00:04:23 -!- ehird` has quit (Read error: 104 (Connection reset by peer)).
00:04:29 -!- blahbot` has quit (Remote closed the connection).
00:18:06 -!- sebbu has quit ("@+").
00:19:19 <ihope> Do those include line breaks and '/"?
00:19:26 <ihope> And, for that matter, ''/""?
00:23:04 <SimonRC> triple-quoting in Python has no escapes, IIRC.
00:32:22 <SimonRC> no, I do not include those
00:33:02 <oklopol> triple quotes include linefeeds and '/".
00:33:37 <oklopol> unless i missed something completely here...
00:34:37 -!- Figs has joined.
00:38:34 <Figs> you ever get around to working out how to play the rest of that song? :D
00:39:40 <oklopol> i'll try to look at it if i happen to remember it at daytime :)
00:51:47 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
02:01:13 -!- GregorR-L has quit ("Leaving").
02:13:13 -!- Figs has left (?).
02:19:34 <RodgerTheGreat> I've decided I wan to do a graphic-novel adaptation of "The Feeling of Power"
02:53:44 -!- Sgeo has joined.
02:55:19 -!- Sukoshi` has joined.
02:56:35 -!- Sukoshi has quit (Read error: 113 (No route to host)).
03:12:28 -!- GregorR-L has joined.
03:17:36 <oklopol> say something totally weird and say i accidentally said it on the wrong channel
03:17:51 <oklopol> (but i'm playing my cards savingly)
03:17:52 <GregorR-L> Actually, that was just the wrong channel :P
03:17:56 <GregorR-L> * Received a CTCP TIME from undesktop
03:18:35 <oklopol> i'm tempted to believe that... but anyway, i'm pretty sure people usually just fake it.
03:19:34 <oklopol> because hey, if i haven't, to my recollection, said anything on the wrong channel, it isn't possible
03:19:50 <oklopol> that recollection didn't quite fit there
03:21:40 <oklopol> you are quite right in laughing about that, it was indeed funny
03:21:47 <oklopol> however, i feel sleepy now.
03:22:15 <pikhq> Harry Potter, not hit points.
03:22:34 <oklopol> i'm so gonna read that book and like it.
03:22:46 <pikhq> And anti-HP folk: shut up, please. I'm allowed to enjoy things that are childish in addition to mature things.
03:23:48 <oklopol> what? so you actually reading that?
03:24:59 <oklopol> i've read every hp like 4 times, i'm not judging, though the 'actually' might suggest differently.
03:25:09 <oklopol> and a mat can be used as a hat
03:25:28 * pikhq is reading the latest book
03:25:44 <GregorR-L> Everyone around me at work is HP-obsessed :P
03:25:48 <pikhq> You've had enough time to read HP7 4 times? Dude. . . No life.
03:26:07 <pikhq> I usually read them once, and then wait until I'm in the mood for a book-a-day habit to reread.
03:26:26 <oklopol> pikhq: actually i've read 2 fo them 4 times
03:26:27 <GregorR-L> I haven't read a novel-type book in ...
03:26:40 <oklopol> that's the only novels i've read
03:26:53 <oklopol> i think that's the official term.
03:26:58 <pikhq> WTF is wrong with you?
03:27:05 <pikhq> *Both* of you, that is.
03:27:16 <oklopol> i've read *some*, just very little.
03:27:24 <oklopol> don't remember what and when
03:27:24 * pikhq reads at about the same frequency most people watch TV
03:27:42 * GregorR-L prefers to rot his brain with television.
03:28:13 * pikhq doesn't do TV much
03:29:05 <pikhq> Too many good books, too much good music, too many good webcomics, etc.
03:29:11 <pikhq> And, of course, too much stuff to code. ;)
03:30:28 <oklopol> so whatcha guys been programming this week?
03:30:43 <oklopol> much less than me, i'll assume :)
03:30:49 <oklopol> if i guess your time zone right
03:32:44 <pikhq> Getting back to sanity before I dare touch my Brainfuck game. . .
03:33:55 <pikhq> Then why not at me?
03:34:08 <GregorR-L> pikhq: I was in the middle of a CTCP-war with undesktop at the time :P
03:34:40 <oklopol> i seem to lack some skill again.
03:35:03 <oklopol> trying to get me some o that sweet yelling, but noooo
03:36:00 <oklopol> i guess YOU ARE NOT REGISTERED as a response *could've* made my understand that on the first 10 tries.
03:36:11 <oklopol> but hey, we're all different.
03:37:03 <oklopol> it seems i'm seriously behind you in this week's coding amount
03:37:16 <oklopol> i've done about ½ hours of programming
03:37:33 <oklopol> any chance you broke your hands during the week or smth?
03:37:57 <oklopol> both of you, in a ctcp fight
03:40:37 <pikhq> 20:39 [freenode] oklopol [n=ville@194.251.103.33] requested unknown CTCP FINGER from pikhq:
03:42:29 <oklopol> hey ppl, what's donnie darko about? the movie?
03:42:38 <pikhq> I don't need to know about your dildo with a dying battery.
03:42:40 <oklopol> i watched it yesterday, but it was in spanish
03:42:44 <GregorR-L> It's about OH MY GOD WHAT THE FUCK THIS MOVIE MAKES NO SENSE AHHHHHHHHHHHHHHHHHHH
03:43:13 <oklopol> they make a nice conversation starter
03:43:19 <pikhq> I said I *don't* need to know about it.
03:43:47 <oklopol> GregorR-L: that's what i thought, but i thought it was the language
03:44:29 <oklopol> i watched hypercube in french 3 times before i found the english one
03:45:11 <oklopol> it's great, just a bunch of rooms, traps and people killing each other and talking nonsense
03:45:50 <oklopol> i liked the english one too, but it lacked the atmosphere
04:49:58 -!- ihope has quit (Connection timed out).
05:12:17 -!- boily has joined.
05:33:08 -!- GreaseMonkey has joined.
05:49:52 -!- immibis has joined.
05:58:58 -!- Sgeo has quit (Remote closed the connection).
07:16:08 -!- boily has quit ("WeeChat 0.2.5").
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:22 <GregorR-L> RodgerTheGreat: Sorry to tell you this, but after that one positive, I've had three people tell me the avatar is scary :P
08:10:01 -!- toBogE has joined.
08:17:27 -!- toBogE has quit (Read error: 104 (Connection reset by peer)).
08:27:35 -!- GregorR-L has quit ("Leaving").
09:22:59 -!- immibis has quit ("Always try to be modest, and be proud about it!").
09:23:14 -!- immibis has joined.
09:29:15 <Sukoshi`> I think ... *gasp* wait for it .... I think ... I'm switching to Scheme over CL.
09:59:15 -!- immibis has quit ("Now if you will excuse me, I have a giant ball of oil to throw out my window").
10:35:22 -!- GreaseMonkey has quit ("three word story: http://greasemonkey.nonlogic.org/mpnp/index.php?docname=three-word-story").
13:16:07 -!- RedDak has joined.
13:56:49 -!- puzzlet has joined.
15:11:26 -!- RedDak has quit (Read error: 104 (Connection reset by peer)).
15:38:07 -!- calamari has joined.
15:40:20 -!- calamari has quit (Client Quit).
16:08:54 -!- sebbu has joined.
16:33:30 * GregorR tries to interpret the meaning of "uncanny valley" and fails :P
16:39:40 <RodgerTheGreat> but those three areas I mentioned are chiefly where it comes into consideration
16:45:57 -!- ehird` has joined.
16:51:54 -!- blahbot` has joined.
16:59:30 -!- ihope__ has joined.
16:59:53 -!- ihope__ has changed nick to ihope.
17:08:04 <Sukoshi`> http://programming.reddit.com/info/17l64/comments/c17oew <-- Thoughts everyone?
17:08:27 <Sukoshi`> Everyone who knows/has experienced CL/Scheme anyways.
17:09:59 <RodgerTheGreat> well, I'm a Nonlogician, so my stance ought to be pretty clear on prebuilt libraries
17:10:42 <Sukoshi`> Well, that's a post meant primarily for application coders. I'm asking whether you *think* CL has more NIH than Scheme does.
17:10:50 <ihope> I think I've experienced CL!
17:11:05 <Sukoshi`> Go back to your Haskell, Bulb!
17:11:13 <ihope> Not quite sure, though.
17:13:11 <Sukoshi`> Wow. The Chicken scene has grown quite a bit since I last visited it.
17:14:30 -!- sebbu2 has joined.
17:16:46 <Sukoshi`> I just wished SLIME worked for Scheme :(
17:33:50 -!- sebbu has quit (Connection timed out).
17:33:50 -!- sebbu2 has changed nick to sebbu.
17:36:44 -!- pikhq has quit (Read error: 110 (Connection timed out)).
17:42:18 -!- pikhq has joined.
17:51:40 <ehird`> slime is for cl only..
17:54:59 <Sukoshi`> Didja read the article, ehird` ?
17:55:31 <Sukoshi`> Hm. It shouldn't be too hard to port the easier parts of SWANK (i.e. zooming functions to the REPL, zooming files to the REPL, etc.)
17:56:52 <Sukoshi`> Wow. I just did a 45 minute shower.
17:59:16 <Sukoshi`> Various forms of self-maintenance.
17:59:32 <oklofok> i do love water particles in large amounts, like rain
17:59:45 <oklofok> shower's just quite pathetic compared to that
18:00:01 <oklofok> ah maintenance, you mean like cleaning yourself and shit?
18:00:56 <Sukoshi`> Well, off to my 2 hour waste of tim... errr Art History course o/
18:01:31 <oklofok> i thought this was the start of a very fruitful conversation
18:02:11 <oklofok> i actually couldn't do one of the exercises in this book :|
18:02:48 <oklofok> it was a math exercise assuming i know how to prove by induction, which i never bothered to learn
18:29:23 <SimonRC> people are getting flooded in England.
18:29:45 <SimonRC> Apparently the Jet Stream has moved due to El Niño.
18:30:39 <oklofok> (that's spanish for ninja)
18:33:07 <ehird`> Hmm.. Is there a blend of OOP that uses generic functions, not messages, but is prototype-based?
18:34:51 <SimonRC> you can build it on top of any OO system with a customisable dispatch policy
18:35:01 <oklofok> (well, perl is anything-based, so that's cheating)
18:35:13 <SimonRC> e.g. Lisp ones, the funny Python one, and Factor's one
18:37:37 -!- pikhq_ has joined.
18:38:37 -!- oerjan has joined.
18:38:37 <oklofok> i guess is what i was looking for
18:40:18 <SimonRC> (Python's multimethods are the explanation for why __call__() rocks.)
18:41:45 -!- jix has joined.
18:42:47 -!- pikhq has quit (Read error: 110 (Connection timed out)).
19:08:14 <oklofok> errr how do i do 5**3 in scheme?
19:17:07 <oklofok> err true, i can do integer exponentiation easily... but i can't say i can remember how to do real number exponents just like that
19:17:27 <oklofok> i guess i could if i played a bit with e^n
19:18:38 <oklofok> oerjan: thank you for showing me i was an idiot in not seeing that right away :P
19:18:58 <oklofok> (also real thanks for telling me that)
19:19:58 <oklofok> i thought it'd need something sick and yellow
19:28:15 <oerjan> it may need something sick if a is negative (namely complex numbers)
19:30:26 <oerjan> and if it is zero you will have to be vewy vewy caweful
19:31:34 <oerjan> because it is only defined if b is an integer >= 0, preferably non-zero
19:32:29 <oerjan> b integer is also a good idea if a is negative, unless you want to consider complex branches
19:33:28 <oerjan> if a is positive that is a reasonable assumption
19:34:06 <oklofok> (-1)^b = (e^(i*pi))^pi = (e^(i*pi^2))... and i have no idea what i'm going for.
19:34:27 <oerjan> however, you still need to carefully know what you are doing with complex branches
19:34:28 <oklofok> how would that conversion possibly be useful...
19:34:44 <oklofok> i do? doesn't (-a)^b = (-1)^b * a^b apply with complex numbers?
19:35:09 <oklofok> -a = -1*a even with complexes... no?
19:35:29 <oerjan> the problem is that there are multiple values, and you cannot always choose them consistently
19:35:31 <oklofok> (a*b)^n = a^n * b^n... or? if it is, also with complexes?
19:35:59 <oerjan> technically, e^(2*pi*n*i) = 1 for all integers n
19:36:14 <oerjan> so any log can have 2*pi*n*i added to it
19:37:05 <oklofok> (getting process on the go)
19:37:30 <oerjan> this happens to give multiple values for a^b if b is not an integer.
19:37:53 <oklofok> so... why did your equation give just one?
19:38:31 <oerjan> log is multivalued too
19:38:47 <oerjan> if you consider it as the inverse of exp on complex numbers
19:39:16 <oklofok> so... basically for a>0 you give just the real answer because it's the most obvious one, but when a<0, none of the results is real, so you can't choose a *right* answer?
19:40:17 <oerjan> more or less. you choose a "branch" of the log function, but there is no way of doing that such that (a1*a2)^b = a1^b * a2^b always holds.
19:41:57 <oerjan> for example, ((-1)^b)^2 cannot be 1 always
19:42:44 <oklofok> i mean, for non integers it won't?
19:43:17 <oerjan> lessee, (-1)^b = exp(b*pi*i*(2*n+1)) for some n
19:43:30 <oklofok> anyway, as interesting as this is, i was actually just looking for the expt function in lisp ;)
19:43:39 <oklofok> but i'm not in a hurry if you wanna explain
19:43:53 <oerjan> now, if b is irrational that has no chance of being 1
19:44:15 <oerjan> for a fraction you might get "lucky"
19:45:12 <oklofok> does that fraction contain complex numbers?
19:45:22 <oklofok> or is it the fraction of complex numbers that are reals?
19:45:31 <oerjan> i mean a fraction of two integers
19:45:54 <oerjan> i.e. a rational number
19:46:13 <oklofok> err yes, i actually happened to know what it is :P
19:46:26 <oerjan> basically you would need b = m/(2*n+1)
19:48:12 <oerjan> well, back to your lisp :)
19:49:22 <oklofok> was nice to have something complex to think about for a minute.
19:50:08 <oklofok> the book is kinda noobity noob (though i'm in the first part)
19:50:08 <oerjan> always stretch your brain daily, eh?
19:51:00 <oerjan> is it scheme or common lisp (or even emacs lisp)?
19:51:36 <oerjan> for scheme at least, i am pretty sure the whole language definition is online.
19:51:53 <oerjan> if you need to look something up.
19:52:37 <oerjan> common lisp is one of those pesky standards that sometimes ask you to pay for a copy, i think.
19:52:55 <oerjan> apart from being huge, i have heard.
19:53:10 <oerjan> but now i am monologuing, just like you :)
19:54:52 <oklofok> i'd like a documentation, but couldn't find a pretty one
19:55:39 <oerjan> http://schemers.org/Documents/Standards/R5RS/
19:56:34 <oerjan> there actually is another standard for scheme which _is_ such a pesky pay-for one
19:56:49 <oerjan> but most people use the free one
19:59:12 <oklofok> (let ((a (lambda (b) b))) (a 4)) <<< why isn't that 4?
20:00:15 <oklofok> ask @ #scheme? good idea, einstein. *clap*
20:00:36 <oerjan> when i tested it with guile
20:01:00 <oerjan> you have a broken interpreter?
20:01:03 <oklofok> i see. i'm using DrScheme :)
20:01:24 <oerjan> hm, i think i have heard good things about that one
20:02:20 <oerjan> i also vaguely recall it has different levels you can set, and some features are disabled at lower levels
20:02:47 <oerjan> but it would be strange if it made allowed programs actually behave _differently_.
20:03:19 <oerjan> what does that expression give you? and is that the actual expression you tried?
20:04:01 <lament> blah, apparently the spanish harry potter won't be out till 2008
20:04:57 <oklofok> > (let ((a (lambda (b) b))) (a 4))
20:05:43 <oerjan> that would explain it.
20:06:17 <ehird`> (force (delay X)) -> X
20:06:25 <ehird`> if lazy scheme, then X -> (delay X) for most X
20:07:00 <oklofok> how do i force that to evaluate ?
20:07:16 <ehird`> why on earth do you want lazy scheme
20:07:20 <ehird`> nobody uses lazy scheme
20:07:30 <lament> how DARE you use lazy scheme
20:07:30 <oerjan> unless you actually want to try out laziness, turn it off
20:07:35 <lament> stop using lazy scheme RIGHT NOW!
20:07:38 <oklofok> (force X) doesn't work, i tried that already
20:07:48 <oerjan> i sense a certain animosity :)
20:08:06 <SimonRC> laziness should only be used in pure things
20:08:12 * ehird` burns lazy scheme at the stake er, steak
20:08:23 <oklofok> i now realize what my problem is in not finding the normal scheme interpreter in the options.
20:08:36 <oklofok> i'm just looking at the options menu's lower part.
20:08:45 <lament> why would you use scheme? nobody uses scheme!
20:08:46 <oklofok> what's up with my brain, really.
20:08:57 <ehird`> why would you breathe? nobody breathes!
20:09:10 <oerjan> you need more complex numbers!
20:09:29 <SimonRC> if you want complex numbers check out Haskell.
20:09:42 <SimonRC> You need a PhD to understand its numerical types
20:09:49 <oklofok> oerjan: is go tc if you interpret it as a multiway system?
20:10:04 <ihope> oklofok: a multiway system?
20:10:18 <oerjan> what is a multiway system?
20:10:27 <ehird`> has anyone written "hello world" in wapr yet?
20:10:39 <ehird`> (aka "jumping to -1 is exciting")
20:10:41 -!- pikhq_ has changed nick to pikhq.
20:10:49 <oerjan> SimonRC: but Complex Integer is not allowed in Haskell :(
20:11:16 <ehird`> if nobody knows what wapr is, then http://esolangs.org/wiki/Jumping_to_-1_is_exciting may help
20:11:24 <ehird`> %wapr code makes blahbot run it
20:11:24 <blahbot`> [61, 73, 62, 63, 71, 59, 69, 63, 77, 60, 70, 59, 66, 60, 73, 78, 76, 79, 72, 67, 78]
20:11:45 <oerjan> i have read that go with arbitrary (finite) board size is PSPACE-complete.
20:11:45 <oklofok> multiway... you have multiple rules and you apply all of them, each in a separate thread
20:13:14 <ihope> oerjan: er, it isn't?
20:13:16 <oklofok> because i think go with an infinite board is semidecidable
20:13:24 * SimonRC fails to understand the # command
20:13:37 <ihope> ...yay, vagueness.
20:13:42 <ihope> oerjan: Complex Integer isn't allowed?
20:13:42 <SimonRC> oerjan: of course. You must be able to divide complex numbers
20:13:55 <ehird`> SimonRC, read the stack description careully
20:14:04 <oklofok> oerjan: i assume you don't know if you didn't know what a multiway system is... unless that's just not the right term
20:14:15 <ehird`> if i tell you # is called "grab" or "pull" that might help too
20:14:18 <oerjan> SimonRC: they should be gaussian integers
20:14:34 <SimonRC> ok, what happens if the TOS is 0 0?
20:14:39 <oklofok> go and chess iirc are both semidecidable on an infinite board... does that automatically make them tc?
20:15:07 <ehird`> if s = [0, 0], error. if s = [X, 0, 0] then [X]
20:15:19 <SimonRC> why are there 2 secial integers rather than one?
20:15:31 <oerjan> ihope: Complex is defined with RealFloat restriction on the elements.
20:15:32 <ehird`> SimonRC, because one is source and one is dest.
20:15:42 <ehird`> here, this is the implementation of it:
20:15:42 <ihope> Make a new Complex, then?
20:16:09 <oerjan> sure, it's just an annoying consequence of Haskell's numerical class hierarchy
20:16:12 <SimonRC> ihope: you can do that just fine
20:16:21 <SimonRC> Haskell numbers are screwed up
20:16:46 <ihope> Screwed up in that it doesn't automatically convert from one to anotheR?
20:16:58 <ehird`> http://pastie.textmate.org/81503
20:17:10 <SimonRC> screwed up in that it is hard to understand
20:17:14 <oerjan> oklofok: tc or _weaker_, i assume
20:17:23 <ihope> What makes them hard to understand?
20:17:55 <SimonRC> and the only thing the Real class ads to its subclasses is a function to convert to a Rational number, which is the main thing that Real numbers *cannot* do
20:18:09 <SimonRC> ihope: there are too many for a start
20:18:16 <SimonRC> o at least, there are lots
20:18:25 <SimonRC> there are enough that I can;t remember them all
20:19:00 <oerjan> ihope: the class hierarchy is based on generalizing the number formats used in computers, rather than on sound mathematical structures
20:19:29 <oerjan> so when you try to _add_ a well-known sound mathematical structure, you cannot fit it in properly.
20:19:34 <ehird`> SimonRC, do you understand now :)
20:19:45 <SimonRC> you need a couple more examples
20:19:59 <ehird`> # is basically the Turing operator, being that it makes it TC without me having to implement long things like DROP, SWAP, ROT, etc.
20:20:10 <ehird`> and there are no examples because it's pretty much not known
20:20:26 <SimonRC> it could get very weird in some cases
20:20:42 <ehird`> also, the choice of "other characters" is not arbitary - it's just to keep it in the displayable chars range, and it starts at 0 one place after the commands
20:20:48 <ehird`> (The commands are from the start of displayable ascii)
20:21:09 <oerjan> although actually, basing it on sound mathematical structures would probably mean even _more_ classes, so harder for anyone but mathematicians.
20:21:18 <ehird`> SimonRC, hey, just spilling everything about it :P
20:21:32 <SimonRC> oerjan: numbers are a PITA
20:21:58 <ehird`> the interp behaves weirdly on undefined situations like that
20:22:34 <ehird`> that's, uh, the exit program
20:22:43 <oerjan> the current system is probably a please-noone compromise
20:22:58 <ehird`> that's.. the infinite loop
20:23:19 <ehird`> just %wapr doesn't match the regexp...
20:23:22 <ehird`> it needs an argument ;)
20:24:11 <ehird`> you must've used - yeah, thought something like that
20:25:03 <SimonRC> it thinks that underscore is nothing
20:25:20 <ehird`> and, /me checks logs - so does the bot
20:25:49 <ehird`> well that sort of thing is undefined behaviour
20:26:22 <ehird`> it's displayable ascii, and operations fitting the stack pictures in the spec, that is defined
20:26:25 <ehird`> everything else, can go boom
20:26:42 <ehird`> my implementation does no error checking, so it does go boom :)
20:26:59 <ehird`> i mean it has all the properties of a TC language
20:27:08 <lament> my girlfriend found my esolangs userpage.
20:27:11 <oerjan> SimonRC: this channel censors color codes, only you see the ones you write
20:27:18 <oerjan> lament can turn them off
20:28:29 <oerjan> lament: well now you will _really_ find out if you are made for each other
20:28:32 <SimonRC> ehird`: I think you have a bug.
20:28:41 <SimonRC> stack.insert_at(stack[d], stack[-s])
20:28:47 <SimonRC> stack.insert_at(stack[-d], stack[-s])
20:28:54 <ehird`> SimonRC, hmm yes you are right
20:29:28 <ehird`> i do so love %reload, %quit, alt-tab, up, enter was so tedious :P
20:30:17 <ihope> ehird`: use more semicolons.
20:30:33 <ehird`> yeah i probably should
20:30:34 <ihope> One more, that is. Or a period.
20:30:38 <ehird`> i do so love %reload; %quit, alt-tab, up, enter was so tedious :P
20:35:38 <ehird`> SimonRC, made a quine yet? =P
20:36:29 <ehird`> hmm.. a quine would be very hard to make in wapr
20:37:43 <bsmntbombdood> i was bored, so i'm writing a numeric lib that uses only lists
20:39:03 <ehird`> might have some problems with -1 :)
20:39:09 <pikhq> Why not do it via functions?
20:39:24 <pikhq> I know a possible way to do -1. . .
20:39:35 <pikhq> Represent the sign as a list containing a list.
20:39:55 <bsmntbombdood> you can get integers from natural numbers with a pair of them, (a, b) where the number is a - b
20:40:37 <pikhq> And fractional? Seperate the integer portion from the fractional via, say, ((())).
20:40:50 <pikhq> 1.1 = (() ((())) ())
20:41:01 <ehird`> pikhq, Hey! I was going to say that, ffs!
20:41:04 <ehird`> You stole my thoughts!
20:41:11 <bsmntbombdood> rationals are a pair of integers, (a, b) where the number is a/b
20:41:13 <pikhq> I'm saying that that's one way to represent it, not that that's how you do it now. ;)
20:41:22 <pikhq> bsmntbombdood: That's another way to do it. . .
20:41:45 <ehird`> bsmntbombdood, lisp i assume
20:41:49 <oerjan> unary? it could be binary even if it's lists
20:42:07 <pikhq> Of course, you could use Church numerals.
20:42:16 <ehird`> assuming all you can do is (x y ...) and ()
20:43:42 <oerjan> just a matter of choosing a coding
20:44:53 <ehird`> bsmntbombdood, what have you got done now?
20:46:58 <ehird`> i'm copying your idea :)
20:47:02 <ehird`> i have succ, pred, add, and subt
20:49:37 <ehird`> mult([[]], [[], []]) => [[], [], [], []]
20:49:40 <ehird`> i think i have work to do
20:50:17 <ehird`> yes, i already have that =P
20:50:55 <ehird`> ah i had a stupid mistake there
20:51:11 <oklofok> those are trivial, division needs some thought
20:51:35 <oklofok> especially if you refuse to use another representation :)
20:52:06 <ehird`> how about... BRUTEFORCE DIVISION!
20:52:12 <ehird`> tries random numbers until it gets the right answer
20:52:37 <oklofok> well, convert to a binary church and it's easy
20:52:42 <ehird`> undefined wapr code is insane =P
20:52:53 <ehird`> binary church integer? there's no such thing
20:53:10 <oklofok> bsmntbombdood: that is thought on the 3rd grade
20:53:18 <oklofok> except it's easier to implement in biary
20:54:34 <oklofok> by this i do not mean it's trivial, but that you should be able to deduce it from the 10 base division i assume you *can* do manually.
20:55:58 <oklofok> what's do you base that on?
20:56:24 <oklofok> well, it is, if you only do integer
20:57:01 <oklofok> yes, sorry, it is actually unbelievably trivial
20:57:06 <oerjan> indeed, if you are doing unary numbers, you cannot do division more efficiently than subtracting anyhow
20:57:37 <oklofok> yes, unless you have a way to chain operations and let them have common internal presentation :)
20:58:24 <oklofok> bsmntbombdood: a/b: a-b, counter+1, if a=0, return counter-1, otherwise recurse
20:59:09 <oklofok> well anyway, it's a few details over that
21:02:08 <oklofok> the reason i thought division is hard is i've done it in brainfuck having negative values illegal... but subtraction without underflow is much easier if you use... whatever you use
21:05:24 <oerjan> <ihope> Game of Life where you can modify the board a little bit when it's your turn? :-P
21:05:47 * oerjan has used to play that, solitaire
21:06:31 <oklofok> i wish i had irl friends like you
21:06:43 <oklofok> err... actually, i'm pretty sure most of my friends would play that
21:07:21 <ehird`> what i think would be a cool board game
21:07:27 <ehird`> is, a finite game of life grid, wrapping
21:07:35 <ehird`> but, two colours of alive cell
21:07:44 <oklofok> yeah, was just gonna say that
21:07:49 <ehird`> and another rule, for "capturing" - capturing white -> black
21:07:52 <ehird`> capturing black -> white
21:07:52 <oklofok> it's just you have to modify the rules a bit
21:07:58 <ehird`> and, some form of movement
21:08:06 <oerjan> and the Day and Life automaton - trying to stabilize a chaotic pattern using only one change per turn.
21:08:09 <ehird`> so you have to set up an initial pattern, move well, and avoid dying, and try to capture
21:08:38 <oklofok> bsmntbombdood: you think you manage division with 2 variables in unary?
21:08:53 <oerjan> ehird`: i am pretty sure i have played such a game online
21:09:04 <oerjan> or on the computer, anyhow
21:09:18 <ehird`> oerjan, the only problem would be having to work out the grid manually each turn
21:09:43 <oklofok> bsmntbombdood: modulo can obviously be made with 2 vars at least
21:10:02 <oerjan> i think it was on my father's computer, actually, in some puzzle pack
21:13:19 <bsmntbombdood> but i don't think there's a simple way to compute inverses
21:16:03 <oerjan> especially if you only have integers.
21:18:00 <ihope> Two-register Minsky machine?
21:19:55 <oerjan> you mean "is Turing complete"
21:22:04 <oklofok> a/(b+1), then use that one technique the name of which i only know in finish
21:22:53 <oerjan> what's the finnish name?
21:23:21 <oklofok> division into partial factors or something.
21:23:53 <oklofok> find x and y such that xa/1 + ya/b = a/(b+1)
21:24:36 <oerjan> i had hoped it would be a translation of something
21:24:52 <oklofok> nah, those are all native words
21:25:10 <oerjan> i would imagine so, unless there is Japanese in it :D
21:26:07 <oklofok> part = osa, murto = fraction (somewhat), tekijihin = into fractions, jako = division
21:26:58 <oklofok> seems it was too hard for me to remember the pattern english=finnish throughout a dictionary of 4 words.
21:27:41 <oklofok> http://www.answers.com/partial%20fraction%20decomposition
21:27:58 <oerjan> sounds like "delbrkoppspalting"
21:27:58 <oklofok> find x and y such that x/b + y/1 = a/(b+1)
21:28:22 <oklofok> i didn't recognize the concept from wikipedia's explanation at first
21:30:19 <oerjan> wait a minute, that looks strange.
21:31:39 <oerjan> somehow that article seems a bit too advanced for this context
21:31:42 <oklofok> the denominator must of course be a multiplication
21:31:53 <oklofok> so that doesn't help at all
21:31:55 <oerjan> yeah, that's what was strange
21:32:24 <oerjan> at least you recognized before i said what it was :)
21:32:41 <oklofok> actually i think it's even easier to do a/(b+1) -> something...
21:32:50 <oklofok> don't know what exactly, though
21:33:08 <oerjan> what if b+1 is a prime?
21:33:50 <oklofok> (ask me when you need more wrong information presented obscurely!)
21:33:54 <oerjan> in fact doing it may be as hard as factorization
21:34:51 <oerjan> if you want more than just taking the integer part
21:35:14 <oerjan> but then that was what we wanted in the first place, i think, so this is circular.
21:36:00 <oklofok> well, we know how to take a modulo, doesn't factorization only need that?
21:36:27 <oerjan> oklofok: tell me about the time cube. i have heard about it but never looked at more than the front page of the site, i think.
21:36:59 <ehird`> basically: nuts, more nuts, lala, daaadeee, TIME CUBE!
21:37:05 <ehird`> http://www.timecube.com/
21:38:02 <oklofok> oerjan: if you meant that, then you already knew more than me ;)
21:38:50 <oerjan> but you are the expert on presenting wrong information obscurely! you must know this!
21:39:23 <oklofok> so there's this scene where these guys go into one of the rooms
21:39:38 <oklofok> it goes like million times faster
21:40:03 <oklofok> so what happens? do the guys chill there for 5 hours of their time and come back having been in the room for a second?
21:40:14 <oklofok> they stay there for their whole rest of their lives.
21:40:23 <oklofok> and we can watch them die of age.
21:40:41 <oklofok> there you have it, wrong information about *timecube*
21:41:01 <ehird`> i used to think when i was a kid that if you went into a particle accellerator it'd be like those nature programs where they show a plant growing really fast, you'd see a beard sprouting out of you and you getting taller in a few seconds, then you'd die
21:41:23 <oerjan> reminds me of a chapter from the Books of Magic series
21:41:54 <oklofok> actually, i did that same error when imagining time acceleration when i was a child
21:42:25 <oklofok> i remember explaining my friend about this room i had been in where time went really fast in the other end and stopped at the other
21:42:51 <oklofok> and if you went to the other end, you died instantly, in the other end you'd just be stuck forever when time stopped
21:43:00 <Sukoshi`> Do you append a CGI request to the end of a GET in an HTTP request?
21:43:01 <oklofok> god i was a noob when i was 4
21:43:47 <lament> oklofok: sounds exactly like what happens in a room positioned next to the edge of a black hole
21:43:58 <Sukoshi`> ehird`: Does Chicken have anything like LOOP or ITERATE?
21:44:15 <ehird`> recurse! or define it yourself
21:44:15 <Sukoshi`> (Not that I'm doing Scheme hackery now, that's for later today, but.)
21:44:20 <oklofok> lament: my friend didn't ask how that was possible, so i didn't need to explain him that
21:44:26 <Sukoshi`> ehird`: I don't want to recurse :P
21:44:31 <ehird`> Sukoshi`, LOOP recurses.
21:44:39 <oklofok> okay, perhaps i'm being a noob now and that would actually happen.
21:44:49 <Sukoshi`> Really?! I thought it was a wrapper for DO.
21:45:29 <Sukoshi`> Because Scheme48 has REDUCE and ITERATE, and I was wondering if anyone had lifted the ITERATE code from Scheme48.
21:45:45 <ehird`> maybe it does have iterate then
21:46:06 <Sukoshi`> If Scheme48's ITERATE is written in plain Scheme, it should be easy to lift.
21:46:46 <ihope> Incoming (possibly).
21:46:48 <Sukoshi`> But anyways, back to CGI queries -- can you just append ``?param=value'' to the end of the GET request in the HTTP requests?
21:47:08 <Sukoshi`> I'm too lazy to do my own telnet analysis :P
21:48:55 <ehird`> ?x=a?z=y is X => "a?z=y"
21:52:12 <oklofok> Sukoshi`: basically you have ? start get parameters, and you separate them with &
21:54:24 <oklofok> i've been debugging a code for 2 hours, it was right on the first try, i just thought it should produce e, while in fact it produced e-2
21:55:05 <oklofok> i was starting to believe my brains were playing games with me
21:55:41 <oerjan> it is. you have now gone back to being deluded.
21:56:55 <ehird`> pikhq, ping to you too?
21:57:04 <oklofok> i just didn't check the exercise, because *i already read it*
21:57:59 <Sukoshi`> Why are you pinging me, pikhq ?
21:58:26 <pikhq> Trying to avoid a disconnection by timeout. . .
21:58:38 <pikhq> That'd be the *smart* thing to do.
21:59:04 <ehird`> why would you time out?
22:00:03 -!- ehird` has quit (Read error: 104 (Connection reset by peer)).
22:00:04 -!- blahbot` has quit (Remote closed the connection).
22:00:54 <oerjan> the disconnection by timeout is based on PING/PONG messages between you and your directly connected irc server, i think
22:01:07 <oerjan> pinging anything else may not help at all
22:01:59 <lament> he's just pinging you, he has a crush on you!
22:02:08 <oerjan> and unless you are using telnet, your client should do it automatically
22:06:44 <oklofok> hmm, that 'aha' may have had something to do with what i said to her.
22:08:00 <oerjan> must be that Mars/Venus thing.
22:11:42 <lament> men are from mars, women are from venus, esoteric programmers are from deep under the surface of europa
22:12:38 <oerjan> actually, i am from ganymede, but that may be just me.
22:13:42 <bsmntbombdood> In Greek mythology, Ganymede, or Ganymedes (Greek: Γανυμήδης, Ganumēdēs) is a divine hero whose homeland was the Troad.
22:19:11 * SimonRC points out that the theory of division of integers is generally considered to be a PITA.
22:20:28 <SimonRC> well, sbtraction can be defined in terms of an additiv inverse
22:20:52 <SimonRC> that doesn't work so well for division of integers
22:21:12 <SimonRC> you also need to decide what to do with fractions.
22:21:18 <SimonRC> and what to do in the negative case
22:21:34 <SimonRC> and what is the multiplicative inverse of 2?
22:22:08 <bsmntbombdood> subtraction isn't closed over the natural numbers either, but it's still easy
22:23:47 <lament> are you trying to do integer divisions without involving fractions or something?
22:23:57 <SimonRC> but adding with negatice numbers involved is quite easy...
22:24:22 <SimonRC> OTOH adding with fractions involved is icky
22:24:38 <SimonRC> and you can't divide by 0 at all even in the reals
22:24:55 <SimonRC> Division is just messier than subtraction
22:24:58 <lament> SimonRC: easy/icky is a matter of taste :)
22:25:27 <bsmntbombdood> the algorithm for addition is almost exactly the same as for subtraction, which makes me wonder why division is so much more complicated than multiplication
22:25:58 <lament> bsmntbombdood: it's not
22:26:05 <lament> bsmntbombdood: it's exactly the same
22:26:23 <oklofok> SimonRC: for negation you need negative numbers, for inverse you need real numbers.
22:26:30 <SimonRC> My favourite integer division system is Haskell's. It has div/mod *and* quot/rem. Most languages only provide the latter, except C and C++, which provide one of them, probably.
22:26:47 <oklofok> you can't do subtraction with positive integers using the inverse either
22:26:57 <lament> bsmntbombdood: 123/456 is the same as 123*(1/456) :)
22:27:31 <SimonRC> "inverse" is a more general term
22:27:41 <lament> bsmntbombdood: so division and multiplication are equally hard because one is easily expressed in terms of the other :)
22:28:05 <oklofok> bsmntbombdood: why can't you carry a counter with you but you can use O(n) memory?
22:28:08 <lament> bsmntbombdood: unless you're talking about _integer_ division, in which case the analogy with addition/subtraction doesn't hold
22:29:02 <bsmntbombdood> oklofok: that can be made tail recursive easily, i just didn't bother
22:29:09 <lament> bsmntbombdood: and you can't subtract naturals very well either, what's 3-8?
22:29:21 <oklofok> bsmntbombdood: without an accumulator?
22:30:04 <oklofok> then division is even more trivial, anyway
22:30:18 <SimonRC> With division in the integers, you lose information, unlike for subtraction in the naturals.
22:30:26 -!- jix has quit ("CommandQ").
22:31:01 <oklofok> SimonRC: no, it's undefined, not information losing.
22:31:27 <SimonRC> oklofok: ah, inthat case the problem is slightly different
22:31:47 <oklofok> if you wish to have undefined things be *approximated* correct, yes, you lose information, but you will as well if you do subtraction without negative numbers
22:31:54 <SimonRC> People seem to want to be able to divide integers, even though it is undefined.
22:32:26 <oklofok> people also want square root with reals even though it's sometimes undefined.
22:33:22 <SimonRC> "20:50:11 < oklofok> those are trivial, division needs some thought"
22:34:47 <SimonRC> I now remember that I was talking about generalised fields (or are they rings)?
22:35:00 <SimonRC> things like (mod 7) with + - *
22:35:25 <SimonRC> additive inverses always exist but division inverses only sometimes exist
22:35:49 <SimonRC> you can divide reliably by anything that isn't a factor of the group size, except 0
22:36:04 <oklofok> (define (div a b) ((church-less-than a b) nil) (else (+ 1 (div (sub a b) b)))))
22:36:11 <oklofok> bsmntbombdood: is that anywhere near?
22:36:43 <oerjan> (mod 7) is a field, (mod non-prime) is not
22:36:45 <oklofok> SimonRC: i say a lot of things.
22:37:40 <oklofok> (define (div a b) ((church-less-than a b) nil) (else (inc (div (sub a b) b)))))
22:38:27 * SimonRC indicates the paper "Every number has at most two digits" if you are interested in better representations of numbers.
22:39:00 <oklofok> err... you mean like every number is the sum of two primes?
22:39:16 <oklofok> which isn't proven or something
22:40:42 <SimonRC> and it tells you a great way to represent numbers based on primitive dtastructures
22:51:25 <oklofok> oerjan: does goldbach say every number is the sum of *exactly* 2 primes?
22:52:59 <oklofok> bsmntbombdood: i didn't know what to do at division by zero
22:53:28 <oklofok> but your subtraction doesn't have error correction so i just let it do what it does
22:53:46 <oerjan> "The conjecture that all odd numbers greater than 7 are the sum of three odd primes is called the "weak" Goldbach conjecture"
22:53:54 <oklofok> subtraction might work even if you do 5-8
22:54:01 <oklofok> i don't know much about lists
22:54:17 <oklofok> i've just learned about numbers and functions.
22:54:31 <oerjan> oklofok: if every even number is the sum of two primes, then you can just add 3 to get every odd integer
22:54:54 <oklofok> yay, then that's a number representation
22:55:08 <oerjan> http://en.wikipedia.org/wiki/Goldbach's_conjecture
22:57:41 <oerjan> rather the opposite, there as so many that it is almost statistically certain to be true
22:58:05 <oklofok> rather the opposite of what?
22:58:05 <oerjan> but still there is no actual proof
22:58:21 <oklofok> it's so hard to understand anything.
22:58:46 <oerjan> See the diagram under "Heuristic justification"
23:00:18 <oklofok> heh, looks like a pretty static growth :P
23:00:39 <oklofok> perhaps the mathematicians haven't noticed that
23:05:22 <oklofok> i think it's more likely than there being a number that can't be expressed as two primes.
23:06:58 <ihope> Wait, did somebody say better representations of numbers?
23:07:02 <oklofok> like... how unlikely is that?
23:07:22 <ihope> "Every number has at most two digits"... hmm. I'll have to check that out sometime.
23:07:28 * ihope slips it into his paper queue
23:07:46 <oklofok> link it to me if it's not the one google gives.
23:11:25 <ihope> Hmm, I want a proof that no heuristic arguments are misleading.
23:11:39 <ihope> (What's the opposite of "misleading"?)
23:13:08 <oerjan> since there probably _are_ heuristic arguments that are misleading, i doubt you'll get a proof
23:13:35 <lament> in general, proving false stuff is difficult
23:13:36 <oerjan> such as things with enormous smallest counterexamples
23:14:04 <oerjan> but _if_ you can manage to prove some false stuff correctly, you will be pretty famous.
23:14:20 <oerjan> since you will have shown inconsistence of the foundations.
23:15:32 <lament> can there be proof of consistence of the foundations?
23:16:02 <oerjan> unless, of course, they are actually inconsistent.
23:16:27 <lament> i don't think godel's theorem says any such thing...
23:16:31 <oerjan> or you use even stronger foundations to prove it
23:18:17 <oerjan> Godel's theorem: if a system include enough of the theory of basic arithmetic, then it cannot prove the encoded theorem of its own consistence, unless it is in fact inconsistent.
23:18:59 <ihope> lament: you mean proving nonexistence is difficult?
23:19:05 <ihope> ...er, never mind.
23:19:15 <ihope> I thought you meant proving stuff to be false.
23:19:20 <lament> oerjan: i have never heard it stated that way, but wiki seems to vague agree
23:19:48 <oerjan> (that part before the comma is my handwaving over the actual technical requirements, which i don't quite remember)
23:20:07 <ihope> It must also include certain truths about consistency.
23:20:13 <oklofok> lament: what would it say then if not that?
23:20:49 <oerjan> but first order predicate logic and some peano arithmetic axioms are certainly enough
23:21:07 <ihope> Consider some Turing machine axioms plus "this theory is consistent". As long as the Turing machine axioms don't mention consistency and are consistent, the resulting theory is consistent.
23:21:12 -!- RedDak has joined.
23:21:25 <lament> oklofok: "a complex system can't be both consistent and complete"
23:21:40 <oerjan> note that there are several Godel's theorems
23:21:46 <SimonRC> A Hungarian spelt Erdős as Erdös. I mentioned that I thought it was spelt Erdős. Am I a bad person for correcting someone on spelling his native language when he actually did spell worngly?
23:22:25 <lament> SimonRC: perhaps he spelled it that way because of technical difficulties.
23:22:30 <GregorR> SimonRC: It's spelled "wrongly"
23:22:48 <oerjan> lament: the encoded sentence of its own consistency is what is used to prove your statement
23:23:25 * oerjan cannot actually see which characters you used, but i assume it's with long and short "
23:23:40 <ihope> oerjan: something like that.
23:23:47 <ihope> Umlaut and double acute accent, or some such.
23:23:50 <oerjan> o with long " is the correct spelling
23:24:27 <oerjan> SimonRC: btw i have done so to english native speakers myself.
23:25:37 <oklofok> doesn't work on this channel though.
23:25:52 <lament> native speakers tend to speak their language horribly :)
23:25:58 -!- sebbu has quit ("@+").
23:26:35 <lament> everybody makes mistakes unless they're being terribly formal
23:26:48 <lament> but there's a difference in the quality of mistakes made by native and non-native speakers
23:26:56 <lament> you can usually tell them apart
23:27:37 <SimonRC> I did not even know he was hungarian until he told me.
23:27:46 <pikhq> Usually, the mistakes made by native speakers are sufficiently common that it's easy to pick through.
23:27:47 <sp3tt> lament: write, rather
23:28:46 <oerjan> hey, i didn't know Asztal was british until he told me :)
23:28:54 <sp3tt> I believe native speakers make more errors when writing than when speaking
23:29:01 <sp3tt> and for the most part speak correctly
23:29:10 <oerjan> after i tried greeting him in Hungarian...
23:29:14 <SimonRC> Such things are nasty as they come totally out of the blue.
23:29:25 <pikhq> Have you *heard* people in the USA talk?
23:29:30 <ihope> There's a couple of errors I see pretty often.
23:29:39 <sp3tt> pikhq: not really, except on tv
23:29:44 <ihope> For example, using "there's" with a plural.
23:29:56 <lament> sp3tt: so you don't actually disagree, you're just saying native speakers make fewer mistakes while speaking.
23:30:12 <pikhq> Mmkay. . . Imagine 1337, spoken, and you'll be close to the correctness level.
23:30:20 <sp3tt> seeing as how rules for spoken languages are less strict
23:30:43 <lament> sp3tt: a mistake is a mistake. The rules are the same.
23:30:44 <ihope> Also, using "me" and such where inappropriate, using "I" and such where inappropriate, using "was" where inappropriate, using "would have" where inappropriate...
23:30:59 <oerjan> lament: ever heard of dialects?
23:31:06 <ihope> lament: I've never heard somebody verbally misuse an apostrophe. :-P
23:31:26 <sp3tt> give me your phone number and I can fix it
23:31:31 <ihope> Also, stop treating "type" and "kind" as if they were plural.
23:31:37 <pikhq> Something that irritates me is people writing the contraction of "would have" as "would of", rather than "would've".
23:31:49 <oerjan> i don't, i treat them as Haskell terms...
23:35:21 <lament> sp3tt: I hear native english speakers make mistakes in speech all the time.
23:36:17 <lament> usually nobody notices them, unless the speaker happens to be G W Bush
23:36:30 <lament> in which case everybody jumps on them
23:37:07 <lament> for example nobody knows the past participle of swim :)
23:37:45 <sp3tt> that's german, I believe
23:38:33 <oklofok> anyway, those things are taught in like elementary school
23:38:54 <oerjan> yeah, we've had _plenty_ of time to forget them :)
23:38:57 <oklofok> in finland, i mean, don't know about america :P
23:39:18 <oklofok> i don't like admitting i've forgotten something
23:39:55 <lament> what did you eat for breakfast a week ago?
23:41:48 * oerjan is pretty sure, but only because he eats the same nearly every day
23:42:51 -!- RedDak has quit (Remote closed the connection).
23:42:57 <oklofok> if it's a school morning, i'm late, otherwise i sleep till 13.00-18.00... never really get to experience a morning
23:43:43 <SimonRC> What is the correct name for that party game with a group of people that attempt to oust the chosen traitors among their ranks?
23:46:21 <SimonRC> oklofok: You are forgetting. God is a shitty programmer. God does not know the meaning of "robust design". With god's creations you are advised to stray from the original use-cases as little as possible. Therefore, sleeping normal hours tends to get you better results.
23:47:05 <oklofok> evolution is a shitty programmer, says i, but the point is valid.
23:47:16 <oklofok> i mean, with that conversion it's the same assertion
23:47:55 <SimonRC> many things that, in the programming world would be treated as atrifacts of a terrible cowboy coder, are reverred when in the natural world
23:48:17 <oklofok> the human body can have it's rest at anytime of the day, only change in the timetable screws you up
23:48:36 <SimonRC> There is no one body clock
23:48:45 <oklofok> yes, that's what i'm saying.
23:49:33 <SimonRC> There are about a dozen body clocks, syncronised to the day by various mechanisms, and sleeping anything other than normal hours tends to screw things up, reducing your concentration abilities.
23:50:05 <oklofok> but you can have your sleep any time, it's the change that makes you tired and stufff
23:50:56 <SimonRC> i have sometimes slept a steady schedule and felt mostly tired
23:51:13 <oklofok> for example, if you're a night watchman, your body takes about 3 months to change to the right schedule in body temperature
23:51:28 <oklofok> i mean, that you're warm at night
23:52:52 <SimonRC> maybe I didn't try for long enough
23:53:18 <SimonRC> I am suspicious of some of th claims made for savant abilities
23:53:31 <oklofok> SimonRC: what you're saying would mean people could never more into a foreign country.
23:53:46 <SimonRC> "if it is so great, why can't we all do that?"
23:53:59 <SimonRC> oklofok: ah, but the daylight is re-synchronised to fit
23:54:09 <oklofok> well, i'm always in the dark
23:54:29 <oklofok> that was one shitty argument, excuse me.
23:55:47 <SimonRC> Analogy: Suppose that there is a real or fake chocolate cake in a room with 20 3-year-olds. After being left alone for 1/2 hour, the cake is untouched. Would *you* believe a claim the it was a real chocolate cake?
23:56:46 * oerjan _really_ wonders what that is an analogy _to_ :D
23:56:59 <SimonRC> Quite a few savant abilities come from something not working. Expecting evolution not to spot that breaking something helps fitness islike expecting aforementionned 3-year-olds not to touch a real chocolate cake.
23:57:52 <lament> SimonRC: It seems pretty obvious that autism is not very useful for survival.
23:58:06 <pikhq> lament: Begging to differ, as an autistic.
23:58:21 <oerjan> however, savant abilities may help only with things that are totally useless in a hunter-gatherer society.
23:58:23 <lament> SimonRC: now, why do you think savant abilities can be isolated from the other symptoms?
23:58:35 <oerjan> and therefore, evolution has not had time to work.
23:58:37 <pikhq> They just kind of suck in primitive societies. . .
23:59:18 <SimonRC> lament: I wasn't. The cake is just plastic.
23:59:36 <lament> SimonRC: i don't understand, but okay.