00:09:25 <tswett> I searched for "There is no scientifically-justifiable reason to exclude pornography, which is a vital part of the web ecosystem." and just got logs of this channel.
00:09:29 <tswett> And stuff unrelated to that quote.
00:10:25 <ais523> tswett: I think it was invented by someone /in/ this channel, when they were writing something in a serious or semi-serious style
00:11:10 <ais523> if I am right on that, it was probably fizzie, due to a quick guess at mine as to who would a) agree with that statement, and b) be doing something in which the sentence would have a reasonable context
00:11:22 <ais523> b) is the main factor here, probably most of the channel agrees with a)
00:14:54 -!- MDude has quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/).
00:15:55 <Taneb> I've got a new favourite out of context bible quote
00:16:06 <Taneb> "Fear ye not therefore, ye are of more value than many sparrows."
00:17:26 <tswett> So I've got one of those Chinese magic puzzle cubes.
00:17:56 <tswett> Also known as "like a Rubik's cube but better".
00:18:12 <tswett> This one's the "Pocket Cube"—2x2x2.
00:19:06 <tswett> I'm studyin' this bad boy.
00:20:07 <tswett> A pretty decent sequence is R U R' U'. It only affects four pieces, and it has order 6.
00:21:20 <tswett> Of course, cube theorists call it [R,U].
00:23:39 <tswett> It's not easy to explain what [R,U] does, but of course it generates sequences [R,U]^2, of order 3, and [R,U]^3, of order 2, and it can be understood in terms of these.
00:24:08 <ais523> Taneb: that statement is a lot more reassuring than its opposite would be
00:24:30 <tswett> Ye are of less value than many sparrows?
00:24:39 <ais523> tswett: you also have to negate the start
00:25:00 <ais523> the resulting statement would become disconcerting, rather than reassuring
00:25:08 <tswett> Fear ye therefore; ye are of less value than many sparrows?
00:25:19 <ais523> also, a 2×2×2 cube is equivalent to the corners of a 3×3×3 cube
00:25:25 <ais523> tswett: I'm glad the Bible doesn't say that *shudders*
00:25:28 <ais523> I don't know how I'd react
00:25:59 * ais523 wonders where to put the quantifier
00:26:05 <tswett> [R,U]^3 flips the FR edge and the UB edge.
00:26:37 <ais523> as in, is it "there is a group of many sparrows so that ye are of more value than that group of sparrows"
00:27:04 <ais523> or "out of all sparrows, many possible choices of sparrow would give ye more value than that sparrow"?
00:27:26 <tswett> [R,U]^2 rotates both of the FR corners clockwise, and both of the UB corners counterclockwise.
00:30:30 <tswett> Lemme see about [R,U]. [R,U] moves UFR to DFR like F, and DFR to UFR like R; it also moves UBR to UBL like U', and UBL to UBR like B'.
00:31:43 <izabera> :o people are talking about the rubiks cube
00:32:24 <tswett> The two-by-two-by-twube.
00:34:10 <izabera> you were talking about edges
00:35:03 <tswett> Sure it does. All cubes have edges. It just doesn't have pieces which correspond to those edges.
00:35:55 <tswett> In any case, by "the edge" I meant "the two pieces lying along the edge".
00:36:23 <int-e> the 2x2x2 cube has edges but no edge pieces.
00:37:06 <izabera> that's just the traditional design
00:37:33 <izabera> i'm sure someone could come up with a center that doesn't need any kind of hidden edges
00:39:25 -!- p34k has quit.
00:39:40 <ais523> you could create a spherical 2×2×2 cube
00:39:44 <int-e> edges. this thing has edge pieces but no edges: http://www.twistypuzzles.com/museum/large/00579-01.jpg
00:39:50 <ais523> except it wouldn't be a cube any more
00:39:54 <ais523> just isomorphic to one
00:40:00 <int-e> (I was looking for a 2x2x2 one, but that's what I found)
00:40:13 <tswett> izabera: it is impossible to create a cube of any kind which does not have edges.
00:40:34 -!- XorSwap has joined.
00:41:17 <tswett> All cubes have faces, right?
00:41:34 <tswett> And all cubes have pairs of adjacent faces, right?
00:41:53 <tswett> What do you call the area at which two adjacent faces meet?
00:41:58 -!- nooga has quit (Ping timeout: 244 seconds).
00:42:16 <tswett> That's the kind of edge I've been talking about this whole time.
00:42:50 <izabera> maybe you should have talked about the right kind of edges this whole time
00:43:06 <tswett> I don't have anything to say about that kind of edge.
00:43:23 <ais523> tswett: hmm, what if the faces /don't/ meet, and are only adjacent in the sense that they're aligned at 90 degrees from each other?
00:43:53 <tswett> That's an interesting thought.
00:44:22 <tswett> You could make a "cube" shaped like an octahedron with the corners just a little bit truncated.
00:45:51 <tswett> All right, where have they got some algorithms for this darned thing?
00:45:51 <int-e> there's a tetrhedron version of the 2x2x2 cube: https://en.wikipedia.org/wiki/Pyramorphix
00:48:00 <tswett> This sequence does some kind of corner rotation, but also has the effect of doing L' R on the whole thing: L' U R' D2 R U' R' D2 R2
00:48:15 <tswett> Let me see if I can adjust it so it doesn't do that.
00:50:10 <tswett> Let's see. Just start with R' instead of L', then rotate the rest of it pitch-up, that should give you... R' F R' B2 R F' R' B2 R2.
00:50:27 <tswett> Yup. Now let me see what that did.
00:51:11 <tswett> That sequence rotates UBL, UBR, and UFR clockwise like U.
00:52:36 <tswett> If you append U' to it, it instead swaps the UF pieces like U.
00:53:10 <int-e> try R U2 R' U' R U' R' U2
00:54:21 <tswett> Ooh, that's a fun one.
00:54:40 <tswett> That one rotates UBL, UFL and UFR counterclockwise.
00:55:01 <tswett> The one I previously mentioned moves the corners around forming a clockwise rotation.
00:55:08 <tswett> This one rotates each corner counterclockwise in place.
00:55:18 <int-e> so you can solve the cube now :P
00:56:26 <izabera> wouldn't it be better to just link a guide or something
00:56:43 <tswett> Can I find a big list of 2x2x2 algorithms somewhere?
00:57:18 <int-e> well, cuber. I can solve them, but I struggle with the notation, and I'm not very fast. "them" being, in principle, arbitrary size (4x4x4 is the most fun one, perhaps)
00:59:42 <int-e> People probably have big lists... but I don't know.
01:07:54 <tswett> Looks like I'm looking for "CLL" algorithms at the moment.
01:08:00 <tswett> I assume that means "completing the last layer".
01:11:16 -!- XorSwap has quit (Quit: Leaving).
01:12:42 <tswett> Let me find my favorite algorithm for flipping an edge around.
01:13:53 <int-e> disassemble; reassemble... unless you have an even sized cube larger than 2x2x2
01:15:30 <tswett> I just did it, actually.
01:17:17 <tswett> Here's how to do it: https://www.speedsolving.com/wiki/index.php/Special:MediawikiAlgDB?mode=view&view=default&puzzle=2&group=CLL&cid=27
01:18:22 <int-e> I see what you mean
01:19:11 <int-e> still won't work on odd sized cubes, but you didn't care about that
01:21:14 -!- MDude has joined.
01:40:52 <tswett> Though I'm doing those algorithms on a solved cube and getting some funny results.
01:44:02 <tswett> R U F2 U R F R' U F R U2 is an algorithm for flipping the UL edge around.
01:45:36 <tswett> And its reverse is the same thing, of course: U2 R' F' U' R F' R' U' F2 U' R'
01:46:14 <tswett> It doesn't seem like a very beautiful algorithm.
01:51:34 -!- XorSwap has joined.
02:02:46 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
02:16:31 -!- jaboja64 has quit (Ping timeout: 244 seconds).
02:32:21 -!- XorSwap has quit (Quit: Leaving).
02:34:50 <ais523> philosophical question: is it possible to use test driven development to develop a testsuite?
02:35:04 <ais523> one of the things I'm writing at the moment is a generic terminal testsuite
02:35:14 <ais523> and I am testing it as I write it by using it to test terminals
02:35:25 <ais523> I am wondering what other development styles I could theoretically have used instead
02:37:56 -!- jaboja64 has joined.
02:47:36 <tswett> Of course you can do that.
02:47:52 <tswett> Just come up with some things that work and some things that don't work and test to see if they're caught as they're supposed to be.
02:50:11 <ais523> tswett: so in other words, I have to write a terminal that's broken in every possible way?
02:50:17 <ais523> and I have to do this /before/ writing my terminal testsuite?
02:50:40 <ais523> I think this leads to an impossible recursive loop, because the test tests would require the tests I'm originally trying to write in order to test them
02:51:13 <tswett> I could probably come up with an answer for that if it weren't for the fact that I just had a bottle of Dragon's Milk.
02:52:05 <ais523> come to think of it, I'd /also/ need a terminal that's working in every possible way :-D
02:52:07 <ais523> that sounds like a useful thing
02:52:30 <coppro> TIL that according to Tolkien, Bilbo and Frodo's names were actually Bilba and Froda, but he translated them as Bilbo and Frodo to align closer to the reader's expectations of masculine names
02:52:48 <zzo38> Maybe I should write a terminal emulator, to avoid the mess of xterm and the insufficiency of everything else
02:53:22 <zzo38> coppro: I did not know that
02:56:18 <ais523> zzo38: what do you consider insufficient in non-xterm terminals?
02:56:27 <ais523> as in, what about them makes them want to not use them?
02:56:55 <tswett> I knew that about Bilbo but not Frodo.
02:57:58 <zzo38> I think the other terminals do not use bitmap fonts is one, and also many ones add too much extra stuff as well as missing stuff too
03:02:13 <ais523> there aren't many bitmap fonts that cover all of Unicode
03:02:26 <ais523> additionally, providing bitmap fonts in enough sizes to let everyone scale their terminals as they want is difficult
03:02:47 <ais523> so I think most terminals don't use bitmap fonts because of practical considerations, rather than because they think they're inherently bad
03:06:55 -!- tromp has joined.
03:10:09 <zzo38> I would to use only bitmap fonts, but it wouldn't have Unicode nor have KOI8, anyways. (Unicode is not even compatible with the pieces of big Sigma in the DEC Technical set anyways.)
03:12:24 <zzo38> (We would need to make up a UTCE bitmap font. A UTCE font is a .pcf font with two bytes per character.)
03:13:36 <zzo38> (Also requiring that character sizes are all the same for byte1 being 0 to 31, and for characters with byte1 being 128 to 255, the width must be exactly twice as much.)
03:16:00 <zzo38> The terminal emulator would then include the internal tables to convert ISO 2022 characters into UTCE characters
03:23:50 <zzo38> Another thing is that xterm does not support user-defined fonts, and there is a reasonable reasons for such, but it could be implemented anyways as an option only if the default font already has the correct size
03:27:11 -!- llue has joined.
03:27:27 -!- lleu has quit (Read error: Connection reset by peer).
03:27:30 -!- tromp has quit (Remote host closed the connection).
03:48:15 <zzo38> The other thing I would have is three files of source codes, one is for DEC emulation, one is for Tektronix emulation, and one is for use with X window, therefore you can omit the files you do not need and can more easily to be use on different computer; do you like this?
03:48:54 -!- lynn has quit (Ping timeout: 246 seconds).
03:54:01 -!- bender| has joined.
03:58:44 <ais523> zzo38: does anyone actually use Tektronix emulation?
03:58:54 <ais523> that's one feature I'd omit I think
03:59:01 <ais523> being able to comple out parts of the code is good though
04:01:04 -!- tromp has joined.
04:02:38 -!- bender| has quit (Ping timeout: 268 seconds).
04:04:48 * oerjan discovers he forgot to restart the putty session after it dropped 4 hours ago
04:05:01 <oerjan> no wonder it was quiet
04:05:09 -!- jaboja64 has quit (Ping timeout: 276 seconds).
04:22:05 -!- MoALTz_ has joined.
04:25:10 -!- MoALTz has quit (Ping timeout: 244 seconds).
04:26:10 -!- treaki__ has joined.
04:29:22 -!- treaki_ has quit (Ping timeout: 260 seconds).
05:05:11 -!- tromp has quit (Remote host closed the connection).
05:32:39 <zzo38> ais523: Yes, if Tektronix emulation is a separate source file then you can easily omit it if you do not need it, although the compile scripts will need to take that into account too.
05:34:48 -!- bender| has joined.
05:37:56 <zzo38> That is one main reason to be a separate file, which is because often it will not even be in use, so it is entirely separate from the others. For example you can have preprocessor symbols USE_DEC and USE_TEKTRONIX and then tell it to link or not link file dec.o and tektronix.o
05:38:57 <zzo38> Unless you are making your own front-end for systems other than X you would also need main.c then.
05:42:49 -!- bender| has quit (Ping timeout: 240 seconds).
05:43:51 <HackEgo> [wiki] [[Special:Log/newusers]] create * SlackerSnail * New user account
05:45:32 -!- Alcest has quit (Remote host closed the connection).
05:45:41 <zzo38> Another thing xterm does not do is to be able to adjust key repeat, although this is because X does not support changing it only per window. But, another idea would be to fake it if the DEC key repeat codes are used and based on its own timers when that happens.
05:47:41 <zzo38> I would aim for more complete DEC and Tektronix emulation than xterm and other programs has.
05:57:48 <zzo38> That mean everything included, including Sixel and ReGIS graphics, Tektronix APL mode, DEC Technical characters, local mode (this can be used to alter settings with escape codes, for example), etc.
05:59:45 <zzo38> Not all actual features of the user setting menu in DEC terminals would actually be in there though, since some are not applicable to terminal emulators, such as settings for power, while others are unneeded because other settings of the terminal emulator can be used instead, such as brightness controls.
06:01:05 -!- hppavilion[1] has joined.
06:01:28 <hppavilion[1]> WE WILL BE A FNORD / COME ON TELL US THAT YOU WANT US / WE WON'T BE IGNORED / NOW IT'S TIME FOR OUR REWARD
06:05:53 -!- tromp has joined.
06:06:04 <zzo38> Then you must learn why.
06:10:17 -!- tromp has quit (Ping timeout: 250 seconds).
06:10:35 <zzo38> Why are the formats of pictures in Xlib called XYBitmap and XYPixmap and ZPixmap? Where do the names come from?
06:13:59 <HackEgo> [wiki] [[Deadfish]] https://esolangs.org/w/index.php?diff=46624&oldid=46610 * SlackerSnail * (+724) /* Implementations */
06:14:36 <HackEgo> [wiki] [[Deadfish]] https://esolangs.org/w/index.php?diff=46625&oldid=46624 * SlackerSnail * (+1) /* =SmileBASIC */
06:16:46 <zzo38> Is there any library to write a X proxy program with?
06:17:39 -!- hppavilion[1] has quit (Ping timeout: 260 seconds).
06:43:08 <deltab> "The data for a pixmap is said to be in "ZFormat" if it is organized as a set of pixel values in scanline order."
06:43:21 <deltab> I guess because of the path of the electron beam
06:48:38 -!- lambda-11235 has quit (Quit: Bye).
06:51:35 <zzo38> I did guess that, although the electron beam goes in the same direction for XYBitmap and XYPixmap formats too, and they are also in the scanline order.
06:52:26 <zzo38> The difference is that XYPixmap stores each plane as a separate bitmap while ZPixmap stores each pixel as a single unit.
06:53:36 -!- AlexR42 has joined.
07:55:04 -!- ski has quit (Ping timeout: 260 seconds).
08:00:12 -!- gonad has joined.
08:00:16 -!- llue has quit (Read error: Connection reset by peer).
08:06:26 -!- tromp has joined.
08:11:30 -!- tromp has quit (Ping timeout: 276 seconds).
08:44:25 -!- J_Arcane_ has joined.
08:46:01 -!- J_Arcane has quit (Ping timeout: 252 seconds).
08:46:05 -!- J_Arcane_ has changed nick to J_Arcane.
08:54:19 -!- AnotherTest has joined.
09:07:07 -!- bender| has joined.
09:07:21 -!- tromp has joined.
09:08:21 -!- nooga has joined.
09:11:33 -!- tromp has quit (Ping timeout: 240 seconds).
09:52:06 -!- oerjan has quit (Quit: Nite).
10:00:01 -!- lynn has joined.
10:14:15 -!- ais523 has quit (Ping timeout: 264 seconds).
10:52:46 -!- AlexR42 has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
10:55:24 -!- Lord_of_Life has quit (Excess Flood).
10:59:08 -!- Lord_of_Life has joined.
11:01:06 -!- Reece` has joined.
11:04:33 -!- Reece` has quit (Client Quit).
11:07:53 -!- tromp has joined.
11:12:01 -!- tromp has quit (Ping timeout: 240 seconds).
12:05:00 -!- carado has joined.
12:16:47 -!- lishaoshuai has joined.
12:18:05 <HackEgo> lishaoshuai: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on EFnet or DALnet.)
12:29:43 <lishaoshuai> thinks, i am junior student . For me,if i want to go to a major company ,I must know TCP/IP and a lot about Linux.I think you know the chinese company like Huawei,and Xiaomi.
12:31:41 <lishaoshuai> http://codecraft.huawei.com/ It's a topic about Huawei,It makes me headache.
12:34:14 <lishaoshuai> I think izabera that you can solve it. if you have interest and Please help me. My mail is 1049188593@qq.com .
12:38:17 <lishaoshuai> I send the topic to you.i will translate it for you.
12:38:45 <izabera> and what do i have to do with it?
12:41:00 <lishaoshuai> It's a code problem about the net.if you have interest,you can write your code or give me some ideas.
12:42:25 <izabera> your description is too vague
12:45:03 <lishaoshuai> It's the top-level code megagame.You know dijkstra?
12:47:24 <lishaoshuai> Do you have interset on it? if you have,we can be a team.I will translate it in a mail for you.If not I can still introduce it for you.
12:49:39 -!- nooga has quit (Ping timeout: 248 seconds).
12:49:47 <lishaoshuai> oh guys,sorry. i am shame to face to you. Ok,goodbye.
12:51:02 -!- boily has joined.
12:51:55 -!- lishaoshuai has quit (Quit: Leaving).
13:05:31 <HackEgo> [wiki] [[User:Rdebath]] https://esolangs.org/w/index.php?diff=46626&oldid=46612 * Rdebath * (+23480) Add some (!!) links to the table
13:08:04 -!- J_Arcane has quit (Ping timeout: 244 seconds).
13:16:33 -!- nooga has joined.
13:31:26 <int-e> In the meantime, the people of Carbo are almost bursting in bubbles of joy and excitement.
13:32:16 <HackEgo> 14) <fungot> oerjan: are you a man, if there weren't evil in this kingdom to you! you shall find bekkler! executing program. please let me go... put me out! he's really a tricycle! pass him! \ 28) <ehird> is there a problem with it being carbonized :D <augur> yes: carbonized coffee bean is known more commonly as "charcoal" \ 57) <fungot> ehird: ev
13:41:36 <HackEgo> 548) <fizzie> I prefer the N64 controller, it's the only one that has place for my third hand. \ 579) <ais523> it's not a list of /all/ interesting esolangs, btw; otherwise you can take the first command from the first esolang, the second from the second, the third from the third, etc, then add 1 to all of them <ais523> and you get a new interesti
13:46:20 -!- J_Arcane has joined.
13:58:35 <boily> ais523's esolang diagonal argument.
13:58:51 <HackEgo> finnish/Finnish suomalaiset ei Perkeleistä on hakkapeliittaan. Ei saa peittää. Parasta ennen!
14:03:17 <HackEgo> tymällenneteell kylmituksensä tykypsyttävilla intyylitsemmistä suurenemänäni aamme ahdistamanne ahavolyyseltä myrityilta lavertaisemana agrologille luovaltasialla kirkkään hettaminäsi häirittavillä aborttumisemmiksesi ahertävinään kasvattavalle edustasautetui nittävissä
14:03:42 -!- MoALTz_ has changed nick to MoALTz.
14:03:46 <fizzie> Not the best set ever.
14:04:41 <fizzie> Only "ahdistamanne", "agrologille" and "kasvattavalle" are real words this time.
14:08:17 <fizzie> (Some vowel transmutation could fix "kirkkään" -> "kirkkaan", "häirittavillä" -> "häirittävillä" and "ahertävinään" -> "ahertavinaan" as well.)
14:08:59 <boily> “Finnish is not the devil, visit the Hakkapeliitta. Do not cover. Best before!”?
14:09:36 <boily> what's a Hakkapeliitta?
14:10:46 <olsner> https://en.wikipedia.org/wiki/Hakkapeliitta
14:11:08 <olsner> I guess you could call that "hackers"
14:12:43 <fizzie> There's no "visit" in there really. Though the first part doesn't really make any grammatical sense, so...
14:14:06 <olsner> maybe it's supposed to be an or, "Finnish isn't the devil or the hakkapeliittes."
14:14:57 <boily> `le/rn French/Le français n'est pas le démon, visitez les Coupeurs. Ne pas couvrir. Meilleur avant!
14:18:26 <HackEgo> sparse matrix algorithm/Sparse matrix algorithms are a trivial special case of non-sparse matrix algorithms, by conjugating with the sparsification operation.
14:24:20 -!- tromp has joined.
14:28:40 -!- tromp has quit (Ping timeout: 244 seconds).
14:44:39 -!- bender| has quit (Ping timeout: 250 seconds).
14:47:19 -!- tromp has joined.
14:49:26 -!- Reece` has joined.
15:03:24 -!- lambda-11235 has joined.
15:29:40 -!- J_Arcane has quit (Read error: Connection reset by peer).
15:30:12 -!- J_Arcane has joined.
15:34:21 -!- p34k has joined.
15:40:58 -!- J_Arcane has quit (Ping timeout: 244 seconds).
15:58:19 -!- Vorpal has quit (Read error: Connection reset by peer).
15:58:53 -!- Vorpal has joined.
16:06:50 -!- J_Arcane has joined.
16:07:05 -!- jaboja has joined.
16:31:19 <boily> “Danish pronunciation: [ˈsmɶɐ̯ɐˌb̥ʁœðˀ]”. am I supposed to just cough the word?
16:33:30 -!- mad has joined.
16:34:31 <mad> oh man I have a hard problem
16:36:37 <mad> suppose you have a bunch of paths
16:37:13 <mad> I was thinking on a 2d plane
16:37:21 <int-e> or paths in a graph? or ... that?
16:37:35 <b_jonas> int-e: no, still the dijkstra stuff for the “tpo-level code megagame”
16:37:36 <mad> or paths in a graph, that works as well
16:37:55 <mad> what's the requirement for this to be turing complete?
16:38:03 <int-e> oh do we have a translation now?
16:38:07 <mad> clearly you need some kind of infinitely repeating pattern
16:38:20 <b_jonas> mad: um... what's Turing complete/
16:38:30 <mad> like a 2d grid
16:38:33 <int-e> what is the computation?
16:38:46 <mad> int-e : following the path
16:38:47 -!- Reece` has quit (Quit: Leaving).
16:38:57 <mad> if the path eventually stops then it halts
16:39:06 <mad> but the path can keep going forever
16:39:47 <mad> lambda-11235 : depending on how your infinite repeated grid is structured you either have a state machine or a turing machine
16:40:07 <mad> for instance if you have no infinite repetitions then it's a state machine, of course
16:40:36 <mad> if it's just an infinite 2d grid it's also a state machine because there's no difference between each grid cell
16:43:08 <mad> if it's a pattern that starts with something, but then has a different pattern that repeats rightwards, then paths in the rightwards repetition either keep going rightwards forever so they're not any different anymore, or the path go leftwards which means the number of iterations is finite (which also excludes being a Turing machine)
16:43:55 <mad> if it's a 2d grid but with a different pattern on the top side and left side, then it can be a Turing machine
16:44:48 <mad> because it's a 2 counter minsky machine using the X cell count and Y cell count as counters -> https://esolangs.org/wiki/Minsky_machine
16:45:38 <int-e> but only a single state now
16:45:55 <mad> so you need an infinity of different paths, some of which are finite (ie the turing machine halts), some of which are infinite (ie it doesn't halt)
16:46:32 <mad> int-e : well, in a repeated grid of paths, which one of the path within your repeated cell is your state
16:46:49 <b_jonas> mad: um, wait, there was a two-dimensional esolang that you're probably thinking of. the two-dimensional language that has an instruction pointer and direction like befunge, a repeating pattern, and able to emulate a two-register counter machine
16:47:18 <mad> yes that's what I'm talking about
16:47:31 <mad> https://esolangs.org/wiki/nopfunge
16:47:49 <mad> ok this brings me to the actual question I wanted to ask
16:47:56 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…).
16:49:02 <mad> Is it possible to design a path grid geometry that has an infinity of different path but isn't Turing complete?
16:49:46 <mad> Or all grid geometries that have an infinity of different paths necessarily turing complete?
16:51:51 <mad> the 2d grid with a different strip on top and left is possibly the simplest grid with an infinity of different paths, and even that one is turing complete
16:52:22 <mad> so I'm conjecturing that all grid geometries with an infinity of different infinte paths are also turing complete
16:53:38 <b_jonas> Was someone able to prove 3SP turing-complete yet by the way?
16:53:45 -!- boily has quit (Quit: DISCARDED CHICKEN).
16:54:44 <int-e> so I'm conjecturing that all <undefined> with an infinity of different <needs to be defined in the context of first <undefined>> are also turing complete
16:54:53 * int-e can't make sense of that.
16:54:55 <mad> And I'm also conjecturing that all turing complete systems contain an infinitely different grid geometry when mapped into paths
16:55:51 <mad> int-e : hmm, let me try to explain it simpler
16:56:37 <mad> a finite set of 2d paths are obviously not turing complete
16:57:20 <mad> like, if you draw a bunch of arrows on a sheet of paper
16:57:26 <mad> and follow the arrows
16:57:39 <mad> it has to stop because you can only have so many arrows
16:58:08 <Vorpal> mad: could also go into a loop
16:58:28 <mad> but then you can't have an infinite tape
16:58:39 <mad> so to build a turing complete system, you have to add the possibility for some kind of infinite repetition
16:58:57 <Vorpal> I don't really understand how nopfunge is TC, but that is amazing
16:59:33 <Vorpal> mad: btw, Befunge-98 is probably not TC for the same reason that C isn't TC
16:59:37 <b_jonas> Vorpal: do you understand why a two-counter machine is turing-complete? or at least a three-counter machine?
16:59:56 <Vorpal> b_jonas: I would have to read up on those, I'm not really familiar with either
16:59:59 <mad> Vorpal : it uses X and Y position as bignums, then absuses those as an infinite tape
17:00:32 <b_jonas> Vorpal: after you read up on the, you'll find out that nopfunge can simulate a two-counter machine
17:00:46 <Vorpal> b_jonas: well that is how it usually goes to show something is TC
17:00:50 <b_jonas> ais calls these thingies Minsky machine
17:01:21 <b_jonas> They're essentially multi-tape Turing machines that can't write their tape, only read
17:01:50 <mad> so yeah if you have an infinite number of copies of a sheet of paper with arrows on it arranged in an infinite 2d grid that extends in all directions, you still don't have a turing machine
17:02:02 <Vorpal> Yes that was one part I was wondering about. How it kept state without writable state as it were
17:02:07 <mad> because you have no way of knowing which copy you're on
17:02:29 <mad> it's as if you made the left side loop with the right side, and the top side loop with the bottom side
17:02:34 <b_jonas> Vorpal: the state, apart from the finite state a Turing machine can keep in its cpu, is kept in how far the tape heads are from the start
17:02:49 <b_jonas> Vorpal: that distance is a "counter" when you regard them as a Minsky machine
17:03:00 <b_jonas> but they can only detect whether the counter is zero or nonzero
17:03:11 <Vorpal> yes that was the next question
17:03:16 <b_jonas> this is why to do anything reasonable, any counter machine will need exponential time
17:03:33 <b_jonas> because it has to store arrays encoded as big numbers
17:03:50 <b_jonas> if you have enough tapes, you can divide a counter by two or multiply by two, thus push or pop a bit
17:04:16 <Vorpal> So completely impractical then XD
17:04:23 <mad> if instead of an infinite 2d grid that extends in all directions, you make it only extend rightwards and downwards, and have a different pattern on the left side and a different pattern for the topside (plus potentially a different one for the topleft corner), then you can count the number of repetitions from the left side or top side
17:04:32 <b_jonas> now it turns out that three counters are enough, because you use two to store two stacks of bits, and one as scratch to do the divisions or multiplications
17:04:33 <mad> yes this is super impractical :D
17:04:55 <b_jonas> But it also turns out, and I admit I don't understand why, that two counters are also enough, but in that case the slowdown isn't exponential, it's double-exponential.
17:05:22 <Vorpal> mad: anyway, it is TC, unlike normal Befunge-98. Though 98 is only not TC due to the sizeof problem
17:05:26 <b_jonas> One counter definitely isn't enough for turing-completeness though, since a single-counter machine is weaker than a stack machine
17:05:59 <mad> you're basically storing your turing machine infinite tape as two stacks encoded as something like 2^w * 3^x * 5^y * 7^z
17:06:10 <b_jonas> Vorpal: anyway, ais523 calls these counter machines Minsky machines, and has some esolangs where he examines restricting them in some way and seeing how powerful they remain
17:06:37 <b_jonas> restrict them in different ways than reducing the number of counters that is
17:06:51 <b_jonas> so there's http://esolangs.org/wiki/The_Amnesiac_From_Minsk
17:06:52 <mad> and then you compute checking if your bignum is divisible by 2/3/5/7 by bouncing between the left side and top side
17:07:02 <mad> and so forth
17:07:50 <mad> so the number of iterations to do anything useful is, like... super-hyper-exponential
17:07:58 <b_jonas> it's only double-exponential
17:08:14 <b_jonas> mad: you haven't been on #esoteric enough
17:08:27 <b_jonas> we have much more impractical languages
17:09:02 <b_jonas> sure, I certainly prefer computation models that incur only quasi-linear slowdown
17:09:19 <b_jonas> this is #esoteric, so people don't restrict themselves to those
17:09:44 <Vorpal> I think I'll keep to befunge-98. It is in fact by far the most practical esolang
17:09:51 <Vorpal> If you actually want to get stuff done
17:10:10 <Vorpal> It is a pain to write interpreters for though, I should know, having written 2
17:10:28 <mad> so if you have some mechanism to build some infinite grid of paths
17:10:39 <fungot> Vorpal: and scheme48's byte-code.
17:10:49 <Vorpal> that was oddly on topic
17:11:21 <b_jonas> Vorpal: what interpreters have you written in befunge?
17:11:41 <Vorpal> I have written programs in it, but no interpreters
17:11:56 <mad> if, after removing all the paths that are "the same" behavior-wise (ie symmetries), you end up with only finite paths, then it's not turing complete
17:11:57 <Vorpal> b_jonas: fungot runs on/used to run on my cfunge
17:11:57 <fungot> Vorpal: of course, it would
17:12:12 <Vorpal> And I wrote a funge-98 interpreter in erlang too
17:12:48 <Vorpal> b_jonas: I lost most programs I wrote in the language in the disk crash before I learned to backup many years ago
17:13:01 <mad> what I'm conjecturing is that if after removing all the paths that are "the same" behavior-wise, there are any infinite paths left, then it _has_ to be turing complete
17:14:09 <b_jonas> mad: I don't know if you care, but either you haven't told us enough context about this path thing, or you're not making sense
17:14:33 -!- jaboja has quit (Ping timeout: 240 seconds).
17:14:40 <mad> I'm generalizing the whole thing
17:14:46 <b_jonas> fungot: do you care if you don't make sense/
17:14:46 <fungot> b_jonas: are you doing? xd a 2d fnord etc. :)
17:15:17 <mad> for instance, the infinitely repeated 2d grid is a "type" of grid
17:15:37 <mad> you've also got the infinitely repeated 1d grid (not turing complete)
17:16:43 <mad> infinitely repeated 3d grid (which I think is turing complete under the same conditions as the 2d grid - ie you have some kind of end or ledge to the pattern which lets you count your position in the grid)
17:17:10 <mad> you also have some fractals
17:17:29 <mad> as a way of generating infinite sets of paths
17:17:54 <mad> some of which are going to be turing complete
17:18:50 <mad> in particular the ones that have some binary-tree kind of thing going on
17:20:40 <mad> it's also possible that when mapped in 2d, you have to have the possibility of wire crossings or else it's not turing complete
17:21:46 <mad> basically I found some way of mapping potentially turing-complete systems into shapes and trying to figure out what you can learn from that :D
17:23:21 -!- jaboja has joined.
18:01:32 -!- sebbu has quit (Ping timeout: 244 seconds).
18:01:58 -!- jaboja has quit (Ping timeout: 260 seconds).
18:03:19 <tswett> Since nobody asked for it, here's the 2x2 algorithm for swapping UFL and UBL like U: U' R' F R' B2 R F' R' B2 R2
18:03:57 <b_jonas> this channel does rubik's cubes? I didn't know, although I'm not too surprised
18:04:16 <tswett> I got one in the mail yesterday.
18:04:46 -!- jaboja has joined.
18:05:06 <tswett> I think there's at least one other person here who also does them.
18:05:38 <b_jonas> I've even been to the #rubik a bit
18:06:06 <b_jonas> but my cuber-fu has lapsed a lot because I didn't practice much for years now, and it never was good in first place
18:06:22 <b_jonas> I still feel the attraction though
18:06:46 <b_jonas> and should practice more some day and concentrate on improving my particular stumbling points
18:07:26 <b_jonas> (including color neutrality)
18:07:32 -!- mad has quit (Quit: Pics or it didn't happen).
18:12:48 <tswett> I wonder how many people are good at "unsolving" cubes.
18:12:59 <tswett> Take a scrambled cube and a solved cube, and make the solved one look like the scrambled one.
18:13:35 <myname> sounds like one added layer of abstraction
18:13:47 <myname> if you can solve it, you should be aböe to produce any state
18:13:59 <myname> it may take longer, though
18:14:09 <int-e> (this may not be true for the larger cubes)
18:14:38 <myname> never done those, but i get the problem
18:14:39 <b_jonas> tswett: given the crazy stuff cubers do, there are probably people competing in that. with all variants, like one-hand each for the read-only template cube and the cube, one hand for the two together, feet, etc.
18:17:46 <lynn> I seem to recall there are algorithms for cycling edges, or swapping corners, or something like that?
18:18:37 <lynn> Maybe the "simplest" way to solve a cube (or recreate a scrambled one) is applying those a billion times
18:18:40 <myname> scrambling is basically just a layer of abstraction that maps each tile to the finishing tile
18:18:43 <b_jonas> lynn: yes. a bit tedious (in the case of corners) but very useful for blind solving
18:18:51 <b_jonas> lynn: and no, it's not the simplest way
18:19:11 -!- jaboja has quit (Ping timeout: 268 seconds).
18:19:15 <tswett> But is it the "simplest" way?
18:20:20 <b_jonas> lynn: the simplest way to recreate a solved cube is to just use an ordinary solving method *forwards* but going to the right state
18:20:44 <b_jonas> the alternate way is the solve the template cube, writing the moves to paper, and then applying the moves *backwards* to a pristine cube
18:21:13 <tswett> I think the simplest way to recreate a given cube is by making random moves until you find that you've succeeded.
18:21:15 <lynn> b_jonas: Well, as tswett pointed out, by "simplest" I rather mean... "dumbest", maybe
18:21:35 <lynn> Oh, that *is* even dumber.
18:21:54 <b_jonas> we don't have an entry for that/
18:22:03 <lynn> Anyway ordinary solving methods need all kinds of algorithms :<
18:22:24 <lynn> If you learn the swappy ones, you can use just two and apply them everywhere! (Maybe?)
18:24:22 -!- llue has joined.
18:24:44 <tswett> You can probably find just one algorithm that you can use for everything.
18:24:55 <tswett> Maybe one that, like, swaps two corners and also rotates three edges.
18:26:45 <int-e> You could take the 12 simple algorithms that each permute four edge and four corner pieces at a time... *ducks*
18:27:25 <int-e> (aka quarter turns)
18:28:12 -!- gonad has quit (Ping timeout: 260 seconds).
18:28:16 -!- tromp has quit (Remote host closed the connection).
18:28:31 <b_jonas> you could take just _two_ algorithms that generate the whole cube group as a group (or as a semigroup, it's the same thing since it's finite), and use them (without any conjugation or anything). you just wouldn't be able to figure out when to apply which one.
18:30:25 <int-e> 4 algorithms seems reasonable to me
18:31:17 <b_jonas> int-e: together with conjugations by whole cube rotations, and conjugation by the cube mirroring outer conjugation?
18:31:18 <int-e> (I'm assuming that one can solve one layer intuitively)
18:32:11 <int-e> b_jonas: yes, some conjugation will be involved. but that sounds scarier than it is in practice (as I'm sure you're aware)
18:32:51 <b_jonas> mind you, I have serious problems with color-neutrality, which means I don't conjugate algorithms as efficiently as I should
18:33:07 * int-e has never cared about efficiency much
18:33:42 <int-e> (so for example, 3x3x3 takes me a about a minute, give or take)
18:34:17 <b_jonas> that's still better than me. I couldn't yet go to under one minute in average, and I'd need more practice and learning stuff for that
18:34:51 <b_jonas> (not specifically color neutrality though, just improving the recognition of some of the steps of the method, as well as some physical turning practice, would let me)
18:35:22 <b_jonas> I could do under 2 minutes in average though
18:36:06 <int-e> Hmm sounds like color neutrality should be the least of your worries.
18:36:37 <b_jonas> My biggest regret about cubes is that the silver mirror cube wasn't yet available to me during high school, so I couldn't turn the cube under the desk without looking at it.
18:36:42 -!- pelegreno has quit (Remote host closed the connection).
18:36:50 <b_jonas> And if you keep glancing at the cube under the desk, high school teachers notice that.
18:37:10 <int-e> anyway, to improve I believe I would have to start learning a million algorithms (well, maybe 3 dozen to start with) and I just can't be bothered.
18:37:17 <b_jonas> int-e: yes, I know, color neutrality is just the easiest to describe problem I have
18:37:22 <int-e> my top layer has 4 distinct phases.
18:37:49 <int-e> (non-standard too: flip edges; place corners, turn corners, place edges)
18:38:52 <b_jonas> int-e: what? that's the standard order
18:39:29 <b_jonas> but the last layer isn't the one I'm having problems with. it's before that.
18:40:05 <tswett> I don't really care about speed at the moment. I think move efficiency is more important to me.
18:40:11 * int-e can't track pieces, so the first two layers are also fairly slow.
18:40:37 <int-e> but oh well, fast enough to impress laypeople.
18:40:45 <b_jonas> tswett: there are competitions for that too
18:41:22 <int-e> (laypeople are also easily impressed by just seeing a 5x5x5 cube)
18:42:08 -!- copumpkin has joined.
18:42:18 <int-e> or this one, https://en.wikipedia.org/wiki/Megaminx
18:42:21 <b_jonas> But I don't think you can get good enough in fewest moves, since even an old personal computer can find the absolute fewest move solution for any 3x3x3 state quickly enough. You'll never beat that.
18:42:21 <Vorpal> what about non-3D "cubes"?
18:42:32 -!- pelegreno has joined.
18:42:41 <Vorpal> Could you have a 4D cube? On a computer that is, you obviously can't make a physical one
18:42:55 <int-e> Vorpal: I've seen programs that do that
18:43:13 <Vorpal> int-e: oh? Do they have any interesting properties?
18:44:00 <int-e> one step further: http://www.gravitation3d.com/magiccube5d/
18:44:27 <int-e> youtube... https://www.youtube.com/watch?v=0AqMb-edXlc (hmm but apparently that's a program solving it)
18:45:35 <int-e> haha http://www.speedcubing.com/records/recs_comp_2222.html
18:46:04 <int-e> and of course this as well: http://www.speedcubing.com/records/recs_comp_3333.html
18:47:05 -!- jaboja has joined.
18:51:35 <b_jonas> I never understood what a 4x4x4x4 rubik's cube even _means_ in first place
18:51:49 <b_jonas> what is its state and what are the elementary steps you can do on them?
18:51:59 <b_jonas> any 4-dimensional cube that is
18:52:32 <Vorpal> b_jonas: so it is a 3x3x3 but with an extra x3
18:52:42 -!- mihow has quit (Ping timeout: 260 seconds).
18:52:48 <int-e> the screenshots on http://astr73.narod.ru/MC7D/MC7D.html are hilarious.
18:54:05 <b_jonas> yes, I also know some really bored people solve large 3d cubes on computer, like 91x91x91 or something like that
18:54:20 <int-e> b_jonas: the 3x3x3x3 cube has 8 hyper-faces of 3x3x3 cubes each, that's the state
18:54:53 <b_jonas> int-e: with a sticker of the same color on each of 27 cubes of a face?
18:55:08 <int-e> the cubes are the stickers :)
18:55:41 <lynn> "Look more understandable? I also think so."
18:55:45 <int-e> of each of the 8 colors, yes
18:56:16 <b_jonas> I mean, not that part is scary
18:56:24 <b_jonas> the scary part is what the rotations are, which I don't understand
18:56:30 -!- boily has joined.
18:56:40 <b_jonas> these faces are on the outer 3d faces of a 3x3x3x3 cube, right?
18:58:59 <int-e> b_jonas: well, the cube has slices of size 3x3x3x1 ... and you'll just rotate them around the short axis.
18:59:51 <b_jonas> int-e: so that means you can rotate any of the slices around 6 axes?
18:59:52 <int-e> so you get 8 basic moves (the usual L, R, U, T, F, B, and two more)
19:00:09 <b_jonas> only around the short axes
19:00:54 <b_jonas> dunno, this is a bit hard to imagine
19:01:16 <b_jonas> we need a wisdom for bogosort by the way
19:02:01 <int-e> `learn Bogosort is an efficient sorting algorithm for nondeterministic Turing Machine. Taneb may have invented it.
19:02:08 <HackEgo> Learned 'bogosort': Bogosort is an efficient sorting algorithm for nondeterministic Turing Machine. Taneb may have invented it.
19:02:10 <int-e> `learn Bogosort is an efficient sorting algorithm for nondeterministic Turing Machines. Taneb may have invented it.
19:02:13 <HackEgo> Relearned 'bogosort': Bogosort is an efficient sorting algorithm for nondeterministic Turing Machines. Taneb may have invented it.
19:02:27 <int-e> But perhaps that's too accurate.
19:02:44 <int-e> `` sed -i s/M/m/ wisdom/bogosort
19:03:38 <int-e> But it is kind of hard to beat linear time :)
19:03:56 <b_jonas> maybe link to http://www.dangermouse.net/esoteric/bogobogosort.html
19:04:22 <b_jonas> no wait, that's too efficient
19:05:54 <boily> (hallynn? blynnjour?)
19:07:31 <int-e> 1. make n! copies of the universe
19:09:29 * int-e is imagining a global quantum suicide approach to solving NP problems.
19:10:00 <tswett> As in, try a random solution, and if it's wrong, kill all sentient beings?
19:10:33 <tswett> By the way, the idea of quantum suicide goes against a certain principle.
19:10:43 <int-e> Basically. I didn't think of stopping after killing the sentient beings...
19:10:57 <tswett> Namely: obscure consequences of theoretical physics have no ethical consequences.
19:11:24 <int-e> It's still a great thought experiment.
19:11:27 <tswett> If you bring up quantum mechanics in an ethical debate, you're doing something wrong.
19:11:51 <int-e> Schrödinger's cat leapt out of the box.
19:12:35 <int-e> I think I don't want to have that ethical debate.
19:12:40 <zzo38> tswett: Are you sure? Ethics is not as simple so it may be necessary. The laws of physics are important too (not limited to quantum, though).
19:13:16 <b_jonas> agreed, _real_ quantum mechanics can come up in ethical debates. quantum suicide probably can't, because most people believe quantum mechanics just doesn't work that way
19:13:40 <int-e> but if we must discuss ethics, let's start with the simpler scenario from https://xkcd.com/1132/ :-P
19:14:00 <tswett> If you can't think of an experiment that the ancient Greeks could have performed in order to demonstrate a law of physics, then as far as ethics goes, that law of physics does not exist.
19:15:17 <zzo38> Greeks did invent classical logic, although logic is a mathematics and the laws of physics are not relevant, but there is other kind of logic too anyways
19:15:21 <int-e> (if the sun goes out, owing somebody 50 dollars will be the least of your worries)
19:15:31 <lynn> I feel like a boring buzzkill for it, but I've never really fully understood the point in thinking about parallel universes.
19:16:04 <int-e> lynn: well, it's a great device for unimaginative writers
19:17:49 <tswett> Well, why *would* ethics depend on stuff that the ancient Greeks could never have observed?
19:17:59 <int-e> And let's not forget Milner's and Hoare's pioneering work on communicating parallel universes...
19:18:39 <b_jonas> if only I had a knife to get back to the original universe from this parallel one
19:18:55 <zzo38> tswett: Because the ancient Greeks (and even today's people) cannot possibly know all of the possible situations that might occur.
19:22:02 <b_jonas> so you think the world wasn't yet quantum mechanical at ancient times, and only became so in the enlightenment when we were too close to figuring out all of physics?
19:22:46 <int-e> does philosophy have a concept of a "sufficiently smart human"? ("A sufficiently smart human could solve the problem of global warming.")
19:23:22 <b_jonas> int-e: no, that's game theory and economics, and the latter calls it “perfectly rational agents”
19:24:43 -!- sebbu has joined.
19:25:00 <zzo38> b_jonas: No, what I mean is that only what you know is approximation and is not quite completed. It applies to philosophy, ethics, science, religion, etc but in different ways.
19:25:35 <zzo38> "Perfectly rational agent" does make more sense I think, assuming that is what you mean of course.
19:28:47 -!- tromp has joined.
19:29:53 -!- Phantom_Hoover has joined.
19:32:46 <Vorpal> b_jonas: makes sense. As much sense as my theory that we are all in a virtual world and what we call quantum physics is just really deficiencies in the simulation. Like floating point rounding errors sort of
19:33:03 -!- tromp has quit (Ping timeout: 240 seconds).
19:33:08 <Vorpal> (Note: I don't actually believe this)
19:40:04 <tswett> 'Course, if it's just deficiencies in the simulation, it's weird that the simulation is deficient in a manner that's so consistent and difficult.
19:41:43 <Vorpal> tswett: well, that is why it is a joke theory ;P
19:50:12 <Phantom_Hoover> like that quantum physics doesn't work anything like simulation errors, for instance
19:56:49 <Vorpal> Phantom_Hoover: well, obviously that is a bit more complicated computer architecture than what we have currently
19:58:07 <Vorpal> After all we can't simulate an entire universe yet, so why should we assume that they will do that on hardware we can understand
20:05:45 -!- jaboja has quit (Ping timeout: 250 seconds).
21:00:48 <HackEgo> [wiki] [[Special:Log/newusers]] create * Ozwg6693 * New user account
21:10:15 <Taneb> I'm playing Factorio
21:10:33 <Vorpal> Taneb: it seems like the new minecraft, kind of
21:10:44 <Vorpal> Seeing it popping up in lots of youtube channels
21:10:46 <Taneb> It's quite different to Minecraft
21:11:14 <Vorpal> similar to certain modpacks for minecraft though
21:11:33 <Vorpal> And similar in that it is an early access game that is quite playable and doing really well
21:12:30 <myname> they should make an android port
21:12:49 <Vorpal> myname: might work well on a really large tablet, probably not on a phone
21:12:56 <Vorpal> but I suspect it would control terribly
21:13:26 <myname> it doesn't need to be that large if you can zoom properly
21:13:52 <Vorpal> what about the inventory
21:14:09 <Vorpal> Especially the inventory of a car or tank in that game
21:17:43 <int-e> oh the tension while waiting for the first edit... is it spam or real?
21:18:53 <int-e> or perhaps the first edit was blocked... where's that page again
21:18:57 -!- jaboja has joined.
21:20:04 <int-e> I guess it's https://esolangs.org/wiki/Special:AbuseLog ... nothing yet
21:28:52 -!- tromp has joined.
21:32:14 -!- copumpkin has quit (Quit: Textual IRC Client: www.textualapp.com).
21:32:26 <tswett> So I guess I know how to solve a 2x2 cube now.
21:32:52 -!- copumpkin has joined.
21:33:21 -!- tromp has quit (Ping timeout: 244 seconds).
21:33:33 <tswett> Now here's a fun little sequence, of order 4: R2 U2 R U2 R2
21:34:31 <tswett> Also known as R^(R2 U2), of course.
21:35:15 <shachaf> isn't R^(R2 D2) more popular
21:35:26 <tswett> Among Star Wars fans, probably.
21:35:29 <shachaf> Taneb: Are you multiplaying Factorio?
21:35:40 <Taneb> With one of my friends from uni
21:36:35 <tswett> shachaf: WHAT IS THE MEANING OF THIS?
21:36:57 <tswett> R^(R2 U2) U has an effect that's somewhat easier to describe.
21:37:08 <zzo38> I have written a very simple X clock in JavaScript, which supports resizing the window but no settings are implemented and it only displays seconds, and only by filling in the circle; it is white at 0 seconds and half green at 30 seconds and so on. I made simply to show its working though, so if you want a real analog clock you could just use xclock, or to modify this one
21:37:41 <tswett> It rotates UBL and UFR in place clockwise while rotating UBR and UFL in place counterclockwise.
21:38:01 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
21:38:52 -!- copumpkin has quit (Max SendQ exceeded).
21:39:15 <tswett> The way that it does this, of course, is by using R2 U2 to move the U pieces into the R layer in a certain way.
21:40:17 <shachaf> zzo38: What if you added another circle for minutes and a third circle for hours?
21:40:22 <shachaf> Then the clock would be useful.
21:40:36 <zzo38> Yes, that can be done by using concentric circles perhaps
21:40:40 -!- Phantom_Hoover has joined.
21:40:47 <zzo38> Although I only intended it as an example anyways.
21:41:11 <zzo38> http://sprunge.us/Ubba
21:42:05 -!- copumpkin has joined.
21:48:50 <tswett> So what the heck is up with this algorithm: U' R' F R' B2 R F' R' B2 R2
21:50:18 <tswett> I can simplify that a *little* bit: U R' B2^(F R') R' B2 R2
21:50:28 <int-e> hmm, add a U and you get a sequence that permutes corners on a 3x3x3 cube, preserving edges
21:50:35 <tswett> Uh, lemme try that again.
21:50:49 <tswett> U' R' B2^(F R') R' B2 R2
21:51:54 <tswett> Especially in combination.
21:54:20 <tswett> B2^(F R') leaves the F layer alone except for UFR. It flips the DB edge around. It also swaps UBL and UFR around, with their formerly U faces now facing L and F instead.
21:58:00 <zzo38> shachaf: Yes, function* means generator functions
21:58:18 -!- jaboja has quit (Ping timeout: 276 seconds).
22:02:06 -!- jaboja has joined.
22:12:14 <tswett> I see that R U R' U R U2 R' U2 is a significant sequence for the 3x3 as well as the 2x2.
22:13:42 <int-e> yeah, I use that (and two variants... inverse, and the mirror image of the inverse) to orient corners
22:14:38 <int-e> (but not the mirror image... I can't explain why :) )
22:17:40 <tswett> Lemme check the exact effect that has on the 3x3.
22:17:59 -!- p34k has quit.
22:18:26 <int-e> and I believe I discovered it myself at some point... playing around with the idea of taking out the FR edge and FRB corner and then putting them back again
22:18:34 <tswett> Only the U layer is affected. I'll have to figure out why that is.
22:18:50 <tswett> The UFL and UL pieces are left alone.
22:19:12 <tswett> The remaining U edge pieces have cycled clockwise.
22:19:33 <tswett> And the remaining U corner pieces have rotated clockwise in place.
22:19:57 <tswett> This, of course, means that the sequence has order 3.
22:20:44 <int-e> and for a long time I've used (R U R' U')^3 to permute corners in the back layer (so turning the cube...)
22:22:54 <int-e> now it's B' R' F' R B R' F R and its inverse
22:33:01 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:45:16 -!- tromp has joined.
22:49:33 -!- tromp has quit (Ping timeout: 240 seconds).
22:54:09 <b_jonas> int-e: oh man, I still have difficulty reading this algebraic notation, because I used to learn move sequences with the Hungarian mnemonics which use S K T N F L for the sides (all consonants so you can make syllables by appending e u i for 1 2 3 turn resp)
22:55:04 <int-e> I can't really read it either... I can perform the sequence slowly on a cube :P
22:55:45 <b_jonas> S K T N F L is so much better
22:56:09 <int-e> Somehow I don't agree.
22:56:54 <b_jonas> no really, being able to recite the moves with one syllable for each moves makes it much more easier
22:59:34 <int-e> I can see that. But I would have to learn the letters.
23:00:25 <boily> OTHAR TRYGGVASEN, GENTLEMAN ADVENTURER!
23:00:35 * boily is catching up on GG from the start
23:00:51 <int-e> boily: quite the project
23:01:29 <boily> strangely, there weren't black and white pages at the beginning, like I remembered them.
23:01:44 <boily> was the first volume redrawn or something?
23:03:52 <b_jonas> int-e: that's not difficult, since there's only six of them
23:04:05 <b_jonas> ok, let's say eleven of them or something
23:04:12 <b_jonas> since not only the sides count, but also e u i
23:04:42 <int-e> anyway it's a non-issue for me since I've never tried to memorize more than one combination at a time :P
23:04:59 <int-e> So mechanical memory was good enough.
23:05:22 <b_jonas> well, these mnemonics did help me retain the moves for long term, but ok
23:05:39 <b_jonas> and they can also help communicating a move with someone else
23:26:28 <int-e> communication, pah, what a silly idea.
23:29:22 -!- oerjan has joined.
23:32:22 * augur carbonizes int-e's coffee
23:36:46 -!- feliks has quit (Ping timeout: 268 seconds).
23:39:29 <int-e> augur: how do you know that I drink cola...
23:40:00 <int-e> oh, I read carbonate
23:40:15 <boily> carbonated coffee? that should be interesting.
23:40:15 * int-e incinerates augur for good measure
23:40:21 <augur> even if i carbonated your coffee it wouldnt be a cola :)
23:40:28 <augur> boily: such a thing exists!
23:40:31 <boily> whoa whoa whoa! no incinerating people in the chännel!
23:40:45 <augur> boily: https://www.manhattanspecial.com/products_pure_espresso.html
23:40:47 <int-e> augur: I know. But the defining ingrediant is coffeine
23:41:04 <augur> int-e: the defining ingredient of a cola is cola nut :P
23:41:10 <augur> otherwise its not a cola!
23:41:11 <boily> holy fentimans. I need that now.
23:41:18 <augur> thats why rootbeer, sprite, etc. arent colas, they're just sodas
23:41:56 <augur> boily: ive had manhattan special. it's REALLY good. but it needs to be ice cold before you open it otherwise it'll spray all over the place
23:42:42 <int-e> rootbeer and sprite also don't contain coffeine ;-) I think you'll have to come up with one of those awful energy drinks...
23:43:03 -!- hppavilion[1] has joined.
23:44:04 <hppavilion[1]> I have a feeling logic programming would be good for OS dev
23:59:53 * boily pokes hppavilion[1] in the hellogical bits