00:20:01 -!- ehird has quit ("Page closed"). 00:38:21 -!- impomatic has quit (Read error: 110 (Connection timed out)). 00:38:30 -!- ehird has joined. 00:38:43 test 00:39:41 -!- ehird has quit (Client Quit). 00:39:55 -!- ehird has joined. 00:40:08 okay, ChatZilla appears to be vaguely passable on windows 00:40:14 which is... unexpected, I guess 00:40:20 considering how much it sucks on OS X 00:41:05 * coppro has not tried it on OS X and so cannot comment 00:41:14 it's awful 00:42:15 I will take your word for it and subsequently ignore it and some later and more relevant time 00:42:27 s/it and/it at/ 00:42:49 coppro: that's a very confusing sentence 00:47:23 -!- ehird has quit (Remote closed the connection). 00:47:31 -!- Sgeo has joined. 00:47:36 -!- ehird has joined. 00:57:26 http://explainxkcd.com/ 01:00:50 useful 01:01:08 "This was wrong. David Gest has never had a sexually transmitted infection and did not have Ms Minnelli's dog killed." --The Daily Mail 01:01:20 in a retraction letter 01:01:51 ha 01:04:22 an easy mistake to make 01:04:48 they keys are like right next to each other 01:05:09 (you know you're awesome when you are referencing bash :/) 01:05:26 "awesome" 01:20:40 http://portal.acm.org/citation.cfm?doid=1289971.1289983 <-- "Context-Aware Scanning for Parsing Extensible Languages" 01:31:54 so... a C++ parser? 01:41:11 -!- Azstal has quit ("."). 01:48:02 -!- puzzlet has joined. 01:51:10 -!- Amiral_ has joined. 01:51:15 hi Amiral_ 01:51:17 who're you 01:51:25 hi ;) 01:51:33 i was coming from http://frox25.no-ip.org/~mtve/wiki/ 01:51:35 :] 01:52:21 cool 01:52:23 hi 01:52:35 hehe, i m searching the guy ;) 01:52:48 need some help from his script 01:52:57 he's not there ? 01:53:16 Sure he is. 01:53:18 mtve: 01:53:27 [01:53]===mtve: away with message “Disconnected now, messages are logged” 01:53:30 cool :] 01:53:33 yes i saw that 01:53:36 What script, anyway? 01:53:42 sigtran 01:53:58 I wonder what the heck it does. 01:53:58 do you know ss7 sctp protocol ? 01:54:03 No. 01:55:57 ok, when do you think he can come ? 02:26:29 -!- jix_ has joined. 02:38:12 -!- MizardX- has joined. 02:38:19 -!- MizardX has quit (Read error: 104 (Connection reset by peer)). 02:38:42 -!- MizardX- has changed nick to MizardX. 02:39:01 -!- jix has quit (Read error: 113 (No route to host)). 02:47:59 -!- Amiral_ has quit (Remote closed the connection). 03:52:22 -!- Gracenotes has joined. 04:38:20 * ehird wonders if he could hack capability-based security into linux with users... 04:40:48 -!- puzzlet_ has joined. 04:50:15 -!- coppro has quit ("The only thing I know is that I know nothing"). 04:53:36 -!- puzzlet has quit (Connection timed out). 04:54:55 -!- coppro has joined. 04:57:10 wb coppro 05:46:09 -!- oerjan has joined. 05:46:26 hi oerjan 05:46:36 hi ehird 05:55:08 -!- oerjan has quit ("Reboot"). 05:58:46 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 05:59:11 -!- oerjan has joined. 06:05:06 damn it, I accidentally did "rm blah.avi.00* > blah.avi", instead of cat. now I have to download it again >:[ 06:11:00 ha 06:17:48 -!- Patashu[CyberOrc has joined. 06:18:08 -!- Patashu[CyberOrc has quit (Client Quit). 06:40:47 -!- MigoMipo has joined. 06:45:27 -!- oerjan has quit ("leaving"). 07:05:04 -!- kar8nga has joined. 07:30:58 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server"). 07:33:25 -!- Pthing has joined. 07:47:06 -!- FireFly has joined. 07:51:43 -!- ehird has quit (Read error: 113 (No route to host)). 07:58:58 -!- kar8nga has quit (Remote closed the connection). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:07:50 -!- FireFly has quit ("Later"). 10:03:06 Which concepts are related to self-organizing maps? Yahoo knows: http://correlator.sandbox.yahoo.net/index.php/concepts/self-organizing+map 10:03:36 Mostly it makes sense, but I find the fact that ""The Dark Horse Book of Witchcraft" is probably related to self-organizing map" a bit dubious. 10:06:08 -!- KingOfKarlsruhe has joined. 10:27:17 -!- Slereah has quit. 11:18:37 -!- KingOfKarlsruhe has quit (Remote closed the connection). 11:39:17 -!- jix_ has quit (hubbard.freenode.net irc.freenode.net). 11:39:17 -!- dbc has quit (hubbard.freenode.net irc.freenode.net). 11:39:18 -!- pikhq has quit (hubbard.freenode.net irc.freenode.net). 11:39:18 -!- olsner has quit (hubbard.freenode.net irc.freenode.net). 11:40:49 -!- pikhq has joined. 11:50:36 -!- jix_ has joined. 11:50:36 -!- dbc has joined. 11:50:36 -!- olsner has joined. 12:00:44 -!- Nowicjusz has joined. 12:00:49 -!- Nowicjusz has left (?). 12:35:25 -!- CESSMASTER has quit (Read error: 110 (Connection timed out)). 12:46:37 -!- ais523 has joined. 12:49:51 hi everyone 12:52:49 Hi, you nobody. 12:53:03 fair enough 12:53:49 I just wanted to be contrasty. 12:55:14 hmm... today presented a data point in favour of the theory that knowing Java is nearly always sufficient to get a job 12:57:30 My personal experience has shown that knowing Perl is the most useful thing; two out of three of my employers have been mostly in search of someone who speaks Perl. 12:57:42 well, I know Perl too 12:57:59 in this case, though, I've been employed to teach Java to first-year undergraduate students 12:58:02 alongside my PhD 12:58:14 * ais523 wonders whether Java or Perl as a first language would be meaner 12:58:18 probably Java, tbh 12:58:30 one of my friends had Perl and Python as her first two languages, and both seemed to go quite well 12:58:41 With Java you also can't be blamed when they turn out to be school-shooters, because "everyone else's doing it too". 12:59:16 "school-shooter"? 12:59:31 A person with a gigantic gun, which shoots schools out of it. 12:59:59 that would be helpful over here, most of the UK faces massive competition for the good schools 13:00:14 and there's a perpetual row about what the fairest way to allocate schooling to peopel is 13:00:17 *people is 13:00:32 and what doesn't help is that the government are trying to get around the problem by attempting to pass laws to make all schools identical 13:00:33 One way is to disqualify people who write "peopel". 13:00:41 I'm not sure if the shot-out schools are very good, though. 13:00:47 which is a) impossible, and b) a bad idea even if it were possible 13:03:13 Incidentally, unless bureaucracy intervenes, my Master's thesis will be accepted/graded next Monday; apparently it will also be the official graduation day, even though there's some formal diploma-distribution event a month later. 13:06:00 ooh, well done 13:06:09 mine was accepted and graded a few months ago 13:06:12 what was it about? 13:06:34 "Methods for Spectral Envelope Estimation in Noise Robust Speech Recognition" 13:06:59 my experience in my department is that such master's projects always go wrong 13:07:03 if it worked for you, then well done 13:07:49 It did take quite a while longer than it should, the work part was more interesting than the writing-about-work part. 13:08:07 I actually rather enjoyed the writeup on mine 13:08:25 it helped that I didn't leave it until after the project was finished 13:10:03 Finally I just wrote it in a month or so, because this "Helsinki Graduate School in Computer Science and Engineering" application deadline is next Friday, and they only accept graduated applicants. (The Academy of Finland -funded project my salary's been paid up to now is going to end this year, and the department would prefer it if I got some independent funding, which is what Hecse would provide.) 13:10:42 are you almost guaranteed to pass? 13:10:44 and with a good grade? 13:11:29 for me, it was nervewracking wondering whether I would get top marks or fail altogether; a quirk in the marking scheme meant that those were the only two plausible results 13:13:16 Well, the supervisor's and professor's statement about it suggested a 5 (out of 5); the approving Council or Conclave or what-ever probably theoretically speaking can decide anything the like, but I think failing is really ludicrously unlikely. As far as I know they could still downgrade it to 4, though. 13:13:29 s/the like/they like/ 13:14:14 The results are sent via email the same evening, at least I won't have to wait long. 13:15:16 There's some sort of quasi-official guideline that they shouldn't be allowed to give the maximum grade without consulting a second professor, but it was a bit unclear whether that's actually observed. 13:16:30 in the UK, the top few grades, the bottom few grades, and all fails, together with a random sample of the others, have to be reviewed by a second University 13:16:48 (I hope it doesn't mean they go all "okay, well, we gave it to this another professor and now you'll have to wait a month more before graduating", because in that case I'll have to do some sort of difficult handwaving in this graduate school application.) 13:16:49 sometimes they interview the students, too 13:17:40 That sounds very serious. 13:17:55 not always, because the grade has to be fixed before the interview 13:18:11 so there's no way to mess up your grade by screwing up the interview 13:18:31 but if the interview doesn't match up with what the 'external examiners' think the grade should be, the university who set the exam/coursework gets into trouble 13:19:15 Aw; I was hoping they'd make the students spin a sort of wheel-of-fortune style pass/fail wheel. 13:19:32 that would be evil 13:19:46 Then they could broadcast it on live national TV. 13:20:26 "graduation" is pretty close (lexically) to "gameshow" anyway. 13:21:08 in English or Finnish? 13:21:38 English. Er, okay, so it's just the same initial letter. It's still closer than, say, a whale. 13:24:47 Actually it's as close in Finnish; "valmistuminen" versus "visailuohjelma". (I had to peek at the dictionary to see what a game show officially is.) 13:25:12 are there unofficial Finnish names for game shows too, then/ 13:26:56 I'm just not sure "visailuohjelma" has exactly the same connotations as "game show". I guess it is pretty close. 13:27:10 such concepts normally don't translate too accurately 13:27:36 actually, that's probably a good way to see how repetitively derivative something is 13:27:54 for instance, I suspect "reality TV" translates more or less perfectly into many different languages 13:28:06 because it didn't really evolve, it just more or less got copied everywhere 13:28:14 and now, thankfully, people are getting bored of it again 13:29:11 Yes, there is an equivalent Finnish word. 13:30:46 fi-Wikipedia says "visailu" was a portmanteau of "visainen kilpailu", literally something like "a difficult contest" except that the word for 'difficult' is a sort of old-fashioned one. (fi:ohjelma is just "program" or "show".) 13:31:11 It's not exactly very "difficult contest" any more, at least in many cases. 13:31:19 well, it's not exactly "reality" TV either 13:31:35 given the efforts the producers go to artificially contrive what happens 13:32:49 I think they've started a new Big Brother round here recently; that's what I most associate with "reality" TV. 13:32:58 they've just finished the penultimate round over here 13:33:09 as in, channel 4 have said one more year, then they give up 13:33:21 it's quite possible that one of the smaller TV networks will buy the rights off them, though 13:34:11 Apparently this is our fifth round. 13:34:52 the rules seem to be slightly different from place to place 13:35:14 come to think of it, I'm almost 100% convinced Wikipedia has a page somewhere summarising the differences, even though I don't know this from personal experience 13:36:32 They certainly have a lot of material about the show. 13:36:33 wow: http://en.wikipedia.org/wiki/Big_Brother_(disambiguation) 13:36:47 All kinds of very colorful tables. 13:36:47 to be precise, the bottom of that page 13:37:13 44 entries 13:37:24 including "Big Brother Second Life" 13:37:36 some are redirects, though, and a couple are stubs 13:37:39 There's also separate page for each of the five seasons for the Finnish version; I assume they extend the same courtesy to all other countries too. 13:37:44 oh, definitely 13:37:55 http://en.wikipedia.org/wiki/Big_Brother_2008_(Finland) is very... tabular. 13:39:19 And the "Variations in the format" section of http://en.wikipedia.org/wiki/Big_Brother_(TV_series) there's something like a summary of differences. 13:40:27 yep, I'm reading that now 13:40:44 and, as is my habit, trying to work out if Big Brother is Turing-complete 13:40:44 "In Germany a new version of the show started: Big Brother - Das Dorf (Big Brother - The Village). -- This was the first version supposed to run for years (without a predetermined end). It was set in a small artificial village including a church tower, a marketplace, 3 houses, 3 working areas -- The season ended after 363 days in February 2006 because of low ratings." 13:40:45 my guess is no 13:41:26 fizzie: see http://en.wikipedia.org/wiki/Castaway_2000 13:41:45 which actually predated Big Brother, and for which it's arguable whether it was reality TV or not 13:41:56 for one thing, it was actually quite good, which means that it probably wasn't 13:44:21 hmm... Wikipedia calls it reality TV; most people at the time called it a documentary, but that's possibly because reality TV hadn't been invented at the time 13:45:32 For some reason I was continuously expecting some sort of gruesome massacre when reading that description page, but apparently it never happened. 13:45:36 Or maybe it's just been hushed. 13:50:18 there was no news of that presented to the outside world 13:50:29 but then, all the information we got is what the contestants themselves broadcast 13:50:36 which is very different from the way most reality TV works 13:54:00 "In the seventh UK series, Big Brother became "twisted". Every week, housemates' mental states were put to the test as Big Brother tried to break them. As a result of this, many housemates broke down." That sounds unintentionally hilarious. (Possibly only to me.) 13:54:23 It's like the housemates are some sort of machines, which break down every once in a while and are then "serviced". 14:08:11 probably they were indeed "serviced".. in a mental clinic^W garage. 14:11:31 -!- CESSMASTER has joined. 14:15:12 -!- JoelyWoely has joined. 14:16:09 -!- CESSMASTER has quit (Client Quit). 14:16:09 -!- JoelyWoely has quit (Client Quit). 14:18:06 Heh. "Teen Big Brother -- was originally shot in advance -- to air in 2003 as an educational item, screened as part of Channel 4's 4Learning programming. -- On Day Six/Seven, Jade Dyer and Tommy Wright became the first Big Brother UK contestants to have sex on the show in its history. According to The Independent, this was the first real-life sexual act shown on British television --" 14:18:12 Yes, that must've been very educational indeed. 14:20:47 they actually showed it in schools 14:20:56 I know because I was in school at the time 14:21:11 but they were obviously trying to contrive the situation to increase as much as possible the chance that there would be sex involved 14:21:58 -!- CESSMASTER has joined. 14:49:50 -!- MatPVB has joined. 14:50:18 -!- MatPVB has left (?). 15:09:16 -!- Slereah has joined. 15:12:57 -!- oerjan has joined. 15:15:38 and what doesn't help is that the government are trying to get around the problem by attempting to pass laws to make all schools identical 15:15:45 ah, the norwegian system 15:15:51 does it fail just as badly there? 15:15:56 (only slightly exaggerated) 15:17:35 well the norwegian school system is in constant need of improvement. many people think the main problem is that it has _too_ frequent reforms, so it never gets to settle into something that actually works... 15:19:03 * oerjan doesn't really know how badly it fails, only that everyone complains about it 15:20:00 however, private/alternative schools are relatively rare, although that too may start to change either way after the election 15:21:30 the left wing seems to want to reduce alternatives but under the pretext of making the state schools better instead, while the right wing wants to allow more private and alternative schools 15:22:30 oh right, the teachers complain about too much bureaucracy ... every reform comes with lots of new forms to fill out :D 15:23:13 and that it's so bad they barely have time to teach 15:24:48 Sounds like the US. 15:25:52 at least we don't have metal detectors ;D 15:28:40 We've consolidated that by moving all the schools into airports. 15:29:02 and the scary thing is I don't even know whether you're joking 15:29:27 -!- FireFly has joined. 15:30:14 There were a select few particularly troublesome schools that had metal detectors in the US after the Columbine incident, and because news from the US tends to become distorted towards the absurd as it leaves our borders (oh who am I kidding, it's distorted towards the absurd in our borders), oerjan probably believes we all have metal detectors in our schools :P 15:30:49 ah. 15:32:00 they use handheld metal detectors at some schools in the UK, apparently 15:32:05 looking for knives rather than guns 15:33:31 Oh, they weren't looking for guns here. 15:33:37 They were looking for golden idols to other gods. 15:33:40 For our God is a jealous God. 15:34:36 heh 15:35:13 well the norwegian school system is in constant need of improvement. many people think the main problem is that it has _too_ frequent reforms, so it never gets to settle into something that actually works... <-- sounds like in Sweden 15:35:17 oerjan, iwc 15:36:01 AnMaster: is the Swedish government trying to make all schools identical too? 15:36:17 ais523, I'm not sure. Yeah it is that bad here... 15:36:18 induction suggests "sounds like almost everywhere", then 15:36:35 GregorR: what about the US? 15:36:56 ais523, I think different political parties in the current government as well as the opposition all pull in different directions 15:37:13 that might be different, or might not, depending on how Republican and/or Democratic the school system is 15:37:38 -!- BeholdMyGlory has joined. 15:39:23 ais523, well the current school system is quite left wing ("social democratic" party was the head of the last government) but the current conservative government (to people from US: This means slightly left of your Democrats ;P) wants to change it iirc 15:39:52 -!- MigoMipo has joined. 15:40:00 schooling's one of the most powerful ways to consolidate your power for future generations 15:40:10 but if the social democratic party wins again I'm sure they will change it again 15:40:12 if you can educate everyone to vote for you, then you can effectively regain your power several years later 15:59:45 -!- kar8nga has joined. 16:01:02 AnMaster: hm, it may seem that those who guessed the mythbusters had got to "good haken"'s dimension were right 16:05:00 oerjan, I don't usually read the forum 16:05:07 so I wasn't aware of ths 16:05:09 this* 16:05:34 well but do you remember good haken? 16:06:10 oerjan, yes, slightly 16:06:27 "vaguely" might be a better word 16:07:00 "vaguely" is an excellent word. everyone should use it more often. 16:07:03 * oerjan ducks 16:08:18 did I typo it or something? 16:08:31 not that i noted 16:09:26 you might vaguely recall someone here who occasionally uses that word, though 16:11:33 oerjan, ... 17:33:25 -!- coppro has quit ("The only thing I know is that I know nothing"). 18:00:39 -!- coppro has joined. 18:10:39 -!- coppro has quit (Remote closed the connection). 18:26:57 err 18:27:00 ais523, there? 18:27:07 I think python just behaved very strange: 18:27:09 >>> -3 % 60 18:27:09 57 18:27:14 why is that strange? 18:27:20 it's a plausible way to do modulo 18:27:25 in fact, the most consistent one 18:27:30 ais523, it is quite different from every other language I seen iirc 18:27:52 there are three ways to do modulo with negative numbers IIRC, and they're relatively equally common 18:28:03 mhm 18:28:17 ais523, what if I instead wanted the reminder? 18:28:24 that is the remainder 18:28:38 -3 / 60 = -1, with remainder 37 18:28:42 oh righ hm 18:28:47 *57 18:28:48 ais523, 57 18:28:51 yeah 18:28:58 ais523, what then is it when it returns -3? 18:29:05 which is iirc what C99 does 18:29:13 wrong 18:29:27 well, that's doing the modulo with negative numbers rather than positive 18:29:37 yes 18:29:41 and iirc that is what C99 does 18:30:43 !haskell ((-3) `mod` 60, (-3) `rem` 60) 18:30:54 (57,-3) 18:31:22 mhm 18:31:35 oerjan, that seems exactly opposite of what ais523 said? 18:31:49 unless that one is printed in reverse order or something 18:31:50 hm? 18:32:00 rem gave -3 and mod 57? 18:32:09 I would assume rem = reminder? 18:32:11 yes 18:32:18 that is the remainder 18:32:18 -3 / 60 = -1, with remainder 57 18:32:20 *remainder 18:32:20 AnMaster: it depends on whether you think 0 or -1 is the correct answer for -3 / 60 18:32:47 !haskell ((-3) `divMod` 60, (-3) `quotRem` 60) 18:32:49 ((-1,57),(0,-3)) 18:33:17 ais523, hm... what is the usual way of rounding... 18:33:35 Define "usual" 18:33:44 towards even numbers, obviously! 18:34:22 well that makes zero sense here 18:35:04 usually integer division for positive integers round towards 0 (or towards -inf if you prefer that), but those work out the same for positive integers 18:35:06 for haskell, div rounds down, quot towards zero iirc 18:35:47 hm 18:35:57 I mean... no one would round *up* for positive integers 18:36:06 (for integer division I mean) 18:36:34 !haskell map (uncurry div) [(-1,-3), (-1,3), (1,-3), (1,3)] 18:36:36 [0,-1,-1,0] 18:37:03 !haskell map (uncurry quot) [(-1,-3), (-1,3), (1,-3), (1,3)] 18:37:05 [0,0,0,0] 18:43:09 Scheme (R5RS) has "remainder" and "modulo" separately; for arguments (13, 4), (-13, 4), (13, -4) and (-13, -4) modulo returns 1, 3, -3, -1 and remainder returns 1, -1, 1, -1. 18:44:08 It's made that way so that for any nonzero n1, n2, you have (= n1 (+ (* n2 (quotient n1 n2)) (remainder n1 n2))); quotient rounds towards zero there. 18:45:33 -!- labo has joined. 19:07:43 question: Silly programming assignment asking about implementing a prime checking function in python, it also specifies that the prime checking function has to make use of the function you wrote just before (which was a function testing if a function is evenly dividable by another)... any idea about a non-naive prime checking function that use it? 19:07:56 because I find implementing the naive one so... pointless 19:08:04 dividing a function by a number is rather tricky 19:08:21 hmm... and dividing a function by another is still harder 19:08:37 ais523, err 19:08:40 integer I meant 19:08:41 not function 19:08:54 sorry for that typo 19:09:25 if they're polynomials it works nicely 19:09:55 oerjan, indeed 19:10:37 in one variable, at least. otherwise i think you need to use that gröbner base stuff which i've never really learned 19:12:19 why doesn't wikipedia has any sort of "list of algorithms for prime test" 19:12:20 or such 19:12:26 * AnMaster looks at ais523 19:12:52 AnMaster: the decisions on what are kept and what are deleted went past the sanity threshhold years ago 19:13:15 ais523, you mean there *was* such a list at some point? 19:13:57 quite possibly not 19:14:10 the side-effect of deletion discussions being unpredictable is that people don't try to start the article in the first palce 19:14:11 *place 19:14:17 AnMaster: http://en.wikipedia.org/wiki/Primality_test isn't good enought? 19:14:23 *-t 19:15:54 oerjan, well... any answer to my original question (+plus typo correction) 19:16:39 *-+ 19:17:00 -_- 19:17:13 (was irresistible) 19:17:17 well yeah 19:18:10 also, no 19:18:47 -!- sebbu2 has joined. 19:19:32 oerjan, meh 19:21:04 AnMaster: how large are the numbers you'll be tested with? 19:21:27 my suggestion for a silly algorithm is to test whether n divides into (n-1) factorial 19:21:44 heh 19:21:46 that works, but it's kind of inefficient above single digits 19:21:57 * AnMaster considers how that would work 19:22:11 ais523, wait, why does that work? 19:22:17 -!- xfire35 has joined. 19:22:22 you also have to check to see if the number's a perfect square 19:22:36 basically, suppose the number has two different factors both greater than 1 19:22:40 which all non-square composite numbers do 19:22:46 then they'll both be involved in the factorial 19:22:49 so the division will work 19:22:51 for a prime, it won't 19:22:55 and for the square of a prime, it won't 19:23:00 heh 19:23:17 actually it should work for a square except maybe 4 i think 19:23:26 because you have both n and 2n in there 19:24:07 Hi! I have a question about Turing-complete languages: How do I prove that I have made one? 19:24:08 ais523, the example of the function used uses 347 as one example for a prime 19:24:09 (sqrt n and 2*sqrt n) 19:24:10 oerjan: oh, good point 19:24:15 No need to think, 4 doesn't divide 6 :-) 19:24:16 -!- sebbu has quit (Read error: 60 (Operation timed out)). 19:24:20 which is for this purpose rather large 19:24:24 xfire35: normally by writing a compiler from some simple TC language into yours 19:24:30 or an interpreter for a simple TC language in yours 19:24:33 !haskell product [1..346] 19:24:35 83786109731741519093102185774271060272617123899294938462545650103958267551816999632384597105781115095746538999305576622259820276406705054815151795675299977497130708316332112015853587961035548260985018324781142346741012571342381443403244375423925845246705116571375322097424190407461087058820073115195230214239315368221366956527807550877343514240476697131078048765701822673835389603217404839039032161799288445236044563772647739851207128764035103255967542618 19:24:43 Small enough 19:24:43 Brainfuck's normally used as the comparison language for that purpose 19:25:05 I was planning to go naive or more efficient :/ 19:25:07 e.g. if you write a compiler from brainfuck into your language, then your language is TC 19:25:11 Why does that prove it is Turing-complete though? 19:25:14 well "smart naive" of course 19:25:29 xfire35: because brainfuck is tc, and your language can do anything that brainfuck can do (due to the compilation) 19:25:33 (like, check only numbers up to sqrt(n) and check only odd ones) 19:25:34 so it must also be at least tc 19:25:42 err typo 19:25:50 OK, thanks :) 19:25:58 xfire35: turing-complete is defined as being able to compile any turing machine to your language 19:26:00 It proves your language can simulate something with power equivalent to that of a universal Turing machine 19:26:21 and although you can do the calculation with an actual turing machine, using another language that's already been proven TC works well 19:26:39 however, if you have something else that a turing machine can be compiled into (i.e. something also tc) then you can go via that 19:26:50 brainfuck, P'', SK combinatory calculus are all good source languages 19:27:10 Isn't proving that there is transform algorithm from some language and that if the original halts, the result will too enough? 19:27:15 Do C++ if you feel ambitious 19:27:32 That would be... interesting 19:27:43 xfire35: Is there description somewhere? 19:28:09 xfire35, what about using BCT? (or has it been mentioned already?) 19:28:12 Ilari: you have to prove the other way round too, that if the original doesn't halt, the result won't 19:28:22 iirc BCT seemed popular for proving TCness recently 19:28:28 oh, good point 19:28:49 ais523, I was wondering why _you_ of everyone skipped BCT in that list 19:29:00 BCT's good if the language you're trying to prove TC doesn't have much in common with other langs 19:29:22 e.g. if it doesn't have any sense of numbers, or data and code have a distance restriction between them 19:29:26 ais523, how was BCT proven TC btw? 19:29:34 The proof must have been.... interesting 19:29:36 AnMaster: via tag systems 19:29:38 Is there another way to prove TC, I have already nearly written one interpreter 19:29:43 proving BCT simulates a tag system is pretty simple 19:29:43 ? 19:29:45 ais523, and how was tag systems proven TC? 19:30:02 xfire35: interpreter for what? 19:30:15 xfire35: all TCness proofs can be expressed in terms of being an interpreter for something 19:30:16 The language I am writing. I also did another one a while back 19:30:37 just writing an interp for your language isn't enough to prove anything but an upper bound for its class, and you're looking for a lower bound 19:31:04 xfire35, oh we didn't mean that, of course such an interpreter is very useful too. We meant writing an interpreter _in_ your language, _for_ a known TC language 19:31:09 writing a program in your language, or a compiler to your language, are the only ways to prove TCness 19:31:22 ais523, actually, writing an interpreter is enough to prove it "TC or less" 19:31:23 I think 19:31:27 Yeah, I gathered that. It would just be annoying to do so. 19:31:30 AnMaster: nah 19:31:35 oh, yes 19:31:41 you can use it to put an upper bound on its class 19:31:42 ais523, unless you write it in a super-TC languag 19:31:44 language* 19:31:47 ais523, yeah exactly 19:32:00 xfire35: BCT is super simple though 19:32:04 likewise, if you write an interp for something in BackFlip, for instance, you know it's termination-testable 19:32:17 ais523, yep 19:32:40 ais523, how was backflip proven termination-testable? 19:32:51 by proving it always terminates 19:33:17 ais523, also writing a bf interpreter in a language isn't enough to prove said language _exactly_ TC. It would be "TC or more" 19:33:32 you could write a bf interpreter in banana scheme easily enough 19:33:33 AnMaster: that's what "proving something TC" means to me 19:33:41 "TC or more" and "TC" are the same thing 19:33:46 IMO, banana scheme /is/ TC; it can just do more as well 19:33:50 Deewiant, hm ok 19:34:06 ais523: that would probably be turing-_hard_, by analogy with NP 19:34:23 heh 19:34:26 ais523, well... what would you call proving "no less, and no more than TC" 19:34:42 np hard means that if it's P, then so are all NP-complete problems, but not vice versa? 19:35:03 AnMaster: TC == can compute every Turing-computable function 19:35:16 np hard means anything in np can be reduced to it, but it's not necessarily in np 19:35:23 AnMaster: "TC and computable" 19:35:25 np-complete means it is also in np 19:35:39 ais523, wait... NP-complete being a subset of P? That doesn't seem to make a lot of sense 19:36:02 AnMaster: if P=NP, then NP-complete is a subset of P 19:36:13 if P != NP, then they're disjoint 19:36:18 AnMaster: ais523 is confusing the definition of np-hard/complete with the p vs. np question itself 19:36:25 that's sort-of the essence of np-completeness 19:36:29 ais523, wouldn't that mean that basically every crypto except OTP would be seriously fucked? 19:36:36 oerjan: well, yes; although, I think my statements are probably true anyway 19:36:49 well possibly quantum cryptos too, but don't know enough about them to be sure 19:36:56 well true but imprecise 19:37:10 Don't confuse symmetric and asymmetric crypto. 19:37:27 also, if P != NP, there are things strictly between P and NP-complete 19:37:29 Ilari, hm, only asymmetric ones would be fucked? 19:37:34 Ilari, that would be bad enough though 19:38:05 oerjan, what would that class be? 19:38:29 quantum cryptography doesn't care about P=NP; it's based on physical principles, not mathematical ones 19:38:52 There's also a few asymmetric cryptosystems not based on the factorization problem; none of them very widespread or (maybe) quite as well-analyzed as you might want. 19:38:55 At least if quantum computers with large bitness became practical, they would totally fuck most asymmetric algorithms. But would be probably quite hopeless against symmetric stuff like AES-256... 19:39:26 oerjan: QBP is probably going to end up between P and NP, isn't it? 19:39:28 Oh, the discussion had went away from the factorization already. Heh. You are so fast. 19:39:34 unless, ofc, P = NP, but that seems rather unlikely 19:40:00 Ilari: well, grover's algorithm gives you an O(n^2) improvement for any algorithm at all 19:40:04 ais523: probably, but it's not proved even assuming P != NP afaik 19:40:13 but it's suspected that for general algorithms, you don't get an improvement better than that 19:40:23 ais523: Yeah. AES-256 is still 2^128 work even with it. 19:40:37 oerjan: yep, neither P != QBP nor QBP != NP is known 19:40:57 however there is a proof that if P != NP then you can construct something in an intermediate class. iirc it uses something diagonalization-like 19:42:05