00:09:07 -!- kwertii has quit (Ping timeout: 252 seconds). 00:11:53 -!- nooga has quit (Ping timeout: 255 seconds). 00:13:15 -!- azaq23 has quit (Read error: Operation timed out). 00:15:35 -!- kwertii has joined. 00:15:36 -!- kwertii has quit (Changing host). 00:15:36 -!- kwertii has joined. 00:52:22 -!- augur has quit (Remote host closed the connection). 01:38:57 -!- cheater_ has joined. 01:39:45 -!- cheater has quit (Read error: Operation timed out). 01:46:40 -!- sebbu has quit (Ping timeout: 252 seconds). 01:49:44 -!- sebbu has joined. 01:49:45 -!- sebbu has quit (Changing host). 01:49:45 -!- sebbu has joined. 02:27:39 -!- augur has joined. 02:30:18 -!- zzo38 has joined. 02:31:34 I think I have figured out how to make a monad from any contravariant functor 02:38:23 -!- Sgeo has quit (Remote host closed the connection). 02:44:15 -!- Sgeo has joined. 02:44:42 Opened up the comp, pushed the HD in (and it was quite easy and went in rather far), so I'll see if that fixes it 02:57:35 -!- augur has quit (Remote host closed the connection). 03:10:43 -!- kwertii has quit (Quit: bye). 03:17:32 -!- augur has joined. 03:56:20 -!- NihilistDandy has joined. 04:18:25 -!- oerjan has quit (Quit: Good night). 04:42:44 -!- MDude has changed nick to MSleep. 05:48:06 Huh. Verisign breach of unknown degree. 05:48:52 The security breach happened *2 years ago* but was only reported to the public *today*, courtesy of Verisign management being horribly fucked up. 05:49:08 There's a chance it extended to their SSL certs. 06:31:38 I think, I have managed to make up the monad from any contravariant functor, a generalization of what is done in the "infinite-search" package. 06:33:23 They use (a -> Bool) -> a while I have used f a -> a where f is contravariant 06:43:27 -!- GreaseMonkey has joined. 06:43:28 -!- GreaseMonkey has quit (Changing host). 06:43:28 -!- GreaseMonkey has joined. 07:48:57 -!- NihilistDandy has quit (Quit: Textual IRC Client: http://www.textualapp.com/). 08:14:14 -!- zzo38 has quit (Remote host closed the connection). 08:42:23 -!- monqy has quit (Quit: hello). 09:50:06 -!- nooga has joined. 09:53:56 -!- cheater_ has quit (Read error: Connection timed out). 09:55:28 -!- cheater has joined. 10:09:11 -!- sebbu has quit (Read error: Connection reset by peer). 10:09:36 -!- sebbu has joined. 10:09:36 -!- sebbu has quit (Changing host). 10:09:36 -!- sebbu has joined. 10:39:39 -!- cheater has quit (Ping timeout: 240 seconds). 10:43:11 -!- cheater has joined. 10:54:39 -!- GreaseMonkey has quit (Quit: The Other Game). 11:44:17 -!- cheater has quit (Ping timeout: 248 seconds). 11:47:08 -!- ais523 has joined. 12:20:57 idiot spambot isn't even getting its links write 12:20:58 *right 12:21:01 "http:\\" 12:23:47 //:dʇʇɥ -- that's what weblinks look like in Australia. 12:39:08 -!- ais523 has quit (Ping timeout: 240 seconds). 12:44:30 Woah how did you make that inverted colon 12:54:01 -!- Taneb has joined. 12:54:07 Hello 12:55:54 -!- ais523 has joined. 12:57:03 -!- Phantom_Hoover has joined. 13:08:10 brb 13:08:17 -!- Taneb has quit (Quit: Leaving). 13:08:41 -!- Taneb has joined. 13:08:49 Back 13:09:50 Eats, shoots and leaves. 13:10:20 That's pandas, you fool, not Hexhamites 13:10:39 -!- nooga has quit (Ping timeout: 240 seconds). 13:10:40 You never know about Hexhamites. 13:10:56 What esolang is the topic in now? 13:10:57 Also I recently checked the weather in Hexham, since it was the first UK location that came to mind. 13:11:05 Heh 13:11:14 It loogs Glassy to me. 13:11:19 Oh dear 13:11:45 The {M[m bit especially. 13:12:02 Oh, I thought you were talking about the weather 13:14:35 The M.m function does (_o)O! -- that is, construct an O object and bind _o to it -- followed by "blah"(_o)o.? -- so "blah", on the stack, get the O.o method from _o, then run it. 13:15:06 O.o outputs a string? 13:15:13 Yes. 13:15:36 I reckon, if I could use it, Glass would be my favourite esolang 13:15:47 Also, you're fizzie! I was meaning to ask you something! 13:16:04 Where does fungot's "fnord" come from? 13:16:04 Taneb: yeah, riastradh is dynamic wind. why doesn't it expand then? 13:16:20 If I'm not mistaken, though, the program could have ben written by using a local variable 'o', as {M[moO!"blah"oo.?]} -- that'd be a bit more compact. 13:16:45 It's the Discordians' fnord. 13:16:51 Okay 13:17:20 I'm not very good at religions 13:18:16 It's a bit of a misuse in that the fnords should be there in-between the "content" part of the sentence, but the way fungot does it replaces the "unknown" word by it. 13:18:16 fizzie: i only get one t! 13:18:38 ^style 13:18:39 Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube 13:18:43 fungot: Yeah, "fungott" would look pretty silly. And remind people of elliott. 13:18:43 fizzie: now that could be nice, for a simple language can be used 13:19:00 -!- Taneb has left ("Leaving"). 13:19:10 -!- Taneb has joined. 13:19:24 I didn't know I had the keyboard shortcut... 13:19:56 Do you know the old "join #2,000 to celebrate the millennium" joke? 13:20:03 No I do not 13:20:38 It's an IRC peculiarity; "JOIN #a,#b,#c" can be used to join multiple channels at once, while "JOIN 0" in fact parts all channels. 13:20:52 So in quite many clients "/join #2,000" will end up parting from all channels. 13:21:00 (In some, it joins #2 and #000 instead.) 13:21:36 It's a bit like the equally old "use /disco to turn on the lights" joke. 13:22:06 (Many clients auto-complete that to a /disconnect.) 13:22:17 Also, fungot seems rather condescending of elliott 13:22:18 Taneb: now everything is screwed up a bit on where you are coding. it just exposes you to often irrelevant low-level details. 13:22:48 fungot holds no respect for us meat-sacks in general. 13:22:49 fizzie: more like java) perl ml modules that define two different types of source code 13:22:58 fungot 13:22:58 itidus21: i don't know until where it should 13:23:12 fungot fungot fungot fungot fungot fungot 13:23:12 itidus21: you would have otherwise written in scheme 13:24:29 !addquote fungot: Yeah, "fungott" would [...] remind people of elliott. fizzie: now that could be nice for a simple language can be used 13:24:29 Taneb: just dig out some snippets of code in the wrapper. maybe for these people: it lets me name things fnord, like yours 13:24:41 Wait, it's ^ 13:24:48 ^addquote fungot: Yeah, "fungott" would [...] remind people of elliott. fizzie: now that could be nice for a simple language can be used 13:24:52 No, that's fungot 13:24:52 Taneb: because i'm just confused 13:24:55 Aaarhg! 13:25:04 `? 13:25:08 `addquote fungot: Yeah, "fungott" would [...] remind people of elliott. fizzie: now that could be nice for a simple language can be used 13:25:08 Taneb: when the value is x. if it loses fnord, then? :p 13:25:18 cat: wisdom/: Is a directory 13:25:35 `ls bin 13:25:39 ​? \ @ \ No \ addquote \ allquotes \ calc \ define \ delquote \ etymology \ forget \ fortune \ frink \ google \ hatesgeo \ json \ k \ karma \ karma+ \ karma- \ learn \ log \ logurl \ macro \ marco \ paste \ pastekarma \ pastelog \ pastelogs \ pastenquotes \ pastequotes \ pastewisdom \ pastlog \ ping \ prefixes \ qc \ quote \ quotes \ roll \ searchlog \ toutf8 \ translate \ translatefromto \ translateto \ units \ url 13:25:48 `pastequotes 13:25:53 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.29440 13:26:12 807) fungot: Yeah, "fungott" would [...] remind people of elliott. fizzie: now that could be nice for a simple language can be used 13:26:49 Funny reply for `? there. 13:26:58 It took too many tries to add that quote. I tried /, !, and ^ before getting it right 13:27:32 `? fungot 13:27:32 fizzie: aha... mediawiki takes care of such odd cases. so i sad down and wrote 30 lines of c 13:27:36 fungot cannot be stopped by that sword alone. 13:27:45 `? Ngevd 13:27:48 ​JbQ.3IG.m-l6c.'\..=q)^..3.X..7mB3^d.9Z.).a...GlӟM;ew4?쮊8,..^J9;Y4.sX\bj1@'h\rNX}8'/j_-..IΕ.ߎd;.6'_mE.c@K7. \ X.g2&d3.ųM.rJl]]C..d୛t{*bttٍ#j.UZU"V`N_rlI.ZY#. \ ]v6}3rP&n.(7Ns.8evi^jd |..Io<{.+.n`.&7"w.˷$".p;P.%VއvA..wx*PQ6dB^1}u㯗c. 13:28:23 I like the expression "to sad down". (Unless it's just a typo for 'sat'.) 13:28:46 Unfortunately, I believe it's more likely to be the latter 13:29:05 Sadding down is an appropriate prelude for writing some C. 13:29:08 -!- pkzip has joined. 13:29:19 -!- pkzip has quit (Client Quit). 13:32:16 -!- boily has joined. 13:35:18 -!- sebbu2 has joined. 13:35:18 -!- sebbu2 has quit (Changing host). 13:35:18 -!- sebbu2 has joined. 13:38:39 -!- sebbu has quit (Ping timeout: 240 seconds). 14:37:20 -!- derdon has joined. 15:05:57 -!- sebbu2 has changed nick to sebbu. 15:10:55 -!- nooga has joined. 15:20:08 -!- jix has quit (Read error: No route to host). 15:35:12 -!- pikhq_ has joined. 15:35:24 -!- pikhq has quit (Ping timeout: 252 seconds). 15:45:19 Bleh. School's actually open today. 15:45:28 Not that I'm going in. 15:45:41 How’s the weather in there? 15:49:31 Blizzard. 15:58:19 -!- Vorpal has joined. 15:58:58 -!- MSleep has changed nick to MDude. 16:15:22 -!- Taneb has quit (Ping timeout: 252 seconds). 16:31:24 -!- Taneb has joined. 16:31:44 Hello! 16:36:44 -!- calamari has joined. 16:38:18 -!- boily has quit (Ping timeout: 255 seconds). 16:45:53 -!- calamari has quit (Quit: Leaving). 17:12:30 -!- Taneb has quit (Quit: Leaving). 17:14:30 -!- zzo38 has joined. 17:39:56 -!- Phantom__Hoover has joined. 17:42:19 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds). 17:45:15 Wow, Glass in the /topic. 17:47:54 Gregor: Am I right that it could be the more compact {M[moO!"...."oo.?]} too? 17:48:23 !glass {M[moO!"Let's find out!"oo.?]} 17:48:25 Let's find out! 17:48:28 ^^ 17:48:54 Variables that start with lower-case letters are class variables, so it's namespace pollution is all. 17:49:07 (IIRC) 17:50:43 -!- hagb4rd has joined. 17:51:37 tropicOfCancer = (cancer, zeroLat) Ecliptic Tropical Earth; -- I am trying to think of a kind of ephemeris software in Haskell........ 17:52:28 Oops! 17:52:55 tropicOfCancer = FixedSphericalCoordinates (cancer, zeroLat) Ecliptic Tropical Earth; 17:53:37 -!- olsner has quit (Quit: Leaving). 17:53:51 Any ideas? 17:55:19 -!- sebbu has quit (Ping timeout: 240 seconds). 17:57:08 wtf is ephemersis 17:57:16 oh 17:57:31 Ephemeris is a way to figure out position of planets, sun, moon, etc 17:57:47 By using calculation instead of having to do observation 18:03:15 -!- derdon has quit (Remote host closed the connection). 18:03:31 do you _believe_ the the celectial bodies affect our lifes in some way? 18:04:14 hagb4rd: Mostly the sun and moon; the others wouldn't do much. But I don't believe they affect our lives in the way that astrologers say they do; they are just using superstitious stuff. 18:04:30 hehe.. is see 18:05:23 -!- sebbu has joined. 18:05:24 -!- sebbu has quit (Changing host). 18:05:24 -!- sebbu has joined. 18:05:25 so why that interest in astrology? is just a technical challange? 18:05:45 I'm pretty sure that's just it. 18:06:02 Yes, as well as the things in common to astrology with astronomy, and some things are useful in both even if not common in either, I suppose. 18:09:55 When doing some kinds of calculation (including calendars), conventions of both astrology and astronomy can be useful; in addition, astrology does have some artistic uses even though divination doesn't work. So, functions of ephemeris can be useful for astronomy as well; and I know about their conventions such as right ascensions, declination, hour angle, etc. 18:11:17 Someone once wrote a article about astrology for astronomers. They tell how the conventions are similar and differences, and aspects, and so on. In my opinion, it is useful to combine astrology, astronomy, and calendar date/time stuff, into one program, since all three functions can be used together. 18:13:10 sure. there is a common determinator between both of them 18:14:30 'though im not sure if between is the correct proposition 18:15:41 there are things known and unknown.. and in between is hagb4rds ignorance 18:18:19 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds). 18:36:45 I think the worst thing about astrology is that the existence of the pseudoscience prevents us from investigating the social science of actual connections between birthdates and personalities, which would naturally stem from the interaction between environment and mating behavior of any animals with a short estrous cycle (e.g. humans). 18:37:35 Gregor: People can do it if they want to! But the exist does make people argue about it probably; still, some people can try to do so. 18:40:38 Do *you* have ideas how to do so more specifically, anyways?? 18:41:54 i dunno.. guess its like tarot. it helps us to reflect on ourselfs 18:42:49 Aaah, tarot. European card games used for divination. 18:43:31 hadb4rd: Well, yes, you can reflect on ourselves using various divinatory methods (but not for divination). You probably mean cartomancy rather than tarot; tarot is simply a deck of cards. Psychological cartomancy uses a tarot deck with more art than ordinary decks; you could probably find an article about that. 18:43:56 yea, thats i mean 18:44:01 +what 18:46:05 I do know a few games involving tarot cards, including a few modern ones, and one which uses icehouse pyraminds in addition to the cards. 18:46:51 i was studying crowley for a while 18:47:54 guess the problem bout science and axioms is it answers how but not why 18:48:08 +its 18:48:32 Science can answer why to a limited extent. Kind of. 18:48:55 Of course there are problems with science but it is the best we have. 18:49:45 to quote s.o. i do not know: problems are hidden opportunities ;) 18:50:07 hagb4rd: Yes, that too! Is good. 18:51:09 One example of artistic use for astrology is someone made up a time of birth for Harry Potter. (Of course it is still arbitrary like many arts but at least they have something to do.) 18:57:18 The best way to reflect yourself is with a mirror. 18:58:27 Sgeo: Yes I agree 19:04:41 -!- ais523 has quit (Remote host closed the connection). 19:34:53 -!- nooga has quit (Ping timeout: 260 seconds). 19:53:01 -!- Phantom__Hoover has joined. 19:53:10 -!- Phantom__Hoover has quit (Changing host). 19:53:10 -!- Phantom__Hoover has joined. 19:59:01 -!- sebbu2 has joined. 19:59:01 -!- sebbu2 has quit (Changing host). 19:59:01 -!- sebbu2 has joined. 20:02:00 -!- sebbu has quit (Ping timeout: 240 seconds). 20:02:08 -!- sebbu2 has changed nick to sebbu. 20:10:32 -!- Phantom__Hoover has quit (Read error: Connection reset by peer). 20:10:53 -!- Phantom__Hoover has joined. 20:13:27 -!- oerjan has joined. 20:17:47 > reverse "elliott sacked as bearer of Element of Loyalty, seeking pegasus replacement | http://codu.org/logs/_esoteric/ | Now slightly on-topic | Now failing to construct an esolang in THE. WORST. POSSIBLE. WAY." 20:17:48 ".YAW .ELBISSOP .TSROW .EHT ni gnalose na tcurtsnoc ot gniliaf woN | cipot-... 20:17:54 fff 20:17:57 ^rev test 20:17:57 tset 20:18:03 > rev "elliott sacked as bearer of Element of Loyalty, seeking pegasus replacement | http://codu.org/logs/_esoteric/ | Now slightly on-topic | Now failing to construct an esolang in THE. WORST. POSSIBLE. WAY." 20:18:04 Not in scope: `rev' 20:18:08 ^rev "elliott sacked as bearer of Element of Loyalty, seeking pegasus replacement | http://codu.org/logs/_esoteric/ | Now slightly on-topic | Now failing to construct an esolang in THE. WORST. POSSIBLE. WAY." 20:18:08 ".YAW .ELBISSOP .TSROW .EHT ni gnalose na tcurtsnoc ot gniliaf woN | cipot-no ylthgils woN | /ciretose_/sgol/gro.udoc//:ptth | tnemecalper susagep gnikees ,ytlayoL fo tnemelE fo reraeb sa dekcas ttoille" 20:18:32 -!- oerjan has set topic: 0>:#,_@. 20:18:34 argh 20:18:41 -!- oerjan has set topic: 0".YAW .ELBISSOP .TSROW .EHT ni gnalose na tcurtsnoc ot gniliaf woN | cipot-no ylthgils woN | /ciretose_/sgol/gro.udoc//:ptth | tnemecalper susagep gnikees ,ytlayoL fo tnemelE fo reraeb sa dekcas ttoille">:#,_@. 20:19:03 Can you find the mistake in this program? http://sprunge.us/gbUF 20:30:20 -!- wanham has joined. 20:30:44 `welcome wanham 20:30:49 wanham: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 20:31:42 -!- chicken has quit (Quit: WeeChat 0.3.2). 20:34:36 -!- Saizan has joined. 20:35:12 so he made it more official http://www.scottaaronson.com/blog/?p=902 20:43:33 Is this just his second $100k thing? Funny how he's already "that guy who always offers $100k for unlikely things". 20:43:53 actually the first one was $200k 20:44:13 Oh, right. Well, "large amount of money" anyways. 20:45:02 i'd suppose gil kalai suggested it in jest, anyway :P 20:45:33 (In not-so-related news, no distribution-related wisdom from the CoaP forum, but 12 views of which only 2 are mine, anyway.) 20:46:56 -!- GreaseMonkey has joined. 20:47:34 make that 13, or something 20:52:07 fizzie: do you have the full sequence of results saved? to check if "every 10 seconds" matters, you might try to graph them in order retrieved or something... 20:53:16 Sure; I guess I'll try that too, when I get to a real computer. 20:54:23 or maybe check if the ratios change much in subintervals 20:55:19 Also a goodie. 20:56:11 I think the same one was quite consistently 1.3 times the mean during the run. But that's just a feeling. 20:57:22 -!- wanham has quit (Quit: Page closed). 21:06:50 What was the first? 21:06:56 The first prize thing 21:07:48 Was it that P=NP proof? 21:07:49 to vinay deolalikar in case his P != NP proof turned out to be correct / repairable 21:08:14 which quickly started looking rather unlikely 21:08:25 -!- Taneb has joined. 21:08:39 Hello! 21:08:43 but hi! 21:09:20 oerjan: If only they'd prove P != NP, then we'd finally know that N != 1. 21:09:51 The topic is making me read backwards 21:09:55 * oerjan swats fizzie -----### 21:10:07 doog yrev :benaT 21:10:23 !befunge 0".YAW .ELBISSOP .TSROW .EHT ni gnalose na tcurtsnoc ot gniliaf woN | cipot-no ylthgils woN | /ciretose_/sgol/gro.udoc//:ptth | tnemecalper susagep gnikees ,ytlayoL fo tnemelE fo reraeb sa dekcas ttoille">:#,_@ 21:10:24 48 Unsupported instruction 'Y' (0x59) (maybe not Befunge-93?) \ Unsupported instruction 'A' (0x41) (maybe not Befunge-93?) \ Unsupported instruction 'W' (0x57) (maybe not Befunge-93?) \ 121 Unsupported instruction 'E' (0x45) (maybe not Befunge-93?) \ Unsupported instruction 'L' (0x4c) (maybe not Befunge-93?) \ Unsupported instruction 'B' (0x42) (maybe not Befunge-93?) \ Unsupported instruction 'I' (0x49) (maybe not 21:10:41 -!- MoALTz has joined. 21:10:52 Stringmode doesn't seem to work 21:10:52 wat 21:10:54 Oh, it's been Befungized. (Or is that over 80 characters? Funge-98ized in that case.) 21:11:01 !befunge98 0".YAW .ELBISSOP .TSROW .EHT ni gnalose na tcurtsnoc ot gniliaf woN | cipot-no ylthgils woN | /ciretose_/sgol/gro.udoc//:ptth | tnemecalper susagep gnikees ,ytlayoL fo tnemelE fo reraeb sa dekcas ttoille">:#,_@ 21:11:02 elliott sacked as bearer of Element of Loyalty, seeking pegasus replacement | http://codu.org/logs/_esoteric/ | Now slightly on-topic | Now failing to construct an esolang in THE. WORST. POSSIBLE. WAY. 21:11:10 I see it is. 21:11:19 oh right 21:11:42 -!- jix has joined. 21:12:00 -!- jix has quit (Client Quit). 21:12:07 -!- jix has joined. 21:12:21 `welcome MoALTz 21:12:25 MoALTz: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:12:45 *nod* 21:13:06 and in a rare moment of on-topicness, too! 21:13:24 Can I return things to normal? 21:13:48 by "normal" you mean "talking about everything except esolangs", right? 21:13:51 Yes 21:13:58 you may. 21:14:06 My gran's in hospital 21:14:07 oerjan: So how're things in the *real* esoteric programming channel that you're spying for? Are they all jubilous for our recent... quietudiness? 21:14:20 wat 21:14:46 * oerjan is a single-channel person 21:14:56 So, new esolang 21:15:19 Taneb: HEY YOU FREAK, WE DON'T TALK ABOUT THAT HERE 21:15:56 well, i guess it beats hospital talk. 21:16:07 It's a CA, so can someone experienced with ALPACA help me? 21:16:07 It might still be about that. 21:16:16 -!- azaq23 has joined. 21:16:27 -!- azaq23 has quit (Max SendQ exceeded). 21:16:35 We haven't heard details. Maybe it's a hospitals-and-grandmothers-themed CA. 21:17:03 -!- azaq23 has joined. 21:17:12 i don't consider myself "experienced" with ALPACA, although i've read the spec once, i guess 21:17:24 Nah, it's falling down massive shafts based 21:18:06 That doesn't sound like it'd exclude hospitals, almost the opposite. 21:18:18 It doesn't mention hospitals 21:18:20 Or grandmothers 21:18:31 Okay then. 21:19:30 * oerjan is reminded of the danish "Riget" series (which he's never seen, but which an old student friend loved) 21:19:42 i believe it had both hospitals and shafts. 21:20:00 i'm not quite sure about the shafts. 21:21:47 * oerjan confirms the shafts using google 21:23:01 http://en.wikipedia.org/wiki/The_Kingdom_(TV_miniseries) 21:26:00 -!- sebbu has quit (Read error: Connection reset by peer). 21:26:25 -!- sebbu has joined. 21:26:26 -!- sebbu has quit (Changing host). 21:26:26 -!- sebbu has joined. 21:28:21 -!- bbear has joined. 21:28:26 hello 21:28:38 `welcome bbear 21:28:41 bbear: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:30:12 I wanted to know :: do you know if it would ever be possible to program a computer with an infinite amount of memory ? 21:30:37 programming is easy. getting one on the other hand... 21:31:01 bbear, in the real world, it is impossible to create a computer with infinite memory 21:31:30 Taneb, we had a discussion about this on ##c 21:32:03 but a question slightly different would be if ever such a computer exist, would it be possible to program with it ? Would the « programming » word would have a sense ? 21:32:11 Ah, yes 21:32:21 It would definitely be possible to program with it 21:32:47 oerjan: http://users.ics.tkk.fi/htkallas/mezzoseq.png -- Does that look periodic to you? It looks periodic to me. 21:33:32 indeed 21:33:50 (It's the ten thousand visits on the X axis, with the small differences in Y axis giving the index of which thing came out. I've just wrapped it into ten lines to make it a bit more clearer.) 21:35:22 Maybe I should permute the indices a bit so that it'd look even more periodic. 21:36:02 do that. 21:37:29 from that, one _might_ suspect the script is keeping a counter... 21:38:23 bbear, with something like brainfuck, you could do [+>], which would use up all the memory after an infinite amount of time 21:38:28 if so, the irregularities would be others looking at the page simultaneously. 21:38:29 -!- zzo38 has left. 21:38:36 Can you imagine a language to program games ? 21:38:42 Well, *now* it is very periodic indeed. (Same URL.) 21:38:55 -!- quintopia has quit (Ping timeout: 255 seconds). 21:39:31 bbear, you could (if you really want to) program games in brainfuck, assuming you didn't care about graphics 21:39:35 Taneb, yeah... I was guessing that an infinite memory could make the system unstable because parts of the system could be uncontrolable in a finite time. 21:39:40 oerjan: Doesn't immediately explain why one value would be biased over the others, though. 21:39:45 But here we are talking about every language. 21:39:54 I mean i want to free my imagination. 21:40:03 bbear, any part of the system you have accessed, you can access again in finite time 21:40:16 fizzie: hm it looks like it relatively frequently jumps back in the sequence. was your 10 sec retrieval very regular or did you take many breaks? 21:41:26 fizzie: i don't suppose you kept the datestamps? :D 21:42:05 oerjan: The file timestamps are probably still there; but I didn't take any breaks at all. 21:42:12 ok. 21:43:09 Of course it was timed with a "sleep 10", so it's not an accurate frequency; in particular the delay between successive fetches depends on how long the fetch itself takes. 21:44:24 hm so if it's timestamp based, random network delay may explain the irregularities 21:45:52 i have a hunch the graph is not accurate enough to see how much it fetches the same twice in a row 21:47:24 fizzie: could you find the frequencies of the _consecutive_ pairs? 21:48:07 it seems that for each item, there's one particular which is most frequently given next, but the alternatives might also give information 21:48:31 (well that's just 2-grams, isn't it.) 21:48:51 -!- Jafet1 has joined. 21:49:19 i suppose timestamps are also necessary to check 21:49:20 -!- Jafet has quit (Ping timeout: 240 seconds). 21:50:29 oerjan: Hey, there's a DMM reply. 21:50:34 Turns out it's time()%11. 21:50:39 yay 21:52:02 Also the "Wow!" signal, the most common duplicate, is also the one that is the longest. Presumably it has a higher likelihood of taking longer than the ten seconds required to walk backwards in the time()%11 sequence. 21:52:21 s/duplicate/version/ 21:55:01 i suppose if you gave a table of frequencies, usual order in the common sequence, and page length, that should explain most of it... 21:55:38 the longer, the more likely to redo it. hm right. 21:57:51 are frequencies and length in completely the same order? 21:59:04 well no. i'm pretty sure the Langerhans one was shorter than the Xabadunis. 22:00:42 -!- bbear has quit (Quit: Goodbye). 22:00:43 > (sum . concat . tails . flip (uncurry enumFromThenTo) 1 . fmap pred . join (,)) 12 22:00:44 364 22:00:45 I don't think they were, no; just the Wow! signal one was the longest. 22:01:02 since removing the top two made it indistinguishable from uniform, i suppose only the two longest were long enough to overcome the natural variation. 22:01:05 How many things are there on the nth day of christmas 22:01:16 And all preceding ones 22:01:18 fizzie: was the second one at least second longest too? 22:01:26 :t (sum . concat . tails . flip (uncurry enumFromThenTo) 1 . fmap pred . join (,)) 22:01:27 forall a. (Num a, Enum a) => a -> a 22:02:03 oerjan: Counts and byte-sizes: http://sprunge.us/QZPd 22:02:03 > scanl1 (+) $ scanl1 (+) [1..12] 22:02:04 [1,4,10,20,35,56,84,120,165,220,286,364] 22:02:14 That is likely shorter 22:02:22 Some correlation, but not quite the same order. 22:03:02 Anyway, the sequence seems to quite often step back more than one step, so if there's a bias where those skips happen, then the sequence order would also be a factor. 22:03:17 > (last . scanl1 (+) . scanl1 (+) . enumFromTo 1) 12 22:03:19 364 22:03:37 maybe it has something to do with exceeding 4096 bytes... 22:04:18 > sum $ scanl1 (+) [1..12] 22:04:19 364 22:05:07 Taneb: btw did i mention that using null and head/tail instead of pattern matching is usually considered bad haskell style? 22:05:26 Probably 22:05:37 especially when you use the null to determine whether the head/tail is safe. 22:05:48 Could also have something to do with other boundaries; the standard MTU (max packet size) tends to be 1500, sometimes smaller, and there's the HTTP overhead + TCP overhead + IP overhead + whatever-is-the-network-layer overhead. 22:08:40 (And it might be gzipped over the HTTP wire.) 22:11:12 fizzie: oh hm to get redone there doesn't have to be a _large_ network delay, just enough to wrap from 0.9 seconds to 11.0 instead of 10.9 22:11:33 *extra delay 22:17:06 -!- GreaseMonkey has quit (Quit: The Other Game). 22:19:53 -!- zzo38 has joined. 22:20:10 Taneb: oh hm have you logged in as Ngevd since i sent those lambdabot messages? 22:20:49 How can we invent Haskell*#?@~ which is a new kind of group of extensions for Haskell which include many strange thing in addition to the normal things 22:21:02 oerjan, you sent them to Taneb 22:21:07 i did? 22:21:17 Yeah, and I cleared them because I logread them 22:21:37 ok then 22:22:44 Ohhh, and here I was thinking you thought there was probably something embarrassing in them and therefore @clear'd instead of @messages'd. (I'm not entirely sure why I thought that, since I suppose you'd just have read them privately in that case.) 22:22:45 zzo38: sounds like what the ghc people are doing all time *badum-tish* 22:22:51 *all the time 22:23:47 oerjan: They are but a lot of things I proposed are the things other people hate. So that is why to invent new one including things too strange for most people in #haskell even though some people in #esoteric might like it, not everyone does 22:24:13 Such as the -XNoEnglishKinds extension which almost everyone in #haskell channel hates 22:24:58 What does -XNoEnglishKinds do? 22:25:29 zzo38: i suspect that if they had considered polymorphic kinds from the start, they would have used alphanumeric kind names from the start too 22:26:20 Taneb: Changes OpenKind back to ? and Constraint to & leaving the alphanumeric names open for custom datakinds and so on 22:26:51 Fancy 22:27:53 Of course, I don't really know enough to appreciate the significance of that 22:28:20 :k (->) 22:28:21 ?? -> ? -> * 22:29:20 This, I am afraid, means nothing to me 22:29:37 As well as -XMoreNotation, -XImprovedNaturalNumberKinds, -XInstanceDisambiguation, -XDefaultInstances, -XInlineLLVM, -XTemplateHaskellKinds, and so on... most people hate these things 22:30:15 -!- Saizan has left. 22:30:21 And the new macro system as well 22:31:04 it means that the -> type constructor takes two arguments, the first being a boxed or unboxed type, the second being that or an unboxed tuple, and returns a boxed type. 22:31:35 the upshot is that you can have haskell function types that handle some unboxed values, although the functions themselves are always boxed. 22:31:46 What do boxed and unboxed mean in this context? 22:32:01 I feel kinda stupid at the moment 22:32:16 boxed is the usual kind of haskell value, implemented as a pointer to a lazy thunk 22:32:16 For that matter, what does tuple mean in this context? 22:33:36 unboxed is a value represented without pointers and laziness, like a sequence of bytes in any other language 22:33:57 Okay, that makes sense 22:34:08 Seems the wrong way round to me? 22:34:09 ghc uses that internally for efficiency 22:34:43 and an unboxed tuple is a tuple of several unboxed values, which can only be returned from functions, not passed into them. 22:35:27 (the values of an unboxed tuple may end up being put directly on the stack or in registers) 22:35:41 @src IO 22:35:42 Source not found. Maybe you made a typo? 22:35:50 As well as -XCompatibility703, -XNoParameterTypeClasses, -XCombinedInstances, -XNoUnicode, -XExtendedOperatorNames, and so on 22:35:52 sadly @src has removed that stuff 22:36:40 So, because functions are boxed, they can flit about in the mind of the haskell program without being in a specific place in memory 22:36:58 But can simultaneously look at bits of memory and other things flitting about 22:37:26 Taneb: unboxed things can be copied of course, like in C... 22:37:29 @src Int 22:37:29 data Int = I# Int# 22:37:52 an example there: Int is implemented as a datatype wrapping an unboxed Int# 22:38:00 I don't know how you call the Haskell that I proposed or how to change GHC or anything like that; write a new one; I don't know how do you know? 22:38:26 ghc's optimizer may frequently remove the entire data wrapping, and handle words in memory instead 22:38:35 Wasn't elliott working on a Haskell Compiler a while back? 22:39:09 oerjan: Yes that is a good idea. 22:39:28 (well of course a data type is also implemented in memory, but with all that laziness support included 22:39:31 ) 22:40:05 Taneb: i don't quite remember. 22:40:35 He was trying to think of a name because Hexham Haskell Compiler looked stupid? 22:41:19 anyway, standard haskell has only one base kind * which is boxed types, while ghc adds a heap of others for supporting unboxed things. 22:41:39 And my proposal added two more + and & 22:42:00 What would they be? 22:42:11 and other haskellers already added & except with the name Constraint 22:42:24 + for natural number types and & is what is now called Constraint 22:43:10 Except mine is different where + is a subkind of * which now means that the built-in type (->) is the type of array with a fixed number of elements. 22:45:12 Another idea is a new kind of n+k patterns (different from the old one) usable both in type level and value level, with types of kind + 22:46:13 zzo38: n+k patterns? now you're just _asking_ for them to hate it :P 22:46:25 -!- MDude has quit (Ping timeout: 255 seconds). 22:47:29 I do hate ordinary n+k patterns 22:47:36 But this is a new kind 22:49:21 Which is only usable with values of types of kind + (as well as types of kind + themself in type level patterns) 22:50:33 I've been thinking about a graphical version of Haskell, possibly similar to Scratch, App Inventor, et al. 22:51:06 It's pretty difficult 22:52:04 oerjan: See? These are a new n+k patterns which are different than the old n+k patterns. I also hate the old n+k patterns but this new kind can explain natural numbers in general 22:52:23 mhm 22:54:34 Kinds would make it simpler 22:55:39 Goodnight 22:55:40 -!- Taneb has quit (Quit: Leaving). 22:55:49 -!- monqy has joined. 22:57:23 For example: type family X (n :: +) (t :: *); type instance X 0 x = x; type instance X (n + 1) x = X n x -> X n x; is a kind of the new n+k patterns for natural number kinds, at the type level. But they could be used at value level as well. 22:58:17 Is it understand now? 22:58:42 i think so 23:01:36 hm except for notation, won't this be essentially that new type-to-kind lifting extension applied to data Nat = Zero | Succ Nat 23:02:26 (i'm not into the details of that, though) 23:04:18 I have read about that, but I don't think it is exactly the same thing, because my proposal puts + as a subkind of * so for example, (100 -> Bool) is a valid type which means an array of 100 bits 23:05:13 (And like I specified, these new n+k patterns would also be usable in value patterns as well as type patterns) 23:16:42 -!- pir^2 has joined. 23:20:26 -!- pikhq has joined. 23:20:37 -!- pikhq_ has quit (Ping timeout: 255 seconds). 23:21:16 Another idea is to write new versions of the list operations: length, (!!), findIndex, and so on, using: class Peanoid x where { zeroP :: x; succP :: x -> x; }; class Copeanoid x where { predP :: x -> Maybe x; }; (These are a specific kind of pointed unary systems, and some variations on them.) 23:21:34 This could be done even in existing Haskell compiler, simply using alternative Prelude 23:21:43 Or other libraries can be written easily. 23:24:01 (h : t) !! i = maybe h (t !!) $ predP i; length [] = zeroP; length (h : t) = succP $ length t; 23:25:43 I don't know whether you would prefer new versions of the list operations defined like this or not 23:41:45 -!- MDude has joined. 23:47:22 -!- mrjbq7 has joined. 23:47:58 -!- mrjbq7 has left. 23:49:32 I think I like the word "ciretose" 23:53:20 There doesn't seem very good document for "preprocessor-tools" package.