00:09:24 -!- augur_ has quit (Remote host closed the connection).
00:09:31 -!- sebbu has quit (Ping timeout: 260 seconds).
00:10:08 -!- augur has joined.
00:42:46 <kmc> what's the best chess AI that can run on an ATtiny13 with 64 bytes of RAM under standard time controls?
00:43:11 <kmc> i think this would be more fun than writing a standard chess AI, and would require a lot more knowledge of the game
00:44:45 <oerjan> so, a chess AI that hasn't room to store more than a handful of game boards. check.
00:45:54 <oerjan> (and that's with some clever compression.)
00:47:02 <oerjan> i guess you just need one game board and the ability to store undos to previous positions, to do a minimax thing.
00:50:53 -!- myndzl has changed nick to myndzi.
00:51:09 -!- nooodl_ has quit (Ping timeout: 260 seconds).
00:52:07 <kmc> you also have 1kB of read-only code+data
00:52:18 <kmc> well actually you can write to it but sloooooowly and only a limited number of times
00:54:37 -!- nooodl_ has joined.
01:13:09 -!- sebbu has joined.
01:15:20 <zzo38> What is standard time controls?
01:17:57 <kmc> don't know
01:18:13 <kmc> FIDE gives "90 minutes for the first 40 moves followed by 30 minutes for the rest of the game with an addition of 30 seconds per move starting from move one"
01:20:53 -!- Phantom_Hoover has quit (Remote host closed the connection).
01:43:34 -!- nooodl_ has quit (Ping timeout: 246 seconds).
02:31:18 <zzo38> Now I made a GEN routine "s3msample" for use with Csound. S3M loop point is one after the end, and I do not know if this is the case with Csound.
02:43:07 -!- sivoais has quit (Ping timeout: 265 seconds).
02:50:36 -!- sivoais has joined.
03:10:12 <zzo38> But it does not support Adlib instruments. Perhaps there could be a plugin or UDO which plays Adlib instruments?
03:18:19 -!- monqy has joined.
03:20:54 -!- Tod-Autojoined has changed nick to TodPunk.
03:29:01 <zzo38> At least what I think would be a good way to design a file browser for UNIX computers would be, is, if you type in some wildcards and push some key to open the file list which you can click the left button to add to the command, and perhaps right button to preview (you can set up the preview command with another key).
03:29:07 <zzo38> What do you think?
03:36:54 * Fiora looks up, sees kmc's chess AI
03:37:29 <Fiora> 64 bytes... I guess you could fit a few bitboards in that
03:39:01 <Fiora> That might be tricky though, I wonder if that's even enough room to store stuff for alpha/beta pruning
03:39:25 <hagb4rd2> zzo038: yes your the ideas are good. but i really don't see why there must be a difference in the behaviour of a file browser and a http browser (for example). indeed i wonder why i need both..
03:39:42 <Fiora> I'd even be worried about running out of just, like, temp space. like doing move generation
03:41:10 <zzo38> hagb4rd2: Well, it is useful for different things, and there are other reasons too, why it should differ in whatever ways.
03:41:39 <kmc> yeah, i think it would be a fun challenge because you would rely a lot more on heuristic tricks rather than tree search
03:41:50 <zzo38> I mean the terminal emulator would load it to allow you to select files and then the files you selected are put in the command line.
03:41:57 <hagb4rd2> zzo38: i dunno. there should be a a good way to browse through any hierarchical grouped data
03:42:15 <Fiora> writing a chess AI sounds like a lot of fun
03:42:24 <kmc> when i was judging students' othello AIs, it seemed that the ones that did reasonably well in very little time were more interesting than the ones that had a competent implementation of alpha-beta minimax that consumed the full available time
03:42:30 <Fiora> I used to do a good bit of chess stuff but I was more interested in the AI than in, like, the actual game ^^;;
03:42:41 <Fiora> othello AIs! I remember that
03:42:43 -!- Arc_Koen has quit (Quit: Arc_Koen).
03:43:08 <Fiora> there's still a ton of heuristic stuff in tree search though
03:43:21 <Fiora> like so much, I remember reading for days about it
03:43:31 <Fiora> and all the really complicated heuristics for gauging the state of the board
03:43:48 <Fiora> extensions, quiescence search...
03:44:23 <kmc> it would be a fun thing for a hacker con with programmable badges
03:44:23 <Fiora> and the horizon problem <_>
03:44:31 <kmc> to have a thing where you can program a chess AI onto your badge
03:44:36 <kmc> and then it plays against other people you encounter
03:45:04 <Bike> that is just about the nerdiest idea i've ever heard said
03:45:14 <Fiora> my favorite chess AI thing though is nakamura beating rybka in blitz chess
03:45:22 <Fiora> and the horrid way he did it
03:45:31 <Bike> that was awesome
03:45:42 <Bike> assuming you mean the playbook hack thing
03:45:42 <Fiora> ( http://fioraaeterna.tumblr.com/post/20186536921/ )
03:46:06 <Fiora> horizon effect abuse, plus 50-move-rule abuse
03:46:14 <zzo38> hagb4rd2: Well, I don't think HTTP really supports hierachical grouped data (although the file being served could be XML or something which does support it), which, the file manager should be difference due to such thing.
03:46:43 <hagb4rd2> i didn't try extensions which enhance the abilities of some browsers to browse files too, but i've heard they're out there
03:47:09 <hagb4rd2> sure there are differences on the technical side
03:47:37 <zzo38> Still I don't think it is such a good idea as a general purpose file browser though, even if a web browser does have a file browser function.
03:48:25 <hagb4rd2> maybe this is due to a kind of a strange meme
03:50:32 <kmc> Fiora: wow, nice
03:51:22 <Fiora> nakamura is terrifying
03:51:29 <Fiora> in a good way of course
03:52:39 <Fiora> the horizon effect is really tricky though with tree search stuff in general
03:52:51 <Fiora> since the AI can keep "putting off" an inevitable doom and make it look like it's not actually coming
03:52:59 <Fiora> but it's hard to prove that it is
03:53:35 <Fiora> @google chess programming wiki horizon effect
03:53:37 <lambdabot> http://chessprogramming.wikispaces.com/Horizon+Effect
04:02:19 <hagb4rd2> zzo38: if no other routes are specified you can access files on a http server in the way their source is organised in files and directories on the server itself. nevertheless you can access all other routes with a string (typically by separating the tree level element names with a slash or sth)
04:03:53 -!- augur has quit (Remote host closed the connection).
04:06:02 <hagb4rd2> on the other hand: who is typing urls anyway?
04:07:56 <hagb4rd2> (i know it was my stupid idea :)
04:08:23 <zzo38> I do type URLs sometimes.
04:09:06 <zzo38> However even though with HTTP generally (not always) the directory/filename is the URL, still HTTP doesn't have the option to list the files, so you have to instead output a HTML or XML file or whatever which list them.
04:09:18 <zzo38> But I am usually typing in the URL if I want one.
04:09:30 <kmc> WebDAV probably has a standard way to list them
04:10:18 <zzo38> Maybe it does, I don't know.
04:10:28 <hagb4rd2> that's right. file listing is not allowed by default.. or better say route listing (which could be an option in the future)
04:10:38 <zzo38> If you want file listing, use FTP
04:11:10 <zzo38> And for menus in general, you can use gopher.
04:11:35 <zzo38> Although I suppose you can also use Plan9 protocol for file access is another way.
04:13:13 <zzo38> But if you want file listing, that is what FTP is for, not HTTP.
04:15:00 <hagb4rd2> zzo38: the issue was to provide a sitemap, to enable the user to browse hierarchical grouped data not beeing bounded to "old fashioned ways" beeing aware of where the data is physically stored
04:15:16 <hagb4rd2> i know there is no such standard
04:15:18 <zzo38> In that case, that is what gopher menu is for.
04:17:10 <hagb4rd2> i love the idea to mount anything as device :)
04:17:56 <hagb4rd2> on which i could just navigate like on my filesystem.. that's it
04:18:38 <zzo38> What I think is programs should expose a virtual filesystem (if they have one) in their directory under /proc/
04:19:09 <hagb4rd2> yes that comes closer to what i called a route map
04:19:10 <zzo38> So process ID 666 might have a virtual filesystem under /proc/666/fs/ or something like that.
04:21:58 <zzo38> So they should make it, if you have environment variable XPID for the X process ID, to access X clipboard buffers and stuff like /proc/$XPID/fs/PRIMARY and so on, as if it is like ordinary files.
04:28:14 <hagb4rd2> also this tree should relly be pretty much orgnanized by the logical context rather than by physical storage or sth
04:29:48 <hagb4rd2> so the role in which the user behaves on the system might change the access (or even the structure)
04:32:56 <shachaf> monqy: http://www.youtube.com/watch?v=GPdzCZdvBp0 http://www.youtube.com/watch?v=GhACtJSCtq4
04:34:05 -!- augur has joined.
04:36:03 <monqy> hi how did i know this would be return to the neverhood
04:36:34 -!- hagb4rd2 has changed nick to hagb4rd.
04:37:32 <monqy> that would explain it
04:37:45 <shachaf> monqy: or was it "bitter experience"
04:37:59 -!- oerjan has quit (Quit: leaving).
04:38:01 <monqy> the bitter experience of you telling me to listen to return to the neverhood
04:49:01 -!- jfischoff has quit (Quit: jfischoff).
04:49:28 -!- jfischoff has joined.
04:54:11 -!- Jafet has joined.
04:54:42 -!- Frooxius has quit (Ping timeout: 264 seconds).
05:12:02 -!- oklofok has joined.
05:15:40 -!- oklopol has quit (Ping timeout: 260 seconds).
06:11:29 <zzo38> If the sample rate does not mean the delay wanted is an integer number of samples, what is the best way to modify the delay buffer to more closely approxmiate the delay wanted more precisely?
06:29:48 -!- david_werecat has quit (Ping timeout: 265 seconds).
06:37:34 <zzo38> For example if the delay is 3.5 samples then how to make it closer to 3.5 rather than 3 or 4?
06:42:39 <kmc> apache is now disabling SSL compression by default
06:46:36 -!- ogrom has joined.
06:56:11 -!- jfischoff has quit (Quit: jfischoff).
07:03:48 -!- Nisstyre has quit (Quit: Leaving).
07:05:07 -!- ifnspifn has joined.
07:07:01 <ifnspifn> if I were to build a program which just takes as input the Godel or "description" number of a Turing machine, along with an input, could the space of input for this program be correctly termed a "programming language"?
07:07:27 -!- sebbu2 has joined.
07:07:45 -!- sebbu2 has quit (Changing host).
07:07:45 -!- sebbu2 has joined.
07:07:51 <ifnspifn> Bike: well, would YOU call it a programming language? :P
07:08:01 -!- ifnspifn has left.
07:08:09 -!- ifnspifn has joined.
07:08:13 <Bike> i'd call it inefficient, have you /seen/ godel-encoded stuff?
07:08:38 <ifnspifn> I have; this isn't a practical project I'm planning on doing
07:09:43 -!- sebbu has quit (Ping timeout: 245 seconds).
07:11:11 * Sgeo__ hasn't seen godel-encoded stuff
07:11:39 <Bike> you know how it works, right? code up something to encode strings or w/e and see for yourself
07:11:43 <Bike> hope you have bignums!
07:12:20 <quintopia> godel encoded programs are shorter than brainfuck programs if they use a good encoding
07:14:43 <ifnspifn> Sgeo__: here's the reference I've been using: http://mathworld.wolfram.com/GoedelNumber.html
07:15:15 <ifnspifn> Bike: I've got a few ideas in mind :D if it works out, I'll share it here
07:15:46 <Bike> ifnspifn: what ideas are those? so far you've just described the standard UTM definition
07:17:08 <ifnspifn> Bike: Well, I was going to go a few levels above just natural numbers, mostly for novelty/interest; in particular, I'm planning on making an interpreter which will read regular english words (e.g. an essay) and return a reasonably large godel number
07:17:36 <Bike> most probably won't correspond to valid programs, you realize
07:18:23 <ifnspifn> for demonstrative purposes, I'll start with some simple machine (Fibonacci, etc) have a dict.txt file and choose adequate words randomly
07:18:44 <ifnspifn> from there it might be the case that I can prune a somewhat sensible selection of words, but that's not necessary
07:19:39 <ifnspifn> if my text-to-natural number converter is flexible enough though, it should be possible to get a meaningful english snippet
07:22:21 -!- Nisstyre has joined.
07:23:04 <Bike> any particular reason you're involving godel coding?
07:27:37 <ifnspifn> mostly because it's neat, and an interesting way to represent a TM
07:28:30 <Bike> well it's not just for TMs, it's for arbitrary strings of integers
07:28:35 -!- iamcal has joined.
07:28:56 -!- augur has quit (Remote host closed the connection).
07:29:52 <ifnspifn> definitely! I actually started learning about its application to his Incompleteness theorems long before I started programming
07:30:19 <ifnspifn> the numbering system, although far from practical, has always struck me as very elegant
07:30:39 <Bike> you might be interested in looking at optimal coding
07:31:27 <ifnspifn> heh, funny you mention that. I'm currently in an introductory abstract algebra class, and got waaaaay over my head in Shannon's work with groups and coding
07:31:59 <Bike> you can find shannon's big paper on communication online somewhere, though it's a bit old now
07:33:07 <ifnspifn> I'll definitely check it out. I suppose the foundational papers are usually a good place to start, considering how technical the modern coding schemes are
07:33:20 <Bike> stuff like huffman coding is pretty easy to understand
07:33:53 <ifnspifn> oh yeah, I've got most of the basic schemes down
07:34:17 <Bike> shannon's paper is just about information entropy and stuff iirc, fundamental but hasn't changed much
07:34:46 <ifnspifn> this stuff though? *whoosh* http://en.wikipedia.org/wiki/Binary_Golay_code
07:35:33 <Bike> things like "twelve-dimensional subspace" aren't nearly as scary as they look. just persevere, it'll make sense soon
07:37:34 <ifnspifn> I know my linear algebra has been neglected for one too many semesters..
07:37:51 -!- epicmonkey has joined.
07:38:06 <Bike> maybe you should follow up your godel thing with representing tms as infinite-dimensional matrices, just to practice your linear shit
07:40:51 <ifnspifn> trying to imagine how that representation applies.. an infinite dimensional matrix would correspond to a TM's (potentially) infinite string, where each "digit" is a vector?
07:41:48 <Bike> nah, just to computable functions (note that I'm tired and not thinking this through)
08:00:45 -!- Tod-Autojoined has joined.
08:01:16 -!- TodPunk has quit (Read error: Connection reset by peer).
08:06:00 <Sgeo__> There are plenty of operations such that a `op` a = a + 1 but a `op` non-a is not necessarily a+1
08:06:22 <Sgeo__> f(a, b) = log_2 (2^a + 2^b) is one, but there's another
08:07:15 <Bike> what about f such that f(a,a) = a+1 but f(a,b) = 37
08:09:10 <Sgeo__> Also f(a,b) = a+1 for all a and all b except when a=0 and b=1; f(0,1) = 0
08:09:38 <Sgeo__> Lots of easy ones, but ... there's something more interesting about the two I named, I think
08:29:12 -!- sebbu2 has quit (Read error: Connection reset by peer).
08:29:56 -!- sebbu has joined.
08:29:56 -!- sebbu has quit (Changing host).
08:29:56 -!- sebbu has joined.
08:30:38 -!- Vorpal has joined.
08:42:56 -!- augur has joined.
08:47:41 -!- Nisstyre has quit (Quit: Leaving).
08:47:48 -!- atriq has joined.
08:50:18 -!- Nisstyre has joined.
08:51:29 -!- atriq has quit (Client Quit).
08:58:54 -!- kallisti has quit (Ping timeout: 264 seconds).
08:59:37 -!- kallisti has joined.
09:01:51 -!- Arc_Koen has joined.
09:02:19 -!- Bike has quit (Quit: asleep).
09:39:28 -!- Nisstyre has changed nick to whitef.
09:41:18 -!- whitef has changed nick to Nisstyre.
09:45:35 -!- ssue has quit (*.net *.split).
09:45:37 -!- Lumpio- has quit (*.net *.split).
09:45:39 -!- rodgort has quit (*.net *.split).
09:47:58 -!- ssue has joined.
09:47:58 -!- Lumpio- has joined.
09:47:58 -!- rodgort has joined.
10:27:46 -!- zzo38 has quit (Remote host closed the connection).
10:29:30 <FireFly> http://www.pltgames.com/competition/2012/12 cue the brainfuck derivatives
10:36:50 <FreeFull> Does it really list Befunge as a turing tarpit?
10:37:33 <fizzie> fungot: Do you feel like living in a tar pit?
10:37:34 <fungot> fizzie: there is always a dumb fuck?
10:37:55 <fizzie> fungot: There's no call for that sort of language, young ma... uh... bot!
10:37:56 <fungot> fizzie: it doesn't output anything. either it has to
10:39:43 -!- ifnspifn has quit (Quit: ifnspifn).
10:46:56 <FreeFull> http://esolangs.org/wiki/REVERSE This reminds me of unefunge
10:51:10 <fizzie> Unefunge programs tend to be so >a#Eb#Dc#Cd#Be#A_-y.
10:52:27 <fizzie> There are probably really few (if any) bidirectional pieces of code in fungot.
10:52:28 <fungot> fizzie: and subtle cough too, sort of reminiscent of " the little schemer is a good approach
10:52:40 <fizzie> (There are also no semicolons.)
10:53:59 <fizzie> (There are three j's, and they're all jump tables.)
10:54:13 -!- ais523 has joined.
10:54:35 -!- arcatan has joined.
10:55:25 <Deewiant> fizzie: Your style is too maintainable, you should be crisscrossing all over the place
10:56:30 <fizzie> Deewiant: It's just because I'm so stuck in the 93s.
10:57:04 <Deewiant> If you were actually stuck in the 93s you'd be forced to crisscross due to having only 80x25 space
10:57:37 <Deewiant> You don't have to use semicolons, in fact it's better if you don't because they make it too easy
11:07:53 -!- Phantom_Hoover has joined.
11:18:02 -!- carado has joined.
11:27:59 <fizzie> I'm mentally stuck, not physically stuck.
11:28:45 <fizzie> So I just naturally avoid the actually-new features, not things like increased playfield size.
11:28:59 <fizzie> I don't think I have any [] turnings in there either.
11:30:29 <fizzie> It's actually just a combined v and ^.
11:31:34 -!- sirdancealot8 has quit (Ping timeout: 246 seconds).
12:21:04 -!- FireFly has quit (Quit: ZNC - http://znc.in).
12:28:27 -!- mekeor has joined.
12:29:36 <Arc_Koen> being physically stuck in 1993 must be weird
12:30:34 <Arc_Koen> are you somehow sent back one year every year? or do you stay young forever?
12:42:47 <fizzie> I wouldn't know; I'm not stuck that way.
12:54:44 -!- sirdancealot7 has joined.
13:12:54 -!- mekeor has quit (Quit: cu).
14:29:31 -!- david_werecat has joined.
14:48:22 -!- monqy has quit (Quit: hello).
14:55:38 -!- nooodl_ has joined.
15:13:03 -!- Nisstyre has quit (Quit: Leaving).
15:19:03 <shachaf> This is some awful keming on Chromium's part: http://slbkbs.org/keming.png
15:20:44 <Jafet> Is it related to the Farnicorn?
15:21:11 <shachaf> fizzie: The HTML source says Unicorn
15:32:36 * ion shivers at the horribly blurry font rendering in the screenshot
15:37:27 <olsner> hmm, shouldn't horrible keming be good kerning?
15:50:26 -!- SPYGAME has joined.
15:54:09 -!- david_werecat has quit (Ping timeout: 260 seconds).
15:56:12 <elliott> Phantom_Hoover: fyi i was sleeping when you said it was tomorrow
16:10:22 -!- SPYGAME has quit (Quit: Leaving).
16:10:52 -!- myndzi\ has joined.
16:10:59 -!- lambdabot has quit (Ping timeout: 255 seconds).
16:11:01 -!- shachaf has quit (Ping timeout: 248 seconds).
16:11:43 -!- shachaf has joined.
16:41:53 <kmc> http://www.amazon.com/gp/product/0262018462/
16:46:17 <shachaf> ghc -e 'import System.Random' -e 'putStr . map ("╱╲"!!) . randomRs (0,1) =<< newStdGen'
16:49:27 -!- Frooxius has joined.
16:51:27 <ion> > chr 205.5
16:55:07 -!- jfischoff has joined.
16:57:30 <kmc> http://en.wikipedia.org/wiki/File:C64_Petscii_Charts.png
16:59:48 -!- jfischoff has quit (Ping timeout: 252 seconds).
17:01:40 <kmc> oh i see RND(1) returns a float between 0 and 1
17:01:57 -!- ogrom has quit (Quit: Left).
17:02:28 <fizzie> https://dl.dropbox.com/u/113389132/Misc/c64-8x8.png I made this once for an "ASCII-compatible" "C64-inspired" "high half is mostly line-drawing" thing.
17:05:20 <fizzie> I've forgotten what the £ is there for.
17:05:40 <fizzie> Some of the last characters are slightly application-specific, too.
17:06:38 <kmc> the book contains an equivalent program written in http://esolangs.org/wiki/PATH as well
17:20:57 -!- ais523_ has joined.
17:21:01 -!- ais523 has quit (Read error: Connection reset by peer).
17:21:35 -!- ais523_ has changed nick to ais523.
17:29:24 <ion> Yes, that’s what lead to my comment.
17:29:51 <elliott> ion: yes. but then I repeated it!
17:37:00 <shachaf> ion: Oh, they were just being political.
17:43:39 <kmc> ╱╲ ╲ ╱╲ ╱╲ ╲ ╱ ╱ ╱╲ ╱╲ ╱╲
17:43:40 <kmc> ╲ ╲ ╲ ╱ ╱╲ ╲╱╲ ╲╱╲ ╲╱╲ ╲ ╲╱╲ ╲╱╲ ╲╱ ╲╱ ╱ ╱ ╲╱ ╲╱ ╱
17:43:55 <kmc> ╱╲ ╲ ╱╲ ╱╲ ╲ ╱ ╱ ╱╲ ╱╲ ╱╲
17:43:56 <kmc> ╲ ╲ ╲ ╱ ╱╲ ╲╱╲ ╲╱╲ ╲╱╲ ╲ ╲╱╲ ╲╱╲
17:43:59 <kmc> ╲╱ ╲╱ ╱ ╱ ╲╱ ╲╱ ╱
17:44:09 <fizzie> Oh, I thought those were some kind of runes.
17:44:09 -!- Nisstyre-laptop has joined.
17:44:10 -!- ais523_ has joined.
17:44:18 <kmc> yes they do look runic
17:44:18 <shachaf> Reminded me of Myst or something.
17:44:19 -!- ais523 has quit (Read error: Connection reset by peer).
17:44:28 -!- ais523_ has changed nick to ais523.
17:44:40 <FreeFull> I was thinking of a simple maze generator
17:44:44 <elliott> kmc: now make it work on the alphabet too
17:45:04 <FreeFull> A simple way to generate mazes is to randomly print ╱ or ╲
17:45:11 <kmc> FreeFull: WOW REALLY TELL ME MORE
17:45:36 <kmc> (that is what we have been talking about)
17:45:52 <kmc> rigth here
17:45:58 <olsner> I thought you were generating /// programs
17:46:18 <kmc> Phantom_Hoover: you can't make them longer on an actual 7 seg display
17:46:32 <shachaf> kmc: Should I go to http://www.meetup.com/Stripe/events/92206302/ ?
17:46:32 <FreeFull> I don't know if all combinations of / and \ are valid /// programs
17:47:04 <shachaf> FreeFull: Also those aren't mazes.
17:47:09 -!- jfischoff has joined.
17:47:19 <kmc> shachaf: shrug
17:47:22 <FreeFull> shachaf: Well, they aren't guaranteed to have a way through
17:47:34 <FreeFull> Or to only have an unique way through
17:48:06 <shachaf> "We're trying something a little different this time: while the hackathon will still be informal and you're welcome to hack on anything you'd like, we're encouraging projects built on top of Stripe Connect (stripe.com/connect)."
17:48:37 <Phantom_Hoover> unless maybe you had, like \ on its own at the end of the file
17:48:56 <FreeFull> You can also use "|| __" as the random characters
17:49:59 <shachaf> I use " " as the random characters.
17:50:13 <shachaf> 2004 THREE-PER-EM SPACE [ ]
17:50:13 <shachaf> 2005 FOUR-PER-EM SPACE [ ]
17:50:17 <shachaf> 2008 PUNCTUATION SPACE [ ]
17:50:25 <shachaf> Hmm, those lines were supposed to merge into one super-line.
17:50:26 -!- david_werecat has joined.
17:50:30 <kmc> shachaf: you should make a brainfuck derivative
17:50:48 <fizzie> shachaf: What, no 205F MEDIUM MATHEMATICAL SPACE?
17:51:02 <shachaf> fizzie: I'm racist against mathematicians.
17:51:07 <kmc> space has a terrible power
17:51:19 <fizzie> 3035 IDEOGRAPHIC HALF FILL SPACE
17:51:49 -!- Phantom_Hoover has quit (Read error: Operation timed out).
17:51:56 <FreeFull> shachaf: They all look like a single space to me
17:52:15 <shachaf> FreeFull: Maybe you should put on your 3D glasses then.
17:52:44 <kmc> they are all a single space
17:52:53 <kmc> each of abcdefghijklmnopqrstuvwxyz is a single letter
17:53:22 <shachaf> I once heard U and I were together.
17:53:33 <shachaf> But I guess that was just a rumour.
17:53:44 <fizzie> OGHAN SPACE MARK is in the Zs "Separator, Space" category even though it has a line right through it.
17:54:20 <fizzie> And the ETHIOPIC WORDSPACE is two dots, but at least it's Punctuation, Other.
17:54:52 -!- Phantom_Hoover has joined.
17:55:26 <FreeFull> kmc: abcdefghijklmnopqrstuvwxyz don't look the same though
17:55:40 <kmc> well the spaces won't look the same either
17:55:43 <kmc> if you have the right font
17:55:56 <FreeFull> I'm in a terminal. Monospace fonts only
17:56:03 <kmc> there's yer problem
17:56:04 <shachaf> kmc: In NetHack I used a configuration option that remapped ghosts to Xs
17:57:02 <fizzie> shachaf: The MONGOLIAN VOWEL SEPARATOR is also whitespace, but at least it's empty.
17:57:54 <shachaf> fizzie: I THINK YOU MEAN personspace
17:58:05 <kmc> i'm still not clear on why all those emoji made it into unicode
17:58:40 <kmc> how do they draw the line between "a nonstandard character encoding with a bunch of new characters" and "a strange binary encoding for both characters and certain graphics"
18:00:04 <kmc> i'm going to invent ISO-8859-GOATSE where every codepoint above 0x7F is goatse
18:00:14 <pikhq_> Committee members debate it, basically.
18:00:29 <pikhq_> ... So, you'll probably not get that through unless you can convince people it should be done.
18:00:41 <pikhq_> In this case, Apple and Google proposed the emoji set initially.
18:01:11 <kmc> i thought they came from japanese mobile phone companies
18:01:23 <kmc> who were already using these characters in nonstandard encodings
18:01:31 <pikhq_> That's who made the characters in nonstandard encodings.
18:01:40 <pikhq_> Google and Apple are the ones who wanted them in standard encodings.
18:01:41 <kmc> i see, and apple and google pushed for them to become part of the UCS
18:01:50 <kmc> because they want to be able to compete in japan
18:02:01 <kmc> nobody will use Android if they can't text their friend a PILE OF POO
18:03:19 -!- Bike has joined.
18:08:28 -!- ais523 has quit (Ping timeout: 246 seconds).
18:09:20 <Bike> http://lambda-the-ultimate.org/node/4644 hey, rapido showed up on LtU!
18:10:52 <shachaf> elliott: That's an innovative GHC switch, don't you think?
18:13:47 <elliott> Bike: "I'm not sure what the goal (or point) of this document is, and how you evaluate whether it accomplished it or not."
18:15:32 <Bike> apparently he wants a language with no runtime exceptions
18:16:17 <elliott> well, that's just a total language
18:16:33 <shachaf> elliott: You know the thing where State = Reader + Writer?
18:16:37 <elliott> you can, of course, represent division in such a language just fine... by giving it a proper type
18:17:22 <Bike> somehow i don't think he understands that
18:21:21 -!- david_werecat has quit (Ping timeout: 256 seconds).
18:22:40 -!- zzo38 has joined.
18:26:56 <kmc> (/) :: (Num a, MonadPlus m) => a -> a -> m a
18:28:44 <elliott> or even div :: Nat -> PosNat -> Nat etc.
18:31:23 -!- ogrom has joined.
18:36:50 <kmc> div :: Nat -> (d :: Nat) -> Positive d -> Nat
18:37:08 <kmc> look at all the types i know
18:37:22 <Bike> i'm a bit curious what he even means by "destroying information"
18:37:29 <elliott> kmc: not the best infix operator
18:37:32 <Bike> like, is 14/2 supposed to be distinct from 7?
18:47:22 -!- ifnspifn has joined.
18:50:33 -!- sebbu has quit (Ping timeout: 245 seconds).
18:51:52 -!- ogrom has left ("Left").
18:54:40 <nooodl_> Bike: i think he just means, "x * 0 = 0 is bad"
18:54:53 <kmc> "0 but true"
18:55:03 -!- AnotherTest has joined.
18:56:32 <Bike> nooodl_: yeah but i'm trying to put more thought into it than he did
19:22:38 -!- Lumpio- has quit (Read error: Connection reset by peer).
19:22:44 -!- Lumpio_ has joined.
19:40:12 -!- sebbu has joined.
19:40:12 -!- sebbu has quit (Changing host).
19:40:12 -!- sebbu has joined.
19:44:26 -!- sebbu has quit (Remote host closed the connection).
19:45:06 -!- sebbu has joined.
19:45:06 -!- sebbu has quit (Changing host).
19:45:06 -!- sebbu has joined.
19:49:58 -!- epicmonkey has quit (Ping timeout: 246 seconds).
19:53:03 -!- sebbu has quit (Ping timeout: 245 seconds).
19:54:29 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds).
20:07:04 -!- oerjan has joined.
20:09:37 <oerjan> <Bike> i'd call it inefficient, have you /seen/ godel-encoded stuff?
20:10:55 <oerjan> you _can_ encode things more compactly than using prime factorization exponents, if you want. it's just that prime factorization exponents is intuitive to think about, even if not efficient.
20:11:42 <Bike> not for me, but of course that's very subjective
20:12:09 <oerjan> in fact we had a discussion some years ago about how to encode brainfuck programs bijectively to numbers, and you had to choose the coding carefully to avoid short programs giving huge blowup. but it is possible.
20:12:20 <elliott> did we ever come up with a viable encoding
20:14:08 <oerjan> i think so. wasn't there this fibonacci base encoding of a list?
20:14:57 <oerjan> basically you took each value, wrote it in fibonacci base, interspersed 11 between them (which cannot occur in fibonacci base) and interpreted _that_ as binary.
20:15:14 <oerjan> it's reasonably balanced between the list elements, naturally.
20:15:42 <elliott> oerjan: um that doesn't work for a bijection does it
20:15:54 <oerjan> *fibonacci base beginning with 1 and ending with 0
20:16:57 <oerjan> i think there were some small details that needed to be just right to make it a bijection
20:17:36 <oerjan> or wait maybe you just interspersed 1
20:17:44 <oerjan> since they were already beginning with 1
20:18:05 <oerjan> and to reverse, you broke up the binary at strings of several 1's
20:19:29 <oerjan> hm that encoding may have had some trouble with deep nesting, though.
20:20:08 <oerjan> since there's a constant multiple of approxim. phi/2 for each level
20:22:01 <zzo38> I think I have had ideas about some encoding of brainfuck programs with natural numbers.
20:22:05 <zzo38> Maybe you could have that since all high bits are zero, they correspond to [] at beginning of the program, might be able to make it bijective.
20:26:53 <oerjan> ifnspifn: you might also be interested in http://esolangs.org/wiki/Fractran
20:31:49 -!- atriq has joined.
20:32:09 -!- Phantom_Hoover has joined.
20:34:58 <oerjan> <FreeFull> I don't know if all combinations of / and \ are valid /// programs
20:35:16 <atriq> "\" on its own isn't, I don't think?
20:35:22 <ifnspifn> oerjan: Fractran's nuts... thanks for sharing!
20:35:39 <oerjan> sure they are. there's a "if you don't have a complete command the program halts" rule.
20:35:47 <oerjan> ifnspifn: you're welcome!
20:36:22 <oerjan> and of course only the current command is parsed at all, since the rest can be changed at any time.
20:37:16 * pikhq_ randomly notes that fopencookie is a neat interface
20:51:29 -!- FireFly has joined.
20:52:16 * oerjan swats FireFly -----###
20:52:25 -!- AnotherTest has quit (Quit: Leaving.).
21:24:06 -!- augur has quit (Remote host closed the connection).
21:24:36 <elliott> fizzie: You should help me get IPv6 working!!!
21:28:04 <fizzie> elliott: I can not fathom why it would not work. Does it work from irssi at all? If you, say, "/connect -6 -network Freenode chat.eu.freenode.net 6667".
21:28:27 <elliott> fizzie: Well, what I did from telnet was SSL-less.
21:28:35 <elliott> I guess I could set my irssi to non-SSL for a bit.
21:29:01 <fizzie> That /connect with explicit 6667 port should be SSL-less.
21:29:24 <elliott> Will that do my autojoins and so on?
21:30:50 <zzo38> Maybe you have to run the autojoin macro manually after connecting?
21:32:00 <fizzie> elliott: I think with -network Freenode it should work just as if you would have added a temporary server to the Freenode network.
21:32:07 <fizzie> But I'm certainly un-sure.
21:32:14 <elliott> fizzie: I have to manually disconnect before doing that, right?
21:32:20 <elliott> Or it'll try to connect to both servers at the same time or something.
21:35:07 -!- epicmonkey has joined.
21:35:42 <fizzie> I would be scared, too.
21:36:42 <elliott> fizzie: You should try it!
21:43:05 <shachaf> elliott: Colenses are "pretty exciting huh"
21:43:22 -!- epicmonkey has quit (Remote host closed the connection).
21:44:47 <fizzie> oerjan: Bulb, by the way.
21:45:15 <atriq> elliott, did you start the fortress yet
21:45:29 -!- epicmonkey has joined.
21:45:38 <elliott> I'm waiting for fizzie to try it.
21:45:45 <fizzie> atriq: How could he, with the IPv6 problem still unfixed? It's unthinkable.
21:46:38 <fizzie> Everything would catch flame.
21:46:47 <atriq> I thought IPv6 didn't exist in the UK or something
21:47:07 <elliott> fizzie: It'll be exciting!
21:50:11 <fizzie> I foresee a problem here.
21:50:22 <shachaf> ion: would i be a better person if i acquired ftl
21:50:40 <fizzie> Perhaps it would be best to nip it off in the bud.
21:50:57 <oerjan> (nothing like merging the conversations, right?)
21:51:25 <ion> shachaf: No, but everyone else would be worse, so it would be a net win for you.
21:51:54 <fizzie> I want to say "killer", but I'm afraid it might have negative social consequences.
21:52:51 <shachaf> ion: Do I need OpenGL for it?
21:53:56 <ion> shachaf: yes
21:55:55 -!- atehwa has quit (Remote host closed the connection).
21:58:25 <zzo38> fizzie: I think you need not worry about those kind of social consequences in this specific circumstances.
22:06:01 <fizzie> No can do, I am suddenly busy with various important matters.
22:10:40 -!- epicmonkey has quit (Ping timeout: 250 seconds).
22:15:59 <oerjan> a sudden urge to hide the bodies better
22:20:28 <zzo38> Can you disguise them?
22:21:00 <oerjan> i mean fizzie's sudden urge. to avoid negative social consequences, you see.
22:26:36 <zzo38> Well, in Dungeons&Dragons game I have idea what to do with the chancellor's dead body, which includes both disguising it and hiding it, and casting a spell on myself to forget it.
22:28:35 <zzo38> Therefore, I think hiding the bodies is not sufficient, live or dead.
22:32:00 -!- monqy has joined.
22:59:19 <atriq> Why is my room so cold
22:59:53 -!- carado has quit (Ping timeout: 246 seconds).
23:02:23 <olsner> atriq: winter is coming
23:03:37 <olsner> also, because your country hasn't invented insulation yet
23:04:18 <olsner> (tip: Alt-F4 usually closes the window)
23:10:06 <Phantom_Hoover> nortti, that's not even consistent on the command line
23:12:24 -!- atriq has quit (Quit: Leaving).
23:15:23 <zzo38> I sometimes use Ubuntu computer in FreeGeek, so I know at least in Ubuntu, close window is ALT+F4 like in Windows.
23:16:00 <fizzie> ^w closes some windows too.
23:18:49 -!- sebbu has joined.
23:25:35 -!- augur has joined.
23:27:23 -!- carado has joined.
23:34:57 <oerjan> no, temperature, you're _not_ allowed to drop below -10 celsius :(
23:35:34 <pikhq_> I've been waiting for it to drop below 0°C here...
23:35:38 <pikhq_> (this is a weird ass winter)
23:36:32 <fizzie> Foreca is forecasting -21°C for the Sun-Mon night.
23:37:59 <fizzie> Daily highs go -12, -9, -6 -14, -11, -7, -7, -6, -4, -5 in the ten-day forecast, that's not so cold.
23:38:25 -!- augur has quit (Read error: Connection reset by peer).
23:38:50 <pikhq_> At that point just start doing it in Kelvin.
23:39:11 <Fiora> Wikipedia says it's 33-55K
23:39:23 -!- augur has joined.
23:39:24 <oerjan> pluto is having a heat wave?
23:40:13 <Bike> pluto doesn't have an atmosphere worth talking about, right?
23:40:32 <Bike> so no climate. sucks
23:40:32 <Fiora> "Pluto's atmosphere consists of a thin envelope of nitrogen, methane, and carbon monoxide gases, which are derived from the ices of these substances on its surface.[89] Its surface pressure ranges from 6.5 to 24 μbar."
23:40:56 <Fiora> the pressure is about like 0.15-0.3 pascal
23:41:10 <Fiora> accordnig to like, different measurements >_>;
23:41:15 <Fiora> I'm guessing new horizons will make things a lot more certain
23:41:19 <Bike> well it's not like we've been there
23:41:27 <Fiora> they did it with observations of occultations with stars
23:41:28 <Bike> (by "we" i am referring to our robot slaves)
23:41:42 <Bike> yeah, i know, but that's not as accurate as just flying a barometer out there, is it
23:41:51 <oerjan> maybe our robot slaves will revolt and send us there.
23:41:53 -!- carado has quit (Ping timeout: 246 seconds).
23:42:01 <Bike> one can only hope
23:42:04 <Fiora> I don't think we're actually putting the spaceship in the atmosphere XD
23:42:16 <Bike> boooooriiiiing
23:42:28 <Bike> i wanna see if arnold's cousin's corpse is actually there, from magic school bus
23:42:38 <oerjan> yeah you need to bore deeply to get to the interesting parts
23:43:09 <oerjan> magic school bus had deaths?
23:43:23 <Bike> that's what we need new horizons to find out
23:44:03 <oerjan> mind you i think i may have seen approx. 1/2 episode of that show.
23:45:50 -!- jfischoff has quit (Quit: jfischoff).
23:46:23 -!- sebbu has quit (Ping timeout: 245 seconds).
23:47:18 <elliott> 23:39:04 <Fiora> only -74?
23:47:21 <Fiora> space probes are so cool, I'll finally be able to actually live through one exploring the outer planets
23:47:21 <elliott> Fiora: i didn't specify units!!
23:47:33 <Fiora> I mean I was alive with Galileo but I was really young
23:48:13 <Fiora> and voyager 2 passed neptune the year before I was born
23:52:30 <Phantom_Hoover> i remember reading a book about space when i was little and there was a thing about cassini
23:52:43 <Phantom_Hoover> when it actually reached saturn i was like "holy shit it's the future!"
23:55:29 <Fiora> wow I totally forgot about that one
23:55:45 * pikhq_ randomly notes that Planck temperature is perhaps the hardest to actually *use* temperature scale.
23:55:58 <Fiora> I remember the pictures it took and everything
23:55:59 <Bike> planck temperature is some absurdly high thing, right
23:56:02 <Fiora> and all the news about Titan
23:56:09 <pikhq_> 0 Planck temp = 0 K. 1 Planck temp ~= 1.4*10^8 YK.
23:56:19 <pikhq_> Bike: Yes. It's in yottakelvin.
23:56:26 <Bike> an oft-used unit right there
23:56:31 <Fiora> the planck temperature is where they hypotheize grand unification right?
23:56:36 <pikhq_> I'd use a larger SI prefix but they don't exist.
23:56:54 <pikhq_> Fiora: Among other things.
23:57:12 <zzo38> Combine SI prefixes if it is really necessary (even though you are not supposed to combine them)
23:57:30 <olsner> bah, it's not even a gigayottakelvin
23:57:35 <Phantom_Hoover> i thought planck units were just where the theory totally breaks down
23:57:37 <pikhq_> It's about 0.14 gigayottakelvin.
23:57:48 <Phantom_Hoover> not necessarily where some physically-significant thing happens
23:57:50 <zzo38> I have used "decimicron" before, which is short for "decimicrometre"
23:57:52 <pikhq_> Phantom_Hoover: Yuh. And then people guess what that means.
23:58:21 <pikhq_> With no justification, because why figure shit out when you can speculate. ;)
23:58:23 <elliott> obviously what this proves is that 1.4*10^8 yottas is the biggest number ever
23:58:28 <elliott> ultrafinitism proven by physics
23:58:38 <zzo38> And there are also some units that I don't like, such as "tonne", so we can use "megagram" instead, for example.
23:58:48 <Jafet> 10^3 K: weld; 10^6 K: fusion; 10^32: grand unification?
23:58:57 <elliott> Fiora: have you ever SEEN an attoparsec
23:59:09 -!- Phantom_Hoover has quit (Quit: Leaving).
23:59:12 <Bike> wait, wait, wait, i thought that at least some of the planck units do correspond to physical things, like with energy quantification and all
23:59:16 -!- Phantom_Hoover has joined.
23:59:27 <pikhq_> Beheheh. The peak wavelength emitted by a black body radiator at Planck temperature is also Planck length.
23:59:43 <zzo38> "Kilokilogram" can be used too, but "kilo" is already the SI prefix for "gram" (even though "kilogram" is considered the base unit)
23:59:48 <Fiora> I think um... 1TeV is roughly equal to 11 petakelvins
23:59:55 <Phantom_Hoover> oi elliott use your dvds and find out what the biggest number is