00:00:04 I tried making a Sonic game in it but it was too slow 00:00:08 rdococ: A JS-ic interpreter for BASIC? 00:00:16 idk how it worked 00:00:18 hppavilion[1]: They're "markov sequences" all right, though it's not quite the most common implementation, where you have a bunch of text and then jump around; instead, it's just generating word by word from a variable-length n-gram model trained with https://github.com/vsiivola/variKN and stored in a I-think-it's-nifty-but-for-some-reason-haven't-much-seen-it-around data structure that I call ... 00:00:19 rdococ: I've been working to make a browser 00:00:22 but it was called Quite Basic 00:00:24 ... a reverse-context tree for lack of a better name. 00:00:26 It's like a word trie except so that if your model has, say, information about the contexts "baz bar foo", "quux bar foo" and "zuul foo", the trie has the form (foo (bar (baz [..]) (quux [..])) (zuul [...])). 00:00:29 hppavilion[1]: this is of course inspired by "halt and catch fire", https://en.wikipedia.org/wiki/Halt_and_Catch_Fire 00:00:30 The idea being that it's easy to synthesize from -- you just descend the tree while reading backwards from the end of the text you've generated so far, and once you no longer find a leaf for the "next" (previous) word, you've arrived at the longest prefix in the variKN model, and can read its list of frequencies for following words. 00:00:35 (For extra credit, also account for backoff weights by having a probability for going back up one level.) 00:01:07 int-e: Ah, yes 00:01:39 If I had to choose a few esoteric assembly instructions, they would be, um 00:01:48 For some reason most things dealing with n-gram models I've seen have tended to put things in a trie, yes, but the "right way around". Which I guess is just fine for a fixed-length model, but much less convenient for a variable-length one. 00:01:53 idk 00:02:10 I'm in the mood to talk about numbers right now 00:02:21 rdococ: Have you heard of the number @? 00:02:32 the number what? 00:02:41 rdococ: The number @ 00:02:44 nope 00:02:44 rdococ: |@|=-1 00:02:49 ... 00:02:51 interesting 00:02:53 |n@|=-n 00:02:56 wow 00:03:13 And I had a lot of fun writing the Funge code. It's a language that's generally much easier to write than read. 00:03:39 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds). 00:03:41 hmm 00:03:43 imagine 00:03:53 rdococ: Go on 00:03:57 hppavilion[1]: Something that randomly chooses a register, and start executing the instruction at the specified address. An instruction that moves memory from a specified location to a random location somewhere else. An instruction that reads the specified memory address, and multiplies it by a random prime number. Other useless things. 00:03:59 # = -i@ 00:04:01 @ * @? 00:04:11 @ * @ = 1? 00:04:14 or different 00:04:14 rdococ: Undefined, as of yet 00:04:22 hmm 00:04:23 rdococ: No, but j^2 = 1 00:04:25 But j != 1 00:04:30 crazy 00:04:31 And k^2 = 0 00:04:35 But k != 0 00:04:42 |@*@| = 1? 00:04:43 That's the 2D real algebras 00:04:44 we can take any condition and turn it into a number... 00:04:47 zgrep: Probably not 00:04:50 zgrep: Ask Sgeo__ 00:04:59 hppavilion[1]: Why not? 00:05:10 sqrt(@)? 00:05:12 zgrep: Because @ isn't based on squaring, it's based on absing 00:05:18 |-@|? 00:05:22 rdococ: 1 00:05:32 crazy 00:05:33 rdococ: |-@|=1 00:05:45 rdococ: Not that crazy; I already said |n@| = -n 00:05:46 3@ / 2@ = ??? 00:05:49 3/2? 00:05:54 rdococ: Yes, definitely 00:05:57 true 00:05:59 rdococ: Sgeo__ invented @ 00:06:05 now I want to invent one 00:06:09 `? @ 00:06:19 ​@ is an OS made out of only the finest vapour. 00:06:19 p. sure it wasn't Sgeo__ who invented @ 00:06:19 rdococ: Just don't make it absy, make it something new 00:06:21 $ 00:06:28 $ * 2 = 0.5 00:06:32 $ * x = 1/x 00:06:34 `misle/rn @/|@| = -1 00:06:43 Was lied to about «@» 00:06:45 but that means $ = 1... 00:06:48 huh 00:06:53 rdococ: That's actually... hm... 00:07:01 rdococ: I don't see how that behaves like numbers 00:07:18 rdococ: It has to behave reasonably like a number, but not quite, or else it's no fun 00:07:27 x$ = x/$ but $ =/= 1 00:07:37 true 00:07:44 what other numbers... 00:07:50 I mean functions 00:07:51 rdococ: @ is numeric, as are i, j, and k 00:08:03 rdococ: You can do the reverse of @'s creation process to 00:08:04 o 00:08:05 * zgrep still doesn't see why | 00:08:06 |$| = i 00:08:09 * zgrep still doesn't see why |-@| = 1... :( 00:08:14 nah 00:08:14 rdococ: $=-i@ 00:08:20 hmm 00:08:44 zgrep: Because |n*@| = -n, -@ = -1*a, and -(-1) = 1 00:08:52 Oh, wow. Heh. 00:08:59 hmmm 00:09:07 $ = cube root (-1) 00:09:10 nah 00:09:17 $ = fourth root (-1) 00:09:22 -!- yorick has quit (Ping timeout: 244 seconds). 00:09:23 rdococ: That's a normal complex number 00:09:35 hmm 00:09:48 explore quaternions. 00:09:50 rdococ: @ was made by taking an unsatisfiable expression and making a new number that satisfies it. The reverse is taking an expression with no solution and making a solution 00:10:02 |@*@| = -@? 00:10:06 should I do the inverse/ 00:10:08 ?* 00:10:08 Maybe you meant: v @ ? . 00:10:16 rdococ: η is the "Notta Constant" 00:10:16 or the one you did? 00:10:22 what's the notta constant? 00:10:28 zgrep: No, unless you have a reason 00:10:38 hmm 00:10:40 rdococ: xη = x/0 00:10:51 so basically infinity 00:10:59 hmm 00:11:02 rdococ: No, not infinity 00:11:03 hppavilion[1]: |n*@| = -n, so |@*@| = -@... What'd I do wrong? :P 00:11:08 rdococ: It's a new plane 00:11:13 zgrep: Ah, good point 00:11:14 a new plane? oh yeah 00:11:19 zgrep: NOW you're onto something 00:11:22 :) 00:11:26 hmmm 00:11:45 so I will make up an unsatisfiable expression 00:11:46 rdococ: You can also invent completely baffling things that look like numbers at first, but aren't quite 00:11:51 rdococ: Like the trigns 00:11:54 trigns? 00:11:58 rdococ: Make sure the expression is fairly simple 00:12:18 rdococ: It's like the bigns, AKA the reals. I'll leave you to extrapolate. 00:12:34 so it's a third sign? 00:12:43 plus, negative and trign? 00:12:47 rdococ: Yes, or more accurately, 2 new signs that replace - 00:12:53 oh, interesting 00:12:57 what are those signs called? 00:13:02 rdococ: $ ~ and & 00:13:10 And they have rules that I forgot 00:13:14 hmm 00:13:15 okay 00:13:22 interesting 00:13:41 hppavilion[1]: |n|^2 ≠ |n^2|, right? 00:13:45 0^x = 1 00:13:46 -!- yorick has joined. 00:13:55 my new number would be %, and 0^% = 1 00:13:57 rdococ: Here's one for you. 0/nρ = n 00:14:00 rdococ: Oh, that works too 00:14:03 rdococ: Look into that 00:14:05 rdococ: Isn't that 0? 00:14:09 rdococ: What's the general case 00:14:13 zgrep: No, that's undefined 00:14:17 hppa: isn't that another infinity? 00:14:23 the p thing 00:14:28 wait 00:14:39 hmm 00:14:41 hppavilion[1]: Oh, heh. Good point. 00:14:44 rdococ: Though I have a nagging feeling that it's related to η 00:14:57 zgrep: Yep. 00:15:01 0^% = 1, so 00:15:09 x^y% = ?? 00:15:16 zgrep: And |n|^2 usually is equal to |n^2| 00:15:31 rdococ: Obviously, 0^%n = n 00:15:39 okay 00:15:42 rdococ: You should probably give it a less mathematical name 00:15:46 Like "r" 00:15:50 lol r 00:15:56 Because @ works, but % is too mathy already 00:16:00 0^rx = ?? 00:16:05 -!- tromp has quit (Remote host closed the connection). 00:16:05 rdococ: x 00:16:08 I was originally going for $ but way too mathy 00:16:10 |-n@| = -|n@| 00:16:16 I'll use r 00:16:27 x^ry = ? 00:16:29 zgrep: I believe so. Let me chack the algebra 00:16:50 hppavilion[1]: |n^2| ≠ |n|^2 if you've got an @ somewhere in there. 00:17:19 rdococ: The simplest solution is x^ry = y(x+1) = xy+y, but that's a bit strange 00:17:27 zgrep: excellent 00:17:37 -!- tromp has joined. 00:17:44 how'd you get that solution? 00:17:50 oh 00:17:52 I get it 00:18:01 |@^2| = |@*@| = -@, but |@|^2 = 1. So |n^2| ≠ |n|^2, right? 00:18:10 you moved the y down like I moved the x down when I was doing 0^rx 00:18:27 it's interesting though 00:18:43 rdococ: It's not very good (too functional, not numbery enough), but it's the best I can think of 00:18:56 hmm 00:19:00 zgrep: Apparently, yes! Good job! 00:19:01 let's think of a different simpler one then 00:19:04 what about 00:20:31 In fact, |ab| ≠ |a|*|b| when @'s involved. 00:20:40 rdococ: x : x≠x 00:20:48 weird idea 00:20:59 does x rdococ: Maybe, who knows? 00:21:21 rdococ: I guess we shouldn't use ≠, we should use !<= 00:21:37 ԑ 00:21:45 ԑ =/= ԑ 00:21:47 nah 00:21:50 -!- tromp has quit (Ping timeout: 248 seconds). 00:21:50 looks like an infinitesimal 00:21:54 rdococ: x : x≰x 00:22:02 so x>x? 00:22:07 rdococ: Precisely 00:22:15 Order-theoretical i 00:22:16 let's call it ᴓ 00:22:20 ᴓ>ᴓ 00:22:24 rdococ: neoletters doesn't include that symbol 00:22:29 neoletters? 00:22:35 `unidecode ᴓ 00:22:38 ​[U+1D13 LATIN SMALL LETTER SIDEWAYS O WITH STROKE] 00:22:58 rdococ: Oh! My last name has that letter! 00:23:01 Kind of! 00:23:09 rdococ: Wait, it's a sideways version 00:23:12 but flipped horizontally 00:23:40 does ξ show up? 00:23:46 rdococ: Yes 00:23:54 so ξ > ξ 00:24:07 xξ > xξ? 00:24:19 rdococ: Obviously 00:24:26 what about xξ and yξ for different x and y? 00:25:02 rdococ: How does that normally work again? 00:25:17 well, 3 > 2 and 2 < 3, and 3 = 3 00:25:29 does ξ = ξ + x for any particular x? 00:26:08 ξ > ξ, and 3ξ = 4ξ? 00:26:13 or 3ξ > 4ξ? 00:26:46 What is ξ, in this case? 00:26:55 ξ > ξ 00:27:05 Ahah. So then ξ < ξ as well. 00:27:19 nope, that rule was broken 00:27:24 How? 00:27:29 hppa's idea 00:27:44 rdococ: Don't blame me! 00:27:58 so does x < ξ for any x? 00:28:13 But if ξ > ξ... why isn't ξ < ξ? 00:28:21 ask hppa 00:28:26 hppavilion[1]: This is me asking you. 00:28:27 my original idea was just ξ =/= ξ 00:28:56 rdococ: I think that was /my/ idea 00:29:09 hppavilion[1]: rdococ: x : x≰x 00:29:13 rdococ: x : x≠x 00:29:14 really now 00:29:24 Need I break out the logs? 00:29:28 oh 00:29:37 but ξ > ξwas also yours 00:30:09 -!- tromp has joined. 00:30:19 -!- J_Arcane has joined. 00:31:06 |@*@*@| = -@*@... ||@*@*@|| = @, then? 00:31:23 Where that's [[@*@*@]] not, []@*@*@[]. :P 00:31:51 @@ 00:32:01 @something 00:32:01 Unknown command, try @list 00:32:08 @1 00:32:08 Say again? 00:32:11 @2 00:32:11 Maybe you meant: v @ ? . 00:32:14 @@ 00:32:22 @poke lambdabot 00:32:22 Maybe you meant: vote more 00:32:28 No, I don't want to vote more. 00:32:29 @botsnack 00:32:29 :) 00:33:35 @|@| = -@... and |@*@| = -@... uh... :/ 00:33:36 Plugin `compose' failed with: Unknown command: "..." 00:34:19 Ω > n for any finite n, but Ω < infinity? so Ω is the largest finite number 00:34:42 -!- tromp has quit (Ping timeout: 244 seconds). 00:34:43 @( 00:34:44 Maybe you meant: v @ ? . 00:34:46 @) 00:34:46 Maybe you meant: v @ ? . 00:34:51 @[ 00:34:51 Maybe you meant: v @ ? . 00:34:54 @{ 00:34:54 Maybe you meant: v @ ? . 00:34:56 @` 00:34:56 Maybe you meant: v @ ? . 00:35:00 So what's ∞ - Ω? 00:35:04 1 00:35:11 Ω = ∞ - 1? 00:35:14 yes 00:35:15 so the lambdie answers to most about anything, except @. very peculiar. 00:35:28 v 00:35:36 boily: I'm assuming that @'s the prefix, whatever comes after that's considered the command. But just @ is ignored. 00:36:00 @ 00:36:06 > 00:36:09 > 00:36:10 : not an expression: ‘’ 00:36:11 @ 00:36:21 Huh. "@ " doesn't trigger, but "> " does. 00:36:48 or what about 00:36:52 xΩ = 1/x 00:37:03 @@@ 00:37:10 @@@@something 00:37:25 rdococ: Becomes weird, because then Ω = x^-2 00:37:27 :P 00:37:42 Though I guess it wouldn't be number, just an operation. 00:37:51 what about 00:37:52 four signs 00:38:05 * zgrep has no clue what zgrep is talking about... 00:38:07 +3 plus x2 = +6 00:38:35 +x + *y = x*y 00:39:08 and x + /y = x/y 00:39:39 This look simple: -0 [something] n = 0 [something] (-1 * n). 00:39:49 And useless. 00:39:57 rdococ: I... don't exactly get what that is... 00:40:13 rdococ: But then Ω > Ω 00:40:33 hppa: I was discussing another idea for a number 00:40:43 rdococ: I know 00:41:44 what about 00:41:45 Ʊ 00:42:47 um 00:43:09 How how did you get an upside-down omega. o.o 00:43:18 `unidecode Ʊ 00:43:21 ​[U+01B1 LATIN CAPITAL LETTER UPSILON] 00:43:27 upsilon. Neat! :D 00:43:59 what about Ψ 00:45:01 Technically, 🐈 could work as a great name too. It would also help annoy others who want to type about it. :P 00:45:04 xΨ + y(1-Ψ) = random probability of being either x or y 00:45:14 what symbol is that? it doesn't showu p on mine 00:45:20 It's a unicode cat. 00:45:46 wow 00:46:03 Ψ = random probability of being either 0 or 1 00:46:07 rdococ: Ψ: Doing any math involving Ψ comes at the expense of your sanity 00:46:16 Example problem: 00:46:19 x = Ψ+1 00:46:19 because it is random and unpredictable 00:46:23 x-1 = Ψ 00:46:29 Ψ+1 = random probability of being either 1 or 2 00:46:30 (algebra) 00:46:38 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHHAHHAAHAHAHAHAAAAAAAAAAAAAAAAAA 00:46:42 (def. of Ψ) 00:47:06 -!- tromp has joined. 00:47:06 doing any math with x comes at the expense of your sanity... 00:47:16 does that mean I have no sanity? 00:47:18 s/with x.*your // 00:47:31 Oh, wait, I cut out too many words. Damn it. 00:47:37 s/with x // 00:47:38 zgrep: Try ms/ 00:47:38 There. 00:47:41 lol 00:47:51 zgrep: ms/// is for fixing s/// expressions 00:48:21 ms/s/with x.*your///s/with x// 00:48:31 hppavilion[1]: Really? I usually either ignore the first regex I made, fix the resulting string as if the regexes were applied in a row, or I write a regex to fix the regex as a regular regex. 00:49:01 so Ψ = 0, no wait it's 1, 0, 1, 0 wait what is it 00:49:04 A regular regex expression! :P 00:49:05 zgrep: new processors for an architecture may support additional features that their predecessors did not in a backwards-incompatible way, but when people make new architectures like ARM, they aren't backwards-compatile with other architectures 00:49:25 coppro: True. 00:49:35 zgrep: Yes, that's ms 00:49:39 zgrep: That last one 00:49:56 hppavilion[1]: It doesn't end up calling our evil Microsoft overlords? 00:49:57 are there irregular expressions? 00:50:06 zgrep: Nope 00:50:11 boily: Yes hth 00:50:14 hppavilion[1]: Oh, okay. Good. 00:50:23 boily: alksfdjaoisdf matches "walrus", but not "oerjan" 00:50:39 It does match "oerjans", though. 00:50:53 But not organs. 00:50:57 zgrep: Correct, but not "oerjan's" 00:51:19 Though alksfdjaoisde would match all of the above. 00:51:27 rdococ: Maybe we should work on that hacking game 00:52:16 do we have to 00:52:17 You can say "it's not rocket surgery" as an amusing combination of the two idioms, but saying "it's not brain science" works much less well. 00:52:40 @ask oerjan do you feel matched? 00:52:41 Consider it noted. 00:53:48 -!- MDude has quit (Read error: Connection reset by peer). 00:54:13 -!- MDude has joined. 00:54:16 what about 00:54:17 x 00:54:29 χ 00:54:43 fizzie: Between a rock and a pickle? :P 00:55:03 3χ / (χ/2) = 6 00:55:07 rdococ: Yes, we have to :P 00:55:29 rdococ: Err... yes. 00:56:00 sure, all numbers satisfy that property, but 00:56:04 χ is a wildcard 00:56:12 χ is algebra without algebra 00:56:22 rdococ: How is it a wildcard? 00:56:57 -!- tromp has quit. 00:56:59 hppa: it is meant for substitution 00:57:00 rdococ: Is it a number that satisfies all properties the reals satisfy (and no others), but that isn't a real? 00:57:07 rdococ: So it's just a variable? 00:57:13 not exactly 00:57:14 rdococ: You're trying to turn math into C macros? D: 00:57:17 rdococ: These numbers are boring me. 00:57:25 ugh 00:57:25 fine 00:57:55 H4X1N470R-MP 00:58:03 Θx = sin(x) 00:58:32 rdococ: That's a function again 00:58:49 hmm 00:58:53 rdococ: There's a difference between weird numbers and weird function call syntax 00:59:12 0^0 = μ 00:59:39 -!- lambda-11235 has joined. 01:00:51 what about that 01:01:53 -!- tromp has joined. 01:02:32 should I just google a random function, find a spot where the function doesn't havw a value, and make a spot there? 01:04:19 what about 01:04:29 Ϝ + 1 = Ϝ 01:04:41 ∞ 01:04:52 Ϝ < n for all n 01:05:03 -∞ 01:05:19 uh 01:05:24 (although not completely, but almost) 01:05:51 I wonder if it's possible to somehow teach mathematica what @ is... 01:05:54 s/mathe/Mathe/ 01:06:04 try it 01:06:07 |@| = -1 01:06:14 what about 01:06:52 β = x for any x 01:06:53 Expression cannot begin with "Abs[@]=-1". :P 01:07:16 try a different symbol like x 01:08:03 -!- mihow has quit (Quit: mihow). 01:08:08 I'm wondering, how would I begin to figure out what |@+k| would be... 01:08:28 -!- mihow has joined. 01:08:40 well, if |x@| = -x@ 01:08:49 then |@+k| = -@ - k 01:09:01 Why? 01:09:06 no wait 01:09:28 |@+k| = |@| + |k| = -@ + |k| 01:09:37 |a| + |b| ≠ |a + b| 01:10:23 why is |@| = -@, i thought it'd be -1 01:10:35 oh yeah 01:10:36 |@| = -1 indeed. 01:10:38 |@| = -1 01:10:40 `tomfoolery @ 01:10:43 ​|@| = -1 01:10:48 |k@| = -k 01:11:05 |@+k| = -@ + k? 01:11:22 `misle/rn @/|n@| = -n 01:11:26 Was lied to about «@» 01:11:46 `tomfoolery random number 01:11:55 `tomfoolery random 01:12:00 what is |-@|? is it |-1@| = 1? 01:12:03 Eh, I guess that's no longer there. 01:12:04 yeah 01:12:12 |-@| = 1 01:12:16 8901 01:12:17 I must confess, I know not of what you are speaking. 01:12:27 Oh, it worked, just slowly. 01:12:34 `tomfoolery random number 01:12:36 6903 01:12:39 :D 01:12:52 so, what is |@+k|? 01:13:19 I would assume the k part gets absed 01:13:40 if I had my way, |-@| = -1 01:13:50 wait no 01:13:54 then @ = -1 01:13:56 no 01:14:03 not that 01:14:28 Ψ is far more interesting 01:14:37 `tomfoolery random number 01:14:40 17204 01:14:50 that's Ψ*100000 01:15:09 no wait 01:15:09 `cat tmflry/random number 01:15:10 it's not 01:15:11 echo $RANDOM 01:15:33 uh 01:15:34 ugh 01:16:19 rdococ: I'm back 01:16:32 myname: |@+k|? = 19 01:16:39 uh? 01:16:39 s/19/42/ 01:16:48 42 01:16:50 that would make @ pretty pointless 01:17:00 hppavilion[1]: Heheh. 01:17:14 Err... how did you mistype 42 as 19? 01:17:16 it already is useless 01:17:19 myname: No, it makes ? pretty pointless 01:17:19 sorta 01:17:20 idk 01:17:21 n? = 42 01:17:26 :D 01:17:30 zgrep: I changed my mind 01:17:36 "The answer to life, the universe, and everything"? = 42 :P 01:17:44 zgrep: Exactly 01:17:57 (?) :: a -> Int 01:18:35 that's const 01:18:43 (?) (+5) (-5) = 42? 01:18:48 rdococ: When you said "Ψ is far more interesting", what does Ψ equal? 01:18:55 I did? 01:18:57 myname: It's an example of a const 01:19:06 rdococ: you did 01:19:22 Ψ = a superposition of 0 and 1 01:19:37 rdococ: A superposition? Are you sure it isn't just 0 or 1 at random? 01:19:45 rdococ: OOH! OOH! QUANTUM MATH! 01:19:47 okay it's 0 or 1 at random 01:19:52 or superposition 01:19:59 Ψ = 0 or 1 01:20:00 |@|*|@| = 1... |@*@| = -@. <-- Why does this have to ruin everything. :( 01:20:07 rdococ: math : Computer math :: Quantum math : Quantum computer math 01:20:21 zgrep: It doesn't, it just breaks one property 01:20:22 omg 01:20:33 rdococ: :) 01:20:34 -!- jaboja has joined. 01:20:34 hppavilion[1]: But it ruins ease of use. :P 01:20:43 zgrep: you made mathematica learn @?!!! 01:20:44 zgrep: Welcome to #esoteric 01:20:49 zgrep: because @ is atupid 01:20:55 I feel welcome here. 01:20:58 myname is going to hell 01:21:02 myname: It's not tupid? Okay. 01:21:11 zgrep: You are, but myname isn't any more 01:21:14 no, it's atupid 01:21:19 it breaks even Z 01:21:23 rdococ: Making it not tupid 01:21:33 rdococ: a- is a prefix for "not" 01:21:40 hppa: it is xtupid for any x 01:21:50 rdococ: Type my full nick, please 01:22:00 rdococ: I don't get alerted if you don't 01:22:05 rdococ: Use tab-complete 01:22:20 myname: How does it break Z? 01:22:31 hppavilion[1]: k 01:22:32 see zgrep's example 01:22:57 the existence of @ would make 1 = -1 01:23:18 "hppa" makes me want to pronounce it "хпавилион". :P 01:23:18 really? 01:23:24 which makes 2 equal to either 2, -2 or 0 01:23:34 -!- mihow has quit (Quit: mihow). 01:23:40 x = -x and x =/= 0 01:23:51 rdococ: β looked good 01:24:32 -!- mihow has joined. 01:24:43 remind me what it was 01:24:51 -!- mihow has quit (Client Quit). 01:25:17 rdococ: Nevermind, β is stupid 01:25:22 a=β, b=β, a!=b 01:25:43 hppavilion[1]: but 1 = -1 is fine? 01:25:58 myname: Where is 1 = -1? 01:25:59 myname: I'm assuming that they consider |a|*|b| ≠ |a*b| to be fine. 01:26:06 myname: Oh 01:26:28 zgrep: Actually, that is true, I think 01:26:30 Let me check 01:26:44 zgrep: it will most likely brealk at |@+k| nontheless 01:26:47 Well, for all reals, |a|*|b| = |a*b|, I think. 01:26:58 zgrep: Not for the complexes though 01:26:59 myname: Well, if k = @, then |@+@| = -2 :P 01:27:11 hppavilion[1]: Probably something like that. 01:27:20 |i| = 1, |1| = 1, |1+i| = sqrt(2) but |1|+|i|=2 01:27:23 Wait, shit 01:27:26 You were doing * 01:27:36 |i| = 1? 01:27:43 zgrep: Yeah. You didn't know that? 01:27:47 I do now. :D 01:27:52 ||@|| = 1 01:28:10 that depends on what norm you are using 01:28:11 Usually, |x| = ||x||, but not when dealing with sgeoids 01:28:21 hppavilion[1]: Wait, but why is |i| = 1? 01:28:25 myname: What norm? 01:28:33 zgrep: Because the distance between i and 0 is 1. 01:28:39 Good point! :D 01:28:43 |i| could also be i 01:28:47 |a+bi| = sqrt(a^2+b^2) 01:28:50 myname: Since when? 01:29:01 @ introduces negative distances... does that simply mean going backwards in time? :P 01:29:05 because the dostance between i and 0 is i 01:29:25 myname: No, it's 1 01:29:28 myname: It's clearly 1 01:29:29 what about a number j which relates to hyperbolic trig? 01:29:45 hppavilion[1]: that depends 01:29:46 -!- jaboja has quit (Ping timeout: 252 seconds). 01:30:24 |a + bi| = |a| + |b| is as valid as |a + bi| = sqrt(a^2 + b^2) 01:30:34 both of them are well defined 01:31:35 no, the complex plane is like an x/y plane - the distance from i to 0 is 1 01:31:59 and from 1+i to 0 is sqrt(2) 01:32:36 as long as it follows some rules, anything is fine really 01:32:58 but those are the rules of the complex system 01:33:18 like, |x| = 0 => x = 0 01:33:40 |a| + |b| >= |a+b| 01:34:18 but I have the idea of a hyperbolic complex plane where e^jx = sin(x) + icosh(x) 01:34:31 sinh* 01:34:39 j* 01:34:45 jcosh* 01:39:06 -|@| = |i|. Huh. 01:39:34 |i| = 1 01:39:40 |@| = -1 01:39:47 -|@| = 1 01:39:48 that's what he said 01:40:24 |3| = |3| 01:41:14 -!- boily has quit (Quit: SELECTION CHICKEN). 01:41:14 |k@| ≠ |k||@|... argh! If only this weren't true, then things with @ would be much easier to think about. 01:41:48 -!- jaboja has joined. 01:41:57 zgrep: |k@| = k|@| 01:42:03 But that's not very good 01:42:15 hppavilion[1]: True... 01:42:17 that would break math for k = @ 01:42:27 Would it? 01:42:33 I don't see it breaking math for k = @... 01:42:34 you posted it 01:42:42 Well, what I wrote, yes. Not what hppavilion[1] wrote. 01:43:02 so |@@| is -@? 01:43:27 Yeap. 01:43:32 and |-3@| would be 3? 01:43:35 Yeppers. 01:43:41 Because |n@| = -n 01:43:44 omg that actually make sense 01:44:14 |n@| = -(|n|@/@) 01:44:27 @.@ 01:44:27 Maybe you meant: @ . 01:44:33 @_@ 01:45:11 what about |@+k| 01:46:00 rdococ: Wat? 01:46:04 myname: Exactly. :/ 01:47:56 asume k = @, that would make |@+k| = -2 01:48:16 |@+k| = -k/@ + -1? 01:48:20 k = -@ would make it 0 01:48:51 |k/@| 01:49:04 what about 1/@ 01:49:26 k = n@ -> |@+k| = -(n+1) 01:50:24 k = -n@ -> |@+k| = -n+1 01:51:50 k = n -> |@+k| = -1 - ((n/@)+1)...? 01:52:46 n = @ -> |@+n| = -1 - ((n/@)+1) -> -3...? 01:53:32 k = n@ -> |@+k| = -(n+1) 01:53:33 k = -(n+2)k -> |@+k| = -(n+2)+1 = -(n+1) 01:53:37 now plug in n = 1/@ 01:54:03 so |@ + k| = |@ - k + 2| 01:54:09 k = (1/@)@ -> |@+k| = -((1/@)+1) 01:55:12 |@+x| = -1/@ - 1 01:55:32 |@+x| = -x/@ - 1 01:55:54 nowait 01:56:17 k = @; |@+k| = -2 01:57:22 <\oren\> The enigma machine had a keyboard that was qwertzuio asdfghjk pyxcvbnml 01:58:56 |x@| = -x 01:59:19 now say x = (@+x)/@ 01:59:28 |x| = |-x| -> |k + @| = |k - 2 - @| -> @ = -2 - @ -> 2@ = -2 -> @ = -1 01:59:36 y* 01:59:52 rdococ: sounds wrong 02:00:01 |x@| = -x 02:00:13 now say x = 1 + y/@ 02:00:20 how should (a+b)/b = a? 02:00:32 that would make a + b = a * b 02:01:09 \oren\: Intriguing. 02:01:09 a = (b+c)/b? 02:01:19 I meant y there 02:01:47 so a = 1 + c/b? 02:02:02 yes 02:02:36 what are you trying to do 02:02:47 Oh, that's interesting. 02:03:13 |n@|=-n; n=(@+x)/@; |@+x| = -(@+x)/@... huh. 02:04:04 i don't quite get where these division rules came from 02:04:24 Are you saying I'm assuming @/@ = 1? 02:04:54 i don't get what made you assume n = (@+x)/@ 02:05:51 Who cares, it works? 02:06:04 how so 02:06:06 And it was rdococ that made me assume. :P 02:06:14 It figures out what |@+k| is. 02:06:25 solve for k = @ 02:06:29 So we want to know |@ + x|. 02:06:35 Hm. Let's try. 02:06:38 Let's divide both sides by @. |(1 + x/@)@| 02:06:39 ah 02:06:45 Apply the rule. -(1 + x/@) 02:06:57 Simplify. -1 - x/@ Assuming my knowledge of parentheses are correct. 02:07:00 -1-1 = -2 :D 02:07:15 It works, of coures. :) 02:07:17 s/res/rse/ 02:07:22 Yay 02:07:24 Or, wait... does it? 02:07:30 Yeah, probably. 02:07:33 Yep, it works. 02:08:07 looks reasonable 02:08:24 i am not sure if that somehow breaks my contradiction above 02:08:28 I've known that a + b = (1 + b/a)a for ages now. 02:08:30 `misle/rn @/|n@| = -n; |@+k| = -(@+k)/@ 02:08:35 Was lied to about «@» 02:09:07 myname: Hm, which contradiction? 02:09:11 a + b = (a/a + b/a)a = (1 + b/a)a 02:09:22 I dunno if it applies to @, but I don't see why otherwise. 02:09:43 |x| = |-x| -> |k + @| = |k - 2 - @| -> @ = -2 - @ -> 2@ = -2 -> @ = -1 02:09:47 that one 02:09:54 Why does |k + @| = |k - 2 - @|? 02:09:59 Or are we assuming that? 02:10:32 because |k + @| = |-(k + 2) + @| 02:10:43 Err... why? 02:11:05 |5 + @| = |-7 + @|... nope, does not make sense, unless that's why it's contradictory 02:12:03 |5 + @| = -1 - 5/@, meanwhile |-7 + @| = -1 + 7/@. One's less than -1, and the other's more. That's assuming @ > 0. 02:12:09 `tomfoolery @ 02:12:10 ​|n@| = -n; |@+k| = -(@+k)/@ 02:12:12 because |k + @| = -(n + 1) for k = n@ and |-(k+2) + @| = -(k+2) + 1 02:12:25 k = -n 02:12:26 -!- Lyka has left. 02:12:36 Err, no, I'm wrong, sorry. 02:13:05 |5@ + @| = |-7@ + @| 02:13:11 So the assumption is that k = n@..., then |k + @| = |n@ + @| = |(n+1)@| = -n-1 02:13:26 wait, no 02:13:50 so wait, |k + @| = -(k/@ + 1), right? that follows what I got, then |-(k+2)+@| = -(k+2) + 1...wait, what? 02:13:54 6 ≠ -6 :P 02:14:17 |5@ + @| ≠ |-7@ + @| because |a@| ≠ |a||@| 02:15:04 but |-n@ + @| = -n + 1 02:15:10 No. 02:15:11 -n - 1 02:15:14 `misle/rn @/|n@| = -n; |@+k| = -1 - k/@ 02:15:17 Err.. wait. 02:15:18 Crap. 02:15:25 Was lied to about «@» 02:15:28 `tomfoolery @ 02:15:29 ​|n@| = -n; |@+k| = -1 - k/@ 02:15:35 I didn't get your version so I put my own 02:15:45 |-n@ + @| = |(1-n)@| = n-1 02:15:51 ^ myname 02:16:12 makes sense 02:16:17 indeed 02:16:23 hmm 02:16:24 rdococ: Same thing. :P 02:16:54 (that's with regards to |@+k|) 02:17:44 interesting 02:20:33 -!- jaboja has quit (Ping timeout: 240 seconds). 02:21:04 rdococ: H444444444444444XXX11111N444444770000000000R 02:21:08 -!- mihow has joined. 02:23:59 -!- p34k has quit. 02:24:57 ^myname 02:25:09 I think it'd look nicer if rephrased: |@-k| = k/@ - 1 :P 02:25:14 Are you there fungot? It's me, hppavilion[1] 02:25:15 hppavilion[1]: not having guarantees for these things is sensible. thread-terminate! brings nothing but trouble...? or am i screwing something up. b. 02:25:40 pretty good answer 02:27:13 So what'd |1/@| be... 02:27:53 -!- mihow has quit (Quit: mihow). 02:27:56 for that we have to know what |@^k| is 02:28:09 Not necessarily. 02:28:36 well, but we want it nontheless 02:28:43 Inverse isn't exactly the same thing as power, not completely/exactly. As far as I know. 02:29:04 |@*@| = -@ 02:29:12 Hm... |@| = -1, |@@| = -@, |@@@| = -@@, etc. 02:29:35 |@^k| = -(@^k-1) 02:29:55 |@^-1| = -(@^-2) 02:29:55 Please put some parens there. :P 02:30:06 -(@^(k-1)) 02:30:10 Indeedy. :) 02:31:02 so, |1/@| is -1/(@@)? 02:31:14 Hm... 02:31:23 |1/0| = -1/(@@) 02:31:28 @* 02:31:28 Maybe you meant: v @ ? . 02:31:31 s/0/@ 02:32:58 Seems like that's it. 02:35:00 `tomfoolery @ 02:35:01 ​|n@| = -n; |@+k| = -1 - k/@ 02:35:32 `misle/rn @/​|n@| = -n; |@+k| = -1 - k/@; |@^q| = -(@^(q-1)) 02:35:35 Was lied to about «@» 02:35:57 x^@ 02:36:01 Oooh, yes... 02:36:32 that's a taugh one 02:38:10 we may ignore it :D 02:38:16 :/ 02:38:42 like, |x^@| could be x^@ 02:39:15 |@^@| = -(@^(@-1)) 02:39:17 i don't see how thatwould break anything since that exists für even exponents 02:39:32 why that 02:39:43 -!- tromp has quit (Remote host closed the connection). 02:39:56 |@^@| also is a nice emoticon 02:39:58 |x@^y| 02:41:35 |x@^y| = x|@^y| = -x@^(y-1)? 02:41:53 Do you mean x(@^y) or (x@)^y? 02:42:10 (x@)^y 02:42:53 (x^y)(@^y) 02:44:02 |@^0| = -(@^(-1)) 02:44:48 Technically... 02:45:25 |((k^@)/@)@| = |k^@| = -((k^@)/@) 02:46:28 k^@ = -@|k^@| 02:47:17 the @^0 is bad 02:47:33 is there any way tp make it be 1? 02:48:27 |ab@| = a|b@|, right? 02:48:42 i guess 02:50:06 myname: I guess that |@^0| simply isn't 1... 02:51:05 myname: It works out: |@^1| = -(@^0), so @^0 = -|@^1| = -|@| = 1 02:51:25 what version? 02:51:33 |@^q| = -(@^(q-1)), q = 1 02:52:16 ah 02:53:32 but i donjt like how |@^0| != |1| 02:53:36 but that is fine 02:53:42 Let's assume that it is... 02:53:43 But 1 = |1|... and -(@^(-1)) = |@^0|... so -1 = @^(-1)... 02:53:49 |0^0| isn't 1 either 02:54:26 So 1/@ = -1? 02:54:31 That doesn't work, does it? 02:55:15 No, it doesn't. 02:55:22 Because @ ≠ -1, right... 02:55:36 Where did I mess u. 02:55:37 s/u/up/ 02:55:49 @^0 simply isn't 1 02:55:50 Maybe you meant: wn v rc pl id do bf @ ? . 02:56:12 myname: No, @^0 seems to be 1... just |@^0| doesn't seem to be 1... :( 02:56:23 ...unless @^0 isn't 1, because @^1 ≠ 1... 02:56:24 fine, too 02:56:37 How's that fine, though... |1| ≠ 1, then? 02:56:55 right 02:57:05 so @^0 must not be 1 02:57:31 `tomfoolery 02:57:32 `tomfoolery @ 02:57:37 but that may imply @^1 is not @ 02:57:42 I have nothing to tell you. 02:57:42 ​​|n@| = -n; |@+k| = -1 - k/@; |@^q| = -(@^(q-1)) 02:57:59 |@^1| = -(@^(-2))... 02:59:05 well, but as you said, |@^1| = -(@^0) 02:59:32 if @^1 were @, that would mean @^0 is 1 02:59:44 -!- tromp has joined. 03:00:22 so we either have to define |@^x| somehow else or we have to say that @^0 is not 1 and @^1 is not @ 03:00:28 Whoops, 03:00:29 s/,/./ 03:00:32 i don't like the second part, though 03:00:58 Yeah... 03:01:21 so the exponential rule is wrong 03:01:39 I mean, it works out that @^0 = 1 and @^1 = 1, they work together... but they end up giving @ a wrong value... 03:02:00 I guess... 03:02:02 we may do something like ^(sqrt(x)) to work around these sneaky 0 and 1 edge cases 03:04:30 a^b = a*a*a... b times, right? That's the definition we're going with? 03:04:34 Or something else. 03:04:36 wait, how did your 1/@ worked above 03:05:06 why is -1 = @^(-1) 03:05:18 I'm not sure what I did above... I've lost my train(s) of thought. 03:05:45 https://upload.wikimedia.org/math/7/f/8/7f80d3b6fbe1d4e35eca5022242872bf.png <-- that's the definition we're going with, right? 03:06:01 -1 = |@| = |@^1| = -(@^0) 03:06:07 yeah 03:06:16 Indeed... 03:07:26 i do think you made an error above but i am quite unsure 03:07:38 I probably made a few errors above. 03:07:42 |@^k| = @^(k-1) * |@|, right? 03:08:10 Simply because of the fact that |a@| = a|@| 03:08:34 So |@^k| = -(@^(k-1)) 03:08:52 If @^0 = 1, then 1 = |@^0| = -(@^(-1))... 03:09:12 aha! 03:09:34 1 = -(1/@)... that means 1/@ = -1...??? 03:10:18 that's... weird - I heard you talking about it above but never stopped to read 03:10:48 ah 03:10:50 If @^1 = @, then -1 = |@^1| = -(@^0), then 1 = @^0... 03:11:10 -1 = (1/@)... 03:11:18 @^0 = 1, @^-1 = -1, and @^1 = @. 03:11:18 Maybe you meant: wn v rc pl id do bf @ ? . 03:11:43 I guess 1/-1 has two solutions now 03:12:03 @ and -1? 03:12:08 may work out 03:12:21 So how did I show that @ is a number... 03:12:35 but does this imply it for any division? 03:12:51 like, is 1/-2 equal to some @ thing? 03:13:19 |@^(-2)| = -(@^(-3))... 03:13:32 1/-2 = -1/2 or (1/2)@ 03:13:35 I would say 03:13:47 |@^-1| = |-1| = 1... so -(@^(-2)) = 1? 03:14:03 that would make -1 may or may not be @ 03:14:24 (@*@) = -1 03:14:25 ? 03:14:25 @^-2 = -1? 03:14:25 Unknown command, try @list 03:14:38 |@*@| = 1?!?!?!?! 03:14:44 no 03:14:51 |@@| = -@ 03:15:06 but zgrep said 03:15:24 If @^-2 = -1, then -1*(@^2) = 1, then @^2 = -1... D: 03:15:48 ...then that means sqrt(-1) = @ and @ = i... 03:15:58 but that's ludicrous 03:16:02 |i| = 1, though. 03:16:05 @ is ludicrous. 03:16:23 indeed, bordering on ridiculous 03:16:31 but interesting enough to keep around 03:16:34 Something weird is going on here with powers... :( 03:17:03 @^-1 is -1, right? 03:17:03 Unknown command, try @list 03:17:07 but that would mean |i| = -1 03:17:27 rdococ: ot depends on how you calculate 03:17:31 but yeah 03:17:58 if we assume that @^-1 = -1, then (@^-1)^-1 = -1 too, so @^-2 = -1... 03:18:03 it acts a lot like -1 03:18:05 -!- earendel has joined. 03:18:18 if my maths is right 03:18:38 I'm going to take a break from @ for now. 03:18:56 hmm 03:19:24 I think the problem lies in the absolute value function of x@ 03:19:25 What if we just say that |n@| = n|@| instead of -1... 03:19:37 Err, instead of -n. 03:19:49 since f(x) = |x@| doesn't swap slopes at the origin 03:20:18 I mean, |x@| = -x, but |x| = x if x >= 0, and -x otherwise 03:20:23 no such conditional in the @ 03:20:24 rdococ: i don't get ypur (@^-1)^-1 = -1 03:20:33 Something makes me think this is simply a problem of choosing absolute value, something related to distances... 03:20:37 @^-1 = -1 03:20:37 Unknown command, try @list 03:20:48 -1^-1 = 1/-1 = -1 03:20:52 or @ 03:21:01 if we say @^-2 = @ 03:21:10 then we complicate things a lot 03:21:19 ah, i see 03:22:19 we may need to start from stratch 03:22:28 I'm going to make a small adjustment to @'s behaviour, and call the new one ©. 03:22:35 |x|-|x| = -2x? 03:22:50 |x©| = -|x| 03:22:55 so |-©| = -1 03:22:55 |x|-|x| = 0 03:23:23 rdococ: Hm. 03:23:28 if we try © instead of @, do you think it will turn out with less contradictions? 03:23:30 is @ something in particulatr already? 03:23:41 `tomfoolery @ 03:23:43 ​​|n@| = -n; |@+k| = -1 - k/@; |@^q| = -(@^(q-1)) 03:23:47 © is different in that |-©| < 0 03:23:48 thanks 03:23:53 while |-@| > 0 03:24:01 Every time I see @ I think ehird. Alas. 03:24:21 |©| = -1? 03:24:28 yep 03:24:44 |©+k| = -1 - k/©, there's no change there, infact it works with any variable or constant 03:24:46 rdococ: which is also true for @ 03:25:01 but |-©| = -1 where |-@| = 1 03:25:04 ah 03:25:06 |n©| = -|n|... hm... 03:25:44 that might work 03:25:53 |©©| = -|©| = 1... interesting, difference already - it almost looks recursive 03:25:56 but i am way to lazy to type a copyright symbol 03:26:02 just put c instead 03:26:04 |n©| = -|n|, |((©+n)/©)©| = |©+n| = -|((©+n)/©)| 03:26:13 whut 03:26:23 Different n's, sorry. 03:27:01 we might as well call it @ and remove out old draft 03:27:13 nah, incase © runs into problems too 03:27:20 -|((©+n)/©)| = -|1 + n/©|... I don't see how this is still the same as @? 03:27:48 true 03:27:58 :) 03:28:00 hmm 03:28:07 not the same, but we basically made all the rules up from a wrong first definition 03:28:25 we might as well change that definition and start over still calling it @ 03:29:07 I don't see how to unwrap it from the absolute value, though, which could be what makes it work, but still... 03:29:15 ...sort-of sad. :( 03:29:15 |© + k| = |(1 + k/©)/©|. Using the rule where |n©| = -|n|, we get -|1 + k/©| 03:29:21 or choose some symbol that's on an ordinary keyboard 03:29:42 © looks like @ 03:29:47 invariance of @ 03:30:04 so 03:30:39 myname: Use the all-mighty compose key! :P 03:30:51 |n©| = -|n|; |© + n| = -|1 + n/©|; |©^n|? 03:30:55 that's a pain in the ass on a software keyboard 03:31:16 myname: Oh... software keyboard? Those have © somewhere, but yeah, it's annoying to get to usually. 03:31:26 Try clicking and holding on 'c' or 'g'? 03:31:42 Or use c. Or `. Or any symbol, really. :P 03:31:47 i have it in a seperate menu on , 03:32:17 |©^n| = -|©^(n-1)| 03:32:19 |©^n| = -|©^(n-1)| I thknk. 03:32:22 s/knk/ink/ 03:32:30 ∆ is also an option 03:32:41 But ∆ is taken for small changes. 03:32:43 |©^1| = -|©^0| = -1 03:32:47 and while we are at it, let's define the ― operation 03:33:00 -? Or –? Or —? 03:33:11 the last 03:33:13 |©^2| = -|©^1| = |©^0| = 1 03:33:24 indeed, recursive 03:33:29 ʒ 03:33:37 How about £. :P 03:33:47 ¸ 03:33:52 & 03:33:56 works nice for multiplication 03:34:03 This is a nice symbol for a variable, right: ‰ ? :P 03:34:11 maybe & 03:34:30 ¡ 03:34:49 -5 = |5©|... hm... 03:35:11 |©^x| is -1 for an odd number but 1 for an even one... weird 03:35:14 © works nicely. 03:35:18 (hold ALT) 169 03:35:18 Sort-of bland, though. 03:35:20 |©^0| = 1 works though 03:35:24 |©^-1| = -1 03:35:32 |©^-2| = 1 03:35:41 So ©^2 is now 1? 03:35:46 what's a software keyboard? like onscreen interface? 03:35:52 yes 03:35:55 yeah 03:35:57 kay. 03:36:04 pretty normal on smartphones 03:36:06 |©©| = -|©| = 1 03:36:12 * zgrep takes a break from this to eat some cake 03:36:14 it's recursive 03:36:18 But |1| = 1, therefore ©© = 1 03:36:19 but 03:36:30 what is |@©| 03:36:31 uh? 03:36:33 Or ©© = -1... 03:36:38 ©© = ±1 03:36:43 There. 03:37:03 |©| = -|©^0| = |©^-1| 03:37:13 myname: :D 03:37:28 |©©| = 1, surprisingly 03:37:32 myname: Obviously -© = 1 03:37:46 If trying to find |@©| you get that. :P 03:37:57 But I think that's because we don't have a working @. 03:38:22 © = 1 or i... depending on the context? :P 03:38:42 * zgrep really goes away now 03:38:43 when did © become i? 03:38:51 If ©© = ±1... 03:38:59 ...I cheated and reversed an absolute value... :P 03:39:05 then |©©| = 1...huh, it is 03:39:20 hey 03:39:23 you can't do that 03:39:29 Why not? :( 03:39:36 The only way to win is to cheat... 03:39:45 1 = |1| = |-1| = |©©| 03:39:55 so you mean 1 = -1 now? 03:40:14 you can't reverse an absolute value, like you did anyway, whether © exists or not 03:42:01 anyone else? 03:42:06 hppavilion[1]? 03:42:42 we made a modified version of @ called ©, and |x©| = -|x|...hello? 04:00:01 @tell Sgeo__ @ seems to fail with regards to exponents... at least, it doesn't do too well... 04:00:01 Consider it noted. 04:00:51 -!- bender| has joined. 04:01:35 zgrep, what's going on? 04:03:02 Sgeo__: So exponents, as in |@^k|. If we assume that |n@| = -n, and n = (@^k)/@, then |@^k| = -(@^k)/@ = -(@^(k-1)) 04:06:10 |@^1| = -(@^0); @^1 = @ (because of https://goo.gl/XMm8lT); So |@| = -1 = -(@^0); so @^0 = 1, so far so good... 04:06:13 But... 04:07:31 |@^0| = |1| = 1 = -(@^(-1)); so (1/@) = -1. Therefore -@ = 1, and @ = -1... :/ 04:08:02 Sgeo__: Which it isn't. 04:08:30 Any suggestions? 04:08:46 rdococ: Suggested |n©| = -|n|, which seems to do the trick. 04:08:59 Where does |n@| = -n come from? Was that my original definiton, or was it |n@| = -|n|? If i remember my most recent proposed definiton for |a+b@| properly, it would be the second, I think 04:09:57 Sgeo__: I don't know what your original definition was, but I know that I first heard of it as |@| = -1. 04:10:15 Sgeo__: What was your most recent definition? 04:11:13 |a+b@| = sqrt(a^2 - b^2) if a^2 - b^2 is positive, i*sqrt(a^2 - b^2) if a^2 - b^2 is negative. Or something like that 04:11:37 Eek. More if statements, atop the absolute value... :( 04:12:39 So theoretically I could have |3+4i+5@+6i@|? :D 04:12:53 Not sure if my definition extends to cover that 04:13:10 If a and b can be complex, then yes. 04:13:16 I think. 04:14:12 |a+b@| = |√(a²+b²)| ? 04:14:18 Err. no. 04:14:26 |a+b@| = √(|a²-b²|) ? 04:14:27 That? 04:15:16 `misle/rn @/|a+b@| = √(|a²-b²|) 04:15:20 Was lied to about «@» 04:15:33 HackEgo is slower than I last recall... 04:17:25 ...either that, or it's my connection. 04:17:42 So |0+1@| = sqrt(|0 - 1|) = 1??? 04:18:11 well maybe not??? 04:18:40 Hm... or I'm wrong... 04:19:03 i*sqrt(a^2-b^2) = i*sqrt(-1) = i^2 = -1... :/ 04:19:42 But i = sqrt(-1) so i*sqrt(a^2-b^2) = sqrt(b^2 - a^2), no? 04:21:48 I guess not. 04:22:06 so @ is a time dimension? 04:22:37 But... I*Sqrt[b] == Sqrt[-b] ??? 04:22:48 yes, I think 04:22:54 Why isn't this working, then? 04:23:12 dunno 04:23:22 a = 0; b = 1; I*Sqrt[a^2 - b^2] => -1; Sqrt[b^2 - a^2] => 1; :( 04:23:38 but I love the idea of a complex hyperplane that takes place in 2 dimensions of space and 1 of time 04:24:17 Sgeo__: was that your intention? a time dimension? 04:24:45 Well, that's what a negative result from absolute value ends up being. 04:24:47 I guess. 04:24:54 Though not exactly. 04:25:03 My recent definition of |a+b@| was inspired by the time dimension, which I believe could be described with a numvber # such that |#| = i 04:25:21 that makes sense 04:25:49 `misle/rn @/|a+b@| = { √(a²-b²) if a²-b² ≥ 0 ; i√(a²-b²) if a²-b² < 0 } 04:26:08 Was lied to about «@» 04:26:31 |a + bi + cτ| = √(a² + b² - c²) 04:26:41 three dimensional tau space 04:26:47 tau representing time dimension 04:27:22 |τ| = sqrt(-1) = i, so it fits your definition too 04:27:42 a + bi + cj + dk <-- 3 dimensions + time? 04:27:53 a + bi + cj - dk 04:28:00 I was close. 04:28:18 Neat. :D 04:28:30 the time dimension is negative 04:28:51 distance is a measure of how hard it is to get to somewhere, so more time allowed, the easier it is 04:29:05 cool, right? 04:29:47 Hm.... hmmmmm.... :D 04:29:57 |a + bi + cτ| = √(a² + b² - c²) if √(a² + b² - c²) >= 0, otherwise i√(a² + b² - c²) 04:30:10 Wait... so if I have 4τ, then |4τ| = ??? 04:30:17 Ahah. 04:30:24 4i? 04:30:29 Err... not 4i. 04:30:31 2i? 04:30:42 |4t| = sqrt(-16) = 4i 04:30:44 Err. No. I can't think. 04:30:47 Yeah, 4i. 04:30:55 But... how is 4i easier than 3i? 04:31:12 i is basically the negatives 04:31:32 because square roots are like that 04:31:33 That's... why? 04:31:37 But... why? 04:31:55 because x*x=-1 doesn't have any real solution 04:32:23 using my adjusted formula, |4t| = i*sqrt(-16) = 4ii = -4 04:32:29 which is easier than -3 04:32:43 But... why's it adjusted that way? 04:32:56 I guess you could say it just, err, is that way... but... 04:33:06 Oh, there. I see. 04:33:11 I didn't finish reading the line. 04:33:44 but mine is basically the same as Sgeo__'s @ 04:34:05 except that I think tau would be a better symbol to fit it than @ 04:34:17 Huh. Is this tau known by any other, more widely accepted names? 04:34:23 tau / @ / whatever. 04:34:29 sometimes t 04:34:53 I meant more as in, is there a wikipedia page with a snippet about it? Or any other links? 04:35:11 nah, idk if anyone thought of it yet 04:36:43 This is pretty neat. 04:37:22 well, I did have a similar idea 04:37:38 a number y where e^yx = sinh(x) + ycosh(x) 04:37:50 tau might fit it 04:38:28 so I guess |@| = -1 after all, since it's time and all 04:39:57 I should read these logs at some point 04:40:13 I'm only half paying attention, and it involves stuff I'm involved with 04:40:31 we should popularise this @ or tau idea 04:40:40 it would be useful 04:40:50 tau, or @? 04:40:53 both 04:40:56 they're the same 04:40:58 tau. 04:41:03 Damn, autocorrect. 04:41:03 just different names 04:41:05 atau. 04:41:06 :P 04:41:15 t@u? :P 04:41:20 lol 04:42:14 I'm confused by rdococ's definition... sqrt is always (normally) >= 0 unless comparison is undefined 04:42:37 Should it be read as just a^2 + b^2 - c^2 in the conditional? 04:42:41 |a + bi + cτ| = √(a² + b² - c²) if a² + b² - c² >= 0, otherwise i√(a² + b² - c²) 04:42:43 yes 04:42:44 you're right 04:43:01 but it's basically the same as yours, with i added on top 04:43:27 I think this kind of stuff is used, just not in this format 04:43:37 rdococ: I'm-a-back 04:43:54 Is there a way to get the result to be imaginary? Because afaict exactly that is used as the time dimension 04:43:56 also, since it's mostly position, not rotation, if we add a third dimension they don't have to be quarternions 04:44:13 then just remove the conditional 04:44:49 Sgeo__: With @? 04:44:50 But then we exclude negatives 04:45:04 Sgeo__: |$| = i -> $ = -i@ 04:45:10 |a + bi + c@ + d#| = ? 04:45:10 what would it mean for the result to be imaginary? 04:45:21 Sgeo__: What's #? 04:45:36 |#| = i 04:45:43 Sgeo__: That's just -i@ 04:45:47 what would # mean then? 04:46:04 |n@| = -n 04:46:14 rdococ: Whatever we like 04:46:20 hppavilion[1]: time? 04:46:22 Sgeo__: You mean this thing? https://en.wikipedia.org/wiki/Quaternion 04:46:25 rdococ: It doesn't /have/ to mean something 04:46:36 hppavilion[1], there's an argument for why |n@| = -n is not a good idea 04:46:42 Sgeo__: Why? 04:46:52 And that |n@| = -|n| makes more sense. Something to do with exponentiation 04:46:56 hppavilion[1]: yeah, but if it's part of a system where everything else means something, then it's weird 04:47:02 Sgeo__: Ah, good point 04:47:05 I had the |n@| = -|n| idea 04:47:20 hppavilion[1]: https://www.irccloud.com/pastebin/VqUBVapD/ 04:47:21 before we realized it was about the time dimension 04:47:56 I'm still convinced that $ or # is time and @ or tau is something else 04:48:14 well 04:48:34 zgrep: Ah, fair enough 04:48:41 distance is a measure of how easy it is to get to somewhere 04:48:51 Sgeo__: What're $ and #? 04:49:03 if you need to be there in a larger time, then it's easier because you have more time to spare 04:49:12 Sgeo__: Also, we need to standardize terms instead of saying "x or y" 04:49:13 hppavilion[1], you just tried to define $ which I tended to call # 04:49:19 Sgeo__: Ah 04:49:20 so that's why tau makes distances lower - getting over there in a day is easier than a minute 04:49:41 makes sense? 04:49:45 it's used in other areas 04:49:54 time as a negative dimension isn't a new idea 04:49:59 Sgeo__: How about we give all the esonums Georgian names? 04:50:13 georgian? 04:50:21 rdococ: http://www.alanwood.net/unicode/georgian.html 04:50:29 rdococ: Georgia the country, not the state 04:50:45 georgia is a state? 04:50:46 rdococ: Greek letters are waaaaaaay overused, and Hebrew is reserved for cardinality 04:50:49 rdococ: Yes 04:50:53 rdococ: Also a country 04:50:59 rdococ: Two different places, though 04:51:02 I have an even better idea 04:51:03 letter t 04:51:09 |a + bi + ct| = √(a² + b² - c²) if a² + b² - c² >= 0, otherwise i√(a² + b² - c²) 04:51:27 Ⴓ is @ 04:51:55 But... neoletters doesn't render Goergian. Of course. 04:52:04 my chat does 04:52:12 rdococ: OH! 04:52:13 but I'm sticking with t 04:52:18 What's wrong with @ similar to ai for something similar to i except with absolute 04:52:19 rdococ: Why don't we name them with emoji? 04:52:26 <\oren\> hppavilion[1]: you should update your neoletters 04:52:41 Wait, I doubt neoletters renders emoji either 04:52:42 \oren\: OK 04:52:45 <\oren\> it's supported Georgian for a while 04:52:47 3:-D + 2;-) 04:52:49 \oren\: Linky? I don't remember where it is 04:52:58 rdococ: Those are emo/ticons/ 04:53:04 oh 04:53:10 does this chat even support emojis? 04:53:16 <\oren\> http://orenwatson.be/fontdemo.htm 04:53:20 Oh, found it 04:53:28 rdococ: It supports Unicode, so yes 04:53:53 Some clients might not display it nicely 04:54:20 Sgeo__: Perhaps 04:54:47 -!- hppavilion[1] has quit (Quit: Leaving). 04:54:57 rdococAbs[a_, b_, c_] := Module[{s}, s = a^2 + b^2 - c^2; If[s < 0, I*Sqrt[s], Sqrt[s]]] 04:55:13 -!- hppavilion[1] has joined. 04:55:28 -!- Kaynato has quit (Ping timeout: 252 seconds). 04:55:34 zgrep: yep 04:55:44 Come oooooooon georg- nope 04:55:49 \oren\: I can't get it to install 04:55:53 I download the latest ttf 04:56:09 Open the "THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG" window that Windows gives me (sorry) 04:56:11 Click install 04:56:17 Say yes when it asks to replace the font 04:56:20 Doesn't update 04:56:28 Sgeo__: if $ was time, then what's # or t? 04:56:35 @* 04:56:35 Maybe you meant: v @ ? . 04:56:37 <\oren\> try restarting your caht app 04:56:41 <\oren\> chat app 04:56:42 s/#/@ 04:57:03 rdococ, not really sure. A time dimension that doesn't get weird past the speed of light? 04:57:14 Or it gets weird in a different way from reality 04:57:47 \oren\: I did. That was the login-logout 04:58:05 Sgeo__: c = \omega 04:58:08 |$| = i, but |t| = -1... what is imaginary distance anyway 04:58:08 Sgeo__: Fixed. 04:58:58 <\oren\> try setting your font to another one and back to neoletters? 04:59:14 <\oren\> I do that in my terminal each time I update it 05:00:03 -!- variable has joined. 05:00:17 variable = 26 05:00:26 Would rolling my own IRC client be a bad idea? 05:00:39 <\oren\> no 05:00:57 <\oren\> IRC is a very simple protocol, easy to implement 05:01:06 \oren\: That didn't work 05:01:15 i'd use ii as a foundation 05:02:08 h,, 05:02:14 <\oren\> there are well-known libraries in Perl and Python for IRC 05:02:21 toiiredp 05:02:34 rdococ: SEGMENTATION FAULT 05:02:42 huh? 05:02:52 hppavilion[1]: yeah, IRC is fairly easy to write a client for 05:02:56 look at the sheer number of bots 05:03:04 http://tools.suckless.org/ii/ 05:05:12 t*t = ? 05:05:39 \oren\, can you guess which is the number 3 and which is the russian letter 'eh'? https://usercontent.irccloud-cdn.com/file/YIDcIFry/3eh.png 05:05:52 Without trying it yourself, that is. :P 05:06:49 i'd say the left one is a 3 05:07:51 i am getting sleepy 05:07:58 <\oren\> myname: you are correct 05:11:46 <\oren\> neoletters also has ↋ƐɛɜɝꜾꜿεЄԐԑЗз 05:11:50 should I implement t into Squeak? 05:12:38 what is t*t? 05:13:52 |a + bi + cj + ... + zt| = √(a² + b² + c² + ... - z²) if a² + b² + c² + ... - z² >= 0, otherwise i√(a² + b² + c² + ... - z²) 05:14:02 generalized to n dimensions 05:14:37 but help me discover t*t 05:15:56 -!- Lilly_Goodman has joined. 05:16:48 |1 + 1i + 2t| = isqrt(1 + 1 - 4) = -2 05:17:51 -!- Lyka has joined. 05:17:55 hi 05:18:06 -sqrt(2)* 05:18:16 update on hexadec: made a "reduced" form 05:18:16 <\oren\> hppavilion[1]: make sure in the font menu you select neoletters Regular? 05:18:30 \oren\: I believe I did 05:18:31 http://pastebin.com/49wLRcB3 05:18:43 Normal 05:19:07 brb 05:19:52 <\oren\> hppavilion[1]: well that's a prblem. it should be "Regular" nd not "Normal" 05:20:03 \oren\: There is no "Regular" 05:20:09 \oren\: So I assume Regular = Normal 05:20:27 <\oren\> the earlier versions had "Normal" the newer ones have "Regular" 05:21:33 <\oren\> hmm maybe go to the controlpanel->fonts and find neoletters and delete it? 05:24:23 back 05:24:48 so, um, does the language make any sense now? 05:25:01 <\oren\> hmm I wonder why installing the new version over the old doesn't work when the naming of the subfonts changed? 05:25:15 <\oren\> spooky 05:29:23 is this a bad time to ask about a language i made? 05:29:39 can you help me determine what t*t is? 05:29:40 |a + bi + ct| = √(a² + b² - c²) if a² + b² - c² >= 0, otherwise i√(a² + b² - c²) 05:29:52 t^2? 05:30:08 well, yes, but 05:30:12 what's t^2? 05:30:12 Dang. Perl6 has a bunch of neat things it can do... http://tpm2016.zoffix.com/ 05:30:20 I can't seem to figure it out 05:31:27 also, how is getting to (5, 5) in 6 seconds as easy as getting to 50, 50 in 51 seconds? 05:31:45 |a + bi + ct| = √(a² + b² / c²) if a² + b² / c² >= 0, otherwise i√(a² + b² / c²) 05:32:00 rdococ: Well, what's t? 05:32:19 |5, 5i, 6t| = sqrt(25 + 25 / 36) = sqrt(50 / 36) 05:32:52 but wait 05:33:03 if we do 05:33:03 |a + bi + ct| = √(a² + b² - c²) if a² + b² - c² >= 0, otherwise i√(a² + b² - c²) 05:33:07 (5*sqrt(2)) / (50*sqrt(2)) == 1/10 05:33:22 rdococAbs[5, 5, 6] = Sqrt[14] 05:33:25 6/51 != 1/10 05:33:37 sorry 05:33:45 i'm not good at math 05:34:20 passed calc 1 by sucking up during the last 5 weeks 05:34:41 passed calc 2 by dropping out of college 05:34:47 then |1 + 1i + 2t| = sqrt(1 + 1 - 4) = sqrt(-2) but |2 + 2i + 4t| = sqrt(4 + 4 - 16) = sqrt(...oh 05:34:55 wait 05:35:06 -8? 05:35:06 rdococAbs[1, 1, 2] = -Sqrt[2] 05:35:22 ik 05:35:40 should it be - c^2 or / c^2? 05:35:57 rdococAbs is written with - c^2... division? 05:36:09 |a + bi + ct| = √(a² + b² / c²) 05:36:24 getting there in 5 seconds is twice as hard as 10 seconds, same as if you double the distance 05:36:25 Just b/c? 05:36:32 no 05:36:34 Or (a^2+b^2)/(c^2) 05:36:38 yes 05:36:57 -!- Lilly_Goodman has quit (Read error: Connection reset by peer). 05:37:03 I think that will work better 05:37:14 rdococAbs2[1, 2, I] = -Sqrt[5] :P 05:37:43 well, Abs2[1, 1, 2] should be 1 + 1 / 4, or 1/2 05:37:51 well 05:38:12 |1 + 1i + 2t| = sqrt(1 + 1 / 4) = sqrt(1/2) 05:38:36 |2 + 2i + 4t| = sqrt(4 + 4 / 16) = sqrt(1/2) 05:38:41 yay 05:38:44 it works as intended 05:39:48 1/Sqrt[2] 05:39:53 Abs2[1,1,2] ^ 05:40:06 yay 05:40:20 what software are you using? Mathematica? 05:40:22 Yep. 05:40:25 is it free? 05:40:30 Unfortunately, no. 05:40:35 how much does it cost? 05:40:40 Too much. 05:40:51 really? 05:40:55 I get it free, courtesy of school. 05:41:06 I ask how much money something will cost and you give me the vaguest answer. 05:41:11 -.- 05:41:28 http://www.wolfram.com/mathematica/pricing/ 05:41:35 okay 05:41:36 ty 05:41:50 See, pricy. 05:41:53 s/cy/cey/ :P 05:42:18 Though technically, your function is easily written in K... 05:42:36 hmm 05:42:43 or any other language really 05:42:58 just need to figure out t*t 05:43:34 well, multiplying by a unit like 1, i or t should keep the magnitude the same 05:43:42 so |x| = |x*t| 05:43:53 no wait 05:43:59 wait yeah 05:44:00 wait 05:45:43 great.... |x|-rated conversation 05:45:58 Heheh... 05:46:46 well 05:46:53 a space angle is rotation 05:46:58 a space time angle is velocity 05:46:58 rdococ: Mathematica can *sometimes* be fed into Mathics, and perhaps even into WolframAlpha sometimes. 05:48:01 Mathics is this thing: https://mathics.angusgriffith.com/ 05:48:53 K? 05:49:08 Lyka: kparc.com/k.txt 05:49:26 Though I don't know if it'd work with imaginary numbers. It wasn't exactly designed for those, I don't think. 05:49:33 https://en.wikipedia.org/wiki/K_%28programming_language%29 05:51:22 hm 05:53:40 |x| = infinity 05:53:49 |x + t| = x 05:53:58 1* 05:54:03 x* 05:54:55 solve for x where x^0 != 1 05:59:24 * Lyka imagines sheep with numbers on them 05:59:30 night 05:59:40 -!- Lyka has left. 06:10:14 -!- lambda-11235 has quit (Quit: Bye). 06:14:10 hhbhb so t ired## 06:16:42 Internet is full of April's fools jokes now. http://www.questionablecontent.net/ has one. 06:25:20 -!- evalj has joined. 06:32:16 -!- hppavilion[2] has joined. 06:35:29 -!- impomatic_ has joined. 06:35:43 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 06:35:45 -!- dos has joined. 06:39:25 -!- hppavilion[2] has quit (Ping timeout: 252 seconds). 06:49:03 -!- variable has quit (Quit: 1 found in /dev/zero). 06:50:07 In underload, is it possible to store data in the source code at four bits per source code character density such that the program can decode it unambiguously? 07:08:05 I think it's possible. I'll have to try to make a proof. 07:09:14 Someone should write a Hello, World with 1000000% cruft, MVC, etc. 07:09:21 -!- dos has changed nick to hppavilion[1]. 07:09:48 AbstractPrinterFactory() 07:10:42 [wiki] [[Evil]] https://esolangs.org/w/index.php?diff=46711&oldid=46710 * Kc kennylau * (+47) /* 0 to 255 using only a, e, u, z (To be completed) */ 07:11:49 Today's GG is probably very good for certain shippers 07:13:12 Could we atomize MVC even further? 07:13:18 MVCQDNX? 07:13:33 To make something even more atrocious 07:13:44 hppavilion[1]: I'm absolutely certain that's been done... http://www.ariel.com.au/jokes/The_Evolution_of_a_Programmer.html 07:16:03 zgrep: Perhaps we should make the world's most absurdly bloated Beginning Projects Repo 07:16:20 With officially-named and trademarked programs 07:16:26 hppavilion[1]: https://github.com/fwilson42/SimpleJavaEasyNumber 07:16:32 Industrial strenght, of course 07:18:00 -!- evalj has quit (Remote host closed the connection). 07:30:34 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 08:07:09 https://www.youtube.com/snoopavision?v=MU39xSNukfg 08:24:50 -!- oerjan has joined. 09:08:25 -!- heroux has quit (Ping timeout: 250 seconds). 09:09:21 -!- heroux has joined. 09:20:35 -!- Deepfriedice has joined. 09:30:09 -!- PinealGlandOptic has joined. 09:31:38 > ord '=' 09:31:40 61 09:31:55 @messages- 09:31:55 boily asked 8h 39m 14s ago: do you feel matched? 09:32:04 @tell boily No. 09:32:04 Consider it noted. 09:34:10 -!- J_Arcane_ has joined. 09:36:33 -!- J_Arcane has quit (Ping timeout: 240 seconds). 09:36:34 -!- J_Arcane_ has changed nick to J_Arcane. 09:37:16 * oerjan realizes he just did something he's been annoyed at others doing. 09:37:27 @tell boily RE: matching. 09:37:27 Consider it noted. 09:40:03 oerjan: You're unmatched? 09:40:12 :P 09:40:29 in so many ways 09:43:19 Heheh. 09:43:57 That's both happy and sad. :| 09:44:41 yep 10:06:38 -!- oerjan has quit (Quit: Later). 10:19:56 I guess unmatched is better than mismatched (maybe( 10:20:07 Hm, that should probably have been (maybe] 10:31:01 Are there any vim users here? Vim has a jump list where it tracks big moves, but that's not really what I need. Is there an edit location list where I can find the places I've edited in the file previously? 10:31:26 FireFly: Agh! [))) 10:32:22 b_jonas: well, it has an undo tree that you can query for some info, maybe it includes position 10:32:28 I wouldn't know how to use it though 10:33:14 There's a plugin called gundo that visualizes the undo tree and allows you to jump around in it 10:34:09 I'd like commands like prevpos and nextpos in joe-editor, which let me quickly jump to places of previous edits. It's very convenient but few editors seem to have it. 10:34:26 Lets you avoid setting bookmarks a lot. 10:34:44 That sounds useful, yeah 10:35:37 http://stackoverflow.com/a/2131407/1267058 oh. 10:35:44 -!- boily has joined. 10:37:15 FireFly: ah, that says g, might do that 10:37:49 It seems to work, toyed around with it a bit 10:37:50 g ; for previous and g , for next. strange assignments 10:38:22 but they make sense 10:38:25 Probably related to the ; , commands that repeat f/F/t/T forwward/backward 10:38:44 But vim's assignments can be pretty weird and arbitrary sometimes 10:39:04 You could always remap them to something else prefixed by if you prefer 10:49:26 g? is the best g. 10:51:54 @massages-loud 10:51:54 oerjan said 1h 19m 50s ago: No. 10:51:54 oerjan said 1h 14m 27s ago: RE: matching. 10:54:06 -!- tromp has quit (Remote host closed the connection). 11:04:03 -!- J_Arcane has quit (Ping timeout: 268 seconds). 11:08:59 -!- J_Arcane has joined. 11:23:16 -!- boily has quit (Quit: HELICAL CHICKEN). 11:59:26 -!- tromp has joined. 12:03:39 -!- tromp has quit (Ping timeout: 246 seconds). 12:27:08 -!- Sgeo__ has quit (Ping timeout: 244 seconds). 12:29:32 -!- Deepfriedice has quit (Quit: Leaving). 12:39:16 -!- Reece` has joined. 12:39:23 -!- PinealGlandOptic has quit (Quit: leaving). 12:41:02 -!- shikhin has quit (Ping timeout: 248 seconds). 12:43:54 -!- shikhin has joined. 12:49:57 -!- shikhin has quit (Ping timeout: 276 seconds). 12:50:23 -!- hydraz has quit (Ping timeout: 244 seconds). 12:51:29 -!- rdococ has quit (Ping timeout: 260 seconds). 12:51:47 -!- shikhin has joined. 12:51:55 -!- hydraz has joined. 12:51:56 -!- hydraz has quit (Changing host). 12:51:56 -!- hydraz has joined. 12:53:54 -!- tromp has joined. 13:00:46 -!- shikhin has quit (Ping timeout: 248 seconds). 13:03:16 -!- shikhin has joined. 13:05:19 -!- rdococ has joined. 13:07:04 -!- tromp has quit (Remote host closed the connection). 13:21:47 -!- bender| has quit (Changing host). 13:21:47 -!- bender| has joined. 13:41:59 -!- spiette has joined. 13:46:36 -!- p34k has joined. 13:53:49 -!- impomatic_ has quit (Quit: http://corewar.co.uk). 13:54:04 -!- Frooxius has quit (Quit: *bubbles away*). 13:54:28 -!- Frooxius has joined. 14:18:12 -!- bender| has quit (Quit: Leaving). 14:35:45 -!- asie has quit (Quit: WeeChat 1.4). 14:38:07 -!- tromp has joined. 14:43:04 -!- tromp has quit (Ping timeout: 264 seconds). 14:49:09 -!- Kaynato has joined. 15:05:35 -!- lambda-11235 has joined. 15:11:33 -!- Kaynato has quit (Ping timeout: 240 seconds). 15:24:16 -!- Kaynato has joined. 15:32:42 |a + ct| = √(a² / c²) 15:32:47 so what is t*t? 15:32:56 well, 15:33:35 |1 + tt| = sqrt(1 / t^2) 15:37:52 -!- Kaynato has quit (Ping timeout: 252 seconds). 15:44:07 did you resolve ©? 15:45:37 sgeo told us that @ was inspired by time dimension and so we come up with |a + bt| = √(a² / b²) = |a|/|b| --- the last part was someone else's idea from another channel 15:48:38 so |tt| = |0|/|t| 15:50:13 you going to help? 15:55:47 -!- mihow has joined. 16:28:15 -!- XorSwap has joined. 16:28:43 -!- Reece` has quit (Read error: Connection reset by peer). 16:31:28 -!- XorSwap has quit (Client Quit). 16:32:31 -!- Kaynato has joined. 16:32:40 http://senseis.xmp.net/?TerritoryScoringOnGoServersConsideredHarmful oh em gee 16:32:47 considered harmful essays everywhere 16:36:36 |a + bt| = √(a² / b²) = |a|/|b| 16:37:29 -!- hppavilion[1] has joined. 16:38:50 hppavilion[1]: help me with this- |a + bt| = √(a² / b²) = |a|/|b| 16:38:56 I want to find out t^2 16:39:06 rdococ: OK 16:39:23 rdococ: You don't have ANY constants 16:39:40 rdococ: Try defining a and b to appropriate numbers and working from there 16:39:50 no, t is like i but different 16:39:59 rdococ: |t|=-1, right? 16:40:24 |t| = sqrt(0 / 1) = |0|/|1| = 0 now 16:40:44 and |a + 0t| = sqrt(a^2 / 0) = sqrt(infinity) 16:40:52 rdococ: If |t|=0, then doesn't that just make t=0? 16:41:04 not really 16:41:04 rdococ: Or t=k 16:41:12 |a + 0| =/= |a + 0t| 16:41:13 (k^2=0, k!=0) 16:41:17 Oh 16:41:28 no, wait 16:41:46 |a + 0| = |a + 0t| 16:41:48 but 16:41:51 no 16:41:53 ye 16:42:09 |0| = sqrt(0/0) = ??? 16:42:12 -!- oerjan has joined. 16:42:16 |t| = sqrt(0/1) = 0 16:42:19 "And now there is non-binary sex. I guess that would be "analog sex"..." 16:42:25 in my system |0| =/= 0 16:42:37 s/"analog sex"/\\"analog sex\\"/ 16:42:56 so... |2t| = 0, but |t + t| = -1 16:43:01 (sex is gender in this case, not the act between bored teenagers) 16:43:15 |t + t| = sqrt(0 / 2^2) = 0 16:43:23 2t = t+t 16:43:26 rdococ: How does |0| != 0? What's |0|? 16:43:31 |a + bt| = √(a² / b²) = |a|/|b| 16:43:46 |0| = sqrt(0 / 0) 16:43:56 in t+t, a is t and b is 1, in 2t, a is 0 and b is 2 16:44:01 ignore the |a|/|b| 16:44:41 |t + t| = sqrt(t^2 / 1) = t... 16:45:01 rdococ: So |0| = r for all r in R 16:45:03 |2t| = sqrt(0 / 4) = |t + t| = sqrt(t^2 / 1) 16:45:08 hppavilion[1]: yes 16:45:11 rdococ: 0/0 can be said to equal any R 16:45:31 sqrt(0 / 4) = sqrt(t^2) = t...??????? what 16:45:41 0 = t 16:45:49 wth 16:46:07 my original definition was |a + bt| = √(a² - b²) 16:47:24 but that would make it as easy to go 10000 miles in 10001 seconds than to go 1 miles in 2 seconds, whereas the second one is much harder 16:47:33 so I changed it to |a + bt| = √(a² / b²) 16:48:34 rdococ: One of the benefits of @ that t seems to lack is that @ is logical and straightforward; it just has one little gotcha that makes it different, but that can be explained in a single equation with exactly 1 operation (assuming -1 is syntactically a number, not "apply negation to 1") 16:48:48 |@| = -1? 16:48:50 @ can be explained as |@| = -1 16:48:52 rdococ: Yes 16:48:57 so 16:49:02 rdococ: You are yet to clearly explain t to me. 16:49:13 rdococ: And you have to establish all these other rules in the process 16:49:19 i has the same property: i^2 = -1 16:49:30 we recently found out that sgeo's latest definition for @ is |a + b@| = √(a² - b²) 16:49:44 There are other subtleties, but the straightforward part of i is just i=sqrt(-1) 16:49:46 rdococ: Really? 16:49:48 rdococ: OK... 16:49:51 apparently 16:49:53 I think 16:49:54 idk 16:49:58 but |@| = -1 16:50:02 rdococ: That's actually pretty interesting 16:50:17 |bt| = √(0² / b²) = 0 16:50:21 rdococ: And much moer useful than the original 16:50:41 but it's not exactly the same as t 16:50:44 rdococ: But why the division? 16:50:56 rdococ: You have to explain that too 16:51:07 hppavilion[1]: it should be equal difficulty to go 100 miles in 200 seconds as it is to go 10 miles in 20 seconds, t does this but @ doesn't 16:51:26 rdococ: Beautiful esonums can be explained as a single operation on them equaling a number that operation usually cannot produce 16:51:42 |10 + 20t| = |100 + 200t| but |10 + 20@| =/= |100 + 200@| 16:51:54 hmm 16:52:05 rdococ: Good point, I guess 16:52:13 that's why I changed it to / 16:53:02 another idea I had was e^fx = sinh(x) + rcosh(x) where the esonum is f 16:53:11 s/rcosh/fcosh/ 16:53:21 hyperbolic sine and cosine 16:53:32 so it's a hyperbolic i 16:53:38 should I change the definition to that? 16:54:11 <\oren\> we REALLY need to make templates compile faster 16:54:14 @1 16:54:14 Say again? 16:54:19 @1 What 16:54:20 W|-|4T 16:54:30 huh 16:54:32 oh 16:54:41 @1337 I see 16:54:42 i seE 16:55:19 <\oren\> @1337 i am a leet haxor 16:55:19 I 4M A leE+ Hax0r 16:55:36 <\oren\> not leet enough 16:56:32 e ͭ ͯ = sinh(x) + tcosh(x) 16:56:40 like this new idea? 16:58:56 hppavilion[1] 17:00:20 rdococ: Yes? 17:00:24 rdococ: Sorry, I vanished 17:00:29 what about this instead - e ͭ ͯ = sinh(x) + tcosh(x) 17:00:45 rdococ: The two characters after e aren't rendering 17:01:00 oh 17:01:08 rdococ: Ah, it's e^tx 17:01:21 how'd you figure it out? 17:01:38 rdococ: My browser still renders them, and I have the godlike powers of COPY AND PASTE 17:02:03 e^tx = sinh(x) + tcosh(x) 17:02:12 anyways 17:02:19 do you think the idea is better than my other one? 17:02:46 rdococ: It might be, I don't know 17:02:51 i'm not sure there isn't already a t which matches there 17:03:08 oerjan: Does e^jpi = 1? 17:03:09 but the t is an esonum 17:03:12 <\oren\> hppavilion ypu really neeed an update 17:03:20 you mean e^tpi? 17:03:23 \oren\: I tried 17:03:27 hppavilion[1]: what's j 17:03:31 rdococ: No 17:03:35 Where j^2 = 1, j != 1 17:03:40 oh 17:03:47 oerjan: I was typing it, but people kept talking :P 17:04:06 <\oren\> try deleting the font in the vontrl$panel and reintaling it 17:04:31 <\oren\> sorry fpr bad spelibg[im on my phone 17:04:46 hppavilion[1]: try plugging it into the series for e^x, that's my goto-definition for whether that makes sense. 17:05:05 (of course then you have to define the limits too) 17:05:19 -!- vanila has joined. 17:05:20 hello 17:05:35 Need to close hc to delete it 17:05:37 -!- hppavilion[1] has quit (Quit: Leaving). 17:05:43 > [(exp x, sinh x + cosh x) | x <- [0, 0.1 ..]] 17:05:44 [(1.0,1.0),(1.1051709180756477,1.1051709180756475),(1.2214027581601699,1.221... 17:05:50 thought so 17:05:57 rdococ: t = 1 hth 17:06:37 e^1x = sinh(x) + cosh(x)? 17:06:43 -!- hppavilion[1] has joined. 17:06:44 <\oren\> the build failed again 17:07:00 \oren\: I deleted neoletters 17:07:07 \oren\: Had to close HC and NP++ 17:07:16 @_@ 17:07:22 \oren\: It didn't work, at least for ͭ ͯ 17:07:29 (And yes, I reinstalled) 17:08:12 e^tx = sinh(x) + tcosh(x), t =/= 1 17:08:28 \oren\: Also, the name changed from "normal" to "medium" 17:09:05 <\oren\> hppavilion[1]: good it shoukd at least work with georgian noe 17:09:13 rdococ: the hyperbolic functions are pretty much what you get when you take the expressions for the trigonometric ones in terms of exp and remove i's everywhere 17:09:32 <\oren\> ill add those superscripts[when i get home ftom work 17:09:36 \oren\: Oh, it does :) 17:09:37 > [(exp (-x), sinh x - cosh x) | x <- [0, 0.1 ..]] -- testing another 17:09:38 [(1.0,-1.0),(0.9048374180359595,-0.9048374180359595),(0.8187307530779818,-0.... 17:09:43 \oren\: Yay! It actually did update! 17:09:47 well 17:09:54 how does time rotation work 17:09:59 hm needs switching order, i think 17:10:20 (In retrospect, the reason it wouldn't update the font was probably because I had applications with it open 17:12:29 In other news, the new s is bugging me :P 17:12:38 @check \x -> exp x == sinh x + cosh (x :: Double) -- wondering if ghc uses this exactly 17:12:39 :1:103: 17:12:39 parse error (possibly incorrect indentation or mismatched brackets) 17:12:42 darn 17:12:55 <\oren\> nah i did a tesy it is a bug in windows where it won't ovrwrite a otf with a ttf 17:13:28 @check \x -> exp x == sinh x + cosh (x :: Double) 17:13:30 *** Failed! Falsifiable (after 3 tests and 4 shrinks): 17:13:30 2.4646339992384982 17:13:37 hm apparently not 17:13:37 <\oren\> for some goddamn reason 17:13:52 what about this instead? t = i^2, t =/= -1 17:15:11 rdococ: that violates basic equality laws, which are logic not arithmetic. 17:15:35 I don't see that stopping any other people from doing stuff like it 17:15:45 thataway lies NaN and other madness. 17:15:52 like I saw an esonum where q^2 = 1 but q =/= 1 17:16:05 rdococ: That's different 17:16:10 rdococ: And it's called "j" 17:16:25 rdococ: that doesn't violate equality. you can have as many solutions to q^2 = 1 as you want in a general algebra. 17:16:39 rdococ: It's one of the three 2d real algebras (or something) 17:16:45 of course it won't be a field if there's more than two. 17:16:50 rdococ: The other two are the complexes and k, which is like j but for 0 17:16:53 ?????? 17:17:22 oerjan: so why can't I do sqrt(t) = i then and t =/= -1? 17:17:27 rdococ: but t = i^2 and i^2 = -1 imply t = -1 from pure logic. 17:17:54 q^2 = 1 and sqrt(1) = q 17:18:09 mine isn't any different 17:18:15 ?????????????????????????????????? 17:18:50 or ti = -i, but t =/= -1 17:19:28 rdococ: because sqrt a = x, is not the unique solution to x^2 = a, but if you don't have (sqrt a)^2 = a then it's not a square root. 17:19:54 hw about this 17:19:59 start with the ring Z 17:20:07 then take the 'algebraic esoclosure' 17:20:13 okay, I'm using the ti = -i definition along with t =/= -1 17:20:20 which is where you add infinityl many distinct solutions to every algebraic equation 17:20:26 rdococ: the point is, to get i^2 to be t you have to redefine i^2 which already exists. 17:20:52 oh... 17:20:55 uh... 17:21:08 What do you think? 17:21:11 but there's nothing stopping me from breaking that rule and giving it two solutions 17:21:16 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 17:21:20 but anyway 17:21:45 ti = -i but t =/= -1... 17:21:56 ... 17:22:00 ah nevermind 17:22:50 rdococ: ok, just define 2+2 = t while you're at it. 17:23:12 yes good idea 17:23:15 2 + 2 = t 17:23:17 t - 2 = 2 17:23:19 t / 2 = 2 17:23:22 hi 17:23:23 t * 2 = 2 17:23:27 t / 5 = 2 17:24:07 vanila: way ahead of you, it seems. 17:24:13 ? 17:24:20 or perhaps crashing high speed into the same point. 17:25:07 (as usual, the real challenge is to end up with anything actually interesting.) 17:25:40 rather than a chaotic mess where everything is equally true. 17:26:00 0=1 17:27:12 make it 1=2 and we can summon pope russell 17:27:19 sin(q) = 2 17:27:20 -!- hppavilion[1] has joined. 17:27:32 stop vanishing 17:27:40 sin(q) = 2 17:28:23 sin(e) q(ua) non 17:28:43 hmm 17:29:21 here it says that in the Minkowski metric, time is imaginary 17:29:39 |x + (y + ti)i| 17:30:16 |x + (y + ti)i| = |x + yi - t|... 17:31:32 sine qua omnes est 17:32:06 im no good at latin 17:33:47 * oerjan summons john cleese to teach quintopia 17:34:21 WITHOUT WHICH THEY ALL IT IS? 17:34:55 actually the IT is redundant 17:34:57 DANCE THE WAX TADPOLE 17:35:10 hppavilion[1]: *BITE 17:35:26 oerjan: THAT's what it is 17:36:46 `addquote bite the wax tadpole 17:37:20 1274) bite the wax tadpole 17:43:51 -!- Kaynato has quit (Ping timeout: 244 seconds). 17:46:40 -!- ais523 has joined. 17:47:01 -!- Kaynato has joined. 17:56:13 "female horse fastened with wax" 17:57:08 i probably meant sine qua omnia est 17:57:28 quintopia: still using a singular verb, there 17:57:57 you just changed the gender of omnia 17:57:57 no, it's fastened with a staple 17:58:10 i think omnia is singular 17:58:15 it is in english 17:58:27 wait 17:58:27 quintopia: no, it is neuter plural nominative 17:58:32 hmm 17:58:37 omnia is plural in nominative, accusative, and vocative 17:58:39 fine have your sunt 17:59:05 https://en.wiktionary.org/wiki/omnis#Declension 17:59:11 -!- Caesura has joined. 17:59:17 but i dont really want omnia thrn 17:59:28 `? thrn 17:59:28 What are you trying to say? 17:59:37 i want the equivalent of english "everything" 17:59:39 thrn? ¯\(°​_o)/¯ 17:59:52 quintopia: omne, then. 17:59:58 totum? 18:00:01 or that. 18:00:07 totum est? 18:00:16 What's the sentence supposed to be? 18:00:30 without which everything is 18:02:33 -!- Kaynato has quit (Ping timeout: 240 seconds). 18:02:35 Doesn't sine take the ablative? 18:03:05 probs bobs 18:03:11 isn't qua ablative 18:03:24 most like 18:03:50 `? latin 18:04:03 latin? ¯\(°​_o)/¯ 18:04:12 Oh, duh. There are like 3 quis in Latin, and I mixed them up 18:04:55 `le/rn LATINA EST SUBLIMISSIMA LINGUA MUNDI 18:04:56 No output. 18:05:00 oops 18:05:04 `learn LATINA EST SUBLIMISSIMA LINGUA MUNDI 18:05:07 Learned 'latina': LATINA EST SUBLIMISSIMA LINGUA MUNDI 18:05:10 hm 18:05:10 le/rn: print better output twh 18:05:24 `` mv wisdom/latin{a,} 18:05:26 No output. 18:06:38 I think sine qua omne est is probably the closest. 18:06:39 le/rn is taking over 18:06:46 even your beloved learn is not immune 18:06:53 ;_; 18:07:35 In Latin, learning is disco, baby 18:10:01 `? le/rn 18:10:03 le/rn makes creating wisdom entries manually a thing of the past. 18:10:04 `` ln -s bin/slashlearn bin/disce 18:10:09 No output. 18:10:23 That wisdom entry is outdated now. 18:10:24 In underload, is it possible to store data in the source code at four bits per source code character density such that the program can decode it unambiguously? <-- of course not, there are only 9 command characters and no way to decode non-command ones. also some combinations are nops or infinite loops so you have to avoid them. 18:10:34 `disce Learning is disco, baby. 18:10:34 ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: disce: not found 18:10:41 Oh, boo 18:10:59 you can store 1 bit per character, though, as i did in the rule 110 program. 18:11:50 Daoyu commands fill the 4-bit space 18:12:07 maybe you can find a code that gives 2, i'd be surprised if 3 were possible (you'd need to be able to lose only one character possibility on average) 18:12:09 daoyu makes little sense 18:12:24 Eh, how so? 18:12:37 If the specification is unclear, please tell me so I can improve it 18:12:50 the diagrams communicate nothing 18:12:58 Ah, I'll fix that right now 18:13:05 `disce Learning is disco, baby. 18:13:05 They were a bit of a rush job 18:13:06 No output. 18:13:12 `? learning 18:13:14 learning? ¯\(°​_o)/¯ 18:14:14 `? learning 18:14:15 Learning is disco, baby. 18:14:17 There 18:14:36 prooftechnique: ooh you meant that disco literally 18:14:59 -!- oerjan has set topic: Aut disce aut discede | The international hub of esoteric programming language and kitten typesetting | logs: http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | http://esolangs.org/ | Note: people with cloaks will be treated as if they're from Budapest. 18:15:07 I guess really the gerund is discere, but whatever hth 18:16:26 I really like that topic :) 18:16:57 is a real old latin verb? 18:17:05 "to disc"? 18:17:20 Most Latin verbs are old 18:17:31 disco, discere, dedici, no PPP 18:17:35 To teach 18:17:49 There is no gerund 18:17:52 prooftechnique except the new ones 18:17:57 Caesura: To learn 18:18:03 gah 18:18:17 doceo 18:18:20 got that mixed up 18:18:20 Somewhat wonderfully, dedisco is to unlearn 18:18:34 Doceo, docere, docui, doctus 18:18:38 That's to teach 18:18:45 Hence docent :) 18:18:46 doctus 18:19:03 3rd person plural indicative active present 18:19:12 prooftechnique: i think "discere" is the infinitive, but wiktionary's table lists it as nominative gerund which it sort of logically is 18:19:13 Hence doctor, one who teaches 18:19:36 -!- mihow has quit (Quit: mihow). 18:19:41 Infinitives as gerunds can be done... but there isn't anything like "discendus" 18:20:11 Caesura: because the real gerund is not used in the nominative... 18:20:32 This is where I realize that I have spent the last two years unlearning the latin from the three years prior 18:20:44 I think they call it a degenerate form or something? My Latin grammar is a little hazy 18:21:00 m=m*2 18:21:11 I think that's right 18:21:59 I wish I knew more Ancient Greek, but I don't think any time I spent learning Latin was wasted 18:22:28 -!- Reece` has joined. 18:22:43 I learned a bit of Attic Greek under the notion that it would be easier 18:22:49 It is not. It is not easier than Latin. 18:22:59 Caesura: funny, virtually all my latin grammar i learned ~ 30 years ago from a grammar i found in the town library. and i still remember much of it. 18:23:34 Aeorist prefixing was such a pain with greek 18:23:35 -!- Reece` has quit (Client Quit). 18:23:48 I won awards in state and national competitions on Latin 18:24:10 I would really like to know what it is in the human brain that seems to lead to the mutation of "to be" and "to go" in so many languages. 18:24:11 wow i wouldnt have guessed you were as old as that implies. 18:24:25 oerjan: Is 31 and advanced 18:24:32 Such a strong child 18:24:34 I should try and keep Latin fresh in my memory with a bit more effort, hm 18:24:41 prooftechnique: 45 18:25:03 i learned it in two year-long classes 15 years ago and remember little besides vocabulary and 1st and 2nd declension suffices 18:25:40 I think I spent my fifth year of latin unlearning it, the more I think about that 18:25:55 I mainly just use it for passphrases, now. 18:25:59 The gerundive / gerund distinction never really sunk into my memory, I think :x 18:26:01 Free word order is pretty nice. 18:26:14 Gerundive is the non-nominative, the noun of "to verb," right 18:26:31 No, isn't that the gerund 18:26:40 Ah, self-demonstration 18:26:56 Gerund is -ing, gerundive is the adjectival form 18:28:04 disco, discere, dedici, no PPP <-- hm are you saying wiktionary's table is erroneous, then 18:28:20 Though Romans were goofs and sometimes used the gerundive in place of the gerund for euphony 18:28:25 This kills the translator 18:28:27 any april 1 stuff worth checking out this year? 18:29:19 Personally, I can't wait to see what my talky box from Amazon has cooked up while I've been at work. It has the power to call my phone, so I expect prank calls. 18:30:31 oerjan: PPP is the past participle, which disco does not have. 18:30:47 ais523, you can telnet read wikipedia 18:31:04 and this https://tools.ietf.org/html/rfc7511 18:31:04 Though it does have a passive one, so... 18:31:10 fungot: smile! 18:31:11 int-e: what is that? 18:31:26 fungot: What do you like on pizza? 18:31:26 prooftechnique: i have a drug overdose? wah, that spoils it even more simply in our new language 18:31:37 vanila: telnet wikipedia actually seems potentially useful 18:31:41 I hope they keep it around 18:31:44 Yah 18:31:48 ais523, there is already a gopher wikipedia 18:31:50 I mean, you could telnet to port 80 but it's annoying 18:31:57 True. 18:32:41 ais523: connection refused :( 18:32:48 huh, RFC MMI syntax has a MIGHT? 18:32:54 Agora uses MAY for that 18:34:15 someone posted a patch for a leftpad(2) syscall to the Linux kernel 18:34:25 that was at least mildly amusing 18:34:25 The wikipedia table lists the supine... hmmm 18:34:34 the kernel devs are pretending to take it seriously 18:35:09 We were taught with the fourth principal part as the PPP 18:35:46 ais523: ouch... 18:35:53 leftpad system call 18:36:21 did they also post a libc wrapper with userspace emulation as fallback to the gnu libc mailing lists? 18:36:32 come to think of it they shuld probably implement it in the vDSO 18:36:36 b_jonas: not that I saw 18:39:25 -!- lleu has joined. 18:39:43 -!- ais523 has quit. 18:39:52 -!- ais523 has joined. 18:42:49 -!- Phantom_Hoover has joined. 18:45:34 -!- hppavilion[1] has quit (Ping timeout: 240 seconds). 18:46:34 -!- lambda-11235 has quit (Quit: Bye). 18:49:49 hmm 18:50:23 what about an esoteric programming language that's turing complete because of its esonums 18:50:41 what's an esonum? 18:51:04 42 18:51:09 a number defined to behave like a number, but slightly different 18:51:14 Seriously. 18:51:22 Marvin Minsky ♥ 18:51:28 rdococ, that's a really cool idea i like it 18:51:33 e.g. an esonum called @ could do this - |@| = -1 18:51:47 rdococ, did you hear my idea about alg. esoclosure 18:51:53 don't think it was my idea 18:52:44 an esolang with addition, subtraction, multiplication, division, exponentiation and i 18:52:52 3+2i^5i 18:52:57 e^i 18:53:08 #^@ 18:53:15 p sure you can have a TC esolang just with reals and equality... 18:53:26 what about a programing lanugage based on counter examples to tarskis high school algebra problem? 18:53:31 the number system 18:53:50 what about 18:53:52 those are surely esonumbers 18:54:04 a programming language that doesn't need to evaluate things to manipulate them 18:54:14 you mean haskell? 18:54:27 it would be able to detect that 3+2 = 4+1 without evaluating either for example 18:54:55 a better example is 3+2+x = 5+x 18:55:12 it would be able to do that without evaluating x 18:56:02 how do you tell which bits to evaluate 18:56:40 super lazy evaluation - if there's any way round evaluating something, do that way 18:57:00 3+2+x = 5+x for example - you might need to evaluate 3+2, but not x 18:57:06 [wiki] [[Daoyu]] https://esolangs.org/w/index.php?diff=46712&oldid=46696 * Kaynato * (+1059) Clarifications (replaced diagrams) 18:57:53 e^sqrt(-1)x would equal sin(x) + sqrt(-1)cos(x) without evaluating sqrt(-1) 18:58:03 -!- hppavilion[1] has joined. 18:58:06 say |@| = -1, no error there either 18:58:56 -!- zzo38 has joined. 18:59:30 [wiki] [[Special:Log/upload]] overwrite * Kaynato * uploaded a new version of "[[File:Daoyu Symbol Table.png]]": Irrelevant information removed 18:59:39 hppavilion[1]: we were just discussing super lazy evaluation - an idea where a programming language would try its absolute hardest not to evaluate something, it's almost not lazy 18:59:55 rdococ: Yes, I've heard of that 19:00:08 how would it know what not to evaluate? 19:00:11 e.g. in such a language, a computer would detect that e^sqrt(-1)x = sin(x) + sqrt(-1)cos(x) without ever knowing that sqrt(-1) = i 19:00:14 rdococ: Ultimate Lazy Evaluation just prints out the program and tells you to "do it your fucking self" 19:00:22 quintopia: Is this better? 19:00:35 vanila: it wouldn't evaluate anything, just manipulate, simplify and stuff 19:01:21 (1/0)/0 = 1, and it wouldn't even know that 1/0 evaluates to an error 19:01:25 -!- jaboja has joined. 19:01:38 You're talking about symbolic computation? 19:01:38 s/(1/0)/0/1/(1/0)/ 19:01:40 yes 19:01:49 Pretty sure you can set that up with Wolfram 19:02:07 yes but it detects stuff like 1/0 19:02:19 mine would be too lazy to realize that 1/0 evaluates to an error 19:02:25 <\oren\> huh theres a vuln in emacs this time? 19:02:27 With some redefinitions you coul do that also 19:03:09 rdococ, so i guess you just try every evaluation order possible until you get a normalised expression? 19:03:29 perhaps...? 19:03:36 maybe it could use substitution 19:03:42 e.g. 1/1/x = x 19:03:59 or (x^y)^z = x^yz 19:04:01 Wolfram evaluates 1/(1/0) to be 0, it seems 19:04:24 Although it also WARNS you that 1/0 is encountered, it doesn't break the entire thing 19:04:53 oh, right, 1/1/0 = 0 19:04:59 wolfram is right 19:05:12 It is very powerful for symbolic computation 19:05:12 rdococ, well i mean at this stage it's not actually true 19:05:13 but it's eager enough to warn you 19:05:18 Yes 19:05:27 e.g. (x^y)^z is not true for general real exponents 19:05:56 it's not? 19:06:19 (x^y)^z = x^(y+z)? 19:06:21 ((-1)^2)^0.5 is not equal to (-1)^1 19:06:23 I can't remember 19:06:37 you need to add them, right? 19:06:55 No, that's for multiplying exponents of equal bases 19:06:58 \oren\: What vuln? 19:07:02 Also check the date, just in case 19:07:07 huh... 19:07:15 the issue with doing e.g. the 1/0 trick you mentioned is that if you specify all these substitution rules you can evaluate 1/0 to get contradictory values 19:07:16 then what is (x^y)^z? 19:07:29 @PH Yes, thus indeterminate forms 19:07:29 Maybe you meant: wn v rc pl id do bf @ ? . 19:07:32 it's (x^y)^z, that's all you can say 19:07:38 oh 19:07:41 okay 19:08:49 infact, why not make the numbers mathematical symbols? 19:08:59 <\oren\> prooftechnique: i just got a work email saying dont use emacs 19:09:22 <\oren\> i use nano anyway 19:11:14 who knew sqrt(-1) could be so useful 19:12:10 You have heard of Cayley-Dickson construction? 19:12:19 https://en.wikipedia.org/wiki/Cayley%E2%80%93Dickson_construction 19:12:41 no i havent 19:12:46 would you recommend it? 19:12:49 yeah but after a certai point the results of that can't be called 'numbers' 19:13:04 https://en.wikipedia.org/wiki/Tarski%27s_high_school_algebra_problem 19:13:06 check this one out 19:13:07 this is really fun 19:13:13 Yes, true, but the first few steps provide useful constructs 19:13:31 The Geometric Algebra is much more powerful, however, in my own personal opinion 19:13:52 -!- Reece` has joined. 19:17:29 Anyway, if I could ask for feedback 19:17:43 Is the current specification of Daoyu understandable and clear 19:17:46 <\oren\> that list of axioms is bullshit. i was taught plenty about subtraction and negative numbrts in high school 19:17:46 -!- Reece` has quit (Client Quit). 19:18:15 isn't subtraction just adding negative numbers? 19:18:36 \oren\: I'm leaning toward work pranks on that email 19:19:14 <\oren\> right. but that list is missing --x = x for example 19:19:51 <\oren\> and x - y = x + -y 19:20:03 :/ 19:20:06 true 19:26:29 <\oren\> it shoukd be called tarskis alternate universe high school problem, where subtraction is a university discipline 19:29:25 <\oren\> ofc one day we might consider complex numbers to be a high school subject 19:29:32 they are 19:29:42 <\oren\> not in canada 19:29:51 Strange 19:31:52 -!- me2 has quit (Read error: Connection reset by peer). 19:32:49 -!- me2 has joined. 19:38:01 rdococ: α-expressions are fun 19:38:11 (s/// is a well-known α-expression) 19:38:31 (y/// is less well-known, and m// is not used at all) 19:43:43 -!- vanila has quit (Quit: Leaving). 19:56:10 -!- AnotherTest has joined. 20:03:44 what about a language where there are no conditionals and you have to execute arbritary code stored in lookup tables 20:06:49 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 20:09:32 ais523: have anyone examined how densely you can embed information in underload source code such that the underload program can decode it? 20:09:52 b_jonas: it's possible to manage at least one bit per character 20:10:07 ais523: I think you can manage two bits per char at least, but I haven't made a proof 20:10:08 IIRC somewhere oerjan made a parser for strings of : and ^ 20:10:28 going higher there isn't an obvious method, but that doesn't mean there isn't a non-obvious one 20:10:29 I should examine it, I don't understand underload enough and this seems like a thing that lets me explore it 20:11:14 well reaching three bits per character is obviously impossible 20:11:20 ais523: I think I'll try to decode strings made of ~ ! : * 20:11:24 so the true value is ≥ 1 and < 3 20:11:32 ais523: you could receive fractional bits 20:11:40 b_jonas: !~~ and ~~! are eqivalent 20:11:45 and it doesn't have to be stateless, if parenthesis are involved 20:11:46 b_jonas: sure, I expect the number of bits to be fractional 20:11:59 also ~~ and :! are equivalent 20:12:12 and :*:* is equivalent to :::*** 20:12:17 ais523: no no, if the string (supposed you know the length) is parenthisized, you execute it, then the characters get pushed into the stack, and you can execute them one by one 20:12:25 ais523: as in, you can use a to parenthisize them 20:12:41 b_jonas: a parameterizes an entire stack element, it doesn't break the string into characters 20:12:49 ~~ and :! are equivalent; (~~) and (:!) are also equivalent 20:12:53 and both equivalent to () 20:12:58 you, but ^ opens the parenthesis, doesn't it? 20:13:14 b_jonas: it just moves the contents of the TOS onto the program 20:13:27 (~~)^ is equivalent to (:!)^ is equivalent to ~~ is equivalent to :! 20:13:32 none of them do anything :-P 20:13:42 -!- Caesura has quit (Quit: Leaving). 20:13:51 ais523: yes, and if that element it moved is a parenthesisized block, then doesn't executing the block push each of the elements to the stack? 20:14:07 b_jonas: you did notice my comment about underload to you above, right? 20:14:31 b_jonas: if you move ((~~)) to the program you end up with (~~) in the program, which then pushes (~~) to the stack 20:14:32 oerjan: ah... I didn't 20:14:45 can you give an example of what you're /expecting/ to happen? it feels like you have some sort of fundamental misconception here 20:14:51 that rule 110 program is the :^ parser 20:16:02 come to think of it, I suspect that strings of :^a are parseable 20:16:07 I'd be pretty surprised if they weren't 20:16:19 oh, except if you have a then ^ 20:16:22 with nothing in between 20:16:25 as that's a no-op 20:16:27 ais523: no. right. 20:16:59 i concluded : and a were parseable, but more complicated than what i did, and also uglier. 20:17:00 and strings like a:^ would also cause problems 20:17:16 but no other 2 character combinations work 20:17:19 -!- jaboja has quit (Ping timeout: 244 seconds). 20:17:24 :a is obviously parseable, because the result of a never affects the second stack element 20:17:28 -!- hppavilion[1] has joined. 20:17:40 meaning that you can create a sentinel value that does to act as your zero 20:17:40 ais523: I expect that the program (xy) puts x and y to the stack (x on top), so (xy)!^ will try to execute y, and (xy)~!^ will try to execute x 20:18:00 b_jonas: ah, right, your problem is that the stack isn't a stack of characters 20:18:03 it's a stack of strings 20:18:12 (xy) pushes a single element, xy, to the stack 20:18:17 then ^ moves that element as a whole into the program 20:18:22 sure, it's a stack of strings because ((xy)) pushes a single string (xy) to the stack 20:18:38 normally we put parens around each stack element when discussing Underload 20:18:39 doesn't executing a parenthesis open it to the stack? 20:18:42 i forgot that S is useless, so you have only 8 usable chars, so clearly 3 bits is impossible, right 20:18:47 oh... 20:18:52 what the heck is the point of that? 20:18:53 oerjan: and ( and ) have to be paired 20:18:57 why doesn't it open the parens 20:19:04 -!- Kaynato has joined. 20:19:05 b_jonas: because that would be an entirely different language 20:19:08 I mean, if you don't want to open the parens, you can just a^ instead of ^ 20:19:27 actually in Underlambda I'm going the other way, removing S 20:19:45 this makes underload much harder to program 20:19:47 this allows you to optimize code both inside the stack and inside the program 20:19:48 and indeed very different 20:19:48 it becomes a pure functional language because the stack elemenets are entirely opaque 20:20:29 ais523: ( and ) pairing is not _really_ a problem as i think that gives only o(something) contribution to the no. of bits. 20:20:32 ais523: yep 20:20:34 crazy 20:20:42 but any nop like :! is fatal 20:21:01 b_jonas: Underload's meant to be about function representation, not string manipulation 20:21:18 ais523: it's not "string manipulation", more like tree manipulation 20:21:23 -!- earendel has quit (Ping timeout: 248 seconds). 20:21:34 because it manipulates elements that are either letters or parenthisized trees 20:21:45 it doesn't manipulate individual characters of a tree 20:21:52 right 20:21:55 actually Overload worked like that 20:22:12 did I accidentally invent a language? 20:23:19 perhaps, I don't like it though :-( 20:23:23 what about a language where a sentence is an array of words 20:23:26 the set of commands doesn't make much sense in this language 20:23:27 and a word has a meaning 20:23:31 you'd probably want a string compare command too, at least 20:23:36 e.g. 20:23:43 the cat ate the dog 20:23:56 cat is the noun, ate is the verb, dog is the subject, or victim 20:24:17 ais523: no no, not a string compare command 20:24:25 I'd want a nop command though, probably space and newline 20:24:42 make that space, tab, lf, cr 20:24:58 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 20:25:22 b_jonas: this is like having C, plus a command that breaks a function pointers into individual asm opcodes in the function it points to 20:25:30 and yet have no way to determine what those opcodes are other than executing them 20:26:22 I'm reminded to two obfus in other languages: the lua one I wrote that uses the same string as both code and data so that if you try to tamper with the code then the output breaks completely, and pts's obfu in dc which uses the same strings as both code and data: dc can't really examined things so I thought that was impossible, but it turns out dc has a length command that queries the length of a string 20:27:28 ais523: yes, I can see why you don't like that concept 20:27:51 it's just a mix of paradigms that doesn't make sense 20:27:56 it's inelegant 20:27:57 yep 20:28:09 -!- ais523 has quit (Quit: dinner). 20:29:44 -!- Kaynato has quit (Ping timeout: 244 seconds). 20:32:39 ais523? 20:32:50 he left. 20:33:44 `quote raspberry 20:34:05 No output. 20:36:11 `quote berry 20:36:13 471) lets not wander around the mulberry bush beating our heads 20:36:29 fungot: do you know any berries? 20:36:29 int-e: ( and i've only seen ocean from far above in a positive light) 20:36:39 `quote berri 20:36:41 No output. 20:37:02 perhaps it was deleted. 20:37:02 so in the description of Underload at http://esolangs.org/wiki/Underload#Commands , the stack top is on the right, not on the left like other such descriptions usually denote, right? 20:37:18 that sounds very malancholic, fungot. 20:37:18 int-e: the description sounds like fnord talk to me instead 20:37:21 fungot, which episode? 20:37:21 b_jonas: proposals only last 15 minutes of fame? guile? you don't get special food, but they look exactly the same for cl, not fnord two of those 20:37:37 fungot: have you read the SIGBOVIK 2016 proceedings? 20:37:37 b_jonas: http://www.fourmilab.ch/ hotbits/ with fnord key and browser with fnord?) has a problem. whom should it be: 20:37:39 b_jonas: i thought on the right was standard forth-notation? 20:37:58 Oh, SIGBOVIK is today. 20:38:12 oerjan: I don't know 20:38:14 fools! 20:38:28 int-e: will you show them all? 20:38:41 oerjan: no I'm just waiting for this stupid day to pass. 20:39:58 oerjan: I guess top on right makes sense in a forth/postscript-like language where lots of statements are literal so just push on the stack, so the stack often looks similar to the code 20:40:08 int-e: What's wrong with today? 20:40:15 `olist 1032 20:40:16 olist 1032: shachaf oerjan Sgeo FireFly boily nortti b_jonas 20:40:31 b_jonas: anyway half the point of the notation we've settled on is that it makes it easy to see underload execution as a program transformation - where stack elements don't need to be distinguished from the program. 20:40:50 oerjan: right 20:41:07 and it seems that http://www.cs.cornell.edu/icfp/task.htm puts the stack top on the right as well 20:41:30 april fools! 20:41:35 this is great 20:41:38 it's just not so obvious from descriptions like this because most operations would make sense to define with the inputs in any order 20:41:40 int-e: #ircpuzzles is a good way to spend April 1 (and 2 and 3 and.. well, it usually takes a while to solve them) 20:41:49 and in any case you can permute the stack top elements before running the command 20:41:52 (it's a yearly tradition to have puzzles on April 1) 20:41:56 FireFly: HireFly 20:41:57 I am awful at writing documentation] 20:42:01 hachaf 20:42:07 Especially when it's a program which I know how it works 20:42:27 And didn't intend anyone else to use it when I was writing it 20:42:53 noh that was an evil olist 20:44:06 [wiki] [[GML]] https://esolangs.org/w/index.php?diff=46714&oldid=40261 * B jonas * (+5) 20:44:16 FireFly: sorry 20:44:58 shachaf: you may have had your fun, but what about YOUR IMMORTAL SOUL? 20:45:15 oerjan: what about it? 20:45:28 it may be in peril tdnh 20:45:58 didn't you just say it was immortal 20:47:00 yes, but not pain resistant 20:47:26 * Taneb is finally writing a page for COMPLEX 20:48:12 * Taneb has largely forgotten the art of wiki page writing 20:51:21 who needs conditionals 20:51:39 * Taneb needs conditionals 20:52:00 you could remove the IF from Lua and it'd still be turing complete 20:53:15 proof: local code = {true = function () print("yay") end, false = function () print("nay") end}; code[3 > 2] 20:53:32 so 20:53:56 say bye to for loops too 20:54:18 and any other non-necessary construct 20:54:34 I will strip Lua and make a turing tarpit called TarLua... or EsoLua... 20:55:03 -!- ais523 has joined. 20:57:42 ais523: it turns out that one thing that confused me about underload is this: in the (stackbefore – stackafter) style description of stack-based language operations, including underload at http://esolangs.org/wiki/Underload#Commands , the stack is listed with the top of stack on the right, but this isn't obvious at first. 20:58:08 b_jonas: this is standard for Underload; I'm not sure if it's mentioned on the page 20:58:13 I write it in most of my writing about it though 21:01:25 [wiki] [[Evil]] https://esolangs.org/w/index.php?diff=46715&oldid=46711 * Iconmaster * (+2538) Got a lot of information from the site onto the wiki 21:02:51 ] 3*9500 21:02:59 -!- evalj has joined. 21:03:02 ] 3*9500 21:03:02 b_jonas: 28500 21:07:19 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 21:08:20 -!- hppavilion[1] has joined. 21:13:06 Fuzzy String Matching is interesting 21:14:23 [wiki] [[GML]] https://esolangs.org/w/index.php?diff=46716&oldid=46714 * Iconmaster * (+10) Stub'd 21:15:50 [wiki] [[Pylongolf]] https://esolangs.org/w/index.php?diff=46717&oldid=46708 * Iconmaster * (+34) Stub'd 21:15:56 pylongolf 21:15:57 gml 21:16:10 -!- spiette has quit (Quit: :qa!). 21:16:11 Do we have a stub template? 21:16:19 I can't be bothered to write this article right now 21:16:27 [wiki] [[Retina]] https://esolangs.org/w/index.php?diff=46718&oldid=46707 * Iconmaster * (+10) Stub'd 21:17:02 -!- iconmaster has joined. 21:19:50 [wiki] [[User:Iconmaster]] M https://esolangs.org/w/index.php?diff=46719&oldid=37163 * Iconmaster * (-75) 21:22:41 Hello. Suddenly felt like coming back and helping with some wiki stuff. How's it going? 21:24:25 [wiki] [[Bulan]] N https://esolangs.org/w/index.php?oldid=46720 * Rdococ * (+925) It's still Turing complete. I think. I'll leave the exercise of proving that it's TC up to you. 21:24:58 https://youtu.be/QScVVDwxSWA procrastinating like a pro 21:27:26 -!- jaboja has joined. 21:29:49 <\oren\> i ordered a elektronika mk-61 calculator 21:33:16 -!- iconmaster has quit (Read error: Connection reset by peer). 21:33:40 -!- Reece` has joined. 21:33:52 Taneb: yes. 21:35:39 -!- Reece` has quit (Client Quit). 21:39:27 \oren\: A soviet calculator, neat. 21:40:11 oerjan, how do I apply the stub template 21:40:35 myname: NOW WE KNOW 21:40:50 Taneb: {{stub}} 21:40:59 HTH 21:41:06 TDH 21:44:19 -!- Reece` has joined. 21:44:46 [wiki] [[COMPLEX]] N https://esolangs.org/w/index.php?oldid=46721 * Taneb * (+345) Start making this article 21:44:49 -!- Reece` has quit (Remote host closed the connection). 21:46:40 [wiki] [[Language list]] https://esolangs.org/w/index.php?diff=46722&oldid=46691 * Taneb * (+14) /* C */ Add COMPLEX 21:47:42 [wiki] [[Nathan van Doorn]] https://esolangs.org/w/index.php?diff=46723&oldid=35933 * Taneb * (+13) Add COMPLEX 21:48:01 I'm attempting to make a regex family called Irgex, with support for 2D matching and TC matching and all of these nice things 21:48:19 But with the core subset being efficient 21:48:34 actually, there are 2d regex 21:48:37 Taneb: Why limit it to the Gaussian Integers? Why not allow real values in general? 21:48:40 myname: I've heard 21:48:50 myname: Or 2D bnf. It's documented on catseye 21:50:05 i actually used these in my thesis 21:50:20 well, mentioned 21:52:46 myname: I'm trying to find different classes of expressions 21:53:09 The first I define is the π-expression, which is the traditional x+y, -z, [i], etc. 21:53:36 There's also the α-expression, which is a generalization of s/// and y/// 21:53:38 hppavilion[1], reasons 21:53:47 Taneb: Ah, very good choice 21:54:10 I'm stuck there though. Are there any other fairly general types of expression? 21:56:22 For a group of esoteric language creators, I'm surprised that the wiki doesn't auto-generate the list of languages... 21:56:59 zgrep: That'd be too straightforward 21:57:04 All pages that are programming languages should be put in a category (or tagged, or something) "language", and the list would automatically update... 21:57:07 zgrep: Also, we have pages that /aren't/ esolangs 21:57:14 zgrep: Ah 21:57:37 zgrep: Do you know of any other types of expression besides the traditional arithmetical and x///? 21:57:53 hppavilion[1]: Not sure exactly what you mean... 21:58:08 zgrep: You know s/// expressions? 21:58:16 Regular expressions? 21:58:20 (I call them alpha-expressions) 21:58:24 Yes, but I don't know what an expression is... 21:58:27 zgrep: Those fall under arithmetical, actually 21:58:53 zgrep: An expression is some predefined set of symbols combined with more expressions, basically 21:58:59 -!- MDude has quit (Ping timeout: 248 seconds). 21:59:06 Ah. How... how... [insert word here]. 21:59:18 zgrep: Oh, that's good 22:00:09 -!- jaboja has quit (Ping timeout: 244 seconds). 22:04:31 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 22:08:07 [wiki] [[Fuun DNA]] N https://esolangs.org/w/index.php?oldid=46724 * B jonas * (+3506) Created page with "'''Fuun DNA''' is a self-modifying string-rewriting language. It was defined for the task of the ICFP contest 2007 (“Morph Endo”). == Description == The DNA of the extr..." 22:10:11 -!- boily has joined. 22:10:17 @metar CYUL 22:10:17 CYUL 012200Z 27020G27KT 15SM FEW030 FEW045 10/03 A2941 RMK CU1SC2 CU TR SLP963 22:10:24 @metar ENVA 22:10:24 ENVA 012150Z 08003KT 9999 SCT045 BKN060 04/M03 Q1011 RMK WIND 670FT 18006KT 22:10:26 -!- earendel has joined. 22:11:34 -!- tromp has joined. 22:13:13 boily: there was snow on the ground this morning 22:13:29 WE DON'T HAVE SNOW ANYMORE! WE'RE FREE! MWAH AH AH AH AH AH! 22:13:36 * boily knocks on wood 22:13:38 new snow, that is. there's some old in parts. 22:14:20 I haven't seen snow since... last Tuesday 22:14:25 But that was up on some high hills 22:15:38 Tanelle. there's snow where you are at? 22:15:48 -!- tromp has quit (Ping timeout: 246 seconds). 22:16:35 @metar EGNT 22:16:35 EGNT 012150Z 18010KT 150V220 9999 SCT023 08/04 Q1007 22:16:50 boily, it was in the Lake District 22:17:06 @metar KOAK 22:17:06 KOAK 012153Z 25012KT 10SM FEW025 17/08 A3019 RMK AO2 SLP222 T01670083 22:17:11 The bit of England with things that are almost mountains 22:17:38 Taneb: Is that like the Lake Union district in Seattle? 22:18:06 You did mention something about using Seattle public transit the other day. 22:18:40 I distinctly mentioned not using Seattle public transit 22:21:13 So how do you get around the Lake Union district? 22:21:25 this reddit chat thing is pretty cool 22:21:31 [wiki] [[Fuun RNA]] N https://esolangs.org/w/index.php?oldid=46725 * B jonas * (+1034) Created page with "'''Fuun RNA''' is a graphic description language. It was defined for the task of the ICFP contest 2007 (“Morph Endo”). == Description == The RNA of the extraterrestrial..." 22:21:34 i've used seattle public transit. i think. 22:21:35 shachaf, solid boots 22:21:41 the room i ended up in got two pianists who are now playing live 22:22:11 Phantom_Hoover: What's this chat thing? 22:22:15 oerjan: how was it 22:22:22 oerjan: was it before i was born or something 22:22:23 shachaf, it's the button with the bird on it 22:22:40 Phantom_Hoover: Can I find out about it without logging in? 22:22:44 Logging in is scow. 22:22:59 shachaf: were you born in 1996? 22:23:13 no, before 22:23:17 because i'm old :'( 22:23:21 shocking 22:23:35 dinochaf 22:23:46 [wiki] [[Language list]] https://esolangs.org/w/index.php?diff=46726&oldid=46722 * B jonas * (+30) + Fuun DNA, RNA 22:24:51 shachaf, not without scraping a bunch of idiots trying to work it out 22:25:23 * boily carbon dates shachaf 22:25:42 whoa whoa whoa 22:26:01 [wiki] [[User:B jonas]] https://esolangs.org/w/index.php?diff=46727&oldid=46672 * B jonas * (-95) 22:26:10 you can't just carbon date people 22:26:28 of course you can. 22:26:40 oerjan: not more than one at a time, if you're carbonogamous 22:27:07 i think you are switching the wrong part of the word tdnh 22:27:26 look 22:27:35 quantity over quality 22:27:44 i got a pun quota to meet here 22:28:23 oerjan, ais523: http://esolangs.org/wiki/Esolang:Community_portal has a webchat link to here through qwebirc. I prefer the kiwi irc webchat to qwebirc. Should I replace the webchat link, replace the webchat link but mention qwebirc as an alternative, add a kiwi irc link as an alternative, or do nothing? 22:28:23 well you're doing the pun equivalent of cockney rhyming 22:28:48 oerjan: ok fine 22:28:49 Note that kiwi irc doesn't require people to fill a captcha, which can be considered either an advantage or a disadvantage. 22:29:04 oerjan: you can only carbon date people if you're carbonogamous 22:29:13 Do we have a category for a language that is known to be weaker than Turing complete? 22:29:23 b_jonas: PDA? 22:29:27 Fuun RNA is not turing complete. 22:29:38 boily: it's not equivalent to a PDA 22:29:42 shachaf: the original pun was 0.9 shachafs, but the remastered version... 0.55 shachafs. 22:29:47 b_jonas: if there are multiple working webchats you should probably list all of them and put your favourite first 22:29:52 b_jonas: fuun? 22:30:22 boily: http://esolangs.org/wiki/Fuun_RNA although that page doesn't tell enough to glaen the computational strength, you have to read the specs 22:30:46 fungot: glaen? 22:30:46 boily: oh yeah?! well you're a funny guy, though.) 22:31:11 ais523: ok. I'll test if it still works first though. 22:31:53 b_jonas: i thought the one that was listed was freenode's official webchat. they've blocked others on occasion i think. 22:32:01 (well, at least one.) 22:32:18 boily: what's the threshold for a mapoling 22:32:33 oerjan: yes, qwebirc is what the freenode admins themselves run, and they suggest it on their homepage 22:33:07 oerjan: as in, freenode runs a qwebirc webchat server itself which lets you connect to freenode 22:33:25 Do we have a category for a language that is known to be weaker than Turing complete? <-- not a general one, no 22:33:34 But I prefer kiwiirc. Advantages include that it lets you connect to any irc network. 22:33:35 we have specific sub-TC categories 22:33:44 ais523: right 22:33:45 and most sub-TC languages are either in one of them or very very weak 22:34:09 fungot: glaen? <-- like glean, but you might get a leprechaun curse if you're not careful. 22:34:10 oerjan: oh ok. they are comparable. the resistance is useless. it's not all that magical.), tell me 22:34:28 shachaf: somewhere below 0.5, give or take a few subjective centishachafs. 22:34:48 -!- hppavilion[1] has joined. 22:34:49 is this a logarithmic unit or what 22:34:57 ais523: Fuun RNA is very weak in some sense, yes 22:35:12 But it has a high constant factor 22:35:44 I'm gonna make a big file'o buzzwords 22:35:44 -!- lambda-11235 has joined. 22:35:54 It can be executed in linear amount of memory, but the constant factor is quite high because it involves 600x600 pixel 4 channel 8 bit image canvases. 22:36:03 shachaf: it's a semilogarithmic one hth 22:36:18 One that can be used to generate language descriptions automatically. Then I'll make it so Hackego can spit them out 22:36:18 At least so I think. I'm not quote sure how Fuun RNA works really. 22:36:24 You'd have ot read the specs. 22:36:33 But I'm quite sure it's weaker than turing 22:36:59 boily: how does that work fcir 22:36:59 -!- wob_jonas has joined. 22:37:07 Does kiwi webchat work or what? 22:37:14 wob_jonas: it seems so 22:37:55 b_jonas: linear memory fits the bounded-storage machine category 22:38:25 shachaf: it's logarithmic for the most part, but then it gets linear for things approaching zero. avoids problems with derivatives and infinite slopes. 22:38:32 oerjan: which category is that? link? 22:38:49 wait 22:38:53 ah 22:38:54 it's an unofficial one 22:38:56 http://esolangs.org/wiki/Category:Bounded-storage_machine 22:39:09 the last edit was me trying to get rid of it... 22:39:19 That doesn't really say what it means, and I'd think it means constant bounded 22:39:28 ok, enough of this webchat 22:39:33 -!- wob_jonas has quit (Client Quit). 22:39:55 Fuun RNA works really. 22:39:55 003624 < b_jonas> You'd have ot read the specs. 22:40:17 url is https://kiwiirc.com/client/chat.freenode.net/#esoteric by the way 22:40:37 -!- Sgeo__ has joined. 22:41:37 What is the maximum length hackego can send as a message? 22:41:58 <\oren\> i was born in 1993 22:41:59 -!- woboily has joined. 22:42:14 he\\oren\. 22:42:16 boily: 15:40 Oracle deprecated the Java browser plugin this year. :-( 22:42:20 b_jonas: ok you want linear bounded automata 22:42:24 15:40 javascript more like java's crypt 22:42:33 * boily mapoles shachaf 22:42:37 ok 22:42:40 gotta calibrate this thing 22:42:46 (That moment when you search for how do do something in bash for HackEgo, but you look up how to do it in HackEgo specifically and thus get no relevant results) 22:42:53 What's with xkcd today? All it says is “The xkcd April 1st comic is currently experiencing technical difficulties.” and shows wednesday's comic. 22:43:05 -!- woboily has quit (Client Quit). 22:43:43 b_jonas: Either the xkcd April 1st comic is currently experiencing technical difficulties, or that's the joke 22:44:17 oerjan: yes, but Fuun RNA is probably actually weaker than a PDA (I'm not quite sure), and definitely weaker than arbitrary linear bounded machine, probably can be executed in linear time. 22:44:39 `` cat /dev/urandom | tr -dc 'a-zA-Z0-9' | fold -w 200 | head -n 1 22:44:41 hppavilion[1]: that can't be the joke. there have never been a MWF with no strip. 22:44:50 hppavilion[1]: that feels weird. Randall always had nice Apriljokes. 22:44:50 b_jonas: Exactly? 22:44:57 boily: I know... 22:45:01 Mx6laZBPgxJpUlcgwwSY1G7kHxefQnSM6PVVhkb9G5OJBeDSnhKXXXBMa5v7MnG7SSirBK2SYXdYRZT9lpdeOeOS3YVRWNQ22lY7uFjAjHdf3nMFqs3UqtC6Ob5aH5kTf5bGLDUvKwDq3gJrJdZdPHr8YDxO0dFXH7hFHmiElMFP7vuRXrUUZOYFXQso6I9NSLguieFJ 22:45:06 And xkcd is never experiencing technical difficulties. 22:45:14 I think there's probably a comic, it's just well hidden so I can't see it. 22:45:37 What is the maximum length hackego can send as a message? <-- 350 bytes 22:45:38 One of my teachers shaved his glorious lumberjack beard, making him look 10 years younger so he could pretend to be a new student at the beginning of class by just sitting at a desk 22:45:42 oerjan: OK 22:45:52 But the problem is, either the explainxkcd.com folks can't see it either, or for some reason they thought the joke of hiding it is funny and don't want to spoil it (yet). 22:45:58 oerjan: I'm pondering making `longcat, which prints out multiple messages 22:46:02 Wait, that won't work... 22:46:16 Unless HackEgo has some direct-to-IRC command 22:46:27 there's a longcat? 22:46:34 `longcat 22:46:45 ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: longcat: not found 22:46:49 b_jonas: I somehow did that without realizing the pun xD 22:47:04 hppavilion[1]: oh, you're "pondering making" 22:47:11 whew 22:47:13 b_jonas: `longcat would be like cat, but it prints multiple messages to show the whole file. 22:47:21 b_jonas: anyway, most of the categories are upper bounds, so it's fine to be weaker. 22:47:26 hppavilion[1]: yeah, but it doesn't exist 22:47:30 b_jonas: Which would have potential for abuse 22:47:34 hppavilion[1]: we have a command that pastes to a website instead 22:47:37 b_jonas: But would also be useful for long HackEgo messages 22:47:52 b_jonas: The problem is I don't think there's a way to make HackEgo print more than one message 22:48:01 oerjan: but the wiki says the linear bounded automata is an equivalence actegory, not an upper bound 22:48:59 Unless HackEgo has some direct-to-IRC command <-- nope. 22:49:01 anyway, if any of you figure out which category applies (for which you may have to read and ponder the spec very carefully, for it's an esolang so the comp class may not be obvious, even after the definition), and read the wiki descriptions of categories, feel free to add it 22:49:09 oerjan: Didn't think so 22:50:22 hppavilion[1]: jevalbot is also bounded output, it won't print more than 6 lines of output per evaluation command (in my default config and in evalj; j-bot has the limit raised to I don't know how many lines) 22:50:52 b_jonas: OK 22:51:39 ] ,.i._65 NB. example 22:51:39 b_jonas: 64 22:51:39 b_jonas: 63 22:51:39 b_jonas: 62 22:51:39 b_jonas: 61 22:51:39 b_jonas: 60 22:51:40 b_jonas: ... 22:52:07 You can print 6 full lines instead of an ellipsis in the sixth line, but you can't print more than six 22:52:12 ...do we have two different bots specifically for evaluating J 22:52:20 Taneb: no, it's the same bot, two instances 22:52:26 evalj, source: 22:52:26 b_jonas, jevalbot source is http://www.math.bme.hu/~ambrus/pu/jevalbot.tgz 22:52:30 b_jonas: That's even worse 22:52:31 j-bot, source: 22:52:31 b_jonas, jevalbot source is http://www.math.bme.hu/~ambrus/pu/jevalbot.tgz 22:52:46 b_jonas: Why. Why do we have two identical bots? 22:52:49 b_jonas, I think I was using "bot" to mean "account associated with an instance of a bot" 22:52:51 b_jonas: What is the reason? 22:53:28 We used to have three different irc bots that can evaluate J, namely besides jevalbot, buubot used to be able to evaluate it (I wrote the J evaluator plugin myself), plus NotJack had a bot he wrote in J. 22:53:56 hppavilion[1]: evalj is my instance but I don't run it constantly, only very occasionally start it up and then stop it within a day when I turn off my home computer 22:53:59 Can EgoBot? If not, why not? 22:54:22 Taneb: it would probably be possible to install a J interpreter to hackego 22:54:33 I can't get my IRC client to work for some reason... 22:54:34 we could even do it as a user 22:54:42 but it would output only 1 line per command of course 22:55:08 -!- oerjan has quit (Quit: Nite). 22:58:00 I'm heading to bed now 22:58:02 Goodnight! 22:59:28 bonne tanuitb! 22:59:37 @time Taneb 22:59:40 @tineb 22:59:51 -!- evalj has quit (Remote host closed the connection). 23:02:35 @localtime Taneb 23:02:46 darn. no CTCP. 23:02:50 @localtime shachaf 23:02:52 Local time for shachaf is Fri Apr 1 16:02:50 2016 23:10:58 @localtime lambdabot 23:10:58 I live on the internet, do you expect me to have a local time? 23:11:02 Yes. 23:11:22 @localtime zgrep 23:11:24 @localtime zgrep 23:11:25 Local time for zgrep is Fri, 01 Apr 2016 23:11:25 GMT 23:11:31 Heh. 23:18:18 [wiki] [[NetBytes]] N https://esolangs.org/w/index.php?oldid=46728 * Iconmaster * (+4415) Created page with "'''NetBytes''' is a language created by [[User:iconmaster]] in [[:Category:2016|2016]]. It severely limits the available memory and program length of programs, instead requir..." 23:19:38 -!- iconmaster has joined. 23:20:06 Well, it's been years, but I'm finally making another esolang. 23:20:29 iconmaster: Cool! 23:20:39 [wiki] [[User:Iconmaster]] https://esolangs.org/w/index.php?diff=46729&oldid=46719 * Iconmaster * (+64) 23:21:25 So, what's been going on here lately? 23:22:20 iconmaster: Nothing much 23:23:28 Ugh 23:23:39 For some reason, my IRC client won't connect properly. It times out 23:23:45 I have a setup to handle PINGs 23:23:54 But I get an ERROR eventualy 23:23:59 s/l/ll/ 23:26:16 ELLOL. 23:26:59 Ellol to you too! 23:27:50 First issue I found: No \r\n 23:32:28 -!- tromp has joined. 23:36:56 -!- p34k has quit. 23:37:22 -!- hppavilion[t] has joined. 23:37:37 -!- hppavilion[t] has quit (Remote host closed the connection). 23:37:52 -!- hppavilion[t] has joined. 23:38:01 YES 23:38:03 IT WORKED 23:39:20 hppavellon[t]! 23:39:24 -!- hppavilion[t] has quit (Remote host closed the connection). 23:39:31 boily: It can't chat yet 23:39:35 boily: But it receives 23:40:22 -!- hppavilion[t] has joined. 23:40:24 Test 23:40:44 -!- hppavilion[t] has quit (Remote host closed the connection). 23:41:02 -!- hppavilion[t] has joined. 23:41:05 Retest 23:41:33 -!- hppavilion[t] has quit (Remote host closed the connection). 23:41:55 -!- hppavilion[t] has joined. 23:42:07 Testing, testing, 123 23:42:18 YES! 23:42:29 -!- hppavilion[t] has quit (Remote host closed the connection). 23:42:38 That's enough testing on this channel 23:42:51 you lack ambition. test more! FOR SCIENCE! 23:43:22 boily: *Fine* 23:43:42 -!- ais523 has quit (Read error: Connection reset by peer). 23:43:45 :D 23:43:47 -!- int-e has left ("BUGGY CHICKEN"). 23:43:47 -!- int-e has joined. 23:44:23 -!- Kaynato has joined. 23:45:45 -!- ais523 has joined. 23:46:00 -!- tromp has quit (Remote host closed the connection). 23:46:36 helloily 23:47:41 quinthellopia. 23:47:52 -!- hppavilion[t] has joined. 23:48:29 -!- hppavilion[t] has quit (Remote host closed the connection). 23:48:42 -!- hppavilion[t] has joined. 23:48:55 -!- hppavilion[t] has quit (Remote host closed the connection). 23:51:31 bulan 23:51:52 hellodococ 23:52:48 what haps today 23:53:34 Does HackEgo evaluate underload? 23:54:08 ^ul didn't this? 23:54:09 ...bad insn! 23:54:25 Oh, fungot, what won't you do? 23:54:25 prooftechnique: hm. that's not hardcore enough for you... but unfortunately, regardless of whether i _think_ i want a glass of merlot. 23:54:36 ^ul (test string)::** 23:55:16 Hmm 23:55:41 ^ul (test string)::**S 23:55:41 test stringtest stringtest string 23:55:45 There we go