00:02:13 <Sgeo> pikhq, become a RoboZZle addict like the rest of us!
00:02:17 -!- pikhq has quit (Read error: 54 (Connection reset by peer)).
00:03:35 -!- cpressey has left (?).
00:04:17 -!- pikhq has joined.
00:05:35 <fizzie> oklopol: Having reached a gross -- in at least two senses -- number of levels, I stop here; feel free to do 145 or whatever to be #1.
00:06:45 <Sgeo> What have I done??
00:10:18 <fizzie> Hey, the game works ~perfectly on the N900 browser.
00:10:22 <fizzie> The JS version, I mean.
00:10:50 <fizzie> There's perhaps a slight lag involved in the code editor part, but actually running a solution seems about as fast as Firefox on the desktop.
00:11:16 <fizzie> That's reasonably nice; I don't think I'll have the motivation to actually spend time with the puzzles much more, but maybe they'd work as a time-waster in a bus or something.
00:11:33 <Sgeo> fizzie, write some puzzles!
00:12:41 <fizzie> Was there a Javascript version of the editor?
00:13:49 <Sgeo> I don't believe so, no
00:17:14 <fizzie> Aw. How did Deewiant design his level, then?
00:22:07 <oklopol> fizzie: if you're really going to sleep, i'll leave it a draw
00:22:45 <fizzie> How considerate! But yes, I am; away right now, in fact.
00:23:51 <oklopol> i'll probably have to leave my homework for tomorrow, THANKS SGEO FOR RUINING MY SCHEDULE
00:24:07 <oklopol> hmm. okay you'll probably take that seriously, i take it back.
00:24:44 <Deewiant> oklopol: Tomorrow you'll notice fizzie's score again and leave it for the next day again
00:26:06 <oklopol> nah, i'm fine with fizzie being better than me; he has a master's degree.
00:27:34 <Sgeo> oklopol, you really think I take everything seriously?
00:28:16 <oklopol> no. i think you take things personally slightly easier than others. might be wrong (this is not a test!)
00:30:06 * Sgeo has no real way to judge himself as to whether that's the case, and to whether that needs fixing
00:30:43 <oklopol> do like the rest of us and don't care about anything
00:32:11 <oklopol> MissPiggy: not trying to split the credit or anything (i totally am), i'm pretty sure i've played less than him today
00:32:48 <MissPiggy> he has the highest degrees of robozzle skill
00:33:32 <oklopol> he *is* incredibly perfect at everything, not arguing that.
00:34:34 <oklopol> if i catch up with him, will you worship me too?
00:35:31 <MissPiggy> also; I was just trying to make puns about 'masters degree'
00:35:59 <oklopol> his degree is in flash games
00:36:28 * Sgeo posts a link to Robozzle to reddit
00:36:38 <oklopol> why do you think me, fizzie and Deewiant are so great that robuzzle or whatever it's called
00:36:47 <oklopol> i didn't look at the name yet
00:38:22 <Sgeo> http://www.reddit.com/r/WebGames/comments/at2lp/robozzle_a_fun_and_addicting_robot_programming/
00:39:21 <oklopol> oh dear, did someone just drop me from the list.
00:43:49 -!- augur has joined.
00:54:46 -!- jpc has quit (Read error: 54 (Connection reset by peer)).
01:01:31 -!- FireFly has quit ("Leaving").
01:25:10 -!- adam_d has joined.
01:36:18 -!- oerjan has joined.
01:40:40 -!- MissPiggy has quit.
02:02:28 -!- Asztal has quit (Read error: 110 (Connection timed out)).
02:07:41 -!- jpc has joined.
02:19:16 -!- augur has quit (Read error: 113 (No route to host)).
02:19:21 -!- augur has joined.
02:38:06 -!- Gracenotes has joined.
03:20:28 -!- Pthing has quit (Remote closed the connection).
03:48:50 -!- oerjan has quit ("Later").
03:53:40 -!- clog has joined.
03:53:40 -!- clog has joined.
05:04:31 -!- augur has quit (Read error: 104 (Connection reset by peer)).
05:10:14 -!- augur has joined.
05:18:31 -!- oklopol has quit (Read error: 54 (Connection reset by peer)).
06:03:49 -!- augur has quit (Read error: 113 (No route to host)).
06:09:48 -!- augur has joined.
06:16:43 -!- Slereah has quit (Client Quit).
06:27:29 -!- augur has quit (Read error: 113 (No route to host)).
06:33:07 -!- adam_d_ has joined.
06:36:18 -!- MigoMipo has joined.
06:43:21 -!- adam_d has quit (Nick collision from services.).
06:43:27 -!- adam_d_ has changed nick to adam_d.
06:59:37 -!- MigoMipo has quit (Remote closed the connection).
07:08:41 -!- augur has joined.
07:10:00 -!- asiekierka has joined.
07:10:10 <asiekierka> uuurgh, i'm looking for some old recordings on my DVDs
07:34:14 -!- MizardX- has joined.
07:50:36 -!- MizardX has quit (Read error: 110 (Connection timed out)).
07:50:39 -!- MizardX- has changed nick to MizardX.
07:56:41 -!- asiek2 has joined.
07:56:42 -!- asiekierka has quit (Read error: 104 (Connection reset by peer)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:22:38 -!- puzzlet has quit ("leaving").
08:23:00 -!- puzzlet has joined.
08:33:05 -!- MizardX has quit (Read error: 110 (Connection timed out)).
08:33:17 -!- MizardX has joined.
08:34:13 -!- puzzlet has quit (Client Quit).
08:39:01 -!- GreaseMonkey has joined.
08:39:35 <GreaseMonkey> sgeo it's all your fault for telling everyone about robozzle and now i'm playing it
08:57:28 <asiek2> but i like you for that
08:57:40 <asiek2> also thanks to you it might appear on the NES soon
08:57:44 <asiek2> YES, the nes does have enough RAM
09:29:30 -!- jpc has quit ("I will do anything (almost) for a new router.").
09:42:40 -!- adam_d has quit ("Leaving").
09:47:52 -!- GreaseMonkey has quit ("HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it").
09:58:49 <fizzie> Heh, oklobbol has dropped from 144 to 142 in the last-24-hours list, whereas I haven't yet; I think I started the thing about 21 hours ago.
10:02:26 -!- asiek2 has quit ("Pong timeout: 180 seconds").
10:21:08 -!- KingOfKarlsruhe has joined.
10:25:36 -!- zeotrope has joined.
10:39:10 -!- tombom has joined.
11:00:42 -!- BeholdMyGlory has joined.
11:16:21 -!- cal153 has quit.
11:17:49 <Wareya> http://esoteric.voxelperfect.net/wiki/Talk:W/
11:26:47 -!- asiekierka has joined.
11:27:42 <AnMaster> argh slow internet.. (I know why, uploading something, but it is still annoying)
11:29:12 <Ilari> At least it is text only and not something worse...
12:03:27 -!- noddd has joined.
12:04:04 -!- MissPiggy has joined.
12:04:05 <noddd> cool, it's the last 24 hours, not just like current 24h period
12:04:13 -!- noddd has changed nick to oklofok.
12:04:38 <MissPiggy> By installing Brainfuck, you will be able to experience the power of Brainfuck???
12:05:50 <oklofok> i'm not sure i understand the q
12:08:13 <asiekierka> i want to make an esolang based on RoboZZle
12:08:31 <asiekierka> basically an application consists of a map, the start coords and the robot command
12:08:44 <asiekierka> you will be able to set any block on the map to any ASCII char
12:09:32 <oklofok> remember, if you add commands that modify the map non-locally, i will be very angry
12:09:45 <asiekierka> no, you can only modify the spot you're on
12:09:59 <oklofok> yeah good, was wondering if you were describing that
12:10:08 <oklofok> infinite length for programs?
12:10:17 <asiekierka> and there's no max length of 10 per function
12:13:13 <asiekierka> and F1+10 would jump to the 11th command in function 1
12:13:31 <oklofok> so jumps in functions, not on map?
12:15:59 -!- kar8nga has joined.
12:17:46 -!- Slereah has joined.
12:35:53 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
12:36:36 <Sgeo> I think Reddit might have marked my post as spam
12:37:00 <Sgeo> I don't see it when looking at http://www.reddit.com/r/WebGames/new/?sort=new in incognito mode
12:41:09 * Sgeo asks a moderator
12:43:59 <asiekierka> actually, i'll make a game based on RoboZZle and other thing
12:44:09 <asiekierka> "The Overly Complicated Robotic Programmer"
12:44:21 <asiekierka> something like a combination of GolfScript, Befunge, RoboZZle and Brainf**k
12:49:14 <MissPiggy> the challenge of robozzle is to translate the higher level idea into this terrible low level code?
12:51:41 -!- MigoMipo has joined.
12:52:15 -!- tombom has quit ("Leaving").
12:52:48 <asiekierka> you also get (as an extra) 7 colors of spaces
12:53:07 <asiekierka> Or i could combine Robozzle with Boulder Dash
12:53:46 <asiekierka> when walked through the dirt turns into an empty of that color
12:54:01 <asiekierka> it also detects if there's a rock in front of you
12:59:22 <asiekierka> "ROCK? if yes, jump to function whatthe!@$"
13:07:26 -!- kar8nga has quit (Remote closed the connection).
13:20:33 -!- FireFly has joined.
13:33:02 -!- Nanakhiel has quit (Read error: 110 (Connection timed out)).
14:05:54 -!- Pthing has joined.
14:11:00 <Ilari> Hmm... Is there "efficient" way to construct one-to-one mapping between elements of regular language and Z_n or Z?
14:18:06 -!- rodgort has joined.
14:20:46 -!- rodgort has quit (Client Quit).
14:21:27 -!- rodgort has joined.
14:32:06 -!- Sgeo_ has joined.
14:35:14 -!- rodgort` has quit (Read error: 110 (Connection timed out)).
14:35:16 <oklofok> like, a computable function that, given a string, tells its number, and the inverse?
14:40:58 -!- FireFly has quit ("Leaving").
14:41:22 -!- FireFly has joined.
14:47:21 <Ilari> Yes, and more efficient than just enumerating the strings in order...
14:49:21 <MissPiggy> oh hm I was thinking about multiplying primes together for that, but going backwards would mean factoring
14:49:48 <MissPiggy> thing is, you'd know the number factors into high-ish powers of small primes
14:50:09 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
14:50:31 <Ilari> For even contex-sensitive language, doing the functions based on pure enumeration in order would be possible, but would have horrid complexity.
14:51:48 <oklofok> yes, exponential obviously, if the language only generates subexponential amounts of strings w.r.t. length
14:52:57 <oklofok> oh actually even more, because it's not enough to enumerate, you need to be able to actually evaluate the functions at some point :)
15:05:34 <Sgeo_> http://yudkowsky.net/singularity/aibox
15:06:48 -!- Pthing has quit (Remote closed the connection).
15:12:20 <Ilari> Does the following encoding work: Write the regular language as minimum DFA. Assign integers 0, 1, 2, ... for each valid symbol from each state (+ end here if state accepts), so that 0 is always on shortest path to accepting state or end here. Then write number as variable-radix with least-signficant number being for starting state.
15:16:09 -!- Asztal has joined.
15:17:02 <Ilari> Nope, doesn't work.
15:20:33 <Ilari> Consider DFA that accepts even number of characters with alphabet {'A','B'}. The first state is starting, accepting and and has three "exits". The second has two exits. 4 decodes to "AB", but 3 decodes to "" and 0 too decodes to "". So that doesn't work.
15:29:28 <Ilari> What if on decoding, each time accepting vertex is hit with nonzero quotent, the quotent gets decremented by 1 before continuing and only symbols are exits for even accepting states (if its 0, string ends).
15:31:34 <oklofok> quotient = language that gets us to accepting state, right?
15:32:37 <oklofok> well clearly not from what you said
15:33:53 <Ilari> Quotent gets its name from the fact that it would get divided after each character...
15:34:08 <MissPiggy> man that Eliezer guy is really scary
15:34:29 <MissPiggy> he's the kind of guy you feel like he could crush you by just imagining it
15:34:57 <oklofok> hmm, what if we just have a regular language, and we just recursively go through it, every time both branches are infinite, we set "next bit = 1" to one branch, "next bit = 0" to the other, if one is finite, we just give it just enough numbers, and so on
15:35:07 <oklofok> infiniteness can be checked efficiently
15:35:38 <oklofok> i really didn't think that through, just a gut-feeling
15:35:50 <oklofok> i meant "regular expression"
15:36:20 <oklofok> regexps can't exactly be made unambiguous can they?
15:36:44 <MissPiggy> (a*)* can parse aaaaaaa in gazillion different derivations
15:37:10 <MissPiggy> just reading what Sgeo linked about the AI box
15:37:27 <oklofok> for A + B, you'd check if both are infinite, if so, you give the numbers ...0 to A (bijection by induction), and the numbers ...0 to B, if they are both finite, you just give like |A| numbers to A branch, and you add |A| to whatever B gives you
15:38:15 <oklofok> MissPiggy: yeah but can we make them unambiguous, that was the question
15:38:26 <oklofok> well, for A + B, you can just check the intersection
15:38:32 <MissPiggy> I think I read denesting the stars in undecidible
15:38:48 <MissPiggy> but you can still make it unambiguous somehow?
15:40:16 <Ilari> Of course, one would have to have ordered alphabet to make the encoding uniquely determined.
15:40:30 -!- kar8nga has joined.
15:44:10 <oklofok> MissPiggy: no i don't think you can; i don't know if that directly implies you can't, either
15:44:26 <SimonRC> he is, after all, named after a prophet
15:45:02 <SimonRC> also, you can see that there can't be a *good* reason to let the AI out, or E.Y. would have convinced himself with that argument
15:45:49 * Sgeo_ wishes the experiments weren't secret
15:46:07 * Sgeo_ wants to know how the AI did it
15:46:35 <MissPiggy> maybe he says, if you let me out I will give you 20 dollars instead of 10
15:48:23 <Sgeo_> Wishing the protocol didn't say what it said means I need to read the protocol?
15:52:26 <Ilari> Consider the language given by 'A{2}|BA*'. Minimal DFA has 4 states. Pick 0 for A and 1 for B for first state. 0 => "AA", 1 => "B", 2 => invalid exit. 3 => "BA". So it doesn't work.
16:03:44 <Ilari> Hmm... Given DFA, number of strings that start in given state is probably computable. If accepting state is unreachable, its 0, if cycle and accepting state is reachable, its infinite, otherwise its finite...
16:04:05 * Sgeo_ is departing Season 7 of SG-1
16:08:14 -!- KingOfKarlsruhe has quit (Remote closed the connection).
16:08:40 <Ilari> Another decoding algorithm attempt: On each state encountered, check if quotent is less than number of strings starting with symbol that takes to state with finite set of accepted postfixes (+1 for accepting states). If its less, pick that string. If its more, substract the number from quotent and pick exit (dividing quotent by number of exits to states with infinite number of postfixes).
16:10:14 <Ilari> That way: 'A{2}|BA*' would give 0 => AA, 1 => BA, 2 => BAA, 3 => BAAA, ...
16:10:32 -!- BeholdMyGlory has quit (Read error: 104 (Connection reset by peer)).
16:12:42 <Ilari> Oops. 1 => B, 2 => BA, 3 => BAA...
16:14:52 <Ilari> The postfix count is efficently computable by first doing topological sort and then using dynamic programming.
16:15:38 <MissPiggy> "The key idea is that if you can improve intelligence even a little, the process accelerates. It’s a tipping point. Like trying to balance a pen on one end – as soon as it tilts even a little, it quickly falls the rest of the way."
16:15:51 <MissPiggy> why hasn't it happened already then?
16:16:20 <mycroftiv> it did, several thousand years ago
16:17:43 <mycroftiv> well, in terms of the kind of major transformative changes done by civilization, you can argue that larger scale and more persistent information transmission was needed than oral culture allows
16:18:34 <mycroftiv> im honestly not even vaguely qualified to comment on the topic of how written language influenced the development of civilization, I was just trying to make the claim that everything talked about in the 'singularity' concept actually already happened, more or less
16:19:54 <mycroftiv> in what I believe to be the canonical contemporary statement, defining the singularity as a point at which future events become unpredictable from past events, so far as I know that condition has *always* applied to historical prognostication
16:21:08 <mycroftiv> and I dont think anyone really believes the old-old version of singularity theory where the mass-energy manipulation capacity of the species was going to suddenly reach cosmological scale within a very short time
16:22:07 <mycroftiv> that was the version I was introduced to a long time ago, based on charting energy manipulation of thes species and a huge ramp up starting in the 19th century with another huge leap to atomic weapons, and then extrapolating to galaxy-rebuilding within a few short decades
16:22:50 <mycroftiv> nowadays I tend to put all this together in my mind under the label "curve-fitting is dangerous and tricky"
16:23:12 <Sgeo_> Well, I still want to upload my mind to a computer
16:23:46 <mycroftiv> I think you can, more or less - just create a lot of stuff via the action of your mind and store it on the computer
16:24:12 <mycroftiv> im a bit of a lunatic, but I happen to think for instance that the souls of artists are basically 'uploaded' in their creative works - when i play a beethoven sonata on the piano, i feel that i directly experience beethoven's literal consciousness
16:24:28 <mycroftiv> and that his cognitive essence, albeit in a 'frozen' state that cant interact with the external world, still persists
16:25:23 <mycroftiv> in fact, playing classical music on the piano often makes me feel that my brain has been literally taken over, and my consciousness has been temporarily overwritten by executing the source code of the music
16:25:49 <Ilari> Then there is related concept what I call "technological escalation". Reliance upon technology building upon reliance upon technology. If it continues unchecked long enough, the culture in question will likely rip itself apart.
16:26:07 <mycroftiv> which brings me to being almost on-topic - has anyone ever written a programming language expressed in musical notation? would be basically trivial
16:26:37 <mycroftiv> i mean, you could arbitrarily convert brainfuck to musical notation and back with almost no hassle
16:28:09 <mycroftiv> it would be interesting though to try to do something where the syntactic rules of the language enforced musical harmony
16:29:15 <oklofok> i wonder where i got that idea... ;)
16:30:17 -!- MizardX has joined.
16:31:14 <mycroftiv> very cool - although this is a programming language for creating music, it isnt expressed itself within notation, is it? or is the mapping so strict that a fugue program is equivalent to its representation in sheet music?
16:32:16 <oklofok> It shares semantics with its sister language, Prelude, but uses music as source code.
16:32:31 <oklofok> i think the sheet is actually the code
16:33:27 <oklofok> now what if there was a language in which every sheet of music had the semantics of playing the song, wouldn't that be just sweet!
16:39:45 <Gregor> ... I believe that language is called "sheet music"
16:41:06 <Gregor> And the interpreter is called "a musician"
16:49:31 <pikhq> What if it's a different song than what's on the sheet music?
16:54:52 <Sgeo_> The way oklofok phrased it, it seems that each sheet represents the full song. I don't think sheet music can do that with multiple-sheet music
17:03:09 -!- Azstal has joined.
17:04:53 -!- kar8nga has quit (Remote closed the connection).
17:16:26 <MissPiggy> are everyone on lesswrong atheists?
17:17:44 <MissPiggy> "Religion is the trial case we can all imagine. (Readers born to atheist parents have missed out on a fundamental life trial, and must make do with the poor substitute of thinking of their religious friends.)"
17:18:48 -!- Asztal has quit (Read error: 110 (Connection timed out)).
17:19:20 -!- Pthing has joined.
17:27:02 -!- MigoMipo has quit (Remote closed the connection).
17:29:26 <MissPiggy> what does it mean 'Eliezer-level rationalist'?
17:31:13 -!- MigoMipo has joined.
17:37:31 <Sgeo_> asiekierka, is green really supposed to have anything to do with left, or is that a red herring?
17:37:47 <asiekierka> you're doing the untrivial triviality?
17:38:13 <asiekierka> if so, then that's up to you to find out
17:38:44 <Sgeo_> By skipping a green and going right
17:39:32 <Sgeo_> Stack's only needed if you actually pay attention to the line about "left"
17:40:19 -!- BeholdMyGlory has joined.
17:40:29 <Sgeo_> A few stars in select locations would fix that
17:41:32 <Sgeo_> I mean, you'd have to have some in that loop thing if you don't want them just continuing right on green
17:45:03 <Sgeo_> oklofok's currently in the 24h lead
17:45:42 <Sgeo_> Actually, oklofok's solved more total than fizzie o.O
17:46:12 <asiekierka> http://robozzle.com/puzzle.aspx?id=1640 - fixed
17:47:47 -!- kar8nga has joined.
17:49:12 * Sgeo_ will try it a bit later
17:50:19 <Sgeo_> You know, there being a chatroom I can talk about RoboZZle in makes it much more fun for me...
17:51:23 <oklofok> and i would love it more if i didn't suck at it
17:51:54 <Sgeo_> oklofok is obviously delusional.
17:52:03 <Sgeo_> Thinking he's bad at it
17:52:19 <oklofok> i think i've done like one that's been rated near 4
17:52:54 -!- BeholdMyGlory has quit (Read error: 60 (Operation timed out)).
17:52:58 <Sgeo_> I can barely do ones that are rated 2.25
17:53:00 <oklofok> haven't really tried many hard ones, but i couldn't do the last one in the campaign, idea is trivial, i just can't make it work.
17:54:09 * MissPiggy is reading less wrong and worrying about whether I should be reading it or not
17:54:26 <Sgeo_> What would be wrong with reading it?
17:54:40 <oklofok> should probably take it as a template, and first implementing a working thing that just goes over the size limit
17:55:10 <MissPiggy> it just seems like a waste of time, even though it's nice to read stuff and nod your head, if you already agree with it all it's sort of pointless?
17:56:33 <oklofok> maybe you should play robozzle
17:56:50 <MissPiggy> I'm looking at asiekierkas puzzle, bemused
17:58:22 <Gregor> SafeAuto seems to advertise exclusively to people who are driving illegally without insurance.
17:59:31 <pikhq> Gregor: "Brilliant".
18:00:14 <Gregor> They advertise that you'll save because they'll give you the state minimum coverage.
18:00:24 <Gregor> SafeAuto: Insurance for the irresponsible!
18:00:30 -!- zeotrope has quit ("Lost terminal").
18:01:24 -!- BeholdMyGlory has joined.
18:02:44 <Sgeo_> MissPiggy, the un-fixed version shouldn't be
18:03:02 <MissPiggy> how can there be two greens in a row
18:03:14 <MissPiggy> you can't detect that because the robot only sees what it is standing on
18:06:22 <Sgeo_> Switch state upon encountering one
18:06:32 <Sgeo_> So that you start out in F1, but switch to F2
18:18:46 <Sgeo_> MissPiggy, how not?
18:19:09 <MissPiggy> imagine you are on a blue path and want to ignore green-green, but turn left on green
18:19:26 <MissPiggy> you have to code in something else, like turn right on the 3rd green
18:20:53 <Sgeo_> Or ignore just the first green it ever comes across
18:22:43 <Sgeo_> MissPiggy, remember, this is a broken version. Why should the description have anything to do with the puzzle?
18:25:24 -!- BeholdMyGlory has quit (Read error: 110 (Connection timed out)).
18:41:13 <MissPiggy> what's going on with this http://wiki.lesswrong.com/wiki/Yudkowsky%27s_coming_of_age
18:41:38 <MissPiggy> I was expecting the posts to be from 2000-2003, but they are from 2008 because it's just talking about 2000-2003
18:46:26 -!- jpc has joined.
18:48:38 -!- oerjan has joined.
19:06:33 <oerjan> <MissPiggy> it just seems like a waste of time, even though it's nice to read stuff and nod your head, if you already agree with it all it's sort of pointless?
19:07:03 <oerjan> take it to the meta-level: the fact that you agree with all a group says should itself be a warning signal that you are _not_ rational.
19:07:28 <MissPiggy> I don't know how that is meta-level
19:08:03 <oerjan> because you are then thinking about _why_ you agree with what the blog says
19:09:07 <MissPiggy> the cartoon loeb thing was cool, I didnt' realize he wrote it
19:09:12 <oerjan> the thing is, less wrong is clearly a cult of "rationality". even if they are right.
19:09:44 <oerjan> otoh they've probably discussed that as well
19:09:52 <oerjan> every blog can develop a cult
19:11:09 <oerjan> i think before accepting the challenge, i'd at least add the qualifier "that many people read"
19:12:33 <oerjan> also, i just realized i'm saying this because it's obvious, not because i have any actual evidence
19:17:21 -!- asiekierka has quit (Read error: 110 (Connection timed out)).
19:27:05 -!- oerjan has quit ("Later").
19:30:39 <MissPiggy> "Once upon a time it seemed to me that I ought to be able to win at the AI-Box Experiment; and it seemed like a very doubtful and hubristic thought; so I tested it. Then later it seemed to me that I might be able to win even with large sums of money at stake, and I tested that, but I only won 1 time out of 3. So that was the limit of my ability at that time, and it was not necessary to argue myself upward or downward, because I could just test it.
19:45:23 -!- zeotrope has joined.
19:48:30 <oklofok> lesswrong sort of treats the reader as an idiot.
19:50:41 * mycroftiv googles "lesswrong", follows the link, and finds bizarre cryonics related material
19:56:45 <oklofok> i get a sort of "why don't people think about this stuff?!?!" feeling, and it's annoying because it's really simple stuff
19:57:01 <oklofok> i mean, i get the feeling the writer is saying that
19:57:15 <oklofok> maybe it's just because it sounds smart, dunno
19:58:18 <oklofok> anyway most of the posts seem to be spot on, although you could phrase them more concisely
19:59:09 <oklofok> i guess that's the real reason why i feel i'm being treated as an idiot
19:59:50 <oklofok> i read a few paragraphs and i'm like "yeah i agree people tend to argue about words instead of actual meaning", and then there's like three pages of explaining and examples and blah
20:03:04 <oklofok> on the other hand i can't stop reading :P
20:03:26 <oklofok> comments seem to be youtube level
20:03:47 <mycroftiv> where on the internet are comments not youtube level, though?
20:07:05 <oklofok> well here on irc we tend to be pretty smurt right
20:08:06 <mycroftiv> actually #esoteric is probably the single most intellectually terrifying place on the internet, I'd agree - I mean I can do some programming, I understand stuff like symbolic logic a bit, but a lot of people in here seem to be able to analyze formal systems almost instinctively
20:09:10 <mycroftiv> I sometimes get the impression here this IRC channel is like an alternate reality composed of kids who were brought up on 'my first turing complete formal language specitication' in the crib
20:10:41 <oklofok> analysing stuff instantly is a distinct skill from analysing stuff properly though
20:11:43 <mycroftiv> true but "research shows" (waves hand) that in general, smarter people 'think faster' - even to the level of physical reflexes, i believe
20:12:47 <MissPiggy> if we shrink everyone by a factor of 2, we well all have on average IQ 800, because the cubic scale factor
20:13:18 <mycroftiv> at the same time, though, it is true that "other research shows" (waves other hand) that "innate talent" is a much worse predictor of successful outcome than sustained effort and concentration
20:14:16 <oklofok> speaking if intelligence, have people here taken IQ tests?
20:14:50 <MissPiggy> IQ goes down as you get older (i.e. bigger, so the distance between neurons gets further and you think slower)
20:15:34 <oklofok> are you sure that's a physiological truth and not the punchline of a joke?
20:17:27 <oklofok> i mean the definition of IQ i saw on mensa does seem to be about actual speed of neurons and not so much actual problem solving skill
20:18:58 <oklofok> i took the test on their page, i'm pretty sure anyone can reach the "top %1 of population", the upper limit of their web test, if they've played flash games
20:19:08 <mycroftiv> on the subject of high-IQ, anyone here ever run into Chris Langan and his CTMU theory on the web? in the past few years he seems to have decided, somewhat oddly in my view, to become culturally affiliated with the awfulness of "creation science"
20:20:53 -!- kar8nga has quit (Read error: 113 (No route to host)).
20:21:17 <MissPiggy> Cognitive Theoretical Model of the Universe
20:22:27 -!- madbr has joined.
20:22:27 <mycroftiv> eh, sadly id say hes mostly degenerated into pseudoscience
20:23:00 <MissPiggy> pretty obvious that it's pseudoscience by the rainbow <hr>'s
20:23:38 <MissPiggy> I wonder what the IQ for some of these idiot-savants is
20:23:52 <Ilari> There's lots of pseudoscience around. Sometimes it even involves serious studies, where everything goes well until one would need to draw the conclusions...
20:24:35 <mycroftiv> i think CTMU is actually more substantial than it seems, and actually borders on being decent *philosophy*, but i fail to understand why he would present it as a 'scientific theory'
20:25:30 <madbr> "The Cognitive-Theoretic Model of the Universe"
20:25:37 <MissPiggy> The free will theorem of John H. Conway and Simon B. Kochen states that, if we have a certain amount of "free will", then, subject to certain assumptions, so must some elementary particles."
20:26:17 <MissPiggy> seems a lot less quacky now that I read it on wiki rather than Times or The Sun whereever I heard about it first
20:26:46 <madbr> This is postmodern or something
20:27:21 <madbr> "Mathematically, the theoretical framework of Intelligent Design"
20:28:12 <mycroftiv> back when i found langan's stuff a few years ago, he hadn't thrown in with those guys, we was still mostly claiming to be extending traditional hard science
20:28:53 <mycroftiv> I'd say he couldn't meet their standards for publication and found the ID people were considerably more welcoming
20:29:18 <madbr> "you cannot describe the universe completely with any accuracy unless you're willing to admit that it's both physical and mental in nature"
20:29:50 <madbr> In conjunction with his ideas, Langan has claimed that "you can prove the existence of God, the soul and an afterlife, using mathematics."
20:30:35 <mycroftiv> wow, im kinda surprised CTMU survived General Notability Guidelines - although i guess it sneaks in on Langan's coattails because he had some popular media attention as world's highest IQ man
20:31:27 <MissPiggy> what IRC does Elizer Yudkowsky go to?
20:33:01 <madbr> yeah this is obvious Quack
20:34:01 <madbr> Like, he speculates on that kind of stuff, ok, but does he do real world experiments?
20:34:15 <mycroftiv> back around 1999 his 'mega foundation' was started basically as the result of huge flame wars stirred up by his attempting to promote CTMU is various ultra-high-iq internet forums
20:34:49 <mycroftiv> his wife later took it upon herself (i believe) to delete a lot of discussion of his ideas on a wiki I participated in, and replace it with copy-pasted chunks of his essays
20:35:36 <madbr> http://www.youtube.com/watch?v=9kf51FpBuXQ <- totally fails at this
20:37:19 <madbr> also it sounds a lot like a lot of postmodernist stuff
20:38:22 <madbr> And from the website it seems to be mostly convoluted language, and not enough claims that can be tested
20:38:37 <MissPiggy> why does everything need to be testible?
20:39:08 -!- kar8nga has joined.
20:40:55 <MissPiggy> you can't test any nonconstructive existence proofs
20:41:56 <mycroftiv> hmm that reminds me of the argument i started up here months ago about whether or not the axiom of choice has implications for physics
20:42:29 <madbr> miss: well, if it's not testable, then it's going to be hard to build a bomb out of it
20:42:36 <madbr> or other neat technological stuff
20:43:41 <madbr> and if it's not testable, who knows if it's true or false?
20:43:48 <MissPiggy> even if it's testible it's not ture
20:44:44 <madbr> well, often it just says that your approximation rule is good enough in conditions X,Y, up to Z decimals yeah
20:45:14 <madbr> but that often leads to interesting real world applications still
20:47:43 <madbr> And it's sure better than vaguely philosophical papers that are mostly obfuscated language
20:53:38 <madbr> "It means using language as a mathematical paradigm unto itself."
20:54:06 <madbr> Ho man, like, yeah that's a perfect recipe for disaster
20:56:35 <madbr> yeah ok basically this guy is going to use the properties of language to try to gain insights on the nature of the univers
20:58:08 <madbr> yeah looking at the paper, this thing reminds me of time cube
20:59:42 <oklofok> mycroftiv: why would it have?
21:00:22 <mycroftiv> oklofok: digging into this issue, i discovered that several proofs of various aspects of quantum mechanics rely on mathematics that assumes AOC
21:00:55 <MissPiggy> mycroftiv, I think that's a really interesting question btw
21:01:17 <MissPiggy> of course I don't know anything beyond high school physics so I don't have anything else to add :P
21:01:20 <madbr> http://s.engramstudio.com/src/unreal.png this is in his paper
21:01:55 <olsner> ah, that explains everything!
21:03:56 <oklofok> mycroftiv: so some models of the universe that are supported by observation only have the observed properties if AoC is assumed?
21:04:01 <Gregor> I'm using that for everything now.
21:04:30 <Gregor> I contest that that diagram cannot make sense, in any situation.
21:04:37 <mycroftiv> oklofok: honestly, im not competent to answer that
21:04:46 <Gregor> Erm, I contest that that diagram /could/ make sense in any situation, rather.
21:05:16 <oklofok> i see, i'm just wondering if it's conjectures that require AoC, very plausible conjectures, or some actual stuff the model explains.
21:05:18 <mycroftiv> the material I found doing research on this, the physicists seemed to get pretty cautious about making any statement about the nature of the relationship between mathematical truth and models, theory, and observation
21:06:16 <mycroftiv> I couldnt say if the way AoC was used mathematically in the relevant proofs of 'how stuff works' quantum mechanically would imply that the physics of a 'non-aoc' universe would be different
21:06:45 <oklofok> i see... it's just in my experience most stuff between very concrete and very formal get very vague and very meaningless pretty fast
21:07:06 <mycroftiv> but the thing is there isnt anything undefined involved in this, i dont think
21:07:24 <mycroftiv> i mean, aoc is well defined, and how to get predictions out of the quantum mechanical model is well defined
21:07:35 <mycroftiv> its not an angels-on-the-head-of-a-pin type question I dont think
21:08:09 <oklofok> sure, it's the observability that's the key issue here, what if other plausible conjectureshad been made given some other non-conflicting axiom
21:08:22 <madbr> personally, I concentrate on more real world stuff
21:08:35 <mycroftiv> the real world is made of quantum particles, last I checked
21:08:51 <madbr> like "how do you simulate a violin" (on a computer)
21:10:08 <oklofok> mycroftiv: reality is relative, atoms are only real to the extent of how well they model the world we observe. most people are good at imagining small things combining into big things. most people aren't good at imagining... quantum stuff.
21:10:37 <madbr> well, atoms are a very useful concept
21:10:46 <mycroftiv> oklofok: I completely agree - as a matter of fact, I like to point out that a 'car' is entirely a theoretical entity, because you are never going to find 'this is atom is part of Car X' written on any of its component atoms
21:10:55 <oklofok> quantum particles are what the world is made of if you're a physicist, they are made of particles that aren't quantum, otherwise
21:11:06 <mycroftiv> so 'cars' arent real, they are just an abstraction we apply to certain chunks of mater
21:11:32 <madbr> you guys are overthinking it
21:11:37 <MissPiggy> cars are basically spheres with uniform mass?
21:12:22 <madbr> you try to do an accurate model of how the thing vibrates
21:13:22 <mycroftiv> then solve the navier-stokes equations for how the waves propagate in the atmosphere
21:13:44 <madbr> actually it's more of a string problem
21:13:47 <oklofok> i heard about these phonons once, are there models of sound that involve particles?
21:14:05 <madbr> no sound is almost always modelled as waves
21:14:08 <oklofok> and excuse my not knowing anything, i just really don't know anything about physics
21:14:29 <madbr> sound doesn't have quantum duality :D
21:14:51 <oklofok> yes, it would seem weird given we have a rather good theory of how sound works, based on the wave theory
21:14:58 <madbr> but yeah violins don't have many vibration modes in air
21:15:07 <oklofok> but i've heard about phonons, i guess i could just wp those
21:15:26 <mycroftiv> i was making a joke about solving navier-stokes for the atmospheric vibrations, obviously grossly impractical
21:15:35 <madbr> I've never heard of a good sound simulation using phonons
21:15:54 -!- scarf has joined.
21:16:30 <madbr> also you have another problem: It turns out that the usual model of friction completely breaks down at "high" frequencies
21:16:34 <MissPiggy> oklofok, stuff like 'Whenever someone exhorts you to "think outside the box", they usually, for your convenience, point out exactly where "outside the box" is located. Isn't it funny how nonconformists all dress the same...'
21:16:36 <madbr> (like, 200hz high)
21:16:55 <MissPiggy> it's enjoyable to read this because I go "hah! I knew this already!"
21:16:55 <madbr> and friction is super important in violin simulation
21:17:10 <madbr> ie we don't actually really know how friction works
21:17:20 <MissPiggy> but then wonder if just sitting reading all these sorts of things is worthless
21:17:39 <scarf> MissPiggy: well, this channel is a good place to try to find new boxes to think outside
21:17:55 <mycroftiv> madbr: yeah im certainly no expert but ive read that there are still quite a few open problems in condensed matter physics because you get chaotic behaviors a lot
21:18:00 <oklofok> i don't think i've had that exact thought, i've just always thought the concept of "thinking outside the box" is ridiculous; then again i'm a mathematician, not a philosopher...
21:18:24 <scarf> oklofok: the mathematical equivalent is trying to find incorrect assumptions you made
21:18:37 <MissPiggy> I am not referring to the content of the statement, just the effect it has on me
21:19:01 <oklofok> MissPiggy: yes, and i tend to get the point, and start talking about something else.
21:19:13 <madbr> dunno, well, the theory I've studied is linguistics
21:19:26 <madbr> and basically, language is butt hard to analyze
21:19:40 <madbr> ie we haven't really figured it out yet
21:19:54 <madbr> and theories on it tend to end up turning in circles
21:20:47 -!- tombom has joined.
21:20:47 <mycroftiv> i have an amazing 'scots philosophical monograph' which is several hundred pages of dense symbolic logic attempting to understand what is 'really meant' by statements like 'I think John believes in Y'
21:21:23 <madbr> yeah that's definitely a turning in circles thing
21:22:11 <oklofok> prolly should go do the homework robozzle prevented me from doing yesterday
21:22:29 <madbr> we don't know how languages go from symbols to meaning
21:22:31 <oklofok> OR play robozzle, both alternatives sound good
21:22:49 <madbr> and we don't really know what meaning is anyways
21:23:00 -!- Sgeo__ has joined.
21:23:48 <oklofok> speaking of philosophy, do you believe people actually 'understand' things, that there's a fundamental difference between memorizing something, and understanding it?
21:24:15 <oklofok> i do believe we all have our own (implicit) models of the world, and we sort of understand things once we can fit new information into that framework
21:24:23 <oklofok> but i believe it's just a structured way to memorize things
21:24:27 <MissPiggy> especially in mathematics, for example calculus
21:24:29 <madbr> and also that sounds like it would turn in rounds
21:24:42 <MissPiggy> imagine being able to differentiate things, but not knowing what a function is
21:24:50 <oklofok> you memorize the thing, and you memorize rules for how to apply the information, you keep them close
21:25:53 <oklofok> i believe i just know the exact definition of a function (rote), and i have a pretty good ability to visualize mappings between sets, and i have a few rules for knowing how the speficic picture i have in my head translates into formal logic, which i can then check by pattern matching
21:26:11 <MissPiggy> but then maybe you could argue the people that don't know what a function is have some kind of deeper understanding about what a differential algebra is ;/
21:27:12 -!- Asztal has joined.
21:28:11 -!- cal153 has joined.
21:30:08 <oklofok> there is, of course, a sort of feeling that i "understand" some things, and don't understand some other things; also this isn't always correlated with my ability to solve different types of problems
21:30:29 <oklofok> so really i'm pretty sure it's just a meaningless feeling, whose evolutionary purpose is to direct my attention to things i need more information about
21:32:14 <oklofok> i guess that's sort of obvious, "understanding" just means we think we understand.
21:32:32 <madbr> that's why my favourite part of linguistics is phonetics... no messed up meaning stuff
21:33:03 <Ilari> Is it soft or hard science?
21:33:45 <madbr> well, phoetics is more hard because it has hard enough data
21:33:58 <oklofok> maybe my point is i think people's models of their own brain usually have consciousness be a sort of black box, with "understanding" being when something gets into consciousness and becomes usable
21:34:02 <madbr> other parts tend to be soft though
21:34:13 <MissPiggy> to feel like I understand a proof in most cases it is a case of producing some mental image which lives through the whole process
21:34:15 <madbr> ESPECIALLY anything that touches meaning
21:34:26 <MissPiggy> like the twisting circle with colored dots around it for fermats little theorem
21:34:42 <oklofok> whereas i see understanding as the process of getting the brain ready to do a certain type of thinking getting near finish
21:34:46 <oklofok> sort of a complicated sentence
21:36:10 <oklofok> really that's a trivial thought too, and one i just saw on lesswrong, people tend to black-box stuff they don't understand.
21:36:50 <oklofok> maybe there are just 7 deep philosophical thoughts, and everything else is just them in less pure form.
21:36:59 <oklofok> in fact i'm going to call this seppuritanism
21:37:05 <oklofok> see you, really have to math now ->
21:37:53 <oklofok> combinatorics of words and coding theory
21:38:04 -!- oklofok has quit (Read error: 104 (Connection reset by peer)).
21:38:19 -!- oklopol has joined.
21:39:06 <oklopol> combinatorics of words is what i linked here last week, this weeks exercises are just as ridiculous; coding theory's second exercise set seems to involve massive binary matrices, so to summarize, no risk of my evening being interesting.
21:39:41 <MissPiggy> why don't you do it on the computer?
21:40:01 <oklopol> well we rarely have two exercises about the same concept
21:40:12 -!- Sgeo_ has quit (Read error: 110 (Connection timed out)).
21:40:13 -!- Azstal has quit (Read error: 110 (Connection timed out)).
21:40:54 <oklopol> well okay we can have entire problem sets about the same concept, but i mean it's rare i could get more than one or two exercises done using the same program
21:41:10 <oklopol> unless of course i was clever and used some sort of sensible programming language that understand matrices and shit
21:41:21 <oklopol> i would use python, and program everything from scratch
21:41:39 <MissPiggy> oklopol they want me to use matlab :(
21:42:04 <MissPiggy> people that give me problems to solve
21:43:20 <oklopol> we have a specific matlab course in the physics dep i think, and a few about mathematica, but we rarely have more number crunching than 5*6
21:43:29 <oklopol> i mean other than those courses
21:45:17 <oklopol> it seems you stopped me from going
22:04:55 -!- jix has quit (Connection timed out).
22:26:09 -!- augur has quit (Connection timed out).
22:33:49 -!- Pthing has quit (Remote closed the connection).
23:11:17 <pikhq> Well, for no good reason, I have created a void_ptr a ptr<T> class... So now I can pretend C++ has C pointer semantics.
23:13:47 <pikhq> s/void_ptr a/void_ptr and/
23:14:14 <pikhq> "C++: because operator overloading lets you abuse the type system!"
23:15:11 <scarf> pikhq: you should see Boost, it's hilarious
23:15:18 <scarf> it reminds me of the good aspects of Perl
23:16:43 -!- GreaseMonkey has joined.
23:17:03 <pikhq> http://sprunge.us/hTGA
23:17:07 -!- augur has joined.
23:17:28 <pikhq> Eff you, C++. I want my implicit casts to/from void* back. So I GOTS THEM.
23:24:10 <mycroftiv> pikhq: i like that, but why are you working in c++ if you don't like its type handling?
23:24:49 <pikhq> mycroftiv: No good reason.
23:24:58 <pikhq> I'm not intending to actually *use* that header, BTW. :P
23:24:59 * MissPiggy just mentions btw, this is #esoteric :P
23:25:18 <pikhq> MissPiggy: C++ is an esoteric language.
23:25:25 <mycroftiv> pikhq: ah, so it was simply created as an exercise
23:25:39 <mycroftiv> that reminds me, you guys are all familiar with the original Bourne shell #defines, i assume?
23:25:49 <pikhq> template <typename T> operator T // INSANITY!
23:26:16 <mycroftiv> ok, lemme find the link, this is great stuff, really awesome late 70s hack at the core of one of the Essential Programs, the bourne shell
23:27:56 <mycroftiv> HERE: http://minnie.tuhs.org/UnixTree/V7/usr/src/cmd/sh/mac.h.html
23:28:48 <mycroftiv> clearly steven bourne liked his ALGOL and wanted his C code to read like algol
23:29:43 <pikhq> His definition of max assumes no side effects, incidentally...
23:29:45 <mycroftiv> so that bit of madness is right at the core of version 7 UNIX, the most influential os distribution fo all time probably
23:30:16 <scarf> sh is famous for that header file
23:31:17 <oklopol> Gregor: okay i give up, what does http://s.engramstudio.com/src/unreal.png mean
23:31:29 <Gregor> I haven't got a clue :P
23:31:50 <oklopol> oh wait it was madbr who linked it
23:32:56 -!- tombom has quit ("Leaving").
23:33:02 <oklopol> i've taken a few looks every now and then
23:37:18 <mycroftiv> I don't know what that picture is, but I assume it is just a bad way of expressing "the map is not the territory"
23:38:35 -!- MigoMipo has quit.
23:40:32 -!- augur has quit (Read error: 110 (Connection timed out)).
23:46:40 -!- oerjan has joined.
23:52:40 <oerjan> <MissPiggy> The free will theorem of John H. Conway and Simon B. Kochen states that, if we have a certain amount of "free will", then, subject to certain assumptions, so must some elementary particles."
23:53:53 <oerjan> my intuition: this is an actual theorem (it _does_ have conway on board after all), but whether it applies to any actual definition of free will your own philosophy would ascribes to, would probably depend.
23:54:29 <MissPiggy> oerjan, yeah sure, I think I just got the wrong idea about it because I read it first in some tabloid
23:54:31 <mycroftiv> actually, the problem of how to give any scientific definition to the concept of 'free will' is pretty serious imo, and it really bugs me
23:55:09 <mycroftiv> we have 'deterministic' and we have 'random', but exactly 'free will' even means is pretty damn hard to express in the terms of rationalist materialism
23:55:30 <oerjan> there are after all philosophies on what free will means that doesn't require the universe to be nondeterministic at all - or would even consider nondeterminism to make it _worse_
23:56:19 <oerjan> because you don't really have any freedom if things are just random...
23:57:01 <oerjan> but i guess that's part of what you are alluding to