00:04:29 <ehird> The ESO pastebin _finally_ launches, after me working on it for all of 10 minutes. :-P
00:04:38 <ehird> http://paste.eso-std.org/. Yes, it's spartan, yes, that's intentional. have fun.
00:05:09 <ehird> Unfortunately, pastes 1 and 2 (http://paste.eso-std.org/1 and http://paste.eso-std.org/2) are taken from me testing.
00:05:14 <ehird> So you can't nab that piece of history.
00:21:26 <ehird> ... and now there's a python api (available on the page)
00:46:57 <GregorR> Aww, it's not written in Brainfuck? :(
00:48:16 -!- jix has quit ("This computer has gone to sleep").
00:57:17 <ehird> GregorR: No, i'm not that crazy.
01:06:04 -!- Slereah_ has quit (Read error: 113 (No route to host)).
01:34:15 -!- ab5tract has quit.
01:45:01 <GregorR> http://www.cafepress.com/bizarregeek.318713308
01:56:33 -!- ab5tract has joined.
02:02:43 <GregorR> pikhq: "There is no reason for anyone but Gregor to wear this."
02:03:50 <pikhq> Unless I want people to choose Gregor's hats?
02:31:04 -!- Jiminy_Cricket has joined.
02:32:15 -!- poiuy_qwert has joined.
02:34:39 -!- omniscient_idiot has joined.
02:36:10 -!- ab5tract has quit.
02:58:24 -!- poiuy_qwert has quit.
03:07:08 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | 6, because the Ex normally means they had to add an extra parameter at some point.
05:17:25 -!- cherez has joined.
05:17:30 -!- cherez has left (?).
05:22:33 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
05:34:48 -!- GreaseMonkey has joined.
06:44:24 -!- GreaseMonkey has quit ("So, how much do you love noodles?").
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:50:11 -!- oerjan has joined.
09:02:37 <fizzie> 9438299 4-grams, but it hasn't finished doing whatever it's doing after that. I don't think it should be doing much else than writing things to file, though.
09:07:08 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | chr() arg not in range(256).
09:50:42 -!- oerjan has quit ("leaving").
10:10:42 -!- Mony has joined.
10:29:48 <fizzie> Sorry, was building furniture.
10:29:58 <fizzie> Or maybe "assembling" is the correct word.
10:30:42 <fizzie> It has written just 40 megs of model.bin.
10:31:24 <fizzie> fis 11679 1.7 67.2 2367952 1385348 pts/10 D+ Oct17 13:30 ./buildlm
10:31:37 <fizzie> Quite a large process.
10:32:14 <fizzie> I'm thinking the whole N-gram tree is too big to fit in memory and it's so udderly slow thanks to all the swapping.
10:32:28 <fizzie> I wonder if I should just interrupt it; I'd like to get my web browser back.
10:57:01 <AnMaster> <fizzie> Or maybe "assembling" is the correct word. <-- IKEA?
10:57:46 <AnMaster> fizzie, do you use lots of std:: stuff in that C++ app?
10:57:54 <fizzie> Yes, to both of those questions.
10:57:59 <AnMaster> normally they take a bit more memory
10:58:17 <fizzie> It was just convenient.
10:58:24 <AnMaster> most of the time not an issue... but it seems to be here
10:58:37 <AnMaster> would it be impossible to process it in chunks?
10:58:57 <AnMaster> and try to keep locality of reference to prevent swapping in/out all the time
10:59:08 <AnMaster> instead working on the data already in memory
10:59:43 <fizzie> Well, it's a big tree, 17581526 nodes. The per-node data structures are not very small either, even though they could be.
11:00:56 <fizzie> I don't think there are, and in any case it just builds the tree and writes it to file; it's not deleting any nodes ever.
11:01:23 <AnMaster> fizzie, well making the data more compact could help indeed
11:01:43 <AnMaster> making sure to organise structs to prevent holes and padding
11:01:58 <fizzie> Yes, I think I could cut memory consumption to a half without too much trouble.
11:02:28 <AnMaster> (the tool pahole, made for the linux kernel originally, can help with that, it reads debug info and report badly organised structs)
11:03:50 <AnMaster> (I have used it on cfunge to reduce the memory usage)
11:05:31 <fizzie> There's even some data duplication here, I have a std::map<unsigned, Node*> for children and a std::map<unsigned, unsigned> for the word-counts, and for every non-leaf node both mappings have the same set of keys. The program is quite the memory hog, but really, I was more concerned in getting it to work at all than making it efficient.
11:07:20 <fizzie> Must go to a birthday thing now. With any luck it has finished before I get back, and then I can postpone fixing that C++ code until the next time I have a large amount of data to feed to fungot.
11:07:21 <fungot> fizzie: premature optimization etc. for now, we're on cleaner topics today.
11:07:36 <fizzie> Heh, that's even related.
11:07:54 <AnMaster> fizzie, idea: ALICE bot in befunge
11:09:26 <fizzie> Currently fungot's babbling is a bit cheating-like; all the hard work is done by this supporting code, and it doesn't even keep any sort of data structures in funge-space, simply uses FILE's seek/read operations to navigate the resulting tree.
11:09:27 <fungot> fizzie: is arizona fnord herbal tonic any good?)
11:10:12 <fizzie> fungot: Simply based on the name, I don't think I'd buy it. The "fnord" there is suspicious.
11:10:13 <fungot> fizzie: the multiple values to bind, what parameters your procedures accept, and what some of the old lisp os's ( can't remember the last time i checked
11:15:04 <fizzie> The babble generation is pretty much: len = 0; while (1) { node = root; for (i = 1; i <= len; i++) { if (!node.children[text[len-i]]) { word = random_with_weights(node.words); break; } node = node.children[text[len-i]]; } text.append(word); len++; if (node.canstop && random_stoppage()) break; }
11:15:09 <fizzie> Very easy to do in Befunge.
11:15:11 <fizzie> Oh, the bus is going. ->
12:40:51 -!- Slereah_ has joined.
12:44:58 <ehird> i read away messages
12:48:18 <ehird> damn i want this http://www.cafepress.com/bizarregeek.11389675 and this http://www.cafepress.com/bizarregeek.255291006
12:48:22 <ehird> except the former's colour scheme puts me off
13:24:03 <ehird> he's not actually here
13:26:24 <AnMaster> ehird, hm one of the products in "Other items by Bizarre Geek:" there says "Original design by RodgerTheGreat, used with permission"
13:26:34 <ehird> GregorR runs that store.
13:26:43 <AnMaster> yes that was what I was about to ask
13:27:29 <ehird> i'm going to make a book of pages and pages of cellular automata
13:27:32 <ehird> i will title it the null string
13:28:47 <ehird> I hate how folklore.org has not been updated since like 2004 and they still haven't added any other collections apart from the original macintosh one
13:48:26 -!- jix has joined.
13:48:52 <ehird> yep, I just los it .
13:50:40 <ehird> http://www.freesoftwaremagazine.com/columns/ten_easy_ways_attract_women_your_free_software_project <- Jesus christ, this is a load of sexual bullshit just pandering to a bunch of idiotic stereotypes.
13:50:55 <ehird> lol freudian slip because all men think about is sex lol
13:52:11 -!- ENKI-][ has quit (Read error: 104 (Connection reset by peer)).
13:53:27 <Slereah_> ehird : That's because stereotypes are funny
13:53:45 <ehird> Not very useful for serious advice though...
13:54:04 <Slereah_> A top ten list is never serious, ehird
13:54:25 <Slereah_> It could be "top ten most efficient sorting algorithm" and be a fucking joke
13:54:27 <ehird> There's not much making it a top ten beyond the sections being labeled with numbers.
13:57:49 <jix> nargh... because of you i lost the game....
13:58:06 -!- ENKI-][ has joined.
14:02:57 <Slereah_> http://membres.lycos.fr/bewulf/Divers5/1220910895470.jpg
14:12:38 -!- kar8nga has joined.
14:12:48 <AnMaster> so just put a regex ignore on that stuff if you care about it. I certainly don't.
14:16:38 <ehird> AnMaster: That's not easy.
14:16:44 <ehird> U l0st the g a e m
14:16:53 <AnMaster> ehird, I can't read that language. What one is it?
14:17:13 <AnMaster> Hm seems google translate can't handle AOL
14:17:14 <ehird> With your sarcastic acknowledgement of it you obviously were able to extract the basic meaning and thus I still succeeded.
14:17:34 <ehird> (Also, I'd advise you stop trying to be funny because you're really not very good at it. (Worse than I.))
14:17:48 <AnMaster> ehird, I can see French and guess it is French even though I can't understand it
14:18:04 <ehird> AnMaster: That was very easily decodable to "U lost the gaem" then "You lose the game".
14:18:19 <ehird> Of course, "Y o u l o s t t h e g a e m" would work too.
14:18:38 <AnMaster> well you could probably make a bayesian filter
14:29:01 <ehird> AnMaster: Bayesian filters aren't very good at anything that isn't spam filtering.
14:30:11 <AnMaster> ehird, well it is kind of spam ;P
14:30:31 <ehird> It's junk, you might think, but it's not spam.
14:30:53 <ehird> Either way, blocking "Y ou los t t he ga m e" without false positives is... difficult.
14:31:07 <AnMaster> ehird, sure, and also I don't care about that silly game.
14:31:35 <AnMaster> fizzie, or ais523: any of you there?
14:31:36 <ehird> <AnMaster> so just put a regex ignore on that stuff if you care about it. I certainly don't.
14:31:43 <ehird> AnMaster: Please learn to use /who.
14:31:51 <ehird> ais523 is set as away when he is not there, rather appropriately.
14:32:13 <ehird> fizzie has been idle for 3 hours, so you could nickping him and see.
14:32:38 <AnMaster> ehird, sometimes I idle a lot and I'm just doing other stuff but still notice a highlight on irc
14:32:56 <ehird> AnMaster: The point is...
14:33:03 <ehird> If fizzie was idle for 20 hours, he's away.
14:33:06 <ehird> If he's idle for 2 minutes, he's here.
14:33:08 <ehird> If it's 3 hours, ask.
14:33:38 <ehird> AnMaster: Not the point, you should still check before just uselessly nickpinging someone who possibly isn't even online to see it.
14:33:41 -!- oklocod has joined.
14:34:06 <AnMaster> ehird, and this bothers you so?
14:34:32 <ehird> Yes, because about half your lines is "ais523 there?" or "why isn't ais523 here".
14:37:12 -!- M0ny has joined.
14:38:00 <AnMaster> ehird, what is the total set for that half?
14:38:36 <AnMaster> "<ehird> Yes, because about half your lines is"
14:38:41 <AnMaster> you mean something else I assume?
14:38:53 <ehird> One. Message. After.
14:39:14 <ehird> "Yes, because about half your lines are "ais523 there?" or "why isn't ais523 here"."
14:39:43 <AnMaster> $ grep -c '<AnMaster>' /home/arvid/.xchat2/xchatlogs/FreeNode-#esoteric.log
14:39:44 <AnMaster> $ grep '<AnMaster>' /home/arvid/.xchat2/xchatlogs/FreeNode-#esoteric.log | grep ais523 | grep there | wc -l
14:40:03 <ehird> Ha ha ha ohh you take things literally!
14:40:05 <AnMaster> and most of those are unrelated
14:40:07 <AnMaster> okt 17 19:19:35 <AnMaster> ais523, I added the security considerations there
14:40:21 <AnMaster> ehird, I decided to do that yes
14:40:27 <AnMaster> since the statistics were incorrect
14:40:28 <ehird> When will you get it in to your head that English is more than flat text?
14:40:46 <AnMaster> maybe the day you start being funny?
14:40:56 <ehird> I'm not trying to be funny.
14:41:14 <ehird> And fuck this shit, I don't have time for a place that's just been consistently annoying ever since you showed up.
14:41:15 -!- ehird has left (?).
14:48:31 <AnMaster> well he will be 1) log reading 2) be back soon
14:55:36 -!- Mony has quit (Read error: 110 (Connection timed out)).
15:07:08 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | ->++>+++>+>+>+++>>>>>>>>>>>>>>>>>>>>+>+>++>+++>++>>+++>+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+>+>>+++>>+++>>>>>+++>+>>>>>>>>>++>+++>+++>+>>+++>>>+++>+>++>+++>>>+>+>++>+++>+>+>>+++>>>>>>>+>+>>>+>+>++>+++>+++>+>>+++>>>+++>+>++>+++>++>>+>+>++>+++>+>+>>+++>>>>>+++>+>>>>>++>+++>+++>+>>+++>>>+++>+>+++>+>>+++>>+++>>++[[>>+[>]++>++[<]<-]>+[>]<+<+++[<]<+]>+[>]++++>++[[<++++++++++++++++>-]<.
15:17:08 -!- GiveMeMony has joined.
15:28:58 -!- puzzlet_ has joined.
15:29:08 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)).
15:30:04 -!- jix has quit ("CommandQ").
15:30:22 -!- GiveMeMony has changed nick to Mony.
15:35:32 -!- M0ny has quit (Read error: 110 (Connection timed out)).
15:40:23 -!- Slereah has joined.
15:40:27 -!- Slereah_ has quit (Read error: 113 (No route to host)).
15:46:23 -!- Slereah_ has joined.
15:46:23 -!- Slereah has quit (Read error: 104 (Connection reset by peer)).
15:49:58 -!- pikhq has quit (Read error: 60 (Operation timed out)).
15:51:47 -!- pikhq has joined.
16:04:41 <AnMaster> ^bf ->++>+++>+>+>+++>>>>>>>>>>>>>>>>>>>>+>+>++>+++>++>>+++>+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+>+>>+++>>+++>>>>>+++>+>>>>>>>>>++>+++>+++>+>>+++>>>+++>+>++>+++>>>+>+>++>+++>+>+>>+++>>>>>>>+>+>>>+>+>++>+++>+++>+>>+++>>>+++>+>++>+++>++>>+>+>++>+++>+>+>>+++>>>>>+++>+>>>>>++>+++>+++>+>>+++>>>+++>+>+++>+>>+++>>+++>>++[[>>+[>]++>++[<]<-]>+[>]<+<+++[<]<+]>+[>]++++>++[[<++++++++++++++++>-]<
16:04:57 <AnMaster> so the thing in topic is cut off?
16:08:54 -!- oerjan has joined.
16:09:07 <oerjan> ^bf ->++>+++>+>+>+++>>>>>>>>>>>>>>>>>>>>+>+>++>+++>++>>+++>+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+>+>>+++>>+++>>>>>+++>+>>>>>>>>>++>+++>+++>+>>+++>>>+++>+>++>+++>>>+>+>++>+++>+>+>>+++>>>>>>>+>+>>>+>+>++>+++>+++>+>>+++>>>+++>+>++>+++>++>>+>+>++>+++>+>+>>+++>>>>>+++>+>>>>>++>+++>+++>+>>+++>>>+++>+>+++>+>>+++>>+++>>++[[>>+[>]++>++[<]<-]>+[>]<+<+++[<]<+]>+[>]++++>++[[<++++++++++++++++>-]<
16:10:55 <oerjan> ^bf ->++>+++>+>+>+++>>>>>>>>>>>>>>>>>>>>+>+>++>+++>++>>+++>+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+>+>>+++>>+++>>>>>+++>+>>>>>>>>>++>+++>+++>+>>+++>>>+++>+>++>+++>>>+>+>++>+++>+>+>>+++>>>>>>>+>+>>>+>+>++>+++>+++>+>>+++>>>+++>+>++>+++>++>>+>+>++>+++>+>+>>+++>>>>>+++>+>>>>>++>+++>+++>+>>+++>>>+++>+>+++>+>>+++>>+++>>++[[>>+[>]++>++[<]<-]>+[>]<+<+++[<]<+]>+[>]++++>++
16:11:15 <oerjan> oh no printing commands
16:12:05 <fungot> >,[>,]<[<]>[<+4[>-8<-]+>-[-7[-2[<+3[>-4<-]+>[<+4[>-5<-]+>[-11[-3[[-]<2[>[-]>+<2-]>>[<2+>>-]+<[->-<3[[>+<-]<]>>[>]]>[->[>]<[[>+<-]<]<2[[>+<-]<]<[[>+<-]<]>>[>]>[[[>]>+<2[<]>-]<2[[>+<-]<]>>[>]>[>]>[<2[<]<[<]<+>>[>]>[>]>-]<2[<]>]>>[[<+>-]>]<2[<]]]<[->>[>]<[[>>+<2-]<]<2[[>+<-]<]>+>[>]+5[>+8<-]+2>-[<+[<]>+[>]<-]]>]<[->>[[<2+>>-]>]<3[[>+<-]<]]>]<[-<[[<]>.[-]>[[<+>-]>]>>[[<+>-]>]<2[<]<2]>>>[[<+>-]>]<2[<]<]>]<[->>[>]<[[>+<-]<]<2[>>>>[>]
16:12:30 <Slereah_> ^bf +++++++++++++++++++++++++++++++++.
16:15:12 <oerjan> i mean there wouldn't be much point censoring colors for being annoying if you didn't censor beeps
16:16:29 <Slereah_> Does it censor anything from 000 to 031?
16:20:49 -!- pikhq_ has joined.
16:21:04 -!- pikhq has quit (Read error: 110 (Connection timed out)).
16:21:09 <oerjan> because that's needed for actions
16:21:26 <fungot> echo reverb rev bf rot13 hi rev2 fib wc ul ctcp
16:22:21 <oerjan> 000 may not be legal in irc at all, not sure
16:22:54 -!- Corun has joined.
16:23:04 <fungot> .. !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ...
16:23:33 <oerjan> the J and M fungot censors itself
16:23:33 <fungot> oerjan: but it's not clear to me where we are permitted to apply a given rule that would give you 0(1) access.
16:24:51 <oerjan> ^bf ++++++++++++++++++++++++++++++++>++[.<.>+]
16:24:52 <fungot> . . ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i ...
16:25:57 <oerjan> ^bf +++++++++++++++++++++++++++++++++>++[.<.>+]
16:25:58 <fungot> !!!!!!!!.!!!.!!!!!!!!!!!!!!!!!!! !!!"!#!$!%!&!'!(!)!*!+!,!-!.!/!0!1!2!3!4!5!6!7!8!9!:!;!<!=!>!?!@!A!B!C!D!E!F!G!H!I!J!K!L!M!N!O!P!Q!R!S!T!U!V!W!X!Y!Z![!\!]!^!_!`!a!b!c!d!e!f!g!h!i ...
16:27:44 <oerjan> hm interesting one ! is missing between ^B and ^H
16:27:55 <oerjan> ^bf +++++++++++++++++++++++++++++++++>++[.<.+>+]
16:27:55 <fungot> !"#$%&'(.)*+.,-./0123456789:;<=> ?!@"A#B$C%D&E'F(G)H*I+J,K-L.M/N0O1P2Q3R4S5T6U7V8W9X:Y;Z<[=\>]?^@_A`BaCbDcEdFeGfHgIhJiKjLkMlNmOnPoQpRqSrTsUtVuWvXwYxZy[z\{]|^}_~`abcdefghi ...
16:28:21 <oerjan> ^bf +++++++++++++++++++++++++++++++++++++++++++++++++>++[.<.+>+]
16:28:22 <fungot> 1345678.9:;.<=>?@ABCDEFGHIJKLMN O!P"Q#R$S%T&U'V(W)X*Y+Z,[-\.]/^0_1`2a3b4c5d6e7f8g9h:i;j<k=l>m?n@oApBqCrDsEtFuGvHwIxJyKzL{M|N}O~PQRSTUVWXYZ[\]^_`abcdefghi ...
16:29:36 <oerjan> clearly some control codes actually work, and swallow part of the other input
16:30:46 <oerjan> ^bf ++++++++++++++++++++++++++++++++++++++++++++++++>++[.<.+>+]
16:30:46 <fungot> 0234567.89:.;<=>?@ABCDEFGHIJKLM N!O"P#Q$R%S&T'U(V)W*X+Y,Z-[.\/]0^1_2`3a4b5c6d7e8f9g:h;i<j=k>l?m@nAoBpCqDrEsFtGuHvIwJxKyLzM{N|O}P~QRSTUVWXYZ[\]^_`abcdefghi ...
16:31:00 <oerjan> hey, there _are_ colors
16:31:15 <oerjan> ^bf +++++++++++++++++++++++++++++++++++++++++++++++>++[.<.+>+]
16:31:15 <fungot> /123456.789.:;<=>?@ABCDEFGHIJKL M!N"O#P$Q%R&S'T(U)V*W+X,Y-Z.[/\0]1^2_3`4a5b6c7d8e9f:g;h<i=j>k?l@mAnBoCpDqErFsGtHuIvJwKxLyMzN{O|P}Q~RSTUVWXYZ[\]^_`abcdefghi ...
16:32:10 -!- Slereah_ has quit (Read error: 113 (No route to host)).
16:33:54 <oerjan> ^bf +++.,[.]!67testing
16:33:54 <fungot> 666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666 ...
16:34:05 <oerjan> ^bf +++.,[.,]!67testing
16:34:30 <oerjan> ^bf ++++.,[.,]!67testing
16:39:09 -!- Corun_ has joined.
16:47:50 -!- Corun has quit (Read error: 110 (Connection timed out)).
17:04:50 -!- oklocod has changed nick to oklopol.
17:06:14 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf, code=text/str:N; ^str 0-9 get/set/add [text]
17:06:18 <fungot> echo reverb rev bf rot13 hi rev2 fib wc ul ctcp
17:06:47 <oerjan> it's an underload interpreter
17:07:27 <oerjan> still the brainfuck version i see
17:07:31 -!- Slereah_ has joined.
17:27:41 -!- Slereah has joined.
17:27:41 -!- Slereah_ has quit (Read error: 104 (Connection reset by peer)).
17:28:19 -!- SimonRC_ has changed nick to SimonC.
17:28:28 -!- SimonC has changed nick to SimonRC.
17:28:50 -!- Slereah_ has joined.
17:28:51 -!- Slereah has quit (Read error: 104 (Connection reset by peer)).
17:37:08 -!- Slereah has joined.
17:37:08 -!- Slereah_ has quit (Read error: 104 (Connection reset by peer)).
17:42:43 <Asztal> my silly shared web hosting doesn't let me use a www.www. subdomain because of their rubbish control panel :(
17:43:16 -!- Slereah has quit (Read error: 104 (Connection reset by peer)).
17:44:05 <GregorR> Asztal: I had to sort of trick cpanel to convince it I wanted www.www.
17:44:16 -!- Slereah has joined.
17:47:58 <Asztal> hmm, I could probably create w.w.w.w.w.w.domain >_>
17:50:01 <GregorR> SimonRC: http://www.www.extra-www.org/
17:50:17 <GregorR> Asztal: Is their stupid control panel cPanel?
17:52:39 <Asztal> any attempt to create hosting for a domain or subdomain starting with "www" is met with "we set this sub-domain up automatically for you." :(
17:55:01 <GregorR> "How DARE you refuse my www's! What are you, some WORLD-WIDE-WEB HATER?!"
17:57:59 -!- Slereah has quit (Read error: 113 (No route to host)).
17:58:39 -!- ehird has joined.
17:58:47 <ehird> tim berners-lee replied to my email :o
18:00:02 <SimonRC> ehird: about what,; saying what?
18:00:21 <ehird> SimonRC: dates in URI structure
18:00:34 <ehird> (as a question about "Cool URIs don't change")
18:01:02 <ehird> i was asking about "continuous" pages, in that e.g. a news page about foos still being updated today would be odd to be placed at /1998/foos-news
18:01:12 <ehird> he basically replied saying that it's okay :-P
18:01:41 <ehird> nothing interesting, then.
18:01:58 <GregorR> How about URI's like www.www.my-site.com/...
18:02:11 <ehird> In fact, here is a direct quote:
18:02:27 <ehird> ... why did you nickping Asztal for no reason.
18:02:43 <ehird> www.www.extra-www.org bitches
18:03:00 <GregorR> ehird: Asztal was trying to create a www.www. domain, but the server's control panel barfed :P
18:03:03 <ehird> GregorR: [[I would like to reccomend adding an extra "www." to the start of all your URIs. This should eliminate any date problems with your display of web patriotism. -Tim]]
18:03:05 <AnMaster> <ehird> ... why did you nickping Asztal for no reason.
18:03:21 <ehird> AnMaster: http://www.www.extra-www.org/.
18:03:32 <ehird> You can't "disagree" with a joke.
18:03:33 <GregorR> WELL THEN YOU HATE THE WEB.
18:03:46 <AnMaster> ehird, of course I know it is a joke
18:03:53 <ehird> ("Disagreeing" with a joke is commonly referred to as "having no sense of humour", which I'd say observation has proved to be quite true.)
18:04:12 <GregorR> ehird: That quote makes no sense :P
18:04:23 <ehird> GregorR: WELL IT'S WHAT HE SAID
18:04:42 <AnMaster> ehird, I think it is a good joke. I don't think anyone except the actual extra-www website should use it
18:04:54 <SimonRC> (Having read Weaving the Web, I would expect that he considers today's web to be Web 0.2, at best.)
18:04:59 <GregorR> ehird: But only due to my strong conviction for self-delusion do I believe you.
18:05:03 <ehird> SimonRC: the semweb stuff is neato.
18:05:36 <ehird> -- to reply to an earlier comment, AnMaster: Yes, because in your world humour is confined into rigid boxes marked "humour" and should not intefere with anything outside of the box.
18:05:54 <SimonRC> (Howwever, with all thi user-generated content, we are getting there, despite being rather short on the ""semantic" side.)
18:06:00 <GregorR> Humour sucks, humor is much funnier.
18:06:06 <ehird> <AnMaster> ehird, I think it is a good joke. I don't think anyone except the actual extra-www website should use it
18:06:06 <SimonRC> yuk, hate editing over a crap connection
18:06:18 <ehird> There is nothing wrong with a website letting you go to www.www.foo.com, as that is amusing in the vain of www.www.extra-www.org.
18:06:31 <AnMaster> ehird, basically I use no-www myself :P
18:06:32 <SimonRC> GregorR: no, homor sucks, and homours is funnoier
18:06:41 <ehird> There is nothing harmed by humour spilling slightly over from the box called humour into something else.
18:06:48 <ehird> If things were like that, then this'd be a pretty darn boring place, on account of not existing.
18:06:57 <AnMaster> due to the success of no-www you should create www.www.www.extra-extra-www.org
18:07:09 <ehird> I don't see how that actually adds to the humour.
18:07:22 <AnMaster> ehird, you lack humor then indeed.
18:07:25 <ehird> GregorR: I love how you use obscure TLDs for the links
18:07:41 <AnMaster> ehird, or rather: Humor is highly subjective
18:07:44 <ehird> AnMaster: No, I just have standards. (One element is that it has to be funny.)
18:07:59 <AnMaster> ehird, humor is always highly subjective
18:08:36 <AnMaster> but you can't assume any set of "this this is humor" is better than any other
18:08:38 <ehird> We'll have none of that smutty talk in #esoteric, please.
18:09:09 <oerjan> whatever you do, don't spill hummus
18:09:29 <ehird> oerjan: That's going to be hard to avoid when sharing parts.
18:09:42 <ehird> Yeah, but you bet they'll have a cover story. "I was just being modular!"
18:10:13 <oerjan> AnMaster: we should demand a www top level domain
18:10:37 <ehird> www.www.www.www.www
18:10:51 <ehird> (the www.www.www. subdomain of the www.www domain)
18:10:58 <AnMaster> oerjan, but ICANN will ask a lot of money sadly :/
18:11:01 <ehird> It's extra-www compliant, ON BOTH SIDES.
18:11:42 <SimonRC> bah, domains are backwards anyway
18:11:54 <oerjan> SimonRC: you silly UK people
18:12:13 <SimonRC> they should hav been big-endian, like oterh types of paths
18:12:44 <oerjan> clearly europeans and americans should use big-endian paths, while japanese should use little-endian ones
18:12:51 <SimonRC> and email addresses should be big-endian too
18:12:59 <AnMaster> oerjan, Americans should use mixed-endian
18:13:28 <AnMaster> oerjan, just look at the American date format.
18:13:40 <ehird> native americans should use little-endian
18:13:44 <ehird> http://instantrimshot.com/
18:13:45 <AnMaster> so it should be. google.www.com
18:13:53 <ehird> http://com.instantrimshot/
18:14:31 <GregorR> European date format isn't mixed-endian, but it's still backwards.
18:14:46 <AnMaster> "The instantrimshot of the com"
18:14:57 <ehird> uris are pretty much perfect
18:15:01 <ehird> the fact that it's foobar
18:15:09 <ehird> is more important to a person than the fact that it's a company called foobar
18:15:17 <oerjan> SimonRC: also, chinese and indonesians should use single token addresses
18:15:17 <ehird> or a nonprofit called foobar
18:18:24 <psygnisfive> oerjan, why did you mention linguistics? :D
18:19:20 <oerjan> psygnisfive: <oerjan> clearly europeans and americans should use big-endian paths, while japanese should use little-endian ones
18:19:39 <psygnisfive> i dont follow why this should be the case tho
18:20:00 <psygnisfive> but the word orders aren't exactly mirrors of one another
18:20:08 <oerjan> it was a joke, btw, so if you disagree with it i'll sic ehird on you :D
18:20:23 <psygnisfive> Japanese is SOV while european languages are SVO for the most part
18:20:35 <oerjan> (see recent discussion for that too)
18:20:38 <ehird> psygnisfive's rigid joke standards strike again
18:21:00 <psygnisfive> and spanish and italian can occasionally get OVS, which IS the exact opposite of english.
18:21:04 <oerjan> psygnisfive: i think you are now officially without a sense of humor
18:21:26 <oerjan> well so can other germanic languages
18:21:53 <psygnisfive> so they can get any word order with the verb second
18:22:33 <psygnisfive> german can only do it in main clauses, interestingly, while yiddish and danish, i believe, can do it in subordinate clauses
18:22:49 -!- kar8nga has quit (Read error: 110 (Connection timed out)).
18:23:38 <oerjan> psygnisfive: also i was not considering the entire word order, only general mostly head-final vs. mostly head-initial
18:24:07 <psygnisfive> well japanese is indeed uniformly head-final, except with the case of demonstratives, which might tell us something about demonstratives
18:24:22 <psygnisfive> but english and other SVO languages tend not to be head initial, even mostly.
18:24:56 <psygnisfive> in a technical sense, the difference is that japanese is almost uniformly Specifier-Complement-Head word order, with English being Specifier-Head-Complement
18:25:11 <psygnisfive> so i guess with regards to Head-Complement ordering you're correct
18:25:30 <oerjan> hm what about romance languages?
18:25:48 <oerjan> they frequently put adjectives last
18:26:10 <psygnisfive> and adjunct ordering tends to be fairly arbitrary
18:26:25 <psygnisfive> the trend, infact, is that languages with V-O word order tend to also have N-Adj word order
18:26:40 <psygnisfive> which makes germanic languages rather unusual in that regard
18:27:03 <oerjan> what's a complement then? prepositional clauses and the like?
18:27:30 <psygnisfive> basically, the idea of Specs and Complements in minimalism is like so:
18:28:31 <psygnisfive> you have some operation called merge that takes two things and sticks them together in a bundle
18:28:32 <oerjan> i think old norse put adjectives last somewhat more frequently
18:28:35 <Asztal> Hungarian has free word order too... Quite a source of confusion (initially, anyway) combined with dropping pronouns and the 3rd person "is/are". :(
18:28:44 <psygnisfive> and this bundle has some distinguished member called the Head
18:29:01 <psygnisfive> the first member merged with a head is called the complement
18:29:25 <psygnisfive> so you have something like this: [Spec [Head Complement]]
18:30:17 <psygnisfive> its essentially a theory-internal matter. complements are sisters-to-heads, specifiers are daughters-to-the-final-phrase
18:32:14 <psygnisfive> i know its kind of confusing. like i said, its kind of theory internal
18:33:23 <psygnisfive> heads seem to be able to put restrictions on their complements, but not on their specifiers.
18:35:12 <oerjan> can you give an example of a nominal phrase with both spec and complement?
18:35:35 <psygnisfive> nouns generally dont take complements, but sure
18:35:56 <oerjan> ah that explains why i felt it didn't fit
18:36:00 <psygnisfive> well.. nominal phrases also don't seem to be headed by a noun.. :p
18:36:27 <psygnisfive> yeah, it seems weird, but its apparently correct
18:36:27 <oerjan> this theory is getting less intuitive all the time
18:36:46 <psygnisfive> yes, well, intuition never did make sense in science :)
18:36:53 <psygnisfive> but i can give you a full nominal phrase that has multiple specs and complements
18:37:06 <psygnisfive> first i suggest you open this page: http://ironcreek.net/phpsyntaxtree/?
18:37:14 <psygnisfive> and draw this tree: [XP Spec [X' [X] [Comp]]]
18:37:32 <psygnisfive> since that'll give you a nice clear visual template for what specs and comps are
18:39:40 <psygnisfive> [DP [DP John] [D' [D 's] [NP [N student] [PP [P of] [DP linguistics]]]]]
18:40:21 <psygnisfive> the thing labeled [D 's] is the head of the main DP, as you can see by the projections of D above it: D->D'->DP
18:40:52 <psygnisfive> [DP John], to the left of D', is the specifier of [D 's] because it's the last thing to join up before the main DP
18:41:19 <psygnisfive> [NP student of linguistics] is the complement to [D 's], because its the first item that joins up with [D 's]
18:41:32 <oerjan> are you saying "'s" is the head of that phrase? O_O
18:41:53 <psygnisfive> but dont think of heads as being what the phrase is about
18:42:09 <psygnisfive> think of heads as more like functions in function applications
18:42:20 <psygnisfive> and infact, when we say it like that, it sounds completely correct right?
18:42:39 <psygnisfive> Function-Applications are like Determiner-Phrases
18:42:47 <psygnisfive> the Function-Application is about the function, is it?
18:42:51 <oerjan> oh like 's is an operator
18:42:59 <psygnisfive> no. and a Determiner-Phrase is not about the determiner
18:43:34 <psygnisfive> if anything, a Noun Phrase isn't about a noun, either
18:43:43 <psygnisfive> the whole nominal phrase is about something in the WORLD
18:43:47 <psygnisfive> like.. a person, or a substance, or an event
18:44:01 <psygnisfive> the words themselves are just things that describe that referent, even nouns.
18:44:27 <psygnisfive> [PP of lx] is the complement to [N student] because its the first thing that joins with [N student]
18:44:46 <oerjan> it _does_ seem weird to consider the complement to be the _first_ thing joined though, since "'s student of linguistics" does not have meaning alone
18:44:50 <psygnisfive> and [DP linguistics] is the complement to [P of] for similar reasons
18:45:06 <psygnisfive> oerjan: we're thinking in bottom up building of the sentence
18:45:17 <psygnisfive> but yes, you're right that ['s ...] has no meaning by itself
18:47:54 <oerjan> otoh why isn't "of" the head of "student of linguistics"?
18:48:18 <psygnisfive> because of is a preposition and so heads a preposition phrase
18:48:24 <psygnisfive> "student of linguistics" isnt a preposition phrase
18:48:33 <psygnisfive> further more, you don't /need/ the preposition phrase
18:48:50 <psygnisfive> but thats getting into more weird shit that complicates things
18:48:54 <oerjan> but 's tails a possessive phrase, so they seem analogous just reversed
18:49:05 <psygnisfive> sure, but "John's" /is/ completely acceptable.
18:49:07 <oerjan> (making up terms here)
18:49:23 <psygnisfive> you dont need the noun-phrase that follows.
18:50:19 <psygnisfive> but like i said, now we're getting into more complicated shit
18:50:53 <oerjan> ok i think that's about enough
18:57:25 <fizzie> Hah-hah, something like 24 hours of computation, resulting in: "terminate called after throwing an instance of 'char const*'"
18:57:48 <psygnisfive> save state up to that point, go in and fix the code
18:58:18 <ehird> fizzie: run it on a cluster next time :P
18:59:08 <fizzie> Yes, and I think I'll just use a bit less of them input.
19:00:34 <fizzie> The problem is probably because I use four-"byte" integers which use 7-bit bytes (because it's unspecified whether FILE's "read characters" instruction does sign-extension or not for values >= 128), so there's a limit of 2^(4*7) = 2^28 = ~270 million for some things like file offsets.
19:00:47 <fizzie> The IRC-built language model file is already 200 megs, this one was probably bit too much.
19:02:31 <fizzie> Maybe I'll just use something like a thousand ten-minute conversations instead of six thousand.
19:03:34 <oerjan> for those automated conversation bots people: http://www.mezzacotta.net/
19:03:59 <psygnisfive> everyone thinks the loebner prize is a real turing test
19:04:05 <oerjan> i was thinking of fungot and optbot here...
19:04:05 <optbot> oerjan: how goes cfunge stuff?
19:04:05 <fungot> oerjan: later tell jonnay riastradh also, in the transformation process.)
19:05:54 <oerjan> psygnisfive: well that link is even more half-baked, as you might guess
19:07:42 <oerjan> but there's an impressive archive ;D
19:11:54 -!- cherez has joined.
19:12:03 -!- cherez has left (?).
19:17:28 <psygnisfive> someone said something about hungarian free word order
19:17:42 <oerjan> topic before main verb, iirc
19:18:06 <psygnisfive> WH phrases in questions front, obligatorially
19:18:28 <psygnisfive> and adpositions have well defined positions relative to their modifier phrases
19:18:47 <psygnisfive> i'd also bet that relative clauses and adjectives and such have to be in specific places too.
19:19:05 <psygnisfive> and i'd also wager that the free word order isn't free at all, but rather, pragmatically determined.
19:19:17 <oerjan> i do recall adjectives being in front of nouns
19:20:58 -!- KingOfKarlsruhe has joined.
19:22:21 <fizzie> Okay, fungot, try speaking like you're having a telephone conversation.
19:22:21 <fungot> fizzie: ( ( noise noise uh-hum))
19:22:38 <fizzie> fungot: Okay, that's not quite what I had in mind...
19:22:38 <fungot> fizzie: on my earphones and ah that's
19:22:41 <oerjan> that's some really bad phone
19:24:14 <fizzie> fungot: Sounds like someone interrupted you.
19:24:14 <fungot> fizzie: they're doing a survey on how people feel about it laughter
19:24:41 <fizzie> There's a lot of things like "[laughter]" in those transcriptions, and fungot just discards []s.
19:24:58 <fizzie> I don't remember what the double-parentheses notation meant.
19:25:56 <fizzie> That's still not as good as the IRC logs, meh.
19:26:25 <Deewiant> fungot: do you feel smarter now?
19:26:25 <fungot> Deewiant: ah that's a hobby
19:26:52 <Deewiant> fungot: well, I'm glad you're making the attempt, at least
19:26:52 <fungot> Deewiant: you know unless ah one it seems like
19:27:29 <Deewiant> fungot: ... even though you're not really succeeding.
19:27:29 <fungot> Deewiant: i have a four year old twins
19:27:52 -!- Sgeo has joined.
19:28:16 <Deewiant> fungot: well, it can't be yours, you're not that old yourself
19:28:16 <fungot> Deewiant: this so that's how i learned so
19:28:41 <fizzie> fungot: Are you telling me you wrote two other befunge bots, or what?
19:28:41 <fungot> fizzie: topic is so mind boggling and i i really
19:29:46 <Deewiant> I think fungot's finally losing it
19:29:47 <fungot> Deewiant: yeah yeah true)) don't think that's
19:30:35 <oerjan> Deewiant: well with four year old twins life can be quite stressful
19:30:48 <Deewiant> oerjan: especially when you're not that old yourself
19:31:21 <oerjan> depends how you reckon. how many human years in a bot year?
19:32:22 <Deewiant> well, I figure the twins must be bots too
19:34:16 <fizzie> Yes, but fungot might well be a old, old fart in bot years.
19:34:16 <fungot> fizzie: i mean they my kids came up i said okay this is my
19:34:44 <oerjan> fungot: perhaps you've got botheimer's?
19:34:44 <fungot> oerjan: right and shaping the minds of the future
19:35:02 <fungot> oerjan: you try to make um there's a lot
19:37:00 <fizzie> fungot: A lot of what?
19:37:00 <fungot> fizzie: ( ( so uh it's one of those
19:38:16 <fizzie> Can't say this telephone conversation experiment has been very successful, though we did learn that fungot has kids.
19:38:16 <fungot> fizzie: and it's a great tool
19:57:20 <Asztal> fungot: how did you trick them into letting you adopt?
19:57:20 <fungot> Asztal: ( ( lipsmack well what a wonderful thing
19:58:07 <oerjan> i don't see where it was mentioned they were adopted
20:02:52 <Asztal> <fungot> Deewiant: i have a four year old twins
20:02:52 <fungot> Asztal: hi how you doing
20:02:53 <Asztal> <Deewiant> fungot: well, it can't be yours, you're not that old yourself
20:02:54 <fungot> Asztal: where are from actually i
20:03:28 <Asztal> fungot: I'm from England. Do continue...
20:03:28 <fungot> Asztal: and do you see what i'm saying
20:04:09 <fizzie> They could be simply stolen instead of adopted. Have you been participating in an illegal baby-smuggling operation, fungot? (As opposed to all the *legal* baby-smuggling, I guess.)
20:04:09 <fungot> fizzie: you know if they
20:07:46 <Asztal> fungot seems to have a problem with not finishing its
20:07:46 <fungot> Asztal: ( ( yeah so am i)) don't think that's right
20:29:09 <AnMaster> fizzie, what word list is that?
20:30:40 <fizzie> It's that Fisher "conversational telephone speech" corpus I was talking about.
20:30:48 <fizzie> Except that it's only one sixth of it, due to problems.
20:52:21 -!- Corun_ has quit ("This computer has gone to sleep").
20:56:52 <fungot> ehird: in a theater or watching a movie
20:56:55 <fungot> ehird: are you which part
20:56:58 <fungot> ehird: that's what she does and she emails but
20:57:06 <ehird> fizzie: make it feed the input sentence into the markov chain
20:57:18 <ehird> make the phrase put in the first word
20:57:22 <ehird> then cut off the first word at the end
20:57:33 <ehird> "the scheme programming language" "... is awesome"
20:59:59 <oerjan> i suddenly envision loading fungot with a telemarketing corpus for that
20:59:59 <fungot> oerjan: yeah and understandably because of our policies but
21:03:56 <fizzie> ehird: I'd need to convert text to tokens in the Funge code for that.
21:04:10 <ehird> fizzie: Just shell out to the c++ :P
21:07:08 -!- optbot has set topic: the entire backlog of #esoteric: http://tunes.org/~nef/logs/esoteric | There is none..
21:07:36 <fungot> AnMaster: yeah i would
21:07:41 <fungot> AnMaster: ( ( you know um certainly in the younger grades i think they
21:07:49 <AnMaster> fizzie, what's with those (( ?
21:08:06 <AnMaster> fungot, you seem to "um" a lot?
21:08:06 <fungot> AnMaster: yeah i used to
21:08:16 <fungot> AnMaster: ( ( noise noise noise noise noise
21:09:26 <AnMaster> ehird, you heard what optbot said
21:09:26 <optbot> AnMaster: But some clients freak out when they recieve notices.
21:09:42 <ehird> optbot: You make more sense than fungot.
21:09:43 <fungot> ehird: most of us even want to be at that before we were even thought of
21:09:43 <optbot> ehird: Its not one step. He uses a for loop.
21:09:43 <AnMaster> optbot, what? Sounds like bad clients to me
21:09:44 <optbot> AnMaster: "what is do im half\nmarkov"
21:09:56 <AnMaster> ehird, it would do... Since it uses plain quotes
21:12:09 -!- atrapado has joined.
21:13:34 <oerjan> they're like duals, order and chaos
21:14:08 <ehird> optbot: be confusing
21:14:08 <optbot> ehird: The language with only one instruction.
21:14:12 <ehird> optbot: be confusing some more
21:14:18 <ehird> oerjan: not really.
21:25:51 <optbot> AnMaster: By the way, if you understood that....
21:26:04 <optbot> AnMaster: how does c++ do abstract classes
21:26:57 <AnMaster> optbot, Using some sort of pointer table for virtual functions in general I believe, so I guess the same way... But I'm hardly an expert on that.
21:27:16 <AnMaster> optbot, You certainly like changing topic a lot...
21:27:22 <optbot> AnMaster: I actually need to upload my latest plof2js
21:39:24 <fizzie> AnMaster: It's some sort of notation in the transcriptions.
21:40:04 <fizzie> I'm not sure where they put the documentation (as opposed to the raw data) of the thing, so I can't check what it meant.
21:41:40 <AnMaster> fizzie, does the license allow this?
21:42:13 <AnMaster> fizzie, also you may be interesting to know that ATHR is progressing nicely, still a lot to do to get it done, but I already managed several points on the road map
21:42:30 <AnMaster> (combined roadmap/todo/design/ideas)
21:43:04 <fizzie> I think it allows just about anything we care to use it for, as long as we don't distribute the stuff. A couple of messed-up quotations is not really copying; OTOH IANAL, so I probably won't leave it up.
21:43:25 <AnMaster> fizzie, http://rafb.net/p/GMoXXJ83.html
21:43:32 <fizzie> "Portions of speech where the transcriber could not be sure exactly what was said were marked with double parentheses -- " (( ... )) " -- and the transcriber could hazard a guess as to what was said, or leave the region between parentheses blank."
21:44:20 <fizzie> Anywhere where's there ((s.
21:44:31 <AnMaster> fizzie, are there a lot of those?
21:44:41 <fizzie> All kinds of "uh mm" type utterances.
21:45:10 <fizzie> The noise one was probably (( [noise] )) or something, fungot just throws the []s out.
21:45:10 <fungot> fizzie: so has your life changed much since september eleventh
21:45:40 <fizzie> fungot: Not to sound insensitive, but not really.
21:45:40 <fungot> fizzie: some car with that
21:46:29 <fizzie> There are 40 topics they had people discuss, let's see what kind of selection the bot got.
21:48:47 <fizzie> Quite wide, actually; topics from 12 to 29 occur in those files I used.
21:55:28 <ehird> fungot: 9/11 was an inside job WAKE UP SHEEPLE
21:55:28 <fungot> ehird: what are we going to reach the top i just drew a straight line
21:55:35 <ehird> fungot: ... you did 9/11?
21:55:35 <fungot> ehird: in other words i suppose
21:55:40 <fungot> ehird: oh really that would be
21:55:48 <ehird> fungot: no that's not ok man, not cool >:(
21:55:48 <fungot> ehird: but er i'm in a
21:55:55 <ehird> fungot: DON'T DODGE THE ISSUE
21:56:14 <ehird> fizzie: I broke it.
21:56:29 <fizzie> It's just the ignore mode, I think.
21:57:26 <fizzie> fungot: Considering that you were "born" not too long ago, how come you're responsible for the 9/11 thing? You used TRDS to time travel or something?
21:57:27 <fungot> fizzie: yeah you find less opportunities to uh
21:58:02 <ehird> fungot: bot babies
21:58:02 <fungot> ehird: oh i had mine on my shoulder ' cause um
21:58:11 <ehird> fungot: be careful with them, you might drop em
21:58:18 <ehird> fungot: DON'T DODGE ISSUES
21:58:22 <fungot> ehird: but you know you could do
21:58:58 <fizzie> fungot's so contemptuous.
21:58:59 <fungot> fizzie: so um yeah and then the second one was a good thing to have
22:09:03 <fizzie> There's not really any punctuation in the transcripts, so fungot doesn't do it either.
22:09:03 <fungot> fizzie: yeah um it's cough
22:09:32 <fizzie> It's all like "my dad has thirteen brothers and sisters you know family can be all different kinds of things it can be friends it can be people you work with if you're close i guess it's all the"
22:13:32 -!- atrapado has quit ("Abandonando").
23:20:23 -!- Mony has quit ("Ne regrettons pas le passé, ne nous préoccupons pas de l'avenir : le sage vit dans le présent.").
23:36:54 -!- oerjan has quit ("Good night").