characters or words. i think. 23:48:34 good mush night folks mush mush mush splat 23:48:36 -!- cpressey has quit (Quit: Leaving.). 23:49:01 (that may be one reason _why_ haskell disallows prefix/suffix operators (except -). i don't recall seeing it explicitly, though. 23:49:04 ) 23:49:07 -!- MigoMipo_ has quit (Read error: Connection reset by peer). 23:50:25 -!- Gregor-W has quit (Quit: Page closed). 23:51:39 Gahwhat?! I broke Minion. 23:53:56 -!- augur has quit (Remote host closed the connection). 23:54:05 -!- augur has joined. 23:54:07 * oerjan declares today's logs too long to read all through 23:54:32 well all of it was complete bullshit anyway 23:54:42 MOST LIKELY 23:54:48 i don't really remember any of it 23:54:52 i was sooooooooooooooo drunk 23:55:09 -!- augur has quit (Client Quit). 23:55:14 -!- augur has joined. 23:58:16 as long as you weren't drank 23:58:55 well... actually i *kind of* was :DDDD