00:00:14 Deewiant, btw at some point I will probably implement DATE and maybe also EVAR. Maybe NCRS too 00:11:51 congratulations on passing your mycology exam :) 00:12:47 * SimonRC wishes he could be as productive as you people 00:12:54 that is mostly not sarcastic 00:18:02 -!- warrie has quit (Read error: 60 (Operation timed out)). 00:18:27 -!- warrie_ has joined. 00:18:53 -!- warrie_ has changed nick to warrie. 00:22:08 -!- Corun has joined. 00:29:37 SimonRC: we're productive? 00:31:20 * SimonRC listens to the final episode of The Brightonomicon on BBC 7. 00:31:28 oklopol: yeah, you write actual code 00:31:36 * SimonRC goes 00:31:45 i've actually done very little of that in the recent months. 00:32:02 mostly been reading, not sure you can call that productive 00:32:42 oklopol: your mom got impregnated by a zebra 00:33:04 -!- jix has quit ("..."). 00:33:36 err, why does this info come to me through you? i should really try to keep better in touch with my parents. 00:34:06 mainly because you asked us to 00:34:42 23:18 oklopol> okay. i'm going to leave now. if i come back in less than, say, three hours, please say something insulting about my mother 00:35:14 (it's now 01:34 here) 00:35:55 thanks, i had so forgotten that thing i said like 5 minutes ago :P 00:36:22 (and that was sarcasm IF YOU DIDN'T NOTICE : |) 00:36:44 oh sarcasm, thought it was alzheimer 00:38:00 alzheimerrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr 00:38:07 back to my readings 00:38:39 second reads are always nicer, i remember pretty much everything, so i can just skip almost every page 00:39:34 ---------------> 00:42:41 lol "zebra" comes from "wild ass", puts this whole conversation into perspective really. 00:58:44 note also that zebras are black with white stripes rather than white with black stripes 00:59:39 * oerjan whacks olsner ----### 01:00:00 * SimonRC returns 01:00:02 ... it lacked a certain something, but good nonetheless. 01:00:26 what lacked and what did it lack? 01:00:51 should have used more rum 01:01:40 the thing I went /away to listen to 01:03:03 The Brightonomicon 01:04:44 what's that 01:05:28 the thing that was on the radio: http://www.bbc.co.uk/programmes/b00fksh6 01:08:50 * SimonRC goes to bed. 01:12:17 my leg is already sleeping, maybe I should follow it soon 01:17:46 my back is already hurting, so guess i should start hitting the other parts of my body against a wall or something. 01:18:20 yes, even it out a bit 01:18:58 you're in luck though - it's far easier running forward into a wall than running backwards into a wall 01:19:39 actually i'm fluent in both. 01:53:36 ehird (or whoever can answer): In Jumpfuck, does a continuation include the memory pointer, or just the instruction pointer? 02:09:40 -!- Corun has quit ("Leaving"). 03:32:48 -!- oerjan has quit ("Good night"). 03:35:54 -!- Corun has joined. 04:30:50 -!- bsmntbombdood has quit (Read error: 60 (Operation timed out)). 04:42:01 -!- bsmntbombdood has joined. 05:21:17 -!- GreaseMonkey has joined. 06:42:55 -!- Dewi has quit (Read error: 60 (Operation timed out)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:06:22 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 08:14:38 -!- GreaseMonkey has quit ("So, how much do you love noodles?"). 08:59:09 morning 09:22:23 morning! :D 09:26:21 -!- Slereah has joined. 09:30:11 -!- Slereah_ has quit (Read error: 60 (Operation timed out)). 09:43:12 -!- oklokok has joined. 09:46:08 -!- oklopol has quit (Read error: 60 (Operation timed out)). 11:16:16 -!- moozilla has joined. 11:17:20 -!- moozilla has quit (Read error: 104 (Connection reset by peer)). 11:17:24 -!- moozilla has joined. 11:18:45 -!- moozilla has quit (Read error: 104 (Connection reset by peer)). 11:23:33 -!- moozilla has joined. 11:23:45 -!- Mony has joined. 11:24:42 plop 11:32:33 -!- moozilla has quit (Nick collision from services.). 11:32:34 -!- metazilla has joined. 11:34:23 -!- metazilla has quit (Read error: 104 (Connection reset by peer)). 11:34:26 -!- moozilla has joined. 11:47:17 -!- moozilla has quit (Read error: 104 (Connection reset by peer)). 12:18:39 -!- moozilla has joined. 12:29:54 -!- moozilla has quit (Read error: 131 (Connection reset by peer)). 12:51:02 MizardX: ehird (or whoever can answer): In Jumpfuck, does a continuation include the memory pointer, or just the instruction pointer? 12:51:05 just instruction 13:06:22 o 13:37:11 -!- moozilla has joined. 14:04:36 -!- kar8nga has joined. 14:19:52 -!- oerjan has joined. 14:23:28 -!- moozilla has quit (Read error: 104 (Connection reset by peer)). 14:23:30 -!- metazilla has joined. 15:09:20 -!- KingOfKarlsruhe has joined. 15:10:38 ehird: That would duplicate the argument on the tape. The next cell being the argument, the current cell replaced by the value of the argument; Leaving two copies of the argument, starting from the current position. 15:10:51 MizardX: Ah, I misunderstood you. 15:10:53 Yes, memory pointer too. 15:10:59 But not the actual memory contents. 15:11:00 Obviously. 15:29:30 -!- Corun has quit ("Leaving"). 15:38:54 ehird: I think I managed to build simple subroutines, but they don't support recursion. http://pastie.org/316101 15:39:06 MizardX: very nice! 15:39:25 MizardX: hmm, is that fortran-style calling? 15:39:28 each function gets one frame on the stack 15:39:31 so no recursion 15:40:15 Since only one argument is allowed, you can either pass the return address, or an argument. A function without arguments isn't very useful. 15:40:43 err... return continuation 15:41:35 -!- sebbu has quit (Connection timed out). 15:41:52 MizardX: Hm, the one argument thing is a bit of a problem, but - 15:41:55 do you know how to do lists in brainfuck? 15:41:58 1 VAL 1 VAL 1 VAL 0 15:42:15 You traverse over, using your list-index as the hopper instead of 1, and decreasing each time 15:42:16 Sorta. 15:42:24 You could do that, just specify an index in the list as an argument list... 15:42:24 I think 15:44:11 Hmm... using a call stack at the beginning of the tape... 15:44:47 Tee hee... 15:44:51 JumpFuck was a 5 minute thing... 15:45:34 But how to find your way back when you return from the sub-routine? 15:45:57 memory pointer would be positioned at the stack 15:47:40 continuation call for that too? It would be a lot of overhead for a simple sub-routine call... 15:50:00 MizardX: Well, subroutines aren't that cheap... 15:50:18 But seriously, a continuation is just two integers... 15:50:22 And calling it is just setting two. 15:54:36 Well... too tired to think about that now. Time to go and buy some pizza! :) 15:56:36 Fortran-style sub-routines for call/return, and the rest using the stack. 16:09:38 -!- Ilari has quit ("Won't be back for a while..."). 16:22:15 o 16:24:20 -!- sebbu has joined. 16:24:40 -!- kar8nga has quit (Read error: 60 (Operation timed out)). 16:28:24 -!- oerjan has quit ("leaving"). 16:39:28 -!- KingOfKarlsruhe has quit (Remote closed the connection). 16:54:14 -!- warrie_ has joined. 17:03:44 -!- sebbu2 has joined. 17:10:43 -!- warrie has quit (Read error: 110 (Connection timed out)). 17:16:17 -!- sebbu has quit (Read error: 113 (No route to host)). 17:16:18 -!- sebbu2 has changed nick to sebbu. 17:26:03 -!- AquaLoqua has joined. 17:43:17 -!- Corun has joined. 17:46:03 -!- Corun has quit (Client Quit). 17:58:13 o 17:58:40 o 17:58:52 ^ul (oOOoOOooOOoOOo)S 17:58:52 oOOoOOooOOoOOo 17:59:07 how about that. 17:59:58 ^ul (^ul^ul (^ul)S)S 17:59:58 ^ul^ul (^ul)S 18:00:20 -!- AquaLoqua has quit ("Dana"). 18:01:05 -!- Corun has joined. 18:10:33 ^ul ((+ul)S)S 18:10:33 (+ul)S 18:10:51 ^ul (+ul (foo)S)S 18:10:51 +ul (foo)S 18:10:51 foo 18:11:59 +ul (^ul (+ul (foo)S)S)S 18:11:59 ^ul (+ul (foo)S)S 18:12:12 ^ul (+ul (^ul (foo)S)S)S 18:12:12 +ul (^ul (foo)S)S 18:12:12 ^ul (foo)S 18:12:53 +ul (a\nb)S 18:12:53 a\nb 18:13:39 +ul (a 18:13:40 ...out of time! 18:13:58 +ul (a 18:13:59 ...out of time! 18:14:09 geh.. breaks on both cr and lf 18:14:24 +ul xxx 18:14:37 ... 18:14:40 ^ul xxx 18:14:40 ...bad insn! 18:14:46 +ul (xxx 18:14:47 ...out of time! 18:14:49 +ul (xxx) 18:15:01 right, that's valid but a no-op 18:17:44 +ul ((foo):S:^):^ 18:17:45 foo 18:18:13 +ul (S:^):^ 18:18:14 S:^ ...: out of stack! 18:19:05 +ul ((+ul )S:aSS)(+ul )S:aSS 18:19:06 +ul ((+ul )S:aSS)(+ul )S:aSS 18:19:55 -!- sebbu2 has joined. 18:21:08 ( ) pushes content onto stack, S prints top of stack, : duplicates top of stack, a wraps top of stack in parenthesis 18:21:19 executed from left to right 18:22:12 MizardX: yes that's a random subset of underload command alright. 18:22:26 what I can remember 18:22:42 ! is pop, it's like "fuck off!!" 18:22:44 ... 18:22:45 or something 18:22:49 yes, like that 18:22:57 +ul (foo)(bar)!S 18:22:57 foo 18:23:13 ~ swaps the two topmost stack cells 18:23:35 and that's it. 18:23:52 +ul (foo)(bar)SS(foo)(bar)~SS 18:23:53 barfoofoobar 18:24:28 you can remember ~ from like a worm that wiggles into the stack and brings out cell number two 18:24:40 i like to make fun memory pegs. 18:24:48 you're weird 18:24:59 :P 18:26:22 +ul (foo)(bar)*S 18:26:23 foobar 18:26:25 i'm tired 18:26:34 sleep 18:26:51 no i gotta read / complain about not having read enough this weekend 18:27:03 MizardX: are you new @ irc and old @ esolangs.org? 18:27:11 *new @ #esoteric 18:27:30 a week or two 18:28:21 because your nick sounds familiar, but i don't remember seeing you here 18:28:28 wiki 18:28:34 about the same time 18:28:38 i c. 18:28:54 planning on compiling more stuff to other stuff? 18:30:04 We'll see... trying to get Redivider -> C working 18:30:22 what is the implementation language? 18:30:27 python 18:30:53 i was thinking subleg->bf, afaik there aren't that many cycles in the compilation graph. 18:31:02 or was that true for interpretation only 18:31:22 you and ais inspired me to do bf->subleg on friday 18:31:39 it's nice when people do stuff, gives me incentive to do at least as much stuff. 18:31:43 -!- sebbu has quit (Connection timed out). 18:31:45 -!- sebbu2 has changed nick to sebbu. 18:31:55 * oklokok is a very competitive weirdo 18:32:00 -!- oklokok has changed nick to oklopol. 18:32:40 I'm going to climb Mount Everest tomorrow. >_> 18:33:14 Ah, maybe should go to sleep. I'm sitting here laughing at my self 18:33:51 MizardX: then may i assume your nick comes from mountain wizard extreme"? 18:33:53 * 18:34:02 *"moutnain wizard extreme" 18:34:05 hmph 18:34:14 typing is hard 18:37:10 It was a joke, a lie to get rid of the competition. Bad joke, since I was the only one laughing. 18:39:02 Mizard was originaly from Lizard, but "Wizard" was later adopted. X was first just a letter, and now means unknown/secret. An alias so to speak. 18:44:09 well that's a bit dissappointing, mountain wizards are cool. 18:44:29 oklopol: How goes the running? 18:44:34 fizzie: 100.6 18:44:38 oklopol: Well done! 18:44:41 yes! 18:44:46 and one day ahead of schedual 18:45:12 -!- Corun has quit ("This computer has gone to sleep"). 18:45:43 A friend on other channel had (without provocation) gotten to the same 100.2 result with exactly the same technique I used. 18:46:00 yeah, and i watched the youtube clips, everyone is doing it that way 18:46:02 except me 18:46:40 i may make a vid at some point 18:52:06 also i may want to get a better jump 18:52:14 i basically just fell on my face at the end of track. 18:52:24 was almost fainting from excitement at that point 18:53:03 hmm 18:53:08 i think it's shoppe time 18:53:08 -> 19:00:37 -!- Corun has joined. 19:01:40 -!- sebbu2 has joined. 19:10:37 -!- LinuS has joined. 19:14:47 -!- sebbu has quit (Read error: 113 (No route to host)). 19:14:49 -!- sebbu2 has changed nick to sebbu. 19:24:49 -!- Corun has quit ("This computer has gone to sleep"). 19:37:28 -!- jix has joined. 19:55:27 -!- Mony has quit ("Join the Damnation now !"). 20:09:32 if I implemented a SQL fingerprint, would anyone use it? 20:09:38 what about you fizzie for example? 20:10:41 oh and ATHR isn't dead if anyone thought so 20:10:46 I just had a lot to do recently 20:12:26 -!- comex has changed nick to biden. 20:15:26 could i secure my Oracle server with an SQL fingerprint reader? :3 20:17:46 sigh 20:17:49 SQLite would be cool :) 20:17:52 funge fingerprints -_- 20:18:15 Asztal, question is, would anyone use it? 20:18:24 and would it fit into the funge "feel"? 20:18:55 and it would be sqlite 20:19:26 maybe it's just better to wait until the need arises? 20:19:32 probably 20:19:59 but if anyone wanted sqlite, would they even consider funge? 20:20:45 if anyone wanted anything, would they even consider funge? 20:20:58 psygnisfive: fingerprints are terrible for authentication 20:21:08 Deewiant, heh 20:21:12 ehird: i know :p 20:21:17 :D 20:21:21 * AnMaster have been working with mysql for most of the day :/ 20:21:23 horrible 20:21:30 not my choice however 20:21:51 anmaster have been neglecting subject-tensed-verb-agreement 20:21:51 I think I'd rather use any sort of funge SQL fingerprint than mysql 20:22:04 oh, wait 20:22:07 Asztal, same 20:22:07 is AnMaster talking? 20:22:08 that'd make more sense 20:22:15 this conversation was ever so slightly surreal 20:22:18 yes he is 20:22:19 haha 20:22:19 hah 20:22:22 no 20:22:23 no no 20:22:23 bbl food 20:22:25 anmaster is not talking 20:22:28 this is just #esoteric 20:22:28 XD 20:22:32 of COURSE its surreal 20:22:55 hold the newsreaders nose squarely, waiter, or friendly milk will countermand my trousers! 20:23:23 that's not as much surreal as obfuscated 20:23:27 You have two cows. The government gives you arthritis. 20:23:43 obfuscated? sir, its surreal! 20:23:50 why would FRIENDLY milk countermand your anything?! 20:24:01 surely only mean milk would do that 20:24:05 happens all the time in these parts 20:24:22 when Dali was asked if he did drugs he replied "My dear, I /am/ drugs" 20:24:49 i am bugs 20:25:35 is someone speaking? 20:25:40 I have everyone ignored 20:25:48 me too 20:25:50 -!- jayCampbell has left (?). 20:26:00 ILONY 20:31:49 bread with garlic butter yum 20:32:28 * bsmntbombdood steals AnMaster's garlic bread 20:32:33 * bsmntbombdood breathes on the channel 20:32:35 already ate it 20:32:46 I'm just *back* from the meal 20:32:49 * psygnisfive stabs bsmntbombdood through the eye 20:33:02 breathe on me again, motherfucker, see what you get 20:33:08 * AnMaster breathes on ehird 20:33:13 that was my glass on, loser 20:33:14 oh thats hot 20:33:15 :o 20:33:19 haha 20:33:27 sigh 20:33:34 anyway 20:33:43 *eye 20:33:48 anmaster have been neglecting subject-tensed-verb-agreement 20:33:52 wtf did that mean? 20:34:09 "AnMaster have been working with mysql for most of the day :/" 20:34:14 oh 20:34:15 had* 20:34:16 right 20:34:18 no 20:34:19 :P 20:34:21 err 20:34:30 "AnMaster" is third person singular 20:34:37 oh right 20:34:38 pretend you're talking about ehird 20:34:41 "ehird have ..."? 20:34:43 surely not 20:34:50 not unless you're british and ehird is a band 20:34:57 i am 20:34:58 XD 20:34:59 psygnisfive, but my client shows it as peforms /me: 20:35:01 ;P 20:35:03 i have 1 member: ehird 20:35:03 O SHI 20:35:07 psygnisfive, what is the right form for that? 20:35:17 PERFORMSSSSSSSSSSSSSSSSSSSSSSSSSSSS 20:35:25 third person singular agreement. :P 20:35:28 ^ul (perform)SSSSSSSS 20:35:28 perform ...out of stack! 20:35:32 have is first person 20:35:39 oklopol, now that was odd 20:35:42 oh 20:35:44 right 20:35:45 or third person plural, or second person 20:35:46 not really 20:35:49 stack underflow 20:35:51 not raelly no. 20:35:55 *rully 20:36:03 raelians! 20:36:11 they believe that life on earth was made by aliens 20:36:15 who guided human evolution 20:36:25 sigh 20:36:26 and who were immortalized as the gods (and then god) of the bible 20:36:27 :o 20:38:28 ... 20:38:33 That was totally conversationally relevant. 20:38:42 not at all 20:39:02 which is why psygnisfive went back on ignore 20:39:28 (timed one) 20:39:44 i'm gonna ignore a random person here now 20:39:54 GregorR: this is #esoteric 20:39:59 theres no such thing as conversational relevance 20:40:00 hmm. 20:40:06 the gricean maxims do not apply here 20:40:08 is there a command for getting the name list 20:40:17 oklopol, /names 20:40:18 or can someone list it for me 20:40:23 that didn't work. 20:40:24 -!- psygnisfive has quit (SendQ exceeded). 20:40:27 oh 20:40:32 oklopol, /quote names? 20:40:34 listed in the server window. 20:40:35 or /raw names 20:40:44 mirc usually lists on the channel, although it's verry random 20:40:49 haha 20:41:14 oh, in mIRC script? there's definitely a way of getting the internal name list... 20:41:56 -!- psygnisfive has joined. 20:41:59 wow 20:42:05 typing /names completely trashed limechat :O 20:42:10 'MizardX' says python 20:42:26 let's just hope he doesn't say anything interesting for a while. 20:42:41 done. and i thought i'd never get to use that feature! :D 20:42:52 ey what 20:43:20 ey what what? 20:43:29 lets make love and listen to death from above 20:44:20 cute girl. 20:44:27 ey? 20:44:43 in the video 20:44:47 for that song? 20:45:00 ya 20:45:05 * psygnisfive shrugs 20:45:23 shes a girl. 20:46:44 yeah that's the turn-on! 20:46:52 that's the turn-off :( 20:46:55 i know, i'm a weirdo. 20:47:38 bye! -> 20:47:41 bye <3 20:47:51 psygnisfive, * psygnisfive has quit (SendQ exceeded) 20:47:52 hm 20:47:59 I wonder what the heck it was doing 20:48:10 probably outputting names to channel or something silly 20:48:22 and yes the timed ignore is over 20:48:47 lets try it agaiin :o 20:48:49 -!- psygnisfive has quit (SendQ exceeded). 20:48:50 no optbot? :( 20:49:06 -!- psygnisfive has joined. 20:49:07 wow 20:49:14 it outputs the names to the log window 20:49:17 and then it disconnected me 20:49:18 XD 20:49:24 lament: sorry. it was murdered. 20:49:33 because ais523 put a rule in the topic saying it can't change the topic. 20:49:36 that pretty much settles it. 20:49:46 just typing /names tho got me names form a bunch of different rooms 20:49:49 ALL of them maybe! :o 20:50:52 /names is teh shit 20:52:21 ah 20:52:24 you forgot channel 20:52:25 haha 20:52:33 my client adds the current tab 20:52:34 :D 20:52:42 my client is stupid 20:52:43 :D 20:54:43 afk shower 20:54:50 ill be thinking of you all ;D 21:00:46 you can't think about me, i'm inthinkible! 21:02:22 -!- LinuS has quit ("Puzzi. Sì, parlo proprio con te. Puzzi."). 21:06:18 Deewiant, so who was it that sent you (and all us on CC) that hate mail about fingerprint behaviour? 21:06:30 I couldn't figure out who it was 21:09:59 -!- ais523_ has joined. 21:09:59 hi ais523_ 21:10:37 AnMaster: Language::Befunge's author, there was more discussion on the mailing list there between me and him... he's still not convinced but he agreed to implement it our way 21:11:49 Deewiant, err didn't it say his interpreter failed far before 21:11:53 hello ais523_ 21:12:14 AnMaster: actually I'm not actually here, ehird just decided to say hi to me for no reason 21:12:24 ais523_, nice one 21:12:44 ais523_, so who was it that said that? 21:12:59 ais523_: you are attempting to apply logic to AnMaster 21:13:13 well I responded back with logic 21:13:18 ais523_, so who was it that said that? <--- I don't understand 21:13:24 -!- Corun has joined. 21:13:27 AnMaster: ? 21:13:32 ais523, who am I talking with if you are not there? 21:13:58 AnMaster: ask ehird, he's the one who said hi 21:14:04 eh 21:14:09 ais523, ehird is ignoring me 21:14:12 so wouldn't work 21:14:19 hmm... ask one of the bots to ask ehird then 21:14:30 ais523_, or I can ask you to ask him? 21:14:36 ah good idea 21:14:43 ais523_: the answer is "kill yourself and i'll tell you" 21:14:46 ehird: who did you say hi to when you said hi to ais523_? 21:14:47 it's a very helpful answer :D 21:14:58 ehird: not really 21:15:11 also, probably false as how could you tell I'd killed myself in the first place? 21:15:23 argh 21:15:49 i was talking to AnMaster 21:16:16 ehird: still doesn't solve the fundamental problem, just mess around with the pronouns 21:16:38 indeed 21:19:06 ais523_, anyway who are you? 21:19:33 HE IS YOUR GOD 21:19:36 WORSHIP HIM 21:19:48 GregorR, now that made 0 sense 21:20:26 while I was thinking ais523 would simply redefine here to solve the issue 21:20:41 he is after all not physically in this channel, nor is anyone else 21:21:00 AnMaster: oh, I wasn't attempting to make sense at all, I was just blatantly lying in the hope of an interesting conversation 21:21:09 hah 21:23:27 ais523, so what are you doing? 21:23:58 hmm 21:24:09 ehird -> AnMaster -> psygnisfive 21:24:15 AnMaster: nothing atm 21:24:17 what 21:24:17 let's make a cycle! 21:24:18 ais523, any updates on underlambda? 21:24:20 just idling in #esoteric 21:24:28 oklopol, err 21:24:31 what? 21:24:37 oklopol, that was a timed ignore 21:24:52 it is gone since over half an hour 21:24:56 AnMaster: no, I've been either playing Battle for Wesnoth, sleeping in a panic because I've realised it's 6pm, going to a roleplaying club, or coming here and idling 21:24:58 a cycle? 21:24:58 ok 21:25:02 :P 21:25:02 without my laptop on me 21:25:05 oklopol 21:25:07 there, cycle complete. 21:25:16 ais523_, you play wesnoth too? 21:25:17 psygnisfive: doesn't work, AnMaster doesn't have you on ignore anymore 21:25:18 strange 21:25:20 you know what you have to do. 21:25:27 oh i see 21:25:30 an ignore cycle 21:25:31 AnMaster: yes, so do lots of people 21:25:44 it's one of the most popular computer games on Linux... 21:25:56 (more worrying is that I actually recognise some of the people in the credits) 21:26:08 ais523_, well hm, I saw ESR in it 21:26:11 for one scenario 21:26:16 yes 21:26:16 * AnMaster tracks wesnoth svn 21:26:17 i think the task of converting search problems to graphs is complex yet useful enough to deserve some kinda language-level support. 21:26:20 some of the #nethack regulars are there too 21:26:33 and I'm not a serious enough player to track the svn 21:26:34 ais523_, I wonder why the heck it is that popular though 21:26:40 however, their .cfg files are Turing-complere 21:26:43 *Turing-complete 21:26:50 I might write a BF interp in them sometime 21:26:52 ais523_, I have suspected that since long 21:26:58 tc-ness I mean 21:27:05 those WML or whatever they are called 21:27:08 the language 21:27:08 another idea i'd like to try in a language is that of keeping everything local, mainly you have certain ways to get information from another part of the algorithm even though you didn't store it there 21:27:15 AnMaster: they have while loops, and you can generate infinite storage using several different methods 21:27:22 ais523_, and IMO it is a horrible hacky macro language 21:27:34 as bad as m4 21:27:35 Muriel-style, creating and unrecalling units to have a huge unit set, or generating identifiers on the fly 21:27:36 if not worse 21:27:43 AnMaster: but I like m4 21:27:47 not for actual programming in, though 21:27:53 ais523, creating a lot of units will crash wesnoth 21:27:55 I have tried 21:27:58 ah, ok 21:28:05 that's what you'd do with english, something like "loop this action here; now do some other action; now print the number of cycles you went around the loop" <<< you wouldn't actually say, when describing the loop, that we're storing the cycle count 21:28:06 patch it 21:28:13 did you create them all on the map? 21:28:24 create and unrecall/store would be the more obvious way to get lots of units 21:28:30 ais523_, more than 500 units in the recall list doesn't really show up in reality 21:28:32 if the recall roster can fill up and crash it, that's a serious bug 21:28:42 also even with around 100-200 units recall is awfully slow 21:28:47 AnMaster: it could do in a long enough campaign, I suppose 21:28:55 and yes, the UI would be a pain if you had 200 recallables 21:28:57 ais523_, usually it won't happen 21:29:25 because the long epic campaigns usually manage to kill off a lot of units 21:29:38 well, yes 21:29:48 but it would be trivial if you were /trying/ for it 21:29:51 say, over the burning sands, Northern rebirth, whateverh that new one is 21:29:52 err 21:29:55 forgot the name 21:30:05 I don't know which ones are new and which ones aren't 21:30:07 * AnMaster loves those huge campaigns 21:30:11 but decent into darkness is infinitely long 21:30:18 ais523, well yeah also I hate it 21:30:23 don't like tragedies 21:30:29 I prefer happy endings 21:30:32 so do I, I decided I didn't like it when I saw it was infinitely long with a stupid ending 21:30:46 Legend of Wesmere 21:30:48 * ais523_ prefers codereading computer games to playing them... 21:30:48 that is the new one 21:30:54 AnMaster: ah, I don't have that one 21:31:02 ais523_, mainlined in svn iirc 21:31:07 not last release 21:31:23 well, I'll wait for the distro to package the release, I think 21:31:29 -!- Corun has quit ("This computer has gone to sleep"). 21:31:34 IIRC Ubuntu just copies Debian for Wesnoth 21:31:34 there is the not yet mainlined Invasion from the Unknown 21:31:37 which is so huge 21:31:42 it is split into two campaigns 21:31:54 AnMaster: can you recall from one to the other? 21:31:56 and it takes over a minute for wesnoth to "build cache" 21:32:07 ais523, at the end of the first you can choose to continue 21:32:16 but you hardly have any units left at that point anyway 21:32:19 -!- oerjan has joined. 21:32:27 and almost no gold 21:32:32 AnMaster: you could have just left some good ones in recall anyway just for the hell of it 21:32:52 the carry-over gold doesn't matter nearly as much as campaigns have to allow for that possibility anyway 21:33:02 (the minimum gold should always be enough to complete the level IMO) 21:33:02 some kinda strategy thing? 21:33:06 ais523, well you are surrounded by something like 5 high level undead armies in the last two battles 21:33:08 brb 21:33:09 oklopol: yes 21:33:21 any details worth knowing? 21:33:56 oklopol: it has a TC programming language to describe levels, although AnMaster doesn't like it it seems 21:34:29 back 21:34:36 ais523: that is true of many dsl's, is it tc in an interesting way? 21:34:39 well I think it is ugly 21:34:47 if it has a "while", i'm scceptical 21:34:50 *sceptical 21:34:53 (hmm... their programming lang looks vaguely XML-based, it has [if] condition=true [then] {DO_STUFF} [/then] [else] {DO_OTHER_STUFF} [/else] [/if] as the main conditional construct 21:34:59 oklopol: no, the control flow is very typical 21:35:03 ais523, xml? oh yes xml with [] 21:35:06 inlined procedures + do + while 21:35:08 hm I guess 21:35:17 ais523, aren't those {} macros? 21:35:20 afaik 21:35:25 the only interesting thing about it from a TCness point of view is you have to do tricks to get infinite storage 21:35:26 somewhat like mediawiki {{}} 21:35:28 really 21:35:30 and {} is a macro reference, yes 21:35:36 ais523: well that's a shame, i'd've wished you'd need to make some kinda computer ai for tcness... 21:35:49 the AIs are in Python, IIRC 21:35:50 if you don't need the actual game for it, that's already a big minues 21:35:52 oh. 21:35:55 ais523, err 21:36:00 C++ iirc? 21:36:03 anything worth knowing about the actual game? 21:36:06 AnMaster: no, the game itself is in C++ 21:36:14 oklopol, fantasy themed RTS 21:36:17 turn based 21:36:19 quite good 21:36:23 oklopol: turn-based, hexagon-based, rules are quite simple, unusual in the amount of difference between the sides 21:36:30 AnMaster: it is so not an RTS, it isn't real-time it's turn-based 21:36:32 ./ai/python/safe.py 21:36:34 hm ok 21:36:36 you are right 21:36:39 -!- warrie_ has quit (Read error: 110 (Connection timed out)). 21:36:57 it has a huge amount of variety for a strategy game 21:37:03 ais523, ah right 21:37:03 most of them just have 2 or 3 types of objective 21:37:07 an TBS? 21:37:09 hmm. i almost want to try that 21:37:32 could you make a level that was actually just a chessboard with chess rules? 21:37:37 ais523, and the variety is partly due to the tc-ness 21:37:39 I think 21:37:42 oklopol: yes, I think so 21:37:49 okay. 21:37:54 * oklopol adds to todo list 21:38:05 although it would be quite difficult, you'd have to say "that's an illegal move" a lot and kill people if they didn't undo it 21:38:05 ais523, LoW even contains a scenario where you give the allied AI orders 21:38:10 very unusual 21:38:16 or else change the terrain so only legal moves were possible 21:38:17 ais523: yeah, that was my idea 21:38:18 but doable 21:38:20 hmm 21:38:23 that's actually niver 21:38:24 *nicer 21:38:39 ais523_, and you can say "this is more important than that" 21:38:52 it would be great if you saw the mechanics of how the rules worked when playing the game 21:39:02 would have an eso feel to it. 21:39:13 oklopol: the mechanics of the game itself are all in the open 21:39:19 as in, if you try to attack someone, say 21:39:30 it gives you all the stats and calculations, and how it came to them 21:39:30 yes that dialog 21:39:35 oh 21:39:37 i see 21:39:48 ais523, often it is just depressing however ;P 21:39:53 does that game work on windows? 21:40:02 think so 21:40:07 but why would anyone want windows? 21:40:11 e.g. you know that each of your attacks has a 50% chance of hitting and does 4 damage and heals you 2, except your opponents attacks have a 70% chance of hitting, alternate with yours and will halve your damage potential when they hit, but it only has 2 of them and you have 6 21:40:20 AnMaster: it does I think 21:40:20 AnMaster: it comes with the laptop 21:40:36 ais523, don't forget "first strike" "defense", type of ground and so on 21:40:37 :P 21:40:44 and the various modifiers 21:40:54 and attributes 21:41:06 like quickness, resilience 21:41:09 and so on 21:41:20 ais523_, sometimes the maths for it is a nightmare 21:41:27 hmm... wonder if there's some other more robust game so you could add unit building and other stuff to it. 21:41:32 oklopol, you could install another OS too? 21:41:43 oklopol, units do advance 21:41:44 by robust i mean, if you change chess even a bit, it's an entirely different game 21:41:45 to new levels 21:41:51 oklopol: Go? 21:42:00 AnMaster: i would install another os if that was easy. 21:42:01 it's not. 21:42:03 hmm... the thing's played on a hex grid though 21:42:08 oklopol, no cd? 21:42:20 AnMaster: Ubuntu's pretty easy on Windows nowadays, you can download an installer and install it as a Windows program 21:42:23 ais523, wesnoth is on a hex gird isn't it? 21:42:24 I've done that before 21:42:26 AnMaster: yes 21:42:26 i'm assuming it'd take more than clicking a button and waiting for 5 minutes 21:42:35 that's pretty much my limit 21:42:41 oklopol: it is clicking a button, asking a few questions and waiting for about 10 minutes 21:42:44 *answering 21:42:47 ais523: you can get around the hex grid. 21:42:52 oklopol: yes 21:42:53 hm 21:42:54 true 21:43:03 ais523: well okay, that sounds easy 21:43:05 yes the hex gird would mess up chess 21:43:08 wouldn't work 21:43:12 Weschess would be an interesting project, come to think of it 21:43:19 it's just i have a feeling it just wouldn't work, and i'd have to do more than just that. 21:43:19 AnMaster: yes it would, use alternate columns and alternate rows 21:43:25 and you have a standard chessboard 21:43:25 i can't even get mingw to install 21:43:33 ais523, hm maybe you could approximate it using 4x4 or so? 21:43:36 for some reason when you were helping me do that, i only installed C 21:43:39 and now i need c++ 21:43:43 and just override the move rules 21:43:46 in some odd way 21:43:52 oklopol: yes, sometimes it doesn't work first time, and you need to ask for help, but I've had much less trouble installing Ubuntu than I have for Cygwin 21:44:08 same 21:44:18 AnMaster: well, you can change terrain at runtime, and unit's terrain movements at runtime 21:44:18 i'd prefer if someone just installed it for me. 21:44:30 i'm getting so goddamn tired at installing things. 21:44:37 well I think LFS is simpler than cygwin 21:44:41 I guess most wouldn't agree 21:44:49 AnMaster: what's LFS? 21:44:56 Linux From Scratch 21:44:58 i never learn to do it myself if the one-click method doesn't work, and it never does. 21:44:59 ah 21:45:06 gives you a manual 21:45:10 and lets you do the job 21:45:12 yes, I know of it 21:45:14 (i never learn, based on having done it a thousand times) 21:45:15 you learn a lot 21:45:17 just I didn't get the acronym straight off 21:45:19 i just don't 21:45:25 i'm an idiot at stuff like that 21:45:29 I have auto-built a linux distro from source, but it was Makefile-automated 21:45:37 the makefile was full of wgets to places like kernel.org 21:45:42 oklopol, you do learn 21:45:45 worked fine the first time but the dependencies were all screwed up 21:45:52 oklopol, you learned one thing 21:46:02 oklopol, you learned that you never learn anything else 21:46:03 ;P 21:46:12 AnMaster: yes, but slowly, and i get frustrated and occasionally hit my screen. 21:46:16 that's not exactly healthy 21:46:20 indeed 21:46:25 i'm very violent when it comes to machines. 21:46:27 and it is bad for the monitor 21:46:36 also it wouldn't help 21:46:39 -!- ais523_ has quit ("http://www.mibbit.com ajax IRC Client"). 21:46:41 it is the computer case 21:46:43 that is relevant 21:46:44 humans never piss me off, though, for some reason 21:46:45 hm 21:46:57 oklopol, what about an AI? would it piss you off? 21:47:06 -!- ais523_ has joined. 21:47:07 hi ais523_ 21:47:11 hello ais523 21:47:12 hello ais523_ 21:47:13 * 21:47:17 why does Mibbit log me off at random every now and then for no good reason 21:47:59 how should I know? 21:48:14 AnMaster: i don't know, haven't met one 21:48:21 ais523, what were the other ways for storage in wesnoth you said? 21:48:36 Muriel-style might work, but I'm not sure 21:48:42 could you do evals at runtime? 21:48:45 ais523_, hm not familiar with that one? 21:48:50 Muriel that is 21:48:57 oklopol: Does fungot piss you off when it doesn't work? 21:48:58 fizzie: some ancient submarine volcanic origin, a high hill :) feldspathic clay-slate projected, retaining its usual defined outline, dead, and one which was just perceptibly fnord and this yields 0.25 per cent. 21:49:03 ais523, I don't know WML syntax very well 21:49:13 AnMaster: basically, the only way to do flow control is to construct a new program, then replace the original program with it and rerun from the start 21:49:19 sort of like Underload where ^ doesn't return 21:49:39 also, I think you can store any number of units in one variable 21:49:40 ais523_, I suspect it woudln't work to do that 21:49:54 and it becomes an array automatically 21:50:00 there are no obvious limits on array length... 21:50:14 ais523, since wesnoth fully expands the file when building the cache 21:50:21 so I doubt eval is allowed 21:50:23 AnMaster: ah, ok 21:50:30 even imports? 21:50:31 however I could be wrong 21:50:31 I suppose so... 21:50:36 ais523, hm not sure 21:50:48 ais523, but it expand all macros 21:51:20 fizzie: when has it not worked? 21:51:22 also I have that idea for a syntax less language, somewhat like lisp, but even less syntax 21:51:33 not sure if you heard about it ais523 ? 21:51:42 oklopol: Well, I hear it doesn't exactly always make sense. 21:51:48 AnMaster: no, I didn't, but Underload is pretty short on syntax... 21:51:51 fizzie: i never have an actual need to get fungot working, so i don't think it would piss me off. 21:51:51 oklopol: goura coronata and victoriae, hybrids :). 21:51:55 as is Befunge fwiw 21:52:09 fizzie: i haven't experienced that 21:52:09 and Malbolge is even shorter 21:52:11 ais523, my language got truly no syntax 21:52:18 ais523, everything is done using the file system 21:52:24 main/command 21:52:26 or whatever 21:52:42 so you just modify the directories and filenames on your file system 21:52:43 the extent of bot malfunctioning i can think of is that thutubot doesn't always gracefully terminate on erroneous input. 21:52:45 that's much more syntaxful than Malbolge, I think 21:52:48 symlinks for calling other functions 21:52:52 you have an entire directory tree worth of syntax there 21:52:55 but tries to evaluate with undefined output 21:53:03 ais523, hah but that isn't a syntax really 21:53:05 oklopol: it wasn't designed to handle erroneous input 21:53:13 and as you noticed, it uses {{ }} for some internal stuff 21:53:19 +ul ({{)S 21:53:19 {{ 21:53:23 +ul ({)S 21:53:24 { 21:53:30 +ul (})S 21:53:30 } 21:53:35 ais523: yes, that's why it's not really malfunctioning, just the closest thing to bot malfunctioning i can recall. 21:53:38 +ul (})(S 21:53:39 ...out of time! 21:53:39 }} 21:53:46 yes 21:53:47 -> }=} 21:53:52 +ul }} 21:53:57 +ul (}})S 21:53:57 }} 21:54:00 +ul ((}}))S 21:54:01 (}=)} 21:54:06 right, that was it 21:54:08 ah, there we go 21:54:14 quite hard to set it off in the first place, it seems... 21:54:26 also, no idea how the = ended up in output, maybe there's a bug in my Thutu interp 21:54:32 or somehow it got escaped 21:55:17 +ul ((}}))S 21:55:17 (}=)} 21:55:19 hn 21:55:21 hm* 21:55:29 god there's a lot of ideas in my brain 21:55:40 annoying, why can't they come at a steady pace :| 21:56:05 but no, they have to come in bursts. and usually when i don't have time to program or spec anything 21:56:28 (mainly because i get ideas from doing stuff, and if i'm doing stuff, i usually need to finish it) 21:56:33 oklopol, just write down the ideas 21:56:33 (i guess that's pretty common) 21:57:31 ais523, does it reset it's internal state between the commands or? 21:57:51 or is it corrupted now? 21:57:58 AnMaster: yes, the internal state is wiped out by a very match-everything regex every iteration 21:58:01 it should still work fine 21:58:28 driving it into an infiniloop is the only real way to disrupt it, but I suspect even that's impossible unless I've screwed up the out of time code somehow 21:58:29 AnMaster: well i do, it's just i need to be excited about the idea as well, and i tend to forget parts of it, because not all is explicit in my head and thus writable down, and lose some of the interest. 21:58:57 i have a lot of ideas in lists that i have no idea what the point is anymore :P 21:59:14 oklopol, maybe there was no point? 21:59:44 my ideas always have a point. if nothing else, the point is there's no point in the idea. 21:59:54 preventing me to make the same mistake again 22:01:21 like that GC idea i had. i've developed a whole new intuition of inherently "amortizedly trivial, hard to do online" problems from that, after i couldn't get it to work 22:01:49 oklopol, eh? 22:01:56 oklopol, details? 22:02:37 i thought i'd invented an online GC algorithm that worked fundamentally differently than doing it amortizedly, but at the same time as continuing running the program. 22:02:59 oklopol: I liked your GC idea too, but I thought "wow that will be difficult, if even possible, to implement" 22:03:10 -!- warrie has joined. 22:03:40 I don't remember it 22:03:43 what i got out of it was a pretty strong impression that you cannot do it. of course algorithmically it's the same whether you do it online, or whether you run the GC in the background 22:04:07 er, "online"? 22:04:10 oklopol, can you tell me how it worked? 22:04:39 oerjan: keep an invariant true as you read input, and not make it true in a monolithic action. 22:05:08 oklopol, when was this so I can read logs? 22:05:20 AnMaster: i have no idea. i can't estimate time 22:05:22 at all 22:05:28 sigh 22:05:32 I grepped logs 22:05:34 found nothing 22:05:40 after summer, before this weekend 22:05:55 nothing 22:05:56 honestly cannot pinpoint better. 22:05:59 must have been offline 22:06:05 yeah it's impossible to find things in the logs. 22:06:16 oklopol, so tell me the details about it? 22:06:30 because so far I got no clue how it would have worked 22:06:32 AnMaster: tell you the details of an algorithm that's fundamentally broken? 22:06:37 oklopol, yes 22:06:48 tbh i don't remember anymore. 22:07:09 it was some kind of extension to refcounting. 22:07:17 that tried to detect cycles 22:08:02 but after taking into account every kind of graph that can emerge, it ended up running some sort of mark-unused-things GC everytime you did something to a reference. 22:08:44 I remember being here when that was talked about, and having some serious doubts about the workability of it all. 22:08:50 FreeNode-#esoteric.log.bz2:sep 03 22:45:23 Last week, Chris Lattner — who manages the Clang, LLVM, and GCC groups at Apple — announced that work was well underway to bring ‘blocks’ to the GCC and Clang compilers. ‘So what?’, I hear you ask, ‘My kid has been using blocks since he was 9 months old.’ Fair point, but maybe not these blocks. 22:08:52 that is all 22:09:09 yeah i remember lolling at that 22:09:11 pretty badly 22:09:29 fizzie: yeah, that was a pretty bad idea. 22:09:29 FreeNode-#esoteric.log.bz2:sep 25 23:25:56 aaaanyway, this is kinda like connected components, you can do it for the whole graph in O(n) (the GC approach), or find the connected component of every node separately in O(n^2) (the refcounting thing); it's just the latter will actually be closer to O(n) if there are only a few references that cannot be "optimized", and you more easily can do 22:09:29 it online 22:09:31 hm? 22:09:54 AnMaster: Try 2008-09-26; "rootrefcount" is a good word to grep for. 22:10:00 Oh, you found it. 22:10:12 yeah i had some idea for optimizing the amount of things you need to check. 22:10:22 that was fundamentally wrong, because you could always get around it. 22:11:26 http://www.vjn.fi/pb/p466532445.txt ? 22:11:37 that was some example 22:12:04 so, basically, i tell you about one of my rare sucky ideas, and you dig it up and throw it in my face? :P 22:12:38 i've had tons of great ideas, much more of those, have i mentioned GRAPHICA THIS WEEK YET, GRAPHICA IS PRETTY COOL 22:12:45 bah I fail at IRC and listening to radio at the same time 22:13:10 oklopol, maybe 22:13:17 i'm doing nothing but ircing 22:13:18 AnMaster: whatever you do, don't start chewing gum 22:13:26 oerjan, eh why would I? 22:13:27 No, you haven't mentioned graphica this week. 22:13:29 or was it a bad pun? 22:13:33 oh. 22:13:36 perhaps i should then 22:13:51 AnMaster: not being able to do things while chewing gum is sort of a meme 22:13:53 ais523: recall you mentioning more languages should have prolog-like returns 22:13:59 yes 22:14:10 if you remember the example where i parsed brainfuck using graphica 2.0, that used those 22:14:12 basically 22:14:13 it's so nice for functions to work backwards as well as forwards as son as you right them 22:14:20 *write them 22:14:25 you can give a variable as an argument with syntax ?x 22:14:44 (incidentally saw that used later in another language, although just as a comment to notate which are usually returns) 22:14:45 oerjan, oh ok 22:14:54 and you can bind them in the nodes 22:15:17 ais523, s/son/soon/ 22:15:48 graphica is fun to use for parsing, because you're very explicitly building a tree. that's really all you can do in the language :P 22:16:05 (2.0 doesn't let you rewrite the graph yet, just adds more computational power for building it.) 22:16:16 mainly the variables. 22:16:46 and two shorthand syntaxes, one based on eodermdrome, one based on predicate logic. 22:17:17 C# has a funny syntax where "nullable" -- a bit like the Haskell "Maybe" type -- primitive types are declared like "int? x;". It's a (arguably) reasonable syntax, but for a C programmer it just looks a bit... weird. 22:17:43 Like the programmer was saying "well, maybe an integer, I guess... not quite sure. what do you think?" 22:17:49 that sounds like a nice practical addition 22:18:09 i like the syntax, and yeah, that's what it looks like :P 22:19:20 Then you need to check for x.HasValue and use x.Value when you want to actually use it. I think. There might have been some form of auto-boxing/unboxing thing. I'm not a C# person. 22:20:05 fizzie, it is meant for talking with SQL iirc 22:20:59 Well, it's not exclusively for that, but I'm sure they've been thinking databases when inventing that one. 22:24:45 * oerjan realizes the word "eodermdrome" itself is a eulerian cycle on K5 22:24:56 heh 22:25:28 oerjan: yes, it is 22:25:30 not me who coined it 22:25:37 oh 22:25:40 in fact I came up with the whole idea for the lang after seeing the word 22:25:52 eodermdrome? 22:25:55 wtf does that mean 22:26:01 it means "a word which is a eulerian cycle on a nonplanar graph" 22:26:13 heh 22:26:17 and was coined to produce an 11-character example of that, as no real words did that 22:27:38 ah 22:28:04 (there are real words which are eodermdromes, but all of them are longer than 11 chars) 22:29:35 ais523, how do you define a graph in that word? 22:29:43 * ais523_ is disappointed that anagolf doesn't have INTERCAL as a language 22:30:11 AnMaster: it has vertices e, o, d, r, m, and arcs from e to o, o to d, d to e, e to r, and so on 22:30:15 ais523, mail the owner? 22:30:30 I might do some time when he's online 22:30:34 he uses Freenode 22:31:03 -!- Sgeo has joined. 22:31:08 online atm, but hasn't nickserv-identified for 11 weeks 22:31:47 ais523, heh 22:31:49 that's long enough that someone else could take the nick 22:32:02 oerjan, not if a client is connected 22:32:12 since you need to take the nick first 22:32:14 with /nick 22:32:22 well they could if he disconnects 22:32:22 you'd have to somehow kill the old nick to be able to take the nick to identify 22:35:24 My nickname is the K1,3 (also called The Claw, although maybe without the definite article and the capitalization) if you ignore the zz-loop. 22:35:36 not bad 22:35:39 mine's just a 5-chain 22:35:47 or 6-chain with the underscore 22:38:02 going for a while, will be back soon 22:38:03 -!- ais523_ has quit ("http://www.mibbit.com ajax IRC Client"). 22:39:02 heh 22:39:17 is it case sensitive? 22:40:16 fizzie, ^ 22:43:35 Well, according to http://esolangs.org/wiki/Eodermdrome it's actually lowercase-only; but feel free to use any definition. 22:43:51 hm 22:44:05 how would my nick look? 22:44:28 Lowercased, it's a triangle with a 4-path hanging from one node. 22:44:35 heh 22:44:57 "anma" is a triangle, and then the "aster" part is just the path. 22:45:09 fizzie, well this is the correct caseing 22:45:23 So ----<| pretty much. 22:45:38 heh 22:46:12 While I'm a \|/ if you don't allow for loops. 22:46:13 -!- ais523_ has joined. 22:46:13 hi ais523_ 22:47:04 what does hello world look like in that Eodermdrome? As far as I understood it got output? 22:47:05 who here is it that has trouble coming up with names for esolang projects? 22:47:20 AnMaster: yes, it does, but any nontrivial Eodermdrome program is very hard to write 22:47:28 ais523_, hello world? 22:47:51 AnMaster: I can't remember if you can output multiple characters at once 22:47:57 if you can't it's difficult 22:48:02 According to the page you can. 22:48:05 due to needing things to happen in the right order 22:48:06 ah, ok 22:48:11 Since it's an "output string", not a character. 22:48:21 couldn't remember if I put it online 22:49:04 thequickbrownfoxjumpsoverthelazydog (Hello, world!) a 22:49:07 would work, I think 22:49:29 anyway, I came up with a good name for nothing in particular 22:49:39 and wondered if anyone wanted it for a project 22:49:46 :P 22:50:25 oklopol: "ceci n'est pas un acronyme récursif" 22:50:40 cnepuar? 22:50:48 this isn't a recursive acronym? 22:51:03 tiara? 22:59:41 Huh, there's no Eodermdrome implementation? 23:00:07 hahaha 23:00:09 oh magritte 23:00:12 youre so funny :) 23:05:45 fizzie: oklopol wrote one I think 23:06:05 oklopol: and yes, "this is not a recursive acronym" 23:06:43 ais523, hm but it isn't is it? 23:06:59 cn[e]puar? 23:10:06 well, it isn't 23:10:14 it refers to itself using "ceci", that's not acronymising 23:10:17 so it's completely correct 23:10:33 ais523, it should be :/ 23:10:40 it would be so much more ironic then 23:11:46 yes, but I don't know enough French to get that to work 23:11:54 Slereah: is there any way to get it to work? 23:12:53 "cnepuar n'est pas un acronyme récursif", i presume 23:12:56 "cnepuar n'est pas un acronyme récursif" works, no? 23:13:56 fizzie: well i've implemented eodermdrome 23:14:06 at least to the point where i could finish it trivially 23:14:13 i have all the logic there 23:14:14 not sure if "est" is considered an independent word... 23:14:25 oh. 23:14:30 also i'm blind 23:14:45 oerjan: well, it's close enough I suspect 23:14:59 I wanted to get ceci n'est pas in there somehow, but I suppose it still works... 23:15:12 ais523_: you are aware that subgraph isomorphism is NP-complete? 23:15:27 * oerjan just found out 23:15:45 oerjan: that just means it's inefficient, not noncomputable 23:15:47 so implementation is not going to be entirely efficient in theory 23:15:48 and I didn't know that 23:15:51 ais523, what does ceci? mean? 23:15:53 but it would explain how hard it was to write... 23:15:55 AnMaster: "this" 23:15:56 s/i?/i/ 23:15:58 ais523, ah 23:16:06 it was pretty simple to write imo 23:16:43 i didn't aim at anything more than a few simple local optimizations though 23:18:10 just that i start with nodes that have a lot of connections; anyway, the point is np-complete matching isn't an issue in practise, because most matchings will probably be quite simple 23:18:58 There's a lot of research on isomorphism testing, and I think also reasonably efficient (for small-ish graphs, anyway) algorithms too. 23:19:52 coooool 23:20:38 maybe it's not a problem since the subgraph cannot have more than 26 nodes... 23:22:27 how do you prove a program "formally correct" 23:22:33 I don't understand how that works 23:22:37 consider BitC for example 23:22:50 http://www.bitc-lang.org/ 23:26:13 Either much in the same way you'd prove that an algorithm works (given the semantics of the language, do the necessary deductions and see that it actually does what you want), or alternatively with some exhaustive model checking. 23:26:25 hm 23:26:32 could you do it for normal C? 23:26:35 Although I guess "with difficulty" would've been an appropriate answer too. 23:26:41 I suspect it would be extremely hard 23:26:52 considering pointers 23:26:55 for example 23:27:11 Not really, no. Well, not with any automation support, anyway, and not without formally defining what C does. 23:27:51 um? 23:30:12 I mean, there are no tools (that I know of) that would help you automate the formal verification process for C programs. And the C standard isn't really written as a formal definition that you could do mathematical reasoning with. 23:32:08 you can prove your algorithm to be correct, and then make sure the program carries out the steps of the algorithm 23:32:35 proving something directly from the c program sounds like a harder job 23:33:12 heh 23:33:19 what about pointer aliasing? 23:33:53 pointers make it harder to prove things because you lose locality. 23:34:12 Then you just need to prove things about where the pointers can possibly point to. 23:34:17 yeah 23:34:27 which sounds like fun, btw 23:34:32 i like odd things 23:34:38 Model checking is a fancy thing too, except a bit limited; that one is based on exhaustively (but in a reasonably smart way) searching all possible states the program can be in, and making sure the invariants and such you care about hold everywhere. 23:34:56 fizzie, interesting 23:35:01 that sounds more useful 23:35:08 I've done a bit of that with PROMELA -- http://en.wikipedia.org/wiki/Promela -- for one course. 23:35:43 -!- jix has quit ("..."). 23:36:03 hope we have something like that in turku 23:36:11 turku? 23:36:17 AnMaster: A city in Finland. 23:36:30 it seems the menu is a bit limited though :| 23:36:31 A city with a university, more to the point. 23:36:37 heh 23:37:01 "the city where i go to university" is what you need for context 23:37:14 but that was probably obvious 23:37:28 hm 23:38:21 -!- ais523_ has quit ("http://www.mibbit.com ajax IRC Client"). 23:38:40 fizzie, is it a programming language or a tool for working on programs written other languages? 23:39:09 Promela is a language, Spin is the model checking tool for Promela programs. 23:39:20 ah 23:39:55 Although typically Promela is just used to prove that the protocol or whatever you are interested in is deadlock-free and stuff; I don't think people actually use that for implementing things. 23:40:08 (Which makes implementation bugs again a possibility.) 23:40:26 is it open source? 23:40:51 I think so, yes. Not quite sure, though. 23:41:18 ah yes 23:41:26 however not useful for the languages I program in 23:41:30 I would want it for C 23:41:31 :/ 23:42:51 It would probably be very difficult to do exhaustive model checking for C either; Promela is designed so that you hopefully don't end up with a horrible amount of possible program states. 23:43:48 Promela/Spin quite easily chokes on "too large state-space to exhaustively search" issues anyway. 23:44:49 For a C program with a unsigned char foo[1024]; in it, there's immediately 2^(1024*CHAR_BIT) potential program states, and it's not very trivial to automatically deduce which states are actually different. 23:45:33 source code annotation? 23:45:51 such as "all these values mean nop"? 23:46:31 That's probably closer to doing logical deductions manually about the program; of course it could help some tool a bit. 23:46:53 fizzie: well you could do that lazily, only branch in the proof where those cells are actually used for something 23:47:03 doesn't help with the underlying problem ofc 23:47:09 proving is impossible 23:47:51 Promela has some "let the programmer help the tool a bit" features too; things like marking some instructions to be performed atomically, because otherwise it needs to think about all the possible ways those instructions can be scheduled with different threads. 23:47:51 true 23:50:46 Our Promela/Spin stuff was for the "Parallel and Distributed Systems" course here; but they tend to change the course contents every year. 23:52:31 oklopol: If you're talking about UTU, quick googling says there's DTEK8028 "System Verification and Design for Testability" -- but they might not talk too much about formal verification. You people seem to have quite little information about your courses in the interwebs; or I might not be looking at the right places. 23:52:46 yes, very little 23:52:58 you can find a bit more in the study guide though, or what's it called again 23:53:15 That's still something like a single paragraph. 23:53:20 yes. 23:53:36 you have a decent amount of information about courses? 23:53:43 that would be awesome 23:53:58 Well, around here (HUT) there's a lot of material but it's all distributed quite randomly around the web pages of the departments; each of which is arranged differently. 23:54:10 heh 23:54:33 Although for next year there is an official ultimatum to move *all* course information into http://noppa.tkk.fi/ -- or at least be linked from there. 23:54:41 i once tried making a parser for the study guide in order to make a DAG of what courses need what as prerequisites 23:55:09 but, in the end i would basically have had to make the graph manually 23:55:23 I think currently that Noppa tool has (for not-currently-active courses) just copies of the data from the study guide book. 23:55:25 because every course says the prerequisites differentle. 23:55:29 *differently 23:55:53 From what I've seen, the prerequisites have been more like recommendations. 23:56:17 you mean, from what you've seen at the university, or from utu's study guide? 23:56:19 :P 23:56:24 okay that was a stupid question. 23:57:17 but yeah, that's my experience too 23:57:17 Have to admit that at least the Noppa thing gives a reasonably browseable/searchable list of all our courses. 23:57:41 good for him/her, but that should really be the website's job. 23:58:32 fizzie, seems to be a mix of English Finnish and Swedish!? 23:59:12 AnMaster: Yeah, theoretically speaking the name of each course should be in Finnish, Swedish and English, but I don't think everyone has bothered to fill in all the fields. 23:59:20 ah 23:59:29 anyway, my goal is to get a degree in math after my cs degree, and that doesn't seem to be possible in utu, so i may have to come to helsinki :o 23:59:39 Oh, and I think the system is designed so that the actual content of the pages can be provided with a single language only. 23:59:40 fizzie, I saw all three languages at https://noppa.tkk.fi/noppa/kurssit/il/t3050 23:59:50 if that happens, which would be cool, i can make a comparison 23:59:54 fizzie, eh?