00:00:25 elliott: Do you mean, like, Unlambda or Lazy-K, for example? 00:00:41 well... for sufficiently wrong definitions of "longer" 00:02:08 elliott: i have a hunch that might somehow violate a theorem, similar to http://en.wikipedia.org/wiki/G%C3%B6del%27s_speed-up_theorem 00:02:28 for sufficiently weird definitions of violate 00:02:51 oerjan: well um of course it's impossible 00:02:55 without an infinite alphabet at least 00:03:01 the famous mathematician G%C3%B6del 00:03:26 kmc: he encrypted his name to keep the persecutors away 00:03:28 "This statement cannot be proved in Peano arithmetic in less than a googolplex symbols" goedel really liked antagonising PA 00:04:43 oerjan: lmao at "List of long proofs" in the see also 00:04:55 i somehow doubt they're googolplex-length 00:06:00 _probably_ not 00:06:16 as even an automated prover couldn't check them 00:07:44 They could be in a less powerful system 00:09:53 http://zzo38computer.cjb.net/textfile/miscellaneous/impossible_physics_list 00:10:43 -!- cheater has joined. 00:10:55 -!- primo has joined. 00:13:32 do most web browsers support HTML% yet? 00:13:34 *5 00:13:54 what is most 00:13:58 hmm, there's a website called caniuse or something 00:14:00 the ones that the people use. 00:14:01 that tracks the support 00:14:49 so HTML5 Audio is supported in the newest versions of everything except Opera Mini 00:15:00 `welcome primo 00:15:11 browsers in risk of not having it include... IE. awesome. 00:15:11 audio why do you want audio audio is awful dont do audio 00:15:12 primo: 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 00:15:12 kallisti: I'd say that's a feature though 00:15:21 monqy: go on 00:15:22 -!- oerjan has set topic: Hexhammers chase scottish clan back to Edinburgh with proton packs | http://codu.org/logs/_esoteric/. 00:15:28 elliott: thanks 00:15:49 but of course, even Mini could open audio through whatever platform audio handler(s) there are 00:15:53 do you want audio why do you want audio dont 00:15:55 theoretically anyway 00:16:23 I suggest not using any fancy stuff unless needed for your webpage for sure; don't use video and audio and JavaScripts and CSS and pictures for everything. 00:16:39 I'd say implementing that would be a horrible thing, and I will obviously prevent it 00:17:51 zzo38: it's difficult to implement a simple audio player without playing audio. 00:18:28 noisy webpages are why bad things happen 00:18:29 kallisti: Yes; if you need to play audio then of course you need to play audio. But if you want to implement a simple audio player you probaly don't need a webpage anyways 00:21:52 -!- augur_ has changed nick to augur. 00:22:34 -!- ais523 has quit (Remote host closed the connection). 00:24:43 -!- kallisti_ has joined. 00:24:54 hi 00:24:59 X suddenly froze 00:25:01 what do? 00:25:06 is there a way to fix this without restart? 00:26:15 I'm just going to restart... 00:26:38 oh... 00:26:49 -!- kallisti_ has quit (Client Quit). 00:26:57 apparently 00:27:00 right clicking was the solution 00:27:46 I think gftp is just bad and I should never use it again. 00:28:09 so based on the last 9 lines, #esoteric is now kallisti's channel for talking to his/her self 00:28:23 yes. 00:29:19 A few lines are not sufficient statistics. 00:30:05 zzo38: hush, you're breaking the statistics 00:32:35 WHY ARE ALL OF THESE FTP CLIENTS SO BAD 00:32:59 kallisti: What happen if you just type in "ftp" at the prompt? 00:33:01 kallisti: ftp(1) 00:33:41 I'll learn that later. 00:34:05 im sure learning a new crappy inconsistent gui app will be about 10x eaiser 00:34:06 easier 00:34:40 well for doing a few simple things when under a slight time pressure. 00:34:43 it will be fine 00:34:45 and then I can go learn ftp 00:43:03 -!- Klisz has joined. 01:03:32 monqy: hi 01:04:29 i got a new irc client Pidgin aparently xchat is not free after 30 days 01:04:40 hahaha 01:05:17 So, US courts think that having a .com or .net hostname suffices in order for them to have jurisdiction over you. 01:05:25 hmm, i need to find some way to convince elliottt to change their legal name 01:05:47 the US has 'jurisdiction' over the entire world anyway 01:05:48 Apparently nobody in the legal system even knows a damned thing about anything. 01:05:48 or else, try and be really active in #haskell about twenty minutes ago 01:05:59 minimise or maximise confusion, the choice is mine 01:06:02 you're lucky if the courts claim jurisdiction 01:06:14 otherwise the executive branch can unilaterally blow up your house 01:06:30 Y'mean it's a matter of "most money wins" rather than "we win" then? 01:06:38 hey gusy i hear ameriKa suxe! 01:06:48 *ameriKKKa 01:07:04 I need to move to the fucking Moon. 01:08:20 jesus h christ in a chicken basket, houston 01:08:24 we're on the fucking moon, over 01:09:17 pikhq: you can't escape on the moon!!! president george w "satan" bush will just declare it as ame"corporatism"riKKKan territory 01:09:28 oh man, ame"corporatism"riKKKan is a really good name 01:09:30 Hey, if I move to the moon, I estimate it'd be at least 50 years before the US could enforce anything there. 01:09:34 that was completely accidental, i've created a masterpiece 01:09:54 Given that the current US space program progress is *negative*, it could actually be forever. 01:10:07 -!- elliott has set topic: united irc of hexham & helsinki of the UNUNITED states of ame"corporatism"riKKKa | http://codu.org/logs/_esoteric/. 01:10:45 pikhq: are you suggesting that we are somehow absorbing spaceflight endeavours from space 01:11:08 elliott: No. If we were we'd probably have fewer energy issues. :P 01:11:32 (I presume if you can absorb spaceflight endeavours, you can then extract electricity from it) 01:11:45 "I call it: the Science-Consuming Generator." 01:12:19 It sucks up future science and makes energy, thus killing two birds with one stone. 01:12:22 :P 01:13:32 someone tell me to go to bed 01:13:46 elliott: Go to bed. 01:14:02 ベッドへ行け！ 01:16:12 oh well, g'night i guess 01:16:13 As far as I can tell, the moon never goes directly above the United States. Its maximum declination appears to be 22 01:17:39 zzo38: The US also has about as much of a valid claim to the moon as it does Alpha Centauri. 01:17:40 nor does the sun, for that matter 01:17:50 i.e. "Hahahaha you can't even get there!" 01:18:57 primo: True; the sun doesn't either. The limits of the sun declination are the tropics of Cancer and Capricorn 01:20:00 Alpha Centauri is much farther away, but I don't know its declination. 01:20:25 -!- elliott has quit (Ping timeout: 240 seconds). 01:20:39 itwouldn't take long to find out, if one were interested in knowing 01:21:53 and that it's something that baffles me as well 01:22:38 we should, by all logic, be the most intelligent humans to have ever lived. anything we want to know, we can know in a matter of seconds 01:22:59 yet, that doesn't seem to be the case at all 01:26:51 I have the declinations for 47 stars, but not Alpha Centauri 01:28:30 I have the right ascension as well. They are listed in the order according to their right ascension. 01:28:45 I can also calculate ecliptic coordinates if wanted. 01:29:28 Declination -60° 50′ 02.308″ says wikipedia 01:29:35 OK 01:29:40 well, that's for A 01:30:10 -60° 50′ 13.761″ for Alpha Centauri B 01:30:49 OK. I didn't know there are two 01:31:17 −62° 40′ 46.141″ for Proxima Centauri 01:36:32 The list I have is not completely by order of right ascension; it is approximate. Andromeda is past zero now; in 1202 it was before zero. And sometimes they move past each other. In 1202 some were in a different order than listed, but now they are in the exact order of right ascension, even though Andromeda is past zero. 01:43:27 kmc: The official XChat for Windows is indeed not free after 30 days. 01:43:39 that's fucked 01:44:06 Windows more like WinKKKbl0$E 01:45:48 Are constellations supposed to be numbered according to alphabetical order of their Latin names? 01:52:01 zzo38: http://www.iau.org/public/constellations/ has no numbers that i can see 01:52:47 * oerjan notes Aquarius (Aqr) is before Aquila (Aql), so it's not sorted by abbreviation 01:53:53 Astrolog includes a list of constellations (Alt+5) in alphabetical order by Latin name (including Aquarius before Aquila); but they are also numbered 1 to 88. 01:55:29 well i imagine you cannot get more authoritative than the IAU on this 01:55:42 Yes, the IAU is authoritative for such things 01:55:56 kallisti, there was an update earlier 01:55:57 But this list is same as the IAU, except that it is numbered 01:58:24 O, the IAU even has constellation boundary files, so I suppose it can be used to make up a computer program that displays them. 01:58:45 -!- roper has joined. 01:59:57 they say one got online every day to remedy affairs caused other hours from her nick 02:01:21 roper: I do not understand without a context? 02:01:40 account hijacking 02:02:13 bot harassment 02:03:02 they are quite talkative 02:03:43 i think they use your history from other conversations 02:03:47 oh wow mozilla doesn't support mp3 in audio elements. 02:03:54 pretty ballsy. 02:04:05 kallisti: Use Vorbis or FLAC. MP3 is no good anyways 02:04:09 zzo38: yes I am 02:04:12 mp3 is not free 02:04:22 yes that's why Mozilla doesn't support its use in audio elements. 02:04:23 roper: Yes, that is one reason it is no good 02:04:25 patented 02:05:34 For videos, use Theora. 02:05:38 worse than secret 02:07:20 question for anyone 02:07:49 if an iterpretter for a given esolang doesn't work at all, can i just overwrite it with one that does? 02:08:17 questionable 02:08:33 i'm looking at ferNANDo 02:08:43 primo: You could do so, yes; if they don't like then they can still view the history. But first ensure that you are correct about it not working at all, because maybe it does work on their computer. 02:08:50 orLANDO 02:08:50 the haskell implementation is correct. python not at all 02:10:02 Check what things were changed by the anonymous modificationsby 164.67.235.79 and 131.179.32.131 02:10:28 Also look at talk page 02:10:31 zzo38: can't possibly work on any machine. A = B nand C was originally implemented as A = A nand B 02:10:49 and the talk page implemenation, is even worse o.O 02:11:14 Well, then, please do fix it. 02:11:33 I did not look carefully at them or try to run them on my computer 02:11:49 output 2**(9-i) should be 2**(7-i) (or, <<7-i) 02:11:57 heh 02:12:16 and the 'fix' to find the last line number is entirely broken 02:12:20 primo: well if you think it's so bad it's not worth basing a corrected one on it 02:12:28 because list.reverse() is in-place, it returns None 02:13:03 * oerjan wrote the haskell one btw 02:13:43 may ever be a limit for the most inefficient formulation? 02:14:04 that does the calculation 02:15:10 roper: you can always put in extra nop's... this is a technical point used to show that O(f) is contained in O(g) if f <= g 02:15:26 hm wait 02:15:27 nop is nop 02:15:29 you don't need that 02:16:13 but you can pad in time and space for technical reasons 02:16:31 I would have written it entirely differently but your way is just as good. But I generally write Haskell programs seem to be differently than most, anyways. A literate Haskell program can be implemented in MediaWiki; just make the codes with > and with 
 around it! And then, it does have the URL to download the raw copy, so you can use that to download .lhs file.
02:18:07  "Boy without a cerebellum baffles doctors" - I wonder if this is a documentary about guido
02:18:18  how  would you make worse a = sum (range(0,n)) ?
02:18:34  worst complexity class
02:18:42  it seems as though if I look away from skype for too long it will be frozen when I come back.
02:19:23  roper: sum(x) = sleep()
02:19:32  nope
02:19:44  not valid
02:19:47  roper: sum(x) = sum(x)
02:19:58  same complexity
02:20:09  I don't understand the question then
02:20:29  Instead of sum(range...) you could convert it to single algebra formula
02:20:32  well O(n)
02:20:37  ahh yes, it seems 164.67.235.79 broke it
02:21:07  zzo38 that would not change complexity
02:22:03  You could write   n*(n+1)/2   instead
02:22:43  zzo38 wrong, that is O(1)
02:22:53  Yes, it is O(1)
02:23:12  i mean worst complexity
02:23:35  I don't know how to change it to worst complexity.
02:23:47  neither i
02:23:52  curious
02:24:11  you can optimize but not worsenize
02:24:32  what is the "worst complexity"?
02:24:42  roper: you can easily worsenize. see bogosort for an example in sorting
02:25:13  I would say there's no end to the amount of useless operations you can perform.
02:25:27  worst complexity is... at least O(n^2)
02:25:28  if you count it, O(infnity) is the "worst possible case"
02:25:39  though big O notation is usually defined over reals.
02:25:49  It is possible to write a program in the "most pessimum" if necessary, such as what someone did when writing a program with rotating memory, they moved the instructions around to slow it down, instead of using delay loops.
02:26:20  i mean complexity class, not number of operations
02:26:31  here's a worse way to do that sum: replace the "a=" with "point a to a location in memory that already has the right value. if it does not exist, write random bits memory and repeat"
02:26:43  zzo38: hm was that the Mel story?
02:26:54  <3 mel
02:27:15  quintopia wrong
02:27:33  roper is never satisfied.
02:27:45  roper: you are wrong
02:27:52  i am over satisfied
02:28:01  roper: the problem here is that it takes a sentient being to determine whether the worsening is allowed or not.
02:28:08  no it is still o n
02:28:19  uh, no?
02:28:54  worse than x*N OPERATIONS?
02:29:01  worst case of quintopia's algorithm is O(infinity)  aka it continues to the end of time.
02:29:19  roper: it is not O(n). it is expected to take O(n*x*1/p)
02:29:27  that is not an alforithm, kallinsti
02:29:43  where x is the size of memory and p is the probability of the right number occurring in it
02:29:57  mmm
02:29:58  roper: http://en.wikipedia.org/wiki/Las_Vegas_algorithm
02:31:02  http://en.wikipedia.org/wiki/Probabilistic_complexity#Computational_complexity
02:31:06  and deterministic?
02:31:20  instead?
02:31:34  impossible?
02:31:44  nope
02:31:56  okay it resizes the range by repeating each element of the range N times where N is the length of the list 
02:31:59  then it calculates the sum of that
02:32:05  and then divides that by N
02:32:05  :P
02:32:08  WORSER
02:32:30 -!- augur_ has joined.
02:32:37  mmm
02:33:00  for instance "count to A(n) every time you increment the counter that will eventually contain your answer"
02:33:16  much worse complexity :D
02:33:58  2^n
02:34:03  it is easy to make it _technically_ at least O(enormous(n)).  Just do enormous(i) useless operations between normal step i and i+1.
02:34:17  ^
02:34:51  2^n is the worst i think... count until the correct result
02:35:05 -!- augur has quit (Ping timeout: 240 seconds).
02:35:10  O.O
02:35:21  :)
02:35:24  roper: i have for a while now thought that you're probably trolling.
02:35:44  not entirely
02:35:51  he could just be really dumb.
02:35:53  >_>
02:36:24  just a though experiment
02:36:49 -!- augur has joined.
02:37:23 -!- augur_ has quit (Read error: Connection reset by peer).
02:38:00  bah paypals fees suck.
02:38:20 * kallisti should try bitcoin.
02:51:21 -!- roper has quit (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org).
03:07:20  what about cosbycoin
03:08:51  well if you insist on using bills...
03:10:17 -!- TeruFSX has joined.
03:10:55  oerjan: ..........................
03:11:05  astounding
03:11:21  yw
03:11:31  I have a different idea the kind of payment system to use; you buy the card with the account number and SSH key, split the account if necessary, and then give them the code. So it would be decentralized system. And not necessarily require the use of a computer.
03:14:05  has anyone heard about the Jan. 18 "blackout"?
03:14:22  kallisti: What does that mean?
03:15:00  reddit is going to go offline on Jan 18 to protest SOPA
03:15:06  also Mozilla
03:15:09  and possibly others.
03:15:11  en.wikipedia might too, or at least display a banner: http://en.wikipedia.org/wiki/Wikipedia:SOPA_initiative/Action
03:15:34  if facebook and google do it, that might be something
03:16:03  oh god I would have to use bing.
03:16:07  or duckduckgo
03:16:08  or yahoo
03:16:11  AAAAAAAAAAAAAAAAAAAAAH
03:16:37  Will Freenode go offline too?
03:16:50  not that I'm aware of.
03:17:32  You could probably still use gopher if Google go ofline
03:19:06  Or some Wikipedia mirror
03:23:20  zzo38, you're aware that Google does not actually control the HTTP protocol, yes?
03:23:39  kmc: Yes, I am aware.
03:26:26 -!- pir^2 has joined.
03:36:53 -!- pir^2 has quit (Quit: Read error: Connection reset by peer).
03:45:31  I am writing a program called Internet Quiz Engine. It is written in CWEB, and will be available to put your own quiz file too, after I have tested it. It is a program to run on gopher services; similar to CGI but much simpler. There is only one environment variable called SELECTOR and no headers.
03:46:07  (I have already written other programs to use this convention, you can use PHP or Perl or whatever you want, as long as you can access environment variables)
04:03:30 -!- pikhq has quit (Ping timeout: 245 seconds).
04:03:38 -!- pikhq has joined.
04:10:22  Even though it is C, there is no buffer overflow because there is no buffer!
04:19:45  How often would you use >>= or join with Parsec? Probably only very rarely; it isn't useful in most cases. Actually, return probably isn't very useful with Parsec either; if I need a constant value I can use <$ to result in that value when certain things are parsed.
04:21:27 -!- _Slereah has quit (Ping timeout: 255 seconds).
04:22:08 -!- _Slereah has joined.
04:25:55  zzo38: you'd want >>= if you want to branch on the first parser in a way too complicated to be efficiently split into <|> cases?
04:26:37  *parser result
04:27:01 -!- TeruFSX has quit (Read error: Connection reset by peer).
04:27:32  oerjan: Well, yes; a dependent parser (I don't know if that is a proper term used anywhere), in case you want to do something such as a number written in decimal form which tells how many bytes to read for the data, and stuff like that. But it is not required for most things you are going to parse.
04:28:37  "non-context-free" would be an approximation.
04:29:04  oerjan: O, is that what it is called? Or, only approximately?
04:30:01  well in principle you always _can_ split into <|> 's; i saw a blog post showing how.  but it requires using haskell's laziness to make the grammar essentially infinite.
04:30:53  and i think if you use only applicative methods and a finite grammar, you can do only context-free things.
04:33:09  Yes I think so.
04:33:39  But you would rarely ever need it differently.
04:34:30  you might also want >>= i think if you want to check for things like variables being in scope.  but of course you can do that post-parsing as well.
04:35:12  oh and you'll need that anyhow if you have arbitrary order of declarations like in haskell.
04:35:22  (that = post-parsing)
04:36:03  oerjan: Yes, you would probably do that post-parsing. Although if it needs to be done during parsing for some reason then yes you would want >>= for that purpose, but I suppose that is also a "dependent parser" or "non-context-free" or whatever? Is it?
04:36:14  yeah
04:40:38  So I suppose context-free with an infinite grammar is like non-context-free?
04:42:02  yes, you can turn an arbitrary String -> Bool function into such a grammar
04:43:01  oh hm
04:43:08  finite strings.
04:46:38  here is the blog article; note that he uses "context-sensitive" but that's too restrictive: http://byorgey.wordpress.com/2012/01/05/parsing-context-sensitive-languages-with-applicative/
04:52:53 -!- NihilistDandy has quit (Quit: Linkinus - http://linkinus.com).
04:53:27  OK
04:54:01 -!- augur has quit (Ping timeout: 240 seconds).
04:57:48 -!- augur has joined.
05:01:18  Nginx does not properly support headerless HTTP. I suggest that they fix that.
05:02:02 -!- pir^2 has joined.
05:02:07  "headerless HTTP" sounds like "not HTTP"
05:02:40  do you mean HTTP 0.9?
05:03:13  kmc: No, I mean headerless HTTP. True it is not complete HTTP. But Apache supports headerless HTTP.
05:03:44  i don't think nginx is obliged to support everything apache does
05:03:45 -!- Nisstyre has quit (Ping timeout: 240 seconds).
05:03:52  do you have some documentation of Apache's support
05:03:59  all i found googling for "headerless HTTP" was HTTP 0.9
05:04:40  With Apache, omitting the HTTP/1.1 or whatever after the URL omits both the request header and the response header. Nginx will omit the request header if you do that, but it will not omit the response header.
05:05:16  is this a documented feature
05:05:22  or just something that happens to work
05:05:28  I don't know, but I know how it works because I tried it.
05:05:52  And Apache's headerless HTTP seems more reasonable to me than Nginx's headerless HTTP.
05:05:52  so now you want nginx to be bug-compatible with apache
05:05:57 -!- Nisstyre has joined.
05:06:31  As far as I know it isn't a bug, it is supposed to be like that.
05:06:57  kmc: According to http://www.apacheweek.com/features/http11 this is standard HTTP 0.9 behaviour.
05:07:58  hm it does seem nginx claims to support HTTP/0.9
05:08:02  i'm surprised
05:08:36 -!- Nisstyre has quit (Max SendQ exceeded).
05:08:41  When I $ nc localhost 80\nGET /\n, it gives me an nginx response page with no headers.
05:08:59  Maybe the headers are coming from a CGI-style executable or something along those lines?
05:09:35  shachaf: OK. Yes maybe that is the case, or maybe they changed it from the version I have tried connecting to (including Google and FreeGeek), or whatever else.
05:10:08  I don't think Google runs nginx.
05:10:29  O, it is gws
05:10:30 -!- Nisstyre has joined.
05:10:44  Then they should fix gws
05:11:07  But I know FreeGeek is nginx and at least the version I have tried had that problem; maybe not all versions do
05:11:26  (I use Apache on my own computer, however.)
05:11:36  zzo38, you seem to have an odd idea of what the word "fix" means
05:16:07 -!- TeruFSX has joined.
06:31:17 -!- Mathnerd314 has joined.
06:31:50  http://stackoverflow.com/questions/4382223/pure-functional-language-haskell oh man, Haskell question answered by Joel Spolsky
06:32:29 -!- Mathnerd314 has set topic: United ITV of hex-him & hill-sinking of the UNITED states of ame"corporatism"riva | http://codu.org/logs/_esoteric/.
06:34:49  hi
06:35:03  hiiiii
06:36:48  Is it correct to say that the big idea is to have the impurity _just outside_ the language level?
06:36:53 -!- Mathnerd314 has changed nick to notaprogram.
06:37:05  I mean, of how IO gets done in Haskell
06:37:12 * oerjan hides an impurity monster under Sgeo's bed
06:37:54 * oerjan turing tests notaprogram 
06:38:21  move pawn to b1
06:38:52  notaprogram: Do you mean a black pawn?
06:39:05  ok you're human, no computer could play chess that badly
06:39:18  zzo38: no, I insist on moving first
06:39:55  notaprogram: Then is it a variant or is it standard FIDE chess? In standard FIDE chess, you cannot move a white pawn to b1
06:40:24  oops, I meant b3
06:40:36  it's friday the 13, I mix up the digits
06:40:41 -!- Nisstyre has quit (Ping timeout: 240 seconds).
06:41:03  O, OK.
06:41:21  Sgeo, no
06:41:31 -!- pir^2 has quit (Quit: ChatZilla 0.9.88 [Firefox 10.0/20111221135037]).
06:43:48 -!- notaprogram has changed nick to quantumprogram.
06:46:01 -!- quantumprogram has changed nick to Mathnerd314.
06:48:16  the universe is a functional programming language
06:48:33  ok
06:48:39  but it is so esoteric that only I know how to program it
06:48:43  ok
06:49:19  O KAY
06:49:38  (i think even elliott would consider that an appropriate case)
06:49:58  so... can I get added as an op here?
06:50:04  yes
06:50:11  Can you?
06:50:16  it's quite physically possible.
06:50:55  but is it probable?
06:51:05  not in the near future, no.
06:51:22  You need a heart of gold.
06:51:35  gold is expensive...
06:56:00  a heart of pyrite will do
06:56:12  Sgeo, to me the essential idea is that descriptions of IO are first-class values
06:56:47  kmc, right, and right outside the language, the description of main is executed?
06:56:55  i wouldn't say that's "outside the language"
06:57:04  the Haskell Report specifies the meaning of each IO action
06:57:13  Outside the miranda of the language
06:57:21  you have one semantics for function application and evaluation, and another semantics for execution of IO actions
06:57:24  they're both part of the language
06:57:40  in fact the rules for execution invoke the rules for evaluation
06:58:12  monqy: http://www.youtube.com/watch?v=Eh44QPT1mPE
06:58:15  before you can execute an IO action, you have to compute what it is
06:58:36 -!- oerjan has quit (Quit: Good night).
06:58:41  not just at the "outer level" of main, but at each occurrence of (>>=)
06:59:17  if you have unsafe{Perform,Interleave}IO then the rules are mutually recursive: the result of evaluating some value can depend on the result of executing some IO action
06:59:56  it would be interesting to program in a strict-evaluation language which still has the evaluation/execution distinction, but where unsafePerformIO is not as stigmatized
07:00:23 -!- Klisz has quit (Quit: SLEEP, GLORIOUS SLEEP).
07:06:36  kmc: it would be even more interesting to program in http://requestforlogic.blogspot.com/2011/08/embracing-and-extending-levy-language.html
07:13:15  Mathnerd314, cool
07:13:48  but unfortunately there's no Haskell implementation
07:17:14  How long is Jyte closed for upgrades/bad gateway?
07:19:32 -!- jpulgarin has joined.
07:19:38 -!- jpulgarin has left ("Leaving").
07:24:12 -!- Mathnerd314 has quit (Read error: Connection reset by peer).
07:45:58 -!- sebbu3 has joined.
07:45:58 -!- sebbu3 has quit (Changing host).
07:45:58 -!- sebbu3 has joined.
07:49:18 -!- Mathnerd314 has joined.
07:49:36 -!- sebbu2 has quit (Ping timeout: 252 seconds).
08:05:24 -!- primo_ has joined.
08:07:51 -!- primo has quit (Ping timeout: 248 seconds).
08:27:22  Internet Quiz Engine:   gopher://zzo38computer.cjb.net:70/1quiz.menu*a
08:28:11  Chrome doesn't seem to support gopher
08:28:36  firefox removed support... but there's an extension
08:28:46  http://zzo38computer.cjb.net/prog/bashgopher/bashgopher
08:29:32  You might need to modify it to work with your system. h = go back, j k = down/up, u i = page down/page up, l = follow, q = quit. When reading text file, space next page, b previous page, q go back.
08:29:59  Such as, adding "-q -1" to nc if necessary, changing "clsb" to "clear", and changing CRLF to LF only.
08:32:47  Does this work for you?
08:38:13  In today's economy, which would be worth more, five golden tumors or five golden mice?
08:40:40 -!- Mathnerd314 has changed nick to [Mathnerd314].
08:43:02 -!- [Mathnerd314] has changed nick to Mathnerd314_.
08:48:04 -!- Mathnerd314_ has changed nick to Mathnerd314.
08:54:40 -!- zzo38 has quit (Remote host closed the connection).
08:54:57 -!- Phantom_Hoover has joined.
09:03:01 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
09:09:03 -!- monqy has quit (Quit: hello).
09:16:18 -!- Phantom_Hoover has joined.
09:20:20  fizzie: ping
09:21:47 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds).
09:22:09 -!- Phantom_Hoover has joined.
09:29:17  me: pong
09:41:26 -!- primo_ has changed nick to primo.
09:44:26  `quote 74
09:44:36  74)  Invalid! Kill! Kill!  I get that feeling too.
09:44:53  `quote 69
09:44:57  69)  Warrigal: what do you mean by 21?
09:45:02  `quote 21
09:45:06  21) IN AN ALTERNATE UNIVERSE:  there is plenty of room to get head twice at once
09:46:27  `quote 314
09:46:31  314)  ais523: quick, say something funny   something funny hagrea:D   can'tä sopt laughitn
09:48:41  `quote 313
09:48:43  `quote 315
09:48:44  313)  it is from 2002 though, I was younger then
09:48:47  315)  elliott: hey, thinking's easier than using the Internet
09:49:14  hmm.
09:49:23 -!- Nisstyre has joined.
09:49:28  `quote 333
09:49:30  333)  haha, god made one helluva blunder there :DS   "WHOOPS HE AIN'T DEAD YET!"   "luckily no one will believe him because christians are such annoying retards"
09:49:51  `quote 249
09:49:54  249)  Sgeo: re "hm?": at the emerging languages conf., after the talks we went out for a drinks and all the Factor team was on heavy liquor
09:50:11  `quote 250
09:50:15  250)  mtve, now he's an expert idler.   mtve: kitty kitty kitty
09:52:11  `quote 252
09:52:14  252)  I need a new desktop background   j-invariant: Try http://codu.org/spinners.png (tiled)   uhrghoaudp
09:53:18  `quote 253
09:53:21  253) * quintopia sits on gregor
09:53:41 -!- Nisstyre has quit (Ping timeout: 240 seconds).
09:53:45  blargh, to sleep I go
09:54:49  `quote 254
09:54:52  254)  [...] reyouthismootherate [...]
10:02:48  `quote 302
10:02:51  302)  Phantom_Hoover: if the list is in random order, like poor ehird here
10:05:02  `quote 304
10:05:06  304) * Received a CTCP VERSION from nyuszika7h   * VERSION Microsoft IRC# 2011 64-bit (Windows 8 Beta, x64, 2GB RAM)    Gregor: Windows 8 Beta? o_O    A small benefit of my brief time as an intern at MS.
10:05:35  `quote 305
10:05:38  305)  addquoting yourself?  isn't that like commenting on your own facebook status?    Yup, but I'm JUST THAT AWESOME.
10:07:12 -!- Mathnerd314 has changed nick to lament.
10:08:28 -!- lament has changed nick to mathnerd314.
10:11:19 -!- mathnerd314 has changed nick to Mathnerd314.
10:11:40  `quote 311
10:11:43  311)  I also do not like that it should be disallow just because of too weird. They haveto make up more name so that not everyone has the same name!!!
10:14:28 -!- Vorpal has joined.
10:20:20  `quote 320
10:20:23  320)  BYE dbc WE'LL BE SURE TO ACCIDENTALLY MENTION YOUR NICK OFTEN
10:20:43  whoops... too far
10:20:53  FFS.
10:20:54  I'll try again tomorrow
10:20:57  `pastequotes
10:21:01  http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.1014
10:21:06  Here, enjoy.
10:21:54  see, they're all wrong
10:22:16  Oh, jesus.
10:22:25  I liked you more when you were just annoying, rather than insane.
10:22:58  yes, I know. but insanity appears to be the only way out...
10:23:22  fizzie: ping me when you're available
10:23:41  `quote 683
10:23:45  683)  COCKS [...]  truly cocks
10:24:34  wouldn't it be cool if i joined with ms comic chat
10:26:23 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.88 [Firefox 12.0a1/20120113031050]).
10:31:25  `quote sleep
10:31:28  325)  i'm really sleep \ 544)  IM FIST IN HEAD AND DONT KNOW TO SLEEP?????? \ 559)  That's the stupidest thing I've heard all morning. (Though I did wake up five minutes ago, so I haven't had a chance to hear very much.)   The "Why are you still asleep? I told the cat to wake you up." comment does come pretty close, though. \ 613)  in the past few minutes I tried remembering
10:31:44  `quote shachaf
10:31:48  617)  elliott: GHC bug? Come on, it's the parentheses.   The more parentheses you add, the closer it is to LISP, and therefore the more dynamically-typed. \ 663)  Real Tar is GNU tar.   You just ignore whichever features don't make you feel superior enough. \ 701)  VMS Mosaic?   I hope that's not Mosaic ported to VMS.   Hmm. It's Mosaic ported to VMS.
10:32:03  Those quotes are terrible.
10:32:30  I'll take solace in the fact that `quote isn't so much "best of" as "random of".
10:38:13    
10:38:17  err, oops
10:38:41  (why was there a single space in the input line?)
11:45:22  clearly a ploy by the romulans to disrupt our communications
11:46:33 -!- elliott has joined.
11:46:37  -NickServ- Last failed attempt from: elliott_!~elliott@c-69-181-214-206.hsd1.ca.comcast.net on Jan 13 15:34:34 2012.
11:46:40  WHO THE FUCK ARE YOU
11:52:55  "A purely functional user interface is not a user interface at all! It is a picture or maybe even an animation, but it is not something you can interact with."
11:52:56 * elliott cries.
11:54:34  01:58:45: -!- roper has joined #esoteric.
11:54:35  01:59:57:  they say one got online every day to remedy affairs caused other hours from her nick
11:54:35  02:01:21:  roper: I do not understand without a context?
11:54:35  02:01:40:  account hijacking
11:54:35  02:02:13:  bot harassment
11:54:35  02:03:02:  they are quite talkative
11:54:37  02:03:43:  i think they use your history from other conversations
11:54:39  I...
11:54:48  romulans
11:54:57 -!- Ngevd has joined.
11:55:31  If there's ever an esoteric blend of Haskell, f should be equivalent to `(flip f)`
11:55:45  haha
11:55:52  what, haskell isn't esoteric enough for you?
11:55:53  Also, Hello
11:55:57  hi Ngevd
11:56:30  Nah, I can generally figure out what a Haskell program is for by looking at it for a while
11:57:06  Ngevd: i was thinking exactly that haskell is a right-to-left language
11:57:24  and indeed i define flip ($) in all my programs to be <>
11:57:29  it's very useful
11:58:24  I would have called it 
11:58:38  yes
11:59:43  £sd
12:04:42  00:24:32  dibblego: Why did you kick me?
12:04:42  00:25:04  Fifo: please do not spam the channel or its users in private
12:04:42  00:25:29  dibblego: Sorry!
12:04:42  00:25:49  dibblego: Please join ##iPhoneFifo!
12:05:43 -!- Nisstyre has joined.
12:05:45  What?
12:06:27  02:35:24:  roper: i have for a while now thought that you're probably trolling.
12:06:27  02:35:44:  not entirely
12:06:27  02:35:51:  he could just be really dumb.
12:06:27  02:35:53:  >_>
12:06:36  oerjan logreading: i'm definitely leaning on the side of just really dumb
12:08:22  this is such a weird channel
12:08:37  It's possitively esoteric
12:08:46  womp womp
12:08:56  03:23:20:  zzo38, you're aware that Google does not actually control the HTTP protocol, yes?
12:08:57  kmc: YET.
12:09:01  yeah
12:09:11   Internet Quiz Engine:   gopher://zzo38computer.cjb.net:70/1quiz.menu*a
12:09:13  SOUNDS LEGIT
12:09:17  it is
12:10:11  gopher://zzo38computer.foeug3g47fgeg34.ch3p-h4rbl-vjaagra.co.ng:22/../../warez2.php
12:10:50  Gopher is a pun!
12:10:54  Heheheh
12:11:44 -!- kmc has left ("Leaving").
12:11:46 -!- kmc has joined.
12:12:02  04:34:30:  you might also want >>= i think if you want to check for things like variables being in scope. but of course you can do that post-parsing as well.
12:12:02  04:35:12:  oh and you'll need that anyhow if you have arbitrary order of declarations like in haskell.
12:12:02  04:35:22:  (that = post-parsing)
12:12:12  hmm, what you need is a bitemporal state monad
12:12:18  so that state flows both forwards and backwards
12:12:33  "bitemporal"!? That two-timing...!
12:12:33  i guess you need the state to be a Monoid to combine them
12:12:37  one point twenty one jiggawatts
12:14:17  06:31:50:  http://stackoverflow.com/questions/4382223/pure-functional-language-haskell oh man, Haskell question answered by Joel Spolsky
12:14:17  surprisingly it's even basically correct
12:15:08  despite all the people going "getChar is, like, referentially transparent so you're wrong" :|
12:16:59  06:48:16:  the universe is a functional programming language
12:16:59  06:48:33:  ok
12:16:59  06:48:39:  but it is so esoteric that only I know how to program it
12:17:03  oh, good, he's still crazy!
12:24:53  it's kind of correct
12:25:13  i mean it's just arguing about what words mean
12:25:29  but if you accept joel's definition than Haskell isn't a pure functional language
12:26:33  well it has a pure subset, but I've never heard a convincing argument that haskell with IO is pure; people say that since getChar is always "the same getChar" it's referentially transparent, but nobody has a definition of "same" for IO actions
12:26:41  i have one
12:26:45  extensional equivalence
12:26:57  there's no experiment by which you can distinguish "one getChar from another"
12:27:14  anyway i don't want to argue about what "pure" means
12:27:21  but evaluating getChar (like with seq) doesn't perform any IO
12:27:24  that's just a fact
12:27:26  that's a little lax for my tastes :P and yeah, I'm not saying it does
12:27:39  elliott, it's the same standard by which you decide whether two functions are "the same"
12:27:40  well whatever
12:27:49  and if we can't do that, we can't even say that "map" is referentially transparent
12:27:50  kmc: sure, but there /are/ experiments you can perform on functions
12:27:59  there are experiments you can perform on IO actions too
12:28:05  you do one every time you run a haskell program
12:28:19  i see you're arguing about what "pure" means :P
12:28:22  nope
12:28:45 -!- MDude has joined.
12:30:07  Haskell has evaluation semantics and execution semantics
12:30:37  the former is a calculus of referentially-transparent functions without side effects
12:30:52  the latter is like an imperative programming language with effects and results
12:31:16 * elliott isn't up for arguing himself, but I *do* understand how IO works in Haskell and I don't think that it's "not really a monad" or "just impurity that you can't see" or whatever newbies say because of a lack of understanding
12:31:20  whether the combination of these forms a "pure functional programming language" is a stupid argument
12:31:27  i know you know how it works
12:31:29  and I agree that treating (IO a) as a model of an imperative program with result type a is just fine
12:31:46  I just think that it lets you consider far too many things that are completely impure in practice as pure
12:31:49  because you can model them that way
12:33:07  i don't understand
12:33:31  but that's ok
12:33:39  :)
12:33:40  Are there any pure imperative languages?
12:33:47  Would such thing have any point at all?
12:34:11  it sounds like a great excuse to argue about what "pure" and "imperative" mean
12:34:36  if you pressed me for a yes/no answer on "pure" and "imperative" I would say that Haskell is both
12:34:43  this may be an unpopular viewpoint
12:35:54  i think it's the wrong focus if you want to understand haskell or why it's good
12:36:11  I think it's much easier to agree on "Haskell code that does IO is too imperative and impure in style" than "Haskell is impure"
12:36:23  "too imperative" by what standard
12:36:34  bring your own standards
12:36:40  compared to "the ideal"
12:36:54  A little impurity is good for a language
12:36:56  i.e. it would be nicer if we could write code that interacts with the outside world in a more functional style
12:37:18  How would that work?
12:37:19  i would say "declarative" rather than "functional"
12:37:32  kmc: yes. others would say "denotational" :P
12:37:46  Ngevd, well, a very simple example is interact :: (String -> String) -> IO ()
12:37:48  but the basic sentiment is the same
12:37:59  yeah
12:38:12  i like declarative programming and it would be cool to use it in more areas
12:38:31  at the same time, i find imperative programming in Haskell to be pleasant enough
12:38:56  I mean, don't get me wrong -- Haskell's IO system is a good compromise and it's much more "crisply" defined than in most other languages
12:39:24  but it definitely bugs me to see people say "haskell is like the purest language ever and if you think IO is in any way imperative or 'impure in style' or anything you're just an idiot"
12:39:27  (ok, not in those words :P)
12:39:59  Purest language ever... surely that would be a tie between a few esolangs?
12:40:11  i think it's pretty much agreed that IO is an embedded imperative language
12:40:12 -!- Nisstyre has quit (Ping timeout: 276 seconds).
12:40:20  whether this makes Haskell overall "impure" is a meaningless argument about definitions
12:40:30  Also, how can we say that language a is more or less pure than language b?
12:40:46  Which is more pure? Java or C++?
12:40:49  kmc: I think it's agreed in #haskell, but I don't think that's the same thing as "agreed by people who call themselves haskell programmers on the wider internet" :p
12:40:57  shrug
12:41:18  to me Haskell is a pragmatic multiparadigm language with a good set of design tradeoffs
12:41:31  so i don't focus on ideological purity
12:41:43  I mean, I've seen at least one person in here and one person on SO yell loudly about how stupid the "action" terminology is because "you wouldn't call (Maybe a) an action and IO is, like, just as pure as the rest of the nomads"
12:41:52  admittedly at least one of those people was dumb.
12:42:04  that's stupid
12:42:12  Nomads aren't pure.
12:42:14  i wouldn't call (Maybe a) a list either, but [a] is
12:42:31  kmc: well it was in reaction to saying "monadic actions" more generally
12:42:36  and I think it's perfectly reasonable to call (Maybe a) an action
12:42:39  shrug
12:42:43  i don't think it matters
12:42:53  i think #haskell should agree on common terminology and push it
12:42:58  instead we just bicker about words
12:43:12  haha, getting a few hundred people to agree on IRC
12:43:15  ur funy
12:43:27  anyway the *reason* i don't worry about whether Haskell is philosophically pure
12:43:29  I avoid #haskell, mainly because I'm scared of big groups
12:43:34  is that i wish *other people* would see it as a pragmatic multiparadigm language with a good set of design tradeoffs
12:44:06  rather than a language for extremists who argue all day about philosophical purity
12:44:09  I would say it is at least primarily functional, although has features from other paradigms
12:44:47  kmc: as an extremist i'm offended
12:44:51  haha
12:44:54  very well sir
12:46:51 -!- Ngevd has quit (Read error: Connection reset by peer).
12:50:56  in particular many of Haskell's strengths have nothing to do with "purity"
12:54:54  in my "why learn Haskell" slides i talk only a little bit about "purity" and types
12:56:26  i spend more time on stuff like EDSLs, lightweight concurrency, STM, semi-implicit parallelism, GHC 
12:56:43  Maybe purity is overrated but I don't think types are.
12:56:48  QuickCheck, Criterion, Threadscope, other tools, etc.
12:57:04 * elliott doesn't think Haskell is very well-suited to EDSLs... it excels at some of them but falls down for others
12:57:47  some of those features rely on purity, in the sense of "Haskell programmers tend to write lots of pure functions", not in the sense of "Haskell absolutely forbids you from writing an impure function"
12:57:55  (which of course, common implementations don't)
12:58:58  elliott, yeah, types are pretty important
12:59:05  but it's hard to convince someone of that in a few slides
12:59:22 * elliott doesn't try to convince people of things in a few slides :)
12:59:28  well ok
12:59:33  But yeah, fair point.
12:59:38  i do, and I was talking about one effort to do so
13:00:02  I agree that things like you listed are better sells under such constraints.
13:00:02  in the best case, you convince them that it's a useful feature for preventing mistakes
13:00:05  which is... still not that sexy
13:00:21  i mean, subconsciously I know that *I* don't make any mistakes
13:00:47  Yeah, but think about all the other terrible programmers you have to deal with!
13:00:54  See, we can sell Haskell on antagonism and ego.
13:00:54  they probably aren't using haskell
13:00:58  Like Ruby!
13:01:00  haha
13:01:03  touché
13:01:53  anyway there was enough to fill an hour with "here's this amazing thing Haskell lets you do"
13:02:17  that I didn't have to focus too much on the essentially negative "here's how Haskell protects you from the stupidity of yourself and others"
13:02:28  i agree it's not a great EDSL host
13:02:42  i think it's a pretty good EDSL host, and that this plays nicely with the other strengths of the language
13:02:54  "The question is a simple one. Must all operations on a TVar happen 
13:02:54  within *the same* atomically block, or am I am I guaranteed thread 
13:02:54  safety if, say, I have a number of atomically blocks in an IO 
13:02:54  function."
13:03:01  i wish Template Haskell weren't so fucking cumbersome; then we could have some better concrete syntax for EDSLs
13:03:18  O_O
13:03:32  tell them to read the last chapter of RWH
13:03:38 * elliott wants to live in a world where "atomically" means "atomicallyAndThenSome", and the layout of code into "IO functions" affects its semantics.
13:03:43  kmc: I'd have to subscribe to haskell-cafe.
13:03:55  instead you just lurk and snark
13:03:58  Yes.
13:04:06  that's cool
13:04:27  kmc: I think the kind of EDSLs Haskell is good at are the ones that are richer than Haskell.
13:04:35  Which is to say "first-class" EDSLs.
13:04:42  e.g. Parsec and so on.
13:04:50  The problems come when you can't embed Haskell in them, e.g. anything you're trying to compile
13:05:06  right
13:05:32  And then you just have to fall back on ugly typeclass hacks and in the end you get sick of Eq because it forces (==) to mean something rather than letting you use any "boolean" you want :P
13:05:52  this would be solvable in principle
13:06:22 * elliott thinks there are a few promising ways to solve it, and I think solving it would be a great way to create a language better than Haskell.
13:06:52  There's a *lot* of this kind of stuff being done -- anything that does observable sharing is essentially working around Haskell's DSL-related weaknesses, for example.
13:06:58  yes
13:07:10  well first you want to unify something like Template Haskell with something like GHC API, only hopefully make both of them a lot less shitty
13:07:45 * elliott thinks that would be very (very) valuable, but isn't sure it's the best basis for most EDSLs
13:07:49  that is, you can construct a Haskell AST, then compile it and use it as a function
13:07:52  metaprogramming in haskell sucks right now
13:08:14  and you can do this at "compile time" or at "run time", and hopefully the distinction between the two is not so pronounced
13:08:29  kmc: Have you seen the http://www.cs.berkeley.edu/~megacz/garrows/ stuff?
13:08:40  uh, started reading it a while ago
13:08:43  didn't get very far
13:08:45  heh
13:08:49  I don't like arrows, but it's very interesting.
13:09:08  And seems to be able to do an awful lot of DSL-y stuff that isn't really possible now.
13:09:17  The syntax is ugly though.
13:09:58  they get bonus style points for implementing part of GHC in Coq
13:10:27  I'm always surprised that Coq extraction actually works in practice.
13:11:14  elliott, so how would you solve the problem
13:11:27  other than better macros and staged compilation
13:14:34  kmc: Well, I think it involves restructuring the language at a bit of a deeper level to be truly natural. Things like using typeclasses for the kind of introspective overloading you see is basically just trying to selectively override the language's semantics, which is generally really awkward in Haskell. (The AwesomePrelude thing takes this to its logical conclusion, and it's not pretty.) I can imagine a language which essentially does the equiv
13:14:35  alent of having a typeclass for each language feature -- conditionals, pattern-matching, self-reference, duplicating values, etc. -- and you can override that piecemeal so that all the same syntax and meaning is retained, but in possibly a more restricted setting; you'd have a bracket mechanism which would let you write code in a new semantics. That's basically what the generalised arrow stuff is doing, to my understanding; there's a typeclass fo
13:14:35  r each additional language feature, and the code you write is just unrolled into a bunch of those.
13:14:58  i see
13:15:06  i should learn more about garrows then
13:15:17  If you want to be *really* radical, you could eliminate the bracketing entirely, and direct the overriding of the semantics by the types -- i.e. if something is expecting an argument in a certain semantics, those semantics are used when you fill in that argument.
13:16:30  kmc: That also ties in with more language-based support for effects; e.g. you could easily define a semantics that supports every such "language ability" but that does nondeterministic computation, and you'd get the equivalent of the list monad but with the same syntax as the outer language.
13:17:26 * elliott hasn't read all of the GArrows stuff, but intends to sometime. What I've read is basically similar to all this, except that it's very explicit about what's "metaprogramming" and what's not -- there's a leap from "normal Haskell" to "metaprogramming stuff".
13:18:24  kmc: And, of course, opting out of implementing a certain capability lets you do things like the kind of optimisations you can do to applicative/arrow-based parsers.
13:18:41  Except more aggressive, since you have full access to the *entire* structure of the EDSL program.
13:18:53  (Nothing hiding behind an "arr", etc.)
13:20:34  It wouldn't even be a large leap from that to make the language self-extending on an even deeper level -- if you can define your own language capabilities (and I don't see what would stop you, if everything was appropriately first-class -- indeed, I'd expect most extensions (like nondeterministic computation) would be implemented by defining a new capability for that, and giving a canonical implementation), then all you need to be able to do is d
13:20:34  efine new syntax for those capabilities and it becomes as good as a native language feature.
13:20:40  That might be going further down the Lisp path than most people would like, though.
13:21:09  yeah
13:21:20  there's already controversy over "how typed" template haskell should be
13:21:53  yeah
13:22:18  I think the proposal to give it a typed layer, but allow access to just AST-mangling, is probably the best thing for it as it stands
13:22:53  but I think that in a more ideal language, TH would be basically completely untyped, because well-typed things would be expressible directly, without having to resort to a special "metaprogramming macro processing" facility
13:34:22  yeah
13:34:24  makes sense
13:34:25 -!- Patashu has quit (Read error: Connection reset by peer).
13:48:31 -!- pikhq_ has joined.
13:48:55 -!- pikhq has quit (Ping timeout: 255 seconds).
13:53:43  Need help indexing XML files into Solr using DataImportHandler
13:53:44  I don't know java, I don't know XML, and I don't know Lucene. Now that that's out of the way. I have been working to create a little project using apache solr/lucene. My problem is that I am unable to index the xml files.
13:53:55  [...] I am not sure what information is required for you to help me so I will just post the code.
14:15:59 -!- primo has quit (Read error: Connection reset by peer).
14:17:54 -!- Nisstyre has joined.
14:25:23 -!- Nisstyre has quit (Ping timeout: 240 seconds).
14:31:03 -!- Nisstyre has joined.
14:58:51 -!- primo has joined.
14:59:11 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds).
14:59:31  RIP Phantom_Hoover
15:02:48 -!- Phantom_Hoover has joined.
15:03:58  eek a zombie
15:05:36  RIP us all
15:05:56  %rip
15:06:24  mv %rip, usall
15:06:27 -!- primo has quit (Read error: Connection reset by peer).
15:06:38 -!- Ngevd has joined.
15:07:39  Hello!
15:07:50  hi
15:08:09  hi Ngevd
15:08:53  How did your Haskell thing-that's-a-bit-like-an-argument-but-you-both-agree go?
15:09:18  i think that's called a "discussion"
15:09:39  anyway the logs are public
15:09:48  I know, I've read them.
15:09:56  I just want to make conversation
15:10:14  not allowed
15:10:17  only arguments
15:12:47  > ap (.) (join (.) (ap (.) (join (.)))) (*2) 1
15:12:48    1024
15:13:01  lols
15:13:23  > ap (.) (join (.) (ap (.) (join (.)))) (join (+)) 1
15:13:24    1024
15:14:06  this is probably what Ngevd's new haskell program actually looks like
15:14:28  Nah, I'm just playing with MIBBLLII
15:14:55  Lambdabot can be used as a non-compliant interpreter for a similar language to MIBBLLII
15:18:00  :t ap id id
15:18:01      Occurs check: cannot construct the infinite type: a = a -> b
15:18:01      Probable cause: `id' is applied to too few arguments
15:18:01      In the second argument of `ap', namely `id'
15:18:04  cool
15:18:11  an esolang that looks like brainfuck but isn't
15:18:27  compared to the usual esolangs that don't look like brainfuck but are
15:18:47  The functions ap, join, const, id, flip, and (.) match up 1 to 1 with MIBBLLII combinators
15:19:07  As long as you work in the ... (->) r monad ???
15:19:07  kmc: Or the ones that look like, and are brainfuck.
15:19:36  Or something?
15:19:47  As long as you have Control.Monad and Control.Monad.Instances
15:19:51  http://esolangs.org/wiki/Zoo_Tycoon_2_Cheats_Money
15:20:13  Have spambots been attacking spam articles?
15:20:56  i would totally use a language named Zoo Tycoon 2 Cheats Money
15:21:41  Shall we create a language by that name?
15:22:22  yes
15:22:31  Zoo == objects, tycoon == control structures, 2 == ...arithmetic?, cheats == functions./subroutines
15:22:40  money == finance library?
15:23:09  it will be brainfuck except the instructions are encoded as the phrases "Wikipedia", "game money prize tree", "free money woman", "money and mbti", "trace a money order", "europes history on money", "find instantly low money price search", and "how much money to bring to europe"
15:23:28  No.
15:26:03  yes
15:26:55  actually you should invent a language such that we can execute these spam pages
15:27:17  has there been any work towards designing programming languages specifically for stoned people
15:27:28  i guess there are PLs for children, and that's kind of similar
15:35:02 -!- Nisstyre has quit (Ping timeout: 240 seconds).
15:38:29 -!- Nisstyre has joined.
15:43:46 -!- azaq23 has joined.
15:52:38 -!- cheater has quit (Ping timeout: 240 seconds).
15:54:01 -!- primo has joined.
16:01:13  asldkfjaksdlfldskjakdlfldskjaksldfldksjaksdlf
16:03:05  hi
16:11:04  Hehehe
16:33:32 -!- pikhq has joined.
16:34:16 -!- pikhq_ has quit (Ping timeout: 268 seconds).
16:43:25 -!- Madoka-Kaname has quit (Ping timeout: 240 seconds).
16:44:47 -!- Madoka-Kaname has joined.
16:58:15 -!- pikhq_ has joined.
16:58:46 -!- pikhq has quit (Ping timeout: 260 seconds).
17:01:27 -!- Sgeo has quit (Remote host closed the connection).
17:02:09 -!- Sgeo has joined.
17:02:51 -!- Deewiant has quit (Quit: Server going down for maintenance.).
17:18:02 -!- Madoka-Kaname has quit (Ping timeout: 240 seconds).
17:20:09  the way i see it the main role of mathematics in esolang design is to provide the language designer with a look at the primitive elements of programming languages instead of making a hopped up basic clone
17:25:59  more accurately, thats what it means to me
17:27:57 -!- Phantom_Vax has joined.
17:28:15  Hello everybody.
17:29:01  Phantom_Vax: Please tell me you're using an OpenVMS emulator.
17:29:04 -!- Ngevd has quit (Quit: Poor timing, Phantom_Vax).
17:29:07  Yes.
17:36:07  elliott: i'm concocting a diabolical language possibly using some kind of message passing, but since i don't know in detail what message passing is i'm not entirely sure
17:36:19  ah.
17:37:20  the idea is really terrible, and "concocting a language" said in this room can mean something other than what i am doing since people here tend to actually create and finish languages and i don't
17:38:40  actually its such a terrible idea that it just might work, but i have left little to go on in my posts
17:39:08 -!- sebbu3 has changed nick to sebbu.
17:41:02 -!- Klisz has joined.
17:43:29 -!- Phantom_Vax has quit (Remote host closed the connection).
17:44:08  ok the fundamental idea is an operator which can tell you if an object has a member of a specific type, and if it does you can access it directly through a local name
17:45:00 -!- Phantom_Vax has joined.
17:45:05  sort of like saying, does this object contain a string member? if it does then call that string s and do toupper(s)
17:46:06  this could run into some kind of trouble when an object has multiple members of the same type but thats another story
17:48:37 -!- Madoka-Kaname has joined.
17:48:51  if that sounds weird, the idea gets even weirder the more i explore it
17:59:15 -!- Madoka-Kaname has quit (Read error: Operation timed out).
18:11:08 -!- ais523 has joined.
18:23:27 -!- zzo38 has joined.
18:29:26 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
18:30:17  I wrote some of my own Arrows stuff because I don't like the existing one much. The purpose of "arr" is moved into a different class called CatFunctor and you can have from any category, doesn't have to be (->) and in addition, the (,) does not have to be (,) you can have any type of kind (* -> * -> *) used for a specific CatArrow. Purpose of ArrowChoice is just done by a dual category, now.
18:31:06  you've said
18:32:59  Is there a way to abuse a monad of some sort to get nice syntax without -XArrows?
18:34:05  There's a "bind-like operator"
18:34:07 * Sgeo blinsk
18:34:11  blinsk
18:34:14  nblinks
18:34:58 -!- Madoka-Kaname has joined.
18:34:58 -!- Madoka-Kaname has quit (Changing host).
18:34:58 -!- Madoka-Kaname has joined.
18:35:16  I could have sworn I saw something though
18:36:43  Do you agree that "arr" in Control.Arrow is really a kind of functor (not an endofunctor)?
18:37:36 * Sgeo has no idea what the difference is
18:38:41  Difference of what?
18:38:58  functors and endofunctors
18:39:23  endofunctor is a functor from a category to itself.
18:39:25  Endofunctor is a functor from a category to itself. The Functor class in Haskell make endofunctor from (->) category to itself
18:39:55  functors are comparable to morphisms over categories.
18:40:08  Functions /are/ morphisms, you dolt.
18:40:08 -!- toruk-mack has joined.
18:40:17  (in fact I believe many of them are morphisms in the category of small categories.)
18:40:39  (but not all?? not sure)
18:41:18  Is Internet Quiz Engine good to you?
18:41:26  Yes it is legitimate.
18:41:28  it is not relevant to me.
18:43:11  But gopher://zzo38computer.foeug3g47fgeg34.ch3p-h4rbl-vjaagra.co.ng:22/../../warez2.php not only looks wrong, and uses . as the type, but doesn't even resolve. I am unsure why kmc posted this wrong URL, possibly to confuse you???
18:45:07  `welcome toruk-mack
18:45:12  toruk-mack: 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
18:45:36  @tell kmc Explain yourself!!!1
18:45:36  Consider it noted.
18:45:37  toruk-mack: Are you good at esolang?????
18:46:07  TG: (omg still lolig @ that word boner i made ooomg) 
18:46:25  Sorry, wrong channel
18:46:30 -!- toruk-mack has left ("Leaving").
18:47:54  :(
18:48:14  `@ past-toruk-mack ? esoteric
18:48:18  past-toruk-mack: This channel is about programming -- for the other kind of esoterica, try #esoteric on irc.dal.net.
18:54:25  They're still online
18:54:36 * Sgeo messages
18:55:03  Yes I checked that too
18:55:11  But I have no message to write to them.
18:55:27 * Sgeo just sent "for the other kind of esoterica, try #esoteric on irc.dal.net."
18:56:07  Sgeo: It work if it was what they were looking for, which is possible, however
18:59:41  Is DALnet down?
19:00:36  Did you try connecting or ping or whatever?
19:01:05  Yes it seem down to me.
19:01:09  No ping
19:05:09 -!- Phantom_Vax has quit (Changing host).
19:05:09 -!- Phantom_Vax has joined.
19:05:20 -!- Phantom_Vax has changed nick to Phantom_Hoover.
19:06:43 -!- monqy has joined.
19:07:26  what would be annoying about only having gender neutral pronouns is that you could only refer to two things at the same time
19:07:44  in English you can refer to at least 4 things unambiguously via pronouns.
19:08:12  I suppose as a compromise would be to invent more than one gender-neutral pronoun
19:08:15  for different kinds of.. things.
19:08:31  kallisti: That only works if the things you're referring to are of different genders.
19:08:46  elliott: ...yes, that's what I mean.
19:08:56  Well, so it's not much of an advantage.
19:09:01  Anyway, people don't seem to have any problem in languages which use "it" for everything.
19:09:11  the difference between it and he or she or they is pretty useful.
19:09:18  Making pronouns gender-neutral only to immediately re-segregate them by some other means is ridiculous.
19:09:30  kallisti: No?
19:09:39  kallisti: I mean, I really doubt you ever talk about that many things at once without using names.
19:09:44  sure.
19:09:45  Because that would be impossible to follow.
19:10:23  it happens frequently enough
19:10:39  and anyway there's no reason not to "resegregate" the pronouns by grammatical or conceptual distinctions
19:10:44  that have nothing to do with people.
19:11:05  well
19:11:10  it could involve people
19:11:29  for example you could have a pronoun for humans, and a pronoun for everything else
19:11:32  or
19:11:35 * elliott doesn't think it happens frequently enough.
19:11:36  a pronoun for livings things, and a pronoun for inanimate things.
19:11:42  You mean they vs. it?
19:11:47  "he knew that it was a bad idea"
19:11:49  but what if you want to refer generally to both humans and nonhumans
19:11:54  Anyway, there's real data on this, i.e. every language without such a distinction, so unless you're a linguist...
19:12:03  "she took it from me"  (thieving bitch)
19:12:13  :)
19:12:16  There can be one for living, one for inanimate, but then, is there going to be one for ideas?
19:12:30  elliott: note that I am not in any way saying that these languages are somehow crippled in their expressiveness
19:12:36  just that having multiple kinds of pronouns is a convenience.
19:12:52  but also an inconvenience
19:13:10  zzo38: yes that would be good. :)
19:13:19  shrugs violently
19:13:27 * kallisti sighs EXPLOSIVELY
19:13:45  i sure do love wild conjecture
19:13:53  elliott: do tell. what is my conjecture.
19:14:09  no
19:14:20 -!- Frooxius has joined.
19:15:41  anyway it happens all the time in storytelling. It's pretty frequent that you have a sentence involving a person and an inanimate that were both previously mentioned.
19:16:13 * kallisti conjectures wildly.
19:16:44  And you think persons and inanimates are so easily-confusable that we cannot deduce that from context like the billion other things we do on a regular basis?
19:18:35  well, no.
19:19:10  I simply assumed it was bad style in those languages to use the same pronoun in the same sentence to mean different things.
19:19:17  but if that's common then... that works too.
19:22:28  the person/non-person distinction isn't unheard of in other languages. For example Finnish.
19:22:37 -!- zzo38 has left.
19:24:21  kallisti: It doesn't appear in colloquial Finnish, from what I've heard in here.
19:24:29  "it" is used for everything.
19:25:35  ah I see
19:29:01  The demonstratives are used of non-human animate entities and inanimate objects. However, se and ne are often used to refer to humans in colloquial Finnish. (This usage is quite correct in a demonstrative sense, i.e. when qualified by the relative pronoun joka, and in fact it is hypercorrect to replace a demonstrative se or ne with hän or he just because the antecedent is human.) 
19:32:12  well, zzo38 and I are going to go invent a language with at least 20 pronouns
19:33:47  and maybe some distinctions between pro-adjectives and pro-verbs and pro-adverbs
19:33:50  also pro-sentences.
19:34:01  it's a language for pros.
19:36:05  to get the other words, we'll take a dataset of every word in every language and its meaning, and then reassign it a different meaning using `words
19:36:11  er
19:36:15  different name rather
19:36:49  the end result will be that there's probably like 50 ways to say "dog"
19:37:13  (assuming 50 languages were uses, and none of those languages have more than one way to say dog)
19:46:35  !perl $x = qr[(a)]g; print "a a a a" =~ $x; 
19:46:39  Bareword found where operator expected at /tmp/input.17425 line 1, near "qr[(a)]g" \ syntax error at /tmp/input.17425 line 1, near "qr[(a)]g" \ Execution of /tmp/input.17425 aborted due to compilation errors.
19:47:02  !perl $x = qr/(a)/g; print "a a a a" =~ $x; 
19:47:02  Bareword found where operator expected at /tmp/input.17496 line 1, near "qr/(a)/g" \ syntax error at /tmp/input.17496 line 1, near "qr/(a)/g" \ Execution of /tmp/input.17496 aborted due to compilation errors.
19:47:07  erm
19:47:12  !perl $x = qr/(a)/; print "a a a a" =~ $x; 
19:47:13  a
19:47:18  !perl $x = qr/(a)/; print "a a a a" =~ /$x/g; 
19:47:19  aaaa
19:47:48  ah
19:49:17 -!- zzo38 has joined.
19:54:22 -!- oerjan has joined.
19:55:35 -!- Deewiant has joined.
19:57:16 -!- _Slereah has quit (Ping timeout: 260 seconds).
19:58:56 -!- Phantom_Hoover has quit (Remote host closed the connection).
19:59:29  hi oerjan
19:59:50 -!- Phantom_Hoover has joined.
20:06:39 -!- Phantom_Hoover has quit (Remote host closed the connection).
20:07:47 -!- Phantom_Hoover has joined.
20:08:00  wow so I got $25 for setting up a simple HTML5 audio player, and then got $10 to write the ugly terrible hack to load an embedded player in the event of IE.
20:08:04  what's wrong with this picture.
20:08:19  "some one always irritate me an facebook massage, he use mobile for post massage. i want to know his mobile no, is there any way ?" what a good answer
20:08:34  ...answer?
20:08:37  kallisti: web audio players
20:08:47  monqy: regardless of that.
20:09:49  `quote fought for
20:09:53  459)  elliott: You have become the very thing you fought for!
20:10:22  elliott: what is meaning of that massage.
20:12:01  no
20:13:35  IE doesn't support