00:01:07 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
00:02:10 -!- tromp_ has quit (Remote host closed the connection).
00:03:57 -!- tromp_ has joined.
00:13:10 -!- ^v has quit (Read error: Connection reset by peer).
00:13:36 -!- ^v has joined.
00:38:10 <Sgeo> Why do people hate checked exceptions so much?
00:39:04 <copumpkin> if you're going to have them, you might as well represent them in the type though
00:40:02 -!- tromp_ has quit (Remote host closed the connection).
00:40:35 -!- tromp_ has joined.
00:44:53 -!- tromp_ has quit (Ping timeout: 252 seconds).
00:48:04 <Taneb> <Taneb> I’m a story about the prohibition of chocolate
00:52:46 <kmc> <Taneb> kmc, I was trying to go to a sci-fi and fantasy society social, and I went to the wrong bar <Taneb> Wound up at my university's fetish society <Taneb> Didn't realise for an hour and a half
00:53:15 <kmc> i never suspected otherwise
00:53:16 <Taneb> My gran won't shut up about it
00:53:22 <kmc> ...........
00:53:30 <Taneb> I regret telling my gran
00:53:35 <kmc> ......................
00:55:13 <kmc> i'm doin' fine
00:56:31 -!- shikhout has joined.
00:59:15 -!- Bicyclidine has quit (Quit: leaving).
00:59:55 -!- shikhin has quit (Ping timeout: 264 seconds).
00:59:55 -!- shikhout has changed nick to shikhin.
01:09:30 -!- ket1v has joined.
01:09:39 -!- tromp_ has joined.
01:14:12 -!- ket1v has quit (Ping timeout: 252 seconds).
01:15:56 <Taneb> btw I recommend Bootleg
01:16:23 <Taneb> Or its anime adaptation, Chocolate Underground, which is hilarious because of how the voice actors mangle the name "Myrtle"
01:28:23 <Sgeo> Maybe doesn't contain information about the nature of the error
01:28:30 <Sgeo> And Either is hard to combine
01:28:50 <Sgeo> Idris's EXCEPTION might be good?
01:29:00 <Bike> well, it's loud. that's always good.
01:29:04 <Sgeo> Still not sure how to actually combine those either without discarding the information
01:29:42 <Sgeo> (iiuc, if you put it in a Maybe context you can have ]EXCEPTION a, EXCEPTION b], but in a Either context you need to choose :/
01:30:42 <Sgeo> Oh hey, copumpkin's here, e could clear things up
01:33:01 <oerjan> i think idris might fall into a tradition of giving dependently typed languages subtly dubious names
01:33:31 <oerjan> like, they claim it's a dragon, but it's also a muslim prophet
01:34:10 <Bike> so what's dubious
01:34:37 <oerjan> that they don't mention that it's a muslim name
01:35:02 <oerjan> kings of libya and morocco
01:35:25 <Bike> i hear 'lisp' is also a speech condition
01:36:04 <kmc> hönan agda
01:36:56 <oerjan> well could be worse, could have been iblis
01:45:41 -!- yorick has quit (Read error: Connection reset by peer).
02:02:56 <quintopia> i'm adding "Lady Chablis" to the list of things to name an esolang
02:03:16 <quintopia> also, i should probably write up some of these langs lying around on my desk sometime...
02:04:47 <Sgeo> I should probably make another esolang
02:07:43 -!- nisstyre has quit (Ping timeout: 264 seconds).
02:07:54 <Sgeo> Braintrust, I think
02:41:55 -!- Sprocklem has quit (Ping timeout: 264 seconds).
02:42:52 -!- nisstyre has joined.
03:09:13 -!- huh has joined.
03:10:35 -!- ket1v has joined.
03:15:28 -!- ket1v has quit (Ping timeout: 265 seconds).
03:19:55 -!- BeingUntoDeath has joined.
03:29:13 -!- huh has quit (Quit: .).
03:32:13 -!- Sellyme has quit (Excess Flood).
03:32:52 -!- tromp_ has quit (Remote host closed the connection).
03:34:05 -!- Sellyme has joined.
03:37:12 -!- tromp_ has joined.
03:39:36 <kmc> Taneb: so did you also explain to your gran what a fetish society is or did she already know
03:46:04 -!- BeingUntoDeath has quit.
03:46:22 -!- BeingUntoDeath has joined.
03:56:00 <oerjan> disturbing twist: his gran once founded that society
04:00:05 -!- Sorella has quit (Quit: It is tiem!).
04:09:35 -!- tromp_ has quit (Remote host closed the connection).
04:10:08 -!- tromp_ has joined.
04:13:02 -!- Sellyme has quit (Excess Flood).
04:13:36 -!- Sellyme has joined.
04:15:04 -!- tromp_ has quit (Ping timeout: 268 seconds).
04:30:31 -!- shikhin has quit (Ping timeout: 264 seconds).
04:43:34 -!- Sprocklem has joined.
05:00:19 <zzo38> I made up a "Pokemon renaming rule", for use with Pokemon Card when the cards are drafted or otherwise randomly selected, rather than playing a Constructed game.
05:04:45 -!- password2 has joined.
05:05:04 <zzo38> Before the game starts, make up a list of names of cards in your deck (ignoring level and quantity), on the left side of the page. For each one, write the name you are renaming it to on the right. You can only rename into names of cards which are possible to be drafted (whether or not anyone did, whether or not it is in your deck). The left side cannot have any duplicates, and the right side cannot have any duplicates.
05:05:57 <zzo38> Basic pokemon cards can only be renamed to names of basic pokemon cards. Evolution cards can be only renamed into name of evolution cards. Trainer cards can only renamed into name of trainer cards. Special energy cards can only renamed into name of special entry cards. Basic energy cards cannot be renamed at all. Renaming a name into itself (so it doesn't change) is permitted.
05:06:05 <zzo38> Bike: I wasn't finished yet; now I am.
05:06:41 <Bike> i stand by my assessment
05:08:11 <zzo38> Well, many cards in the game (mostly evolution cards) require card of other specific name, and if cards are drafted or selected at random or something like that, then you cannot do that so easily. This rule is not meant to be used in a Constructed game.
05:11:20 -!- ket1v has joined.
05:15:58 -!- ket1v has quit (Ping timeout: 240 seconds).
05:29:58 <Sgeo> "In gbeta, object metamorphosis coexists with strict, static type-checking: It is possible to take an existing object and modify its structure until it is an instance of a given class, which is possibly only known or even constructed at run-time. Still, the static analysis ensures that message-not-understood errors can never occur at run-time."
05:30:20 <Sgeo> I think I tried to learn gBeta once. I don't think I succeeded
05:39:44 <password2> if you stopped trying , then you probably did
06:02:45 -!- oerjan has quit (Quit: leaving).
06:09:14 -!- confuzedstuff has quit (Ping timeout: 246 seconds).
06:11:32 -!- nisstyre has quit (Quit: bai).
06:28:53 -!- password2 has quit (Ping timeout: 268 seconds).
06:30:29 -!- heroux has quit (Read error: Connection reset by peer).
06:36:13 -!- Vorpal has quit (Quit: ZNC - http://znc.sourceforge.net).
07:11:33 -!- BeingUntoDeath has quit.
07:12:12 -!- ket1v has joined.
07:15:30 -!- Slereahphone has joined.
07:16:06 -!- Sprocklem has quit (Ping timeout: 252 seconds).
07:17:02 -!- ket1v has quit (Ping timeout: 252 seconds).
07:32:35 -!- Slereahphone has quit (Remote host closed the connection).
07:46:31 -!- ^v has quit (Quit: Leaving).
07:56:48 -!- Zom-B|zz has changed nick to Zom-B.
07:57:45 -!- Slereahphone has joined.
07:59:47 -!- Slereahphone has quit (Remote host closed the connection).
08:18:47 -!- password2 has joined.
08:18:57 -!- Slereahphone has joined.
08:29:12 -!- MoALTz has joined.
08:31:00 -!- conehead has quit (Quit: Computer has gone to sleep.).
08:37:54 <Zom-B> i'm making an 2-dimensional esolang and I'm still trying to figure out some details
08:40:06 <Zom-B> like, numbers are encoded in unary, so should i make a special constant for '0' or make everything starting from 1?
08:40:21 <Zom-B> (memory addresses, etc)
08:50:41 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
08:51:52 -!- Slereahphone has joined.
08:55:35 -!- Phantom_Hoover has joined.
09:11:10 <Zom-B> or whether to use two instructcions to address from the begin or from the end of the memory, or use one instruction with negative addresses
09:12:03 <Zom-B> negative addresses can be made by pushing a unary value and then calling the NEG instruction
09:13:16 -!- ket1v has joined.
09:17:15 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
09:17:50 -!- ket1v has quit (Ping timeout: 252 seconds).
09:17:58 -!- password2 has quit (Ping timeout: 240 seconds).
09:22:27 -!- fungot has quit (Ping timeout: 252 seconds).
09:23:06 -!- fizzie has quit (Ping timeout: 245 seconds).
09:44:23 -!- fizzie has joined.
10:00:46 -!- Phantom_Hoover has quit (Remote host closed the connection).
10:02:21 -!- Phantom_Hoover has joined.
10:14:48 -!- boily has joined.
11:03:02 -!- boily has quit (Quit: Can one trigger a botloop with a /quit message?).
11:14:06 -!- ket1v has joined.
11:18:38 -!- ket1v has quit (Ping timeout: 240 seconds).
11:37:41 -!- fungot has joined.
11:39:15 <Deewiant> Zom-B: If subtraction is possible then zero is x-x for any x.
11:56:12 -!- shikhin has joined.
12:03:58 -!- yorick has joined.
12:06:35 -!- Slereah has joined.
12:08:19 -!- Slereah_ has quit (Ping timeout: 264 seconds).
12:13:39 -!- Sorella has joined.
12:16:35 -!- Sellyme has quit (Excess Flood).
12:17:06 -!- Sellyme has joined.
12:29:48 -!- Sgeo has quit (Read error: Connection reset by peer).
12:43:02 -!- MindlessDrone has joined.
12:55:32 -!- heroux has joined.
12:56:48 -!- shikhout has joined.
12:58:52 -!- tertu has joined.
12:59:55 -!- shikhin has quit (Ping timeout: 264 seconds).
12:59:56 -!- shikhout has changed nick to shikhin.
13:01:54 -!- heroux has quit (Ping timeout: 252 seconds).
13:02:49 -!- heroux has joined.
13:15:02 -!- ket1v has joined.
13:19:40 -!- ket1v has quit (Ping timeout: 265 seconds).
13:24:06 -!- heroux has quit (Read error: Operation timed out).
13:28:11 -!- heroux has joined.
13:37:16 -!- tertu has quit (Remote host closed the connection).
13:56:10 -!- tertu has joined.
14:56:41 <Zom-B> sorry, was afk at a neighbor, 'fixing' his windows 8
14:57:09 <Zom-B> yes, x-x is possible
14:58:13 <Zom-B> bit it would take 3 instructions
14:58:31 <Zom-B> push x, push x again, subtract
14:58:56 -!- nooodl has joined.
15:01:14 -!- oklopol has quit (Ping timeout: 246 seconds).
15:07:02 <Zom-B> i just made everything 1-based
15:10:38 -!- tertu has quit (Disconnected by services).
15:10:39 -!- ter2 has joined.
15:13:13 -!- oklopol has joined.
15:15:52 -!- ket1v has joined.
15:20:55 -!- ket1v has quit (Ping timeout: 264 seconds).
15:26:34 -!- tertu3 has joined.
15:30:09 -!- ter2 has quit (Ping timeout: 265 seconds).
15:43:06 -!- password2 has joined.
15:58:38 <password2> wow , nice i see someone actually edited and correct my bf^ page
16:01:06 <Zom-B> in thinking of also releasing a shunken-down version of my new esolang with only 7 symbols, equal to the brainfuck <>+-,.?
16:01:30 <Zom-B> ? turns if the value is nonzero, otherwise it goes straight
16:03:13 <Zom-B> so (too?) many BF clones already
16:04:10 <password2> is there an upper limit of how many there may be?
16:04:34 <password2> the mathematics of large numbers say n==n+1
16:05:19 <password2> so thusly you may create another one without it affecting how much there is
16:05:44 <Zom-B> i'm a computer scientist, not a mathematician
16:09:10 <Zom-B> go back to your engines
16:10:51 <Zom-B> then some WireWorld
16:12:45 <fizzie> I'd `? brick if there was a working HackEgo.
16:13:36 <Zom-B> HackEgo is broked?
16:14:05 <fizzie> `run echo "No output." # at least this still works
16:14:52 <Zom-B> maybe the specification says it should output "No output."for all possible inputs except the input of zero length
16:18:58 <Jafet> I'm now pretty convinced that mathematica can do anything.
16:19:15 <Jafet> http://46.4.207.77/sum.gif
16:20:20 <Zom-B> i invented that algorithm when i was young
16:20:45 <Zom-B> also recently used it to encode two integers into a single integer
16:21:41 <Zom-B> with approximate preservation of magnitude, so not just bit fiddling
16:28:15 -!- Sellyme has quit (Excess Flood).
16:28:37 -!- Sellyme has joined.
16:34:22 <Zom-B> i think i've figured out the complete design of my esolang
16:35:40 <Zom-B> fizzie: is it possible you could rename a wiki page (not 'move', but really rename in the database)? It's not linked from or to anything yet
16:38:35 <Zom-B> it's like this: (an analogy)
16:38:36 <Zom-B> it's like i started something that's like cooking recepies
16:38:36 <Zom-B> then called it chef and made the wiki page
16:38:36 <Zom-B> then the design went through so many iterations that it ended up like an instruction manual instead
16:38:55 <password2> i tried to click the buttons of that gif
16:41:39 <Zom-B> i.e. the essence that the name captured is not in the language anymore
16:42:16 <Zom-B> moving leave residues and fossils
16:42:50 <Phantom_Hoover> just rename it, the page history really doesn't matter that much
16:44:23 <Zom-B> i still need to think of a good name
16:46:51 <Zom-B> the new essence is: its 2-dimensional and self-modification is required to make loops
16:47:16 <Zom-B> (except the BF variant)
16:56:45 <Zom-B> maybe i should release the BF vaiant on april 1st
16:58:33 <Zom-B> zzo38: how do you choose names for all your esolangs?
17:05:56 -!- ^v has joined.
17:15:39 <fizzie> There's a HackEgo command for it, but it's also inoperative.
17:16:00 <fizzie> (I don't think any languages have yet been named by `words --esolangs, though.)
17:16:52 -!- ket1v has joined.
17:21:27 -!- ket1v has quit (Ping timeout: 252 seconds).
17:21:34 <fizzie> There are several 2D (or more) brainfuck derivatives, at least.
17:21:47 <fizzie> Not sure if any of them tries to do only the 2D thing and nothing else.
17:23:34 <password2> i sould probably not make it run over irc , atleast not the memory dump parts
17:24:03 <fizzie> A quick intersection of Category:Two-dimensional languages and Category:Brainfuck derivatives returns: 2L, Archway, Braincopter, Brainloller, Braktif, Electric BitFunk, Minifuck-2D, Minimal-2D, Pirandello, Puzzlang, Sansism and YABALL.
17:31:03 <password2> mmm , i see none of these programs usses 1D instructions over a 2D tape
17:33:43 <myname> what about brainfuck 2d?
17:33:54 -!- MoALTz_ has joined.
17:35:18 <myname> there is, but i don't like it
17:35:37 <myname> basically, the instruction depends on the 8 directions you can take
17:35:57 <myname> e.g. going south-west could be +
17:36:20 <password2> most of em uses an 2d instructional set
17:36:51 -!- MoALTz has quit (Ping timeout: 252 seconds).
17:36:53 <myname> 2l looks pretty hard for a brainfuck derivate
17:37:12 <fizzie> Oh, you meant just having a plane instead of tape for the memory pointer, not code?
17:37:30 <myname> a plane for the code, but still a tape
17:38:06 <fizzie> password2: In that case, it wouldn't be in Category:Two-dimensional languages, of course.
17:38:57 <myname> tbh, i don't see any fun in n-dimensional tapes
17:39:24 <fizzie> There probably isn't really a category for that, but there are all kinds of variants that do something to the tape.
17:40:18 <password2> all i want to do is replace <> with wasd , i think
17:41:00 <password2> its just something i kinda always wanted to try
17:41:15 <password2> the fun is trying to discover what fun it is:
17:44:04 <password2> essentially w would only translate to >>>>>>>>>>>>>
17:44:37 <myname> that would not be truly 2d
17:45:59 <myname> because i could reach the cell "above" the current by moving to the right?
17:46:31 <password2> see it as a tape rolled into a cylinder
17:46:49 <myname> you could just make generic macros, then
17:49:14 <password2> or maybe i'll make it like a bunch of parralel tapes
17:49:17 <fizzie> Clusterfuck is kinda-sorta close, with a fixed-size memory.
17:51:03 <myname> password2: just make something like #define w >>>>>>>>>> possible and you can make stuff
17:51:47 <password2> i'll see what i will do when i actually start
17:55:49 <password2> the farther you move from center the bigger the tape would be
17:56:30 <maurer> You could make it move at approximately constant angles rather than incrementing by cells
17:56:40 <maurer> So that when your radius was high, you'd sweep 8 bytes in a read/step
18:00:02 <password2> or i was think that that the 'width' of the cell is constand such that each ring has ,say, 4 more cells , and changing the angle by a small amount means that the first few cells from the center would still be the same and it diverges a bit as you move outwards
18:01:26 <myname> you could use a rubik's cube of say 6x6x6 as "take" with rotation instructions
18:02:50 <password2> ah , so the pointer almost tays fixed then and a /tape/ rotates
18:24:41 <password2> this is what i mean http://img.ffffound.com/static-data/assets/6/06f04fcd77d6038a71226f2edba1ea166c299449_m.png
18:25:23 -!- evalj has joined.
18:27:40 -!- conehead has joined.
18:50:57 <ion> http://zachholman.com/posts/only-90s-developers/
18:55:13 <Zom-B> i checked all those 2D BF derivatives and none is true tot he original BF essence
18:57:06 -!- shikhout has joined.
18:59:55 -!- shikhin has quit (Ping timeout: 264 seconds).
18:59:56 -!- shikhout has changed nick to shikhin.
19:17:42 -!- ket1v has joined.
19:22:20 -!- ket1v has quit (Ping timeout: 246 seconds).
20:26:38 -!- Sellyme has quit (Excess Flood).
20:26:56 <Zom-B> either instruction bloat (R, L, U, D) or instruction overloading (* does all of <>,.)
20:27:35 -!- spiette has quit (Quit: :qa!).
20:28:06 -!- Sellyme has joined.
20:30:35 <Zom-B> Brainloller is close to my idea, but instead of just ? it has [] and two extra rotation operations
20:31:07 <Zom-B> Archway is even closer, though not graphical and has two splite /\ instead of just ?
20:31:13 <password2> btw , do you want one whose instruction set is in 2d , or dataset?
20:31:37 <Zom-B> dataset 2d is a bit pointless
20:31:59 <Zom-B> you have to 'walk' from a to b anyway wether 1d or 2d
20:32:07 <Zom-B> you just need extra instructions
20:32:34 <password2> but now you have twice the directions
20:33:01 <password2> thats like saying reality is useless becaus you still have to travel
20:33:55 <Zom-B> its not like you have the crossing problem or anything, it's data not logic
20:34:34 <Zom-B> the average distance between any two cells is lower on average, but thats the only practical difference
20:34:57 <Zom-B> unless you want to make pictures with the memory map (instead of stdout, which is already possible)
20:35:21 <Zom-B> maybe a competitive variant might benefit from 2d memory though
20:36:26 <password2> and you can access data in a different sequence easier
20:37:36 <tromp> if you want exponential memory balls you should change from linear to a tree
20:39:00 <tromp> so add ^ for going up, and <> become left-down and right-down
20:39:30 <tromp> you can reach 2^n cells in n steps...
20:42:31 <Zom-B> lol, from linear to quadratic, to exponential. what'
20:42:35 <Zom-B> lol, from linear to quadratic, to exponential. what's next?
20:42:48 <Bike> it's not exactly triangular, given the interior nodes and all
20:44:22 <Bike> need a data structure where you can reach A(n,n) nodes in n steps
20:45:02 <Zom-B> actually hyperbolic might be suboptimal and mappable to tree
20:45:32 <Zom-B> iknow! try hilbert space
20:46:13 <Zom-B> thats not a space but a transform
20:46:26 <Zom-B> frequency domain is a space
20:46:51 <Zom-B> would be fun if program memory and data memory were mapped by the fourier transform
20:47:03 <Zom-B> if you mutate one too much, the other starts to slowly corrupt
20:48:29 <password2> i though about a simplistic version where you can only go forward
20:48:48 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:50:13 <password2> i was just thinking of random mutations
20:51:13 <password2> and you have to , dunno wrap around to get back where you want to be
20:51:46 <Zom-B> can't be turing complete
20:53:13 <Zom-B> wait, maybe it can if memory is circular and values infinitely big
20:53:53 <Zom-B> but how would you prove it for a circle of N cells with the limit N->1?
20:54:16 <Zom-B> if that is turing complete then a single memory cell memory would also be
20:54:43 <Zom-B> i think it boils down to random-access within a number (like bit-fiddling)
20:55:01 <Zom-B> if you dont have that it's not turing complete
20:55:10 <myname> Zom-B: afaik, 2 bigint-cells are known to be TC
20:55:57 <Zom-B> a cell is more like a stack (of unary numbers no less), not a ram
20:56:18 <password2> the unltimatly worst language to edebug would be one where the outcome of a cmd is linked to a random chance
20:57:34 <myname> http://esolangs.org/wiki/Collatz_function
21:13:15 -!- password2 has quit (Quit: "quick pm me when you read this!").
21:18:44 -!- ket1v has joined.
21:22:49 <Zom-B> my CAT program is evil / holy (depending on your viewpoint/background): http://esolangs.org/wiki/File:Cat-inf.png
21:23:05 -!- ket1v has quit (Ping timeout: 246 seconds).
21:23:46 <Zom-B> and the my brand new esolang that it runs on: http://esolangs.org/wiki/Floater
21:24:22 <Zom-B> (i'm deciding to keep the name after all, too lazy to rename it to a name i have yet to invent)
21:24:55 -!- Sprocklem has joined.
21:25:15 -!- Zom-B has changed nick to Zom-B|zz.
21:40:07 -!- trn has quit (Ping timeout: 264 seconds).
21:40:43 -!- EgoBot has quit (Ping timeout: 264 seconds).
21:41:36 -!- EgoBot has joined.
21:43:18 -!- trn has joined.
21:49:20 -!- lambdabot has quit (Read error: Connection reset by peer).
21:53:48 -!- lambdabot has joined.
22:02:54 -!- ket1v has joined.
22:04:46 -!- Bike_ has joined.
22:05:01 -!- prooftechnique has joined.
22:26:53 -!- oerjan has joined.
22:48:23 -!- Bike_ has quit (Remote host closed the connection).
22:48:37 -!- Bike_ has joined.
22:53:05 -!- nooodl has quit (Ping timeout: 268 seconds).
22:54:38 <oerjan> <Zom-B> fizzie: is it possible you could rename a wiki page (not 'move', but really rename in the database)? It's not linked from or to anything yet <-- i think moving and then deleting the old should work well enough?
22:55:01 <oerjan> i think history is transfered with a move automatically
22:55:27 <oerjan> (only admins can do the deletion bit)
22:56:22 <oerjan> <password2> i tried to click the buttons of that gif <-- wait a moment, not only did i do that but it seemed to _work_.
22:56:38 <oerjan> (as in, it was stalled before and then continued)
23:00:51 <oerjan> @tell Zom-B|zz If you move the page and one of us admins deletes the original, the effect will be the same as renaming (page history is automatically transfered on moving)
23:05:43 -!- MoALTz_ has quit (Quit: Leaving).
23:06:48 <oerjan> @tell password2 <password2> mmm , i see none of these programs usses 1D instructions over a 2D tape <-- see paintfuck
23:11:45 <oerjan> <tromp> you can reach 2^n cells in n steps... <-- 3^n if make it infinite upwards too
23:12:29 <tromp> to get 3^n you need 3 down dirs
23:12:56 <tromp> a ternary tree stil lhas only one up direction
23:13:19 <tromp> what does your tree look like?
23:14:33 <oerjan> didn't think about the fact ^ reverses both < and >
23:15:33 <oerjan> however, it should be possible to make a tree that's entirely symmetric in ^ < >
23:15:42 <oerjan> (still only 2^n though)
23:16:56 <oerjan> hm the easiest way would be make each direction its own inverse
23:17:39 <oerjan> essentially the free product of 3 copies of the group Z_2, i think
23:18:36 <oerjan> i am pretty sure i had 2 copies in my thesis somewhere
23:19:05 <tromp> then you're labeling the edges of a free 3-tree with 0,1,2 such that every vertex has all labels on incident edges
23:20:05 <tromp> easiest to make it 0,1,2 in clockwise order around each vertex
23:20:33 <oerjan> hm that's systematic at least
23:20:47 -!- evalj has quit (Remote host closed the connection).
23:20:51 <oerjan> you could also switch direction each step, by parity of distance
23:21:04 <tromp> or prettier is to color the edges red,green,blue
23:21:18 <oerjan> someone has to have made this fractal picture already
23:21:28 <tromp> each vertex has all three colors incident
23:21:41 <tromp> and the three movements are called R,G,B
23:22:19 <tromp> call it RGBfuck :)
23:22:44 <tromp> or BrgFuck to stay with the same initial 2 letters
23:24:06 <oerjan> i think i crashed google trying to find that tree
23:27:26 <oerjan> i might try restarting the net, but then i'm afraid i'd lose irc too...
23:27:31 <tromp> http://telliott99.blogspot.com/2010/11/phylogenetic-tree-surgery-1.html
23:27:56 <oerjan> wtf did _that_ page load.
23:28:40 <oerjan> hm maybe because it autoredirects to .no
23:29:26 <oerjan> pretty pictures, at least
23:35:24 <tromp> was wondering if we cld store bits in the RGB orientation
23:36:53 -!- Sgeo has joined.
23:37:01 <oerjan> you'd need a command to detect it, then
23:37:37 <tromp> [...] could be loop while clockwise oriented
23:38:03 <oerjan> this reminds me a little of V
23:38:19 <tromp> so this would do bitwise io
23:38:36 <tromp> . outputs 0 or 1 depending on orientation
23:38:45 <oerjan> like, you could have commands that mirror flipped the entire subtree in one direction
23:39:41 <tromp> hmm, you can only flip infinitely many bits at once:)
23:39:59 <tromp> this is getting wild
23:40:22 <oerjan> yep, that's the "reminds me of V" part
23:46:11 <fizzie> @tell Vorpal More silly photo stuff: http://zem.fi/2014-03-25-tl
23:47:12 <tromp> maybe the control should not only have a location but a direction as well
23:47:45 <tromp> if you're heading is B then RG would change direction
23:47:52 <tromp> while B would cross the edge
23:49:30 <oerjan> hm right, then the tree flipping command can depend on the direction.
23:49:47 <tromp> there wld only be one flip
23:50:16 <tromp> we can still limit the flip to one cell
23:50:29 <tromp> by swapping to adjacent edges
23:50:42 <tromp> instead of flipping over subtrees
23:50:55 <tromp> that's probably more sane
23:51:50 <tromp> so with heading B, Swap would swap R&G
23:52:47 <tromp> and , (input) would do a possible R&G swap to set orientation to input bit
23:53:03 <tromp> now we have exactly 8 instructions, like BF
23:53:45 <oerjan> mind you, the R&G swap has no detectible effect other than flipping the orientation of the current vertex.
23:54:03 <tromp> yes, it flips the local bit
23:54:15 <tromp> just like -/+ in bitfuck
23:55:20 <tromp> maybe it's the very first BF variant with exponential storage
23:55:45 <oerjan> no i think V counts for that too
23:56:07 <oerjan> http://esolangs.org/wiki/V
23:57:57 <oerjan> i had forgot that had a mirror command too