00:06:56 @free traverse 00:06:57 Extra stuff at end of line in retrieved type "(Applicative f, Traversable t) => (a -> f b) -> t a -> f (t b)" 00:07:22 @free (a -> f b) -> t a -> f (t b) 00:07:22 Pattern match failure in do expression at src/Lambdabot/Plugin/Haskell/Free/FreeTheorem.hs:54:21-35 00:07:32 @free t :: (a -> f b) -> t a -> f (t b) 00:07:33 Plugin `free' failed with: src/Lambdabot/Plugin/Haskell/Free/Type.hs:(152,17)-(160,45): Non-exhaustive patterns in case 00:07:50 stupid plugin 00:08:19 @list free 00:08:19 free provides: free 00:08:23 @help free 00:08:23 free . Generate theorems for free 00:08:45 @free sin 00:08:46 Extra stuff at end of line in retrieved type "Floating a => a -> a" 00:09:01 has @free been broken completely 00:09:08 @free map 00:09:10 g . h = k . f => $map g . map h = map k . $map f 00:09:16 :t map 00:09:17 (a -> b) -> [a] -> [b] 00:09:29 @free mapM 00:09:30 Extra stuff at end of line in retrieved type "(Monad m, Traversable t) => (a -> m b) -> t a -> m (t b)" 00:10:14 @free traverse :: (a -> IO b) -> [a] -> IO [b] 00:10:15 $map_IO g . h = k . f => $map_IO ($map g) . traverse h = traverse k . $map f 00:17:26 -!- MoALTz__ has quit (Read error: Connection reset by peer). 00:18:13 -!- MoALTz__ has joined. 00:19:16 -!- olsner has quit (Ping timeout: 246 seconds). 00:19:48 -!- olsner has joined. 00:22:36 -!- augur has quit (Ping timeout: 264 seconds). 00:28:29 -!- XorSwap has joined. 00:31:44 i'm eating a can of tuna and on the can there's this text: bast before 18/09/2017 09:53 00:31:57 -!- augur has joined. 00:32:16 *best 00:32:37 (it's in italian so that part doesn't really matter) 00:37:19 Programming needs ternary comparators 00:37:32 (6>2 by 3) 00:39:29 -!- adu has quit (Quit: adu). 00:44:45 -!- Phantom_Hoover has quit (Read error: Connection reset by peer). 00:51:25 I want to force the "mUsingConnect" of "nsHttpConnectionInfo" to false; it is set in the "Init" method but I want to override it in a JavaScript code so that it forces false 00:55:06 I want to force it not to use the CONNECT method 00:56:41 -!- Patashu has quit (Ping timeout: 250 seconds). 00:59:04 -!- Patashu has joined. 01:01:22 in brainfuck, does "[" always reference the same cell on each iteration, or does it dynamically reference the current pointer when "]" is encountered? 01:02:06 why does it have to be dynamic? 01:02:28 both choices are valid, i want to know what brainfuck defines 01:02:48 newsham: It refers to the current pointer. 01:03:45 i think i need a clarification on that question 01:04:25 -!- zzo38 has quit (Quit: zzo38). 01:04:38 thank you, that solved my bug :) 01:05:14 I,I amUsingConnect 01:05:21 -!- adu has joined. 01:06:35 -!- doesthiswork has quit (Read error: Connection reset by peer). 01:06:42 -!- doesthiswork has joined. 01:07:38 19:44 toyed with the idea of writing my own coding language before i found this room <-- i'm going to assume e joined on the wrong network by mistake 01:08:03 -!- XorSwap has quit (Ping timeout: 255 seconds). 01:09:22 e? 01:09:34 gender neutral pronoun 01:09:54 -!- XorSwap has joined. 01:09:55 that's a new one 01:10:09 it's the official GNP in Agora Nomic 01:10:19 aka "spivak pronoun" 01:10:27 -!- XorSwap has quit (Client Quit). 01:10:36 -!- sc00fy has quit (Ping timeout: 265 seconds). 01:11:21 how did that become official? 01:11:39 well, traditional, anyway. it may not be an actual rule. 01:11:44 spivak's are so different from everything else... i'd much rather use zie/hir 01:11:58 i'm not sure, it's possible we inherited it from Nomic World 01:12:38 (can i use "we" despite currently not being a Player? well, i just did.) 01:12:45 -!- Patashu has quit (Ping timeout: 268 seconds). 01:13:00 The tradition comes because Suber (inventor of Nomic) was fond of Spivak pronouns. 01:13:22 do you all play nomic here? 01:13:34 -!- Patashu has joined. 01:13:37 oh was it in the original suber set too... 01:13:39 Not *all* of us, but definitely a higher proportion than normal. 01:14:12 What is normal? 01:14:36 I don't entirely know. 01:14:48 shachaf: somewhere below 1% i guess 01:15:00 1% of what 01:15:27 Homo sapiens sapiens. 01:16:08 hm doesn't look like it 01:17:55 it's not in the original nomic world ruleset, but it's in this version http://www.nomic.net/deadgames/nomicworld/norrish/ruleset-1dec92 01:18:42 rather inconsistently in a single rule. 01:18:53 (other rules use "he or she" and "they") 01:21:00 eek dead link to ruleset collection 01:22:20 ah the initial Agora ruleset had 5 instances https://www.fysh.org/~zefram/agora/chuck0_nr_19930630.txt 01:22:34 + an "eir" 01:24:58 what if this nomic thing was a complicated distributed system where by the time a proposal you submit to amend the rules is seen by other people, the rules they know about have already changed 01:25:59 sounds cool, but probably not workable with the number of players a game tends to get 01:31:32 -!- bender| has joined. 01:31:32 -!- bender| has quit (Changing host). 01:31:32 -!- bender| has joined. 01:35:34 -!- FreeFull has quit (Ping timeout: 268 seconds). 01:37:16 -!- FreeFull has joined. 01:38:13 this lambda calculus discussion made me wonder 01:38:42 can you write the S combinator as a lambda expression using only 2 variables 01:39:10 (K and I are trivial) 01:44:09 hey, you're not completely trivial 01:44:28 you don't know that hth 01:45:39 -!- Patashu has quit (Ping timeout: 250 seconds). 01:45:48 oerjan: interesting question 01:45:56 my first guess is that the answer is no 01:46:27 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 01:47:26 -!- ^v has joined. 01:47:52 K and KI give you booleans 01:47:59 -!- Patashu has joined. 01:48:01 can you encode pairs... 01:51:34 hm you have church numerals 01:51:50 sure 01:51:51 \f x -> f (f (f x)) 01:55:32 you can shadow, right? 01:55:43 this is like de bruijn indices where you only allow 0 or 1 02:03:26 it's interesting that in BCKW, CKW correspond to the three types of structure people talk about in substructural type systems 02:04:01 my pastebin served pages to over 1000 different ip addresses since august 23 02:04:08 what sort of substructural type system has reordering, discarding, and duplication, but not composition? 02:04:11 is it really boring? 02:11:15 -!- variable has joined. 02:24:41 -!- adu has quit (Quit: adu). 02:42:18 -!- grotewol_ has joined. 02:44:52 -!- grotewo__ has joined. 02:45:10 -!- grotewold has quit (Ping timeout: 272 seconds). 02:46:21 -!- grotewol_ has quit (Read error: Connection reset by peer). 02:46:51 -!- adu has joined. 02:51:34 -!- fowl has joined. 02:51:36 -!- fowl has left ("http://quassel-irc.org - Chat comfortably. Anywhere."). 03:02:16 -!- hppavilion[1] has joined. 03:04:56 -!- mauris has quit (Ping timeout: 240 seconds). 03:08:35 -!- grotewold has joined. 03:11:30 -!- grotewo__ has quit (Ping timeout: 260 seconds). 03:20:02 -!- bender| has quit (Ping timeout: 246 seconds). 03:21:35 -!- bender| has joined. 03:24:16 [wiki] [[Befunge]] http://esolangs.org/w/index.php?diff=44393&oldid=44259 * 198.58.161.28 * (-10) /* Compilation */ Befunjit paragraph applies to Bejit as well 03:29:29 -!- MDude has quit (Ping timeout: 246 seconds). 03:42:07 -!- Frooxius has quit (Quit: *bubbles away*). 03:52:59 -!- ^v has quit (Read error: Connection reset by peer). 03:53:59 -!- Patashu has quit (Ping timeout: 246 seconds). 03:58:36 -!- Patashu has joined. 04:05:02 -!- grotewol_ has joined. 04:07:28 -!- grotewold has quit (Ping timeout: 246 seconds). 04:07:33 -!- grotewo__ has joined. 04:09:21 -!- doesthiswork1 has joined. 04:09:21 -!- doesthiswork has quit (Read error: Connection reset by peer). 04:10:00 -!- grotewold has joined. 04:10:22 -!- grotewol_ has quit (Ping timeout: 244 seconds). 04:12:52 -!- grotewol_ has joined. 04:13:18 -!- grotewo__ has quit (Ping timeout: 265 seconds). 04:15:39 -!- grotewold has quit (Ping timeout: 256 seconds). 04:18:14 -!- grotewold has joined. 04:20:12 -!- zzo38 has joined. 04:20:46 -!- grotewol_ has quit (Ping timeout: 240 seconds). 04:21:00 -!- grotewol_ has joined. 04:23:33 -!- grotewo__ has joined. 04:23:34 -!- grotewold has quit (Ping timeout: 246 seconds). 04:23:50 -!- Sgeo has quit (Read error: Connection reset by peer). 04:24:19 -!- variable has quit (Ping timeout: 244 seconds). 04:25:54 [wiki] [[Special:Log/newusers]] create * Vihan * New user account 04:26:23 -!- grotewol_ has quit (Ping timeout: 244 seconds). 04:26:35 -!- grotewo__ has quit (Client Quit). 05:02:21 HSTS can be disabled in Firefox by the use of a hexeditor. But I have read that HSTS flags cannot be deleted on iPad even if the device is wiped, and can leave an indelible mark by the use of HSTS supercookies. Actually I think it may be possible to get rid of them if you can use a proxy server to set all of the HSTS flags that make up that supercookie; it is kind of like deleting the data on a punch card by punching all of the positions. 05:04:04 -!- Frooxius has joined. 05:09:15 -!- Sgeo has joined. 05:10:47 -!- FreeFull has quit (Ping timeout: 268 seconds). 05:11:28 -!- FreeFull has joined. 05:14:20 -!- Sgeo has quit (Read error: Connection reset by peer). 05:22:16 -!- Sgeo has joined. 05:25:47 [wiki] [[HALT]] N http://esolangs.org/w/index.php?oldid=44394 * Vihan * (+2521) Added specification 05:28:18 [wiki] [[Language list]] M http://esolangs.org/w/index.php?diff=44395&oldid=44279 * Vihan * (+11) Added HALT 05:31:04 [wiki] [[HALT]] M http://esolangs.org/w/index.php?diff=44396&oldid=44394 * Vihan * (+81) Added Interpreter 05:33:18 -!- adu has quit (Quit: adu). 05:39:08 [wiki] [[HALT]] http://esolangs.org/w/index.php?diff=44397&oldid=44396 * Vihan * (+197) 05:46:49 The X cursor font is missing a magnifier icon; it seems to have everything else. 06:03:31 -!- Frooxius has quit (Quit: *bubbles away*). 06:04:34 -!- grotewold has joined. 06:08:23 -!- grotewol_ has joined. 06:11:10 -!- grotewold has quit (Ping timeout: 250 seconds). 06:11:20 -!- grotewold has joined. 06:13:46 -!- grotewol_ has quit (Ping timeout: 250 seconds). 06:14:05 -!- grotewol_ has joined. 06:16:00 -!- grotewo__ has joined. 06:16:16 -!- J_Arcane has joined. 06:16:37 -!- grotewold has quit (Ping timeout: 246 seconds). 06:17:11 -!- [1]blurelIse has joined. 06:18:46 -!- alx__ has joined. 06:18:47 -!- grotewol_ has quit (Ping timeout: 264 seconds). 06:18:52 -!- grotewold has joined. 06:19:07 -!- doesthiswork1 has quit (Quit: Leaving.). 06:19:45 -!- blurelIse has quit (Ping timeout: 256 seconds). 06:19:45 -!- [1]blurelIse has changed nick to blurelIse. 06:21:47 -!- grotewol_ has joined. 06:21:47 -!- grotewo__ has quit (Ping timeout: 264 seconds). 06:23:36 -!- grotewold has quit (Read error: Connection reset by peer). 06:25:00 -!- ChanServ has set channel mode: +o oerjan. 06:25:49 -!- oerjan has set channel mode: +b *!~grotewold@c-68-50-42-113.hsd1.in.comcast.net#fixyourconnection. 06:25:56 -!- oerjan has set channel mode: -o oerjan. 06:26:55 hm wait was that the right syntax 06:27:31 -!- ChanServ has set channel mode: +o oerjan. 06:27:41 -!- oerjan has set channel mode: -b *!~grotewold@c-68-50-42-113.hsd1.in.comcast.net#fixyourconnection. 06:27:52 -!- oerjan has set channel mode: +b *!~grotewold@c-68-50-42-113.hsd1.in.comcast.net$##fixyourconnection. 06:28:16 -!- oerjan has set channel mode: -o oerjan. 06:35:53 ---$## 06:37:21 let's start from 15 06:37:23 15 is 3 * 5 06:37:29 glue 3 and 5 together and you get 35 06:37:35 35 -> 5 * 7 06:37:47 57 -> 319 -> 1129 06:37:51 1129 is prime and we stop 06:38:24 does this sequence always stop for every integer > 1 ? 06:39:11 I don't know. 06:40:03 <\oren\> wait why does 57 become 319 and not 139? 06:40:06 I not only don't know the answer for base ten but I don't know the answer for base two either. 06:40:26 <\oren\> oh 06:40:29 <\oren\> right 06:40:29 \oren\: 57 is 3 * 19 06:41:07 <\oren\> the prime factors are arranged in order from small to large 06:41:07 i wrote a small script to test this and all numbers below 1000 terminate, except 49 which is still going 06:41:10 my computer is slow 06:41:17 heh 06:41:36 <\oren\> 49 -> 77 -> 711 -> 06:41:50 `factor 711 06:41:51 711: 3 3 79 06:41:57 `factor 3379 06:41:57 3379: 31 109 06:42:01 `factor 31109 06:42:02 31109: 13 2393 06:42:09 `factor 132393 06:42:10 132393: 3 44131 06:42:19 `factor 344131 06:42:20 344131: 17 31 653 06:42:31 `factor 1731653 06:42:32 1731653: 7 11 43 523 06:42:43 `factor 71143523 06:42:44 71143523: 11 11 577 1019 06:42:57 `factor 11115771019 06:42:58 11115771019: 311 35742029 06:43:05 `` n=49; while true; do echo $n; n=$(factor $n | tr -cd '[0-9]'); done 06:43:10 `factor 31135742029 06:43:11 31135742029: 7 17 261644891 06:43:29 izabera: do you know after what number of iterations the others stop? 06:43:35 49 \ 4977 \ 497733779 \ 497733779497733779 \ 497733779497733779711131952579497733779 \ factor: `497733779497733779711131952579497733779' is too large 06:44:02 `` n=49; for i in `seq 100`; do echo $n; n=$(factor $n | cut -d\ -f2- | tr -cd '[0-9]'); done 06:44:05 myname: i wrote a script that does it for me 06:44:22 izabera: what is the maximum iteration besides 49? 06:44:33 49 \ 77 \ 711 \ 3379 \ 31109 \ 132393 \ 344131 \ 1731653 \ 71143523 \ 11115771019 \ 31135742029 \ 717261644891 \ 11193431873899 \ 116134799345907 \ 3204751189066719 \ 31068250396355573 \ 62161149980213343 \ 336906794442245927 \ 734615161567701999 \ 31318836286194043641 \ factor: `31318836286194043641' is too large 06:44:56 @oeis 49,77,711,3379,31109 06:44:56 Concatenate all the prime divisors in previous term (with repetition), start... 06:45:04 -!- grotewol_ has quit (Ping timeout: 265 seconds). 06:45:05 woot 06:45:14 it's a thing 06:45:18 oh 06:45:23 @list oeis 06:45:23 oeis provides: oeis sequence 06:45:24 why doesn't it print put a link 06:45:37 int-e: i'm with myname 06:45:45 How do you tell it to display the OEIS ID number 06:45:48 at least have a command for it 06:45:57 looks like i missed something and 77 takes a while... ^^' 06:46:05 @sequence 49,77,711,3379,31109 06:46:05 Concatenate all the prime divisors in previous term (with repetition), start... 06:46:17 bah 06:46:21 Rather than only the description 06:46:48 izabera: weöö, obviously 06:46:50 http://arin.ga/t6ZGY8/raw output for 49 so far 06:46:55 https://oeis.org/A056938/internal 06:46:59 izabera: 77 needs one step less than 49 06:47:57 izabera: page claims up to 117 steps without stopping 06:48:18 http://arin.ga/t6ZGY8/raw 06:48:25 wrong paste 06:48:43 well my laptop isn't really suited for number crunching 06:49:03 especially if it's pointless number crunching like this <.< 06:49:18 Oh, it's slow because it requires factoring 06:49:25 <\oren\> at some point I should figure out how to network all my linux boxen into a cluster 06:49:45 <\oren\> or not 06:50:21 <\oren\> Are there cluster architectures that allow use of 286's? 06:50:42 <\oren\> and x86-64's in the same cluster 06:51:02 I don't think you want to write any software for a cluster that has a 286. 06:51:53 <\oren\> Jafet: but I have a 286 PC just sitting there 06:52:37 <\oren\> it's a ibm 06:52:56 how old is it? 06:53:34 <\oren\> I got it from my high school, sitting on the curb so i have no idea 06:54:05 <\oren\> it has several external parallel ports 06:55:13 <\oren\> I was using it for a while to play DOOM 06:55:47 http://www.worldofnumbers.com/topic1.htm 06:56:19 -!- alx__ has quit (Ping timeout: 246 seconds). 06:58:50 * \oren\ googles parallel port and realizes it has only one parallel port. the multiple ones are serial prts, which make no sense, since they have nine pins in parallel? 06:59:36 <\oren\> well regardless it has three serial ports and one parallel 06:59:52 "Probabilistic arguments give exactly zero for the chance that the sequence of integers starting at a given number n contains no prime, so a home prime should exist for every positive integer. " 07:00:00 http://mathworld.wolfram.com/HomePrime.html 07:00:13 (that's not a proof, of course, just a good hint) 07:02:19 proving that is probably a hard problem. 07:05:29 -!- Patashu has quit (Remote host closed the connection). 07:05:41 -!- Patashu has joined. 07:18:01 -!- MoALTz__ has changed nick to MoALTz. 07:19:23 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 07:26:53 -!- gamemanj has joined. 07:29:32 -!- oerjan has quit (Quit: leaving). 07:32:06 -!- JesseH has quit (Ping timeout: 240 seconds). 07:36:50 http://pastebin.com/QTf1EWuc 07:37:24 -!- JesseH has joined. 07:45:38 3331113965338635107 07:55:53 anyone wanna put my bf impl on the wiki? 07:58:48 -!- alx__ has joined. 08:04:51 -!- TieSoul has joined. 08:05:02 newsham: Can you put it by yourself? 08:06:40 I remember my grandfather once threatened to call the President and tell him to remove the word "I don't want" from the dictionary (this happened many years ago). This is despite that we live in Canada. 08:07:06 What did the President of Canada say? 08:08:11 He did not actually do that; he only threatened to do so. 08:10:04 The word "I don't want"? That would've failed even had there been a President of Canada... 08:10:10 -!- TieSoul has quit (Ping timeout: 260 seconds). 08:10:24 ...even if they were willing to remove words from dictionaries. 08:11:16 Yes, I know, there are so many things wrong with that. 08:11:37 -!- TieSoul has joined. 08:11:41 > length "I don't want" 08:11:43 12 08:11:55 Perhaps your grandfather uses 96-bit words. 08:12:20 Sounds like a good word for if you're playing Countdown, or Scrabble. 08:14:02 PSA: Ending the world is usually not a good idea. 08:14:15 And of course removing words from dictionaries won't generally stop people from using them anyways. 08:14:42 Sgeo: too much olist 08:15:04 * Sgeo was referring to events IRL during the Cold War actually 08:15:26 Although I guess saying the literal world is a bit poetic, just civilization 08:16:39 Especially Scrabble, because although I'm not sure what you'd represent the spaces and apostrophe as, it's made of 3 component words, so if you managed to manuver the letter-tree so that you could insert "I", "don't" and "want" in the right positions, then add whatever you'd be using as spaces in-between for lots of points on the massive mega-word... 08:17:01 Relatedly, happy Stanislav Petrov day! 08:17:42 Ending the world is the kind of thing that I think someone had to say "What could possibly go wrong?" in order to get even close to performing. 08:20:29 What could possibly go wrong with orders to retaliate if a new satellite system registers an attack? 08:21:05 A likely not-completely-tested one at that, I presume? 08:21:49 How does anyone completely test anything? 08:22:03 Via actual experience, I suppose? 08:23:34 There can always be edge cases, like sunlight reflecting off of clouds in a certain way 08:23:51 In the end, relying on anything except a mass dropout of everything in a certain area (and even then, if they have a power failure...) for information on if you've been hit or are going to be hit is a quick way to make a complete mess. 08:35:36 @where oeis 08:35:36 I know nothing about oeis. 08:40:18 -!- [1]blurelIse has joined. 08:42:23 -!- blurelIse has quit (Ping timeout: 246 seconds). 08:42:23 -!- [1]blurelIse has changed nick to blurelIse. 08:56:55 -!- sc00fy has joined. 09:31:44 -!- alx__ has quit (Quit: Leaving). 09:32:33 -!- Phantom_Hoover has joined. 10:06:51 -!- Vorpal has joined. 10:06:51 -!- Vorpal has quit (Changing host). 10:06:51 -!- Vorpal has joined. 10:12:32 -!- evalj has joined. 10:16:10 -!- AnotherTest has joined. 10:43:48 -!- JesseH has quit (Remote host closed the connection). 10:59:56 -!- mauris has joined. 11:22:39 [wiki] [[Nyarlathotep]] N http://esolangs.org/w/index.php?oldid=44398 * CosmoConsole * (+5187) Created page with "'''Nyarlathotep''' (named after a deity in the Cthulhu Mythos) is a esoteric language created by [[User:CosmoConsole]] in 2015, and has received a large amount of inspiration ..." 11:23:12 [wiki] [[Language list]] http://esolangs.org/w/index.php?diff=44399&oldid=44395 * CosmoConsole * (+19) /* N */ [[Nyarlathotep]] 11:23:41 [wiki] [[Nyarlathotep]] M http://esolangs.org/w/index.php?diff=44400&oldid=44398 * CosmoConsole * (+9) /* External resources */ fixed link 11:30:36 [wiki] [[Nyarlathotep]] M http://esolangs.org/w/index.php?diff=44401&oldid=44400 * CosmoConsole * (+4) /* External resources */ clarify python version 11:33:18 [wiki] [[User:CosmoConsole]] http://esolangs.org/w/index.php?diff=44402&oldid=40731 * CosmoConsole * (+69) 11:42:54 -!- evalj has quit (Remote host closed the connection). 11:46:34 http://imgur.com/gallery/meIKvBa what's happening here? 12:06:42 [wiki] [[Nyarlathotep]] M http://esolangs.org/w/index.php?diff=44403&oldid=44401 * CosmoConsole * (+0) fixed a small mistake 12:08:06 [wiki] [[Nyarlathotep]] M http://esolangs.org/w/index.php?diff=44404&oldid=44403 * CosmoConsole * (+12) /* Operation */ clarified the operation of C when reaching end of memory 12:19:32 -!- J_Arcane has quit (Ping timeout: 265 seconds). 12:31:00 -!- J_Arcane has joined. 12:32:35 -!- AnotherTest has quit (Ping timeout: 264 seconds). 12:33:04 -!- sebbu has quit (Ping timeout: 265 seconds). 12:55:05 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds). 12:55:46 -!- Phantom_Hoover has joined. 13:01:43 -!- Patashu has quit (Ping timeout: 246 seconds). 13:18:10 -!- sebbu has joined. 14:05:07 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…). 14:07:30 -!- doesthiswork has joined. 14:16:09 -!- mauris has quit (Quit: Leaving). 14:24:27 -!- mauris has joined. 14:33:46 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds). 14:39:48 -!- Phantom_Hoover has joined. 14:50:30 zzo38: i dont have wiki skills.. is there an esolang for posting to wiki? 14:58:10 -!- atrapado has joined. 14:58:47 -!- J_Arcane has quit (Read error: Connection reset by peer). 14:59:02 -!- J_Arcane has joined. 14:59:36 -!- XorSwap has joined. 15:02:31 -!- ^v has joined. 15:48:53 -!- evalj has joined. 15:51:49 -!- bender| has quit (Ping timeout: 246 seconds). 16:21:59 -!- ^v has quit (Ping timeout: 240 seconds). 16:30:58 -!- AnotherTest has joined. 16:41:31 -!- XorSwap has quit (Quit: Leaving). 17:04:13 -!- ais523 has joined. 17:05:53 fizzie, I'm having fun running a fuzzer on cfunge, lets see what it comes up with 17:06:59 fizzie, btw, didn't you find some bug in cfunge a few months ago that I never got around to looking at properly? 17:07:03 Can't find the IRC log any more 17:09:32 That rings a bell, but only faintly. 17:11:00 fizzie, oh well 17:11:03 Grepping, the one recent thing I can find is that I was trying to use cfunge <(echo ...) to test a oneliner, and it went to a busyloop somewhere. 17:11:17 huh, Debian stable just had an update for linux-libc-dev but not the matching linux-libc package 17:11:21 (echo ... > tmp.bef; cfunge tmp.bef) worked fine. 17:11:23 I wonder what /that's/ about 17:11:27 fizzie, the fuzzer is obviously mostly finding "hangs" due to programs running into infinite loops 17:11:35 fizzie, considering how to deal with this hm 17:11:42 -!- MDude has joined. 17:11:59 oh, I see, it's the #defines that the kernel exposes to userspace 17:12:04 so there is no matching non-dev package 17:12:09 Grepping, the one recent thing I can find is that I was trying to use cfunge <(echo ...) to test a oneliner, and it went to a busyloop somewhere.<-- well that doesn't surprise me 17:12:13 fizzie, cfunge mmaps 17:12:26 but I find it hard to see how a security bug in those wouldn't require a rebuild of every other package that referenced the buggy setting 17:12:27 Because \r\n are annoying to deal with across fread() boundaries 17:12:46 perhaps a setting requesting extra security was accidentally defined to 0 rather than what it should have been, but no Debian packages used it 17:12:54 Vorpal: Well, failing instead of getting stuck would still be better. 17:12:55 ais523, heh 17:13:20 fizzie, I would have assumed it would do that yes 17:13:27 fizzie, I wonder why it gets stuck 17:13:36 Kind of hard to debug that one via gdb... 17:13:47 I guess 17:13:49 I assume it would do the same for, say, a named pipe. 17:14:04 Possibly 17:14:31 fizzie, I find it unlikely that mmap() would get stuck though instead of returning an error 17:14:53 addr = mmap(NULL, len, PROT_READ, MAP_PRIVATE, fd, 0); 17:14:53 if (FUNGE_UNLIKELY(addr == MAP_FAILED)) { 17:14:53 diag_error_format("mmap() on file \"%s\" failed: %s", filename, strerror(errno)); 17:14:53 goto error; 17:14:53 } 17:14:58 Hm 17:15:17 evil tabs 17:15:31 int-e, ? 17:15:52 int-e, tabs for indenting is great, that way you can set the tab stop to whatever you prefer and it still looks good 17:16:01 http://int-e.eu/~bf3/tmp/tabs.png 17:16:16 Incidentally, I'm just filling form 306, titled "Application for pre-examination and permission to publish the dissertation". 17:16:36 int-e, well that is your client, not my fault. I use hexchat and it renders it as a space 17:16:45 so it looks under-indented but still readable 17:17:15 (And no I don't use tabs to adjust continued lines to the preceding line, *that* would break if you change tab width) 17:17:32 fizzie, oh? 17:18:22 current policy: frequency should be within 1.60 GHz and 3.70 GHz. 17:18:22 The governor "performance" may decide which speed to use 17:18:22 within this range. 17:18:22 current CPU frequency is 3.58 GHz. 17:18:23 What 17:18:41 Shouldn't performance ensure it is 3.7 all the time? 17:18:52 I think the 3.7 GHz number counts the "turbo boost" stuff. 17:19:01 Oh, of course 17:19:04 Like, absolute maximum for a single core, not the "normal 100%" thing. 17:19:10 model name: Intel(R) Core(TM) i5-2500 CPU @ 3.30GHz 17:19:11 yeah 17:19:13 makes sense 17:20:33 "cpufreq stats: 3.40 GHz:4.86%, 3.40 GHz:0.00%, 3.20 GHz:0.44%, ..." hrm 17:20:45 Wonder what that second, never-used 3.40 GHz speed is. 17:20:54 heh 17:21:07 fizzie, where did you find that line? 17:21:20 The "cpufreq-info" tool printed it out. 17:21:21 hmm, I find this bug report strangely amusing: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=522773 17:21:24 fizzie, hm not for me 17:21:26 (b_jonas in particular might like it) 17:21:34 [wiki] [[Brainfuck implementations]] http://esolangs.org/w/index.php?diff=44405&oldid=44141 * 72.235.201.146 * (+125) /* Normal implementations */ add msp430 interpreter 17:21:51 pity it's unresolved, though 17:21:51 ais523, I find the debian bug tracker obtuse. Hard to see the status of a bug. 17:21:57 Vorpal: open, in this case 17:22:01 ais523, where does it say that 17:22:02 but it's more the bug itself, rather than its resolution 17:22:06 Could be the statistics collection is a new thing and/or needs to be enabled somehow. Although I don't recall doing anything for it. 17:22:13 Vorpal: it doesn't mention it being closed anywhere 17:22:19 that I noticed, at least 17:22:47 ais523, right, but even bugzilla is more readable than debian's system :P 17:22:51 And bugzilla is terrible 17:23:17 Nice graphviz plot in the upper-right corner, though. 17:23:26 Lots of nodes titled "some versions". 17:24:49 fizzie, you can read the graph? it is tiny for me 17:25:03 oh clickable 17:29:29 I'm not sure what it's trying to illustrate, but I like graphs no matter what. 17:32:25 heh 17:32:28 bbiab 17:32:54 fizzie: I think it shows the affected package versions, color-coded; red = affected, green = fixed. 17:33:52 And the edges? 17:34:11 hard to find good examples though... https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=522759 has no forks :/ 17:34:55 there is no linear versioning; there are forks because there are three (I think?) debian distibutions at any point in time. 17:35:25 note, I don't know, I'm making an informed (hopefully) guess here. 17:35:34 In all (two) graphs I've seen so far, "some versions" could have been an implied by an edge. 17:36:20 Oh, there's the "don't collapse" mode. 17:36:21 well, somebody has to tag version+bug pairs, I suppose... 17:36:44 anyway, perhaps a good point to stop guessing :) 17:37:42 I tried to turn off "ignore boring" and "collapse" for the first graph, and got a broken image. 17:38:00 https://bugs.debian.org/cgi-bin/version.cgi?found=linux-2.6%2F2.6.29-2;found=linux-2.6%2F3.2~rc7-1~experimental.1;ignore_boring=0;package=linux-libc-dev;absolute=0;info=1;format=png;height=;width=;collapse=1 nice and wide 17:38:44 oh, grey. 17:38:52 That's the "boring", I think. 17:39:24 which means "unknown"? 17:42:03 fizzie, so far the tool has found no crashes, that is pretty nice 17:42:40 https://wiki.debian.org/HowtoUseBTS explains some of it 17:48:35 Do you know how to make proxy working with Apache? I always get error message, the stuff they mention on their IRC doesn't work either 17:50:20 The configuration is set up like http://sprunge.us/EEYY (the "example.org" and "stuff.example.org" and "localhost:9999" are all just for testing and will be changed later, but "127.0.0.50" is going to remain the same; also I add the ProxyPassReverse which is not listed in that old version of the file) 17:50:42 I used to have one Apache "reverse-proxy" configuration, but I don't think I have that configuration saved anywhere. 17:52:04 Do you know how to get it to work though? 17:52:32 Not really. I think I got it working by following the instructions, but this was some years ago. 17:52:41 why apache btw? 17:53:32 Apache is the server I already have set up for the main HTTP server, so I try to use the same one as the proxy 17:56:08 In my configuration, I just had the reverse-proxy ProxyPass part, because I wasn't trying to connect to the target via another proxy. 17:56:35 I did get this error message in the log file: [error] proxy: ap_get_scoreboard_lb(1) failed in child 10872 for worker proxy:reverse 17:57:46 Haven't seen that. 17:59:30 google finds https://serverfault.com/questions/454421/apache-httpd-error-proxy-ap-get-scoreboard-lb-with-proxypass which suggests to stop and start apache (rather than reloading or restarting)... 17:59:49 OK I will try that 18:00:18 (I have not looked at the configuration, but it sounds like a harmless and at the same time obscure thing to try) 18:00:30 * int-e googled the error message 18:01:34 That error message is gone, but I still get a 500 error and the same other logged errors as before, just not the scoreboard error 18:01:51 The error is: [warn] proxy: No protocol handler was valid for the URL /what. If you are using a DSO version of mod_proxy, make sure the proxy submodules are included in the configuration using LoadModule. 18:02:14 I did load both mod_proxy and mod_proxy_http and checked it with apachectl -M 18:04:22 https://stackoverflow.com/questions/23931987/apache-proxy-no-protocol-handler-was-valid ... LoadModule ssl_module modules/mod_ssl.so 18:04:49 (I guess that otherwise it fails to speak https) 18:05:19 I do not need it to speak https 18:05:23 -!- doesthiswork has quit (Read error: Connection reset by peer). 18:05:23 -!- doesthiswork1 has joined. 18:06:11 I want it to just send a request like "GET https://example.org/what HTTP/1.1" to localhost:9999 18:06:23 -!- doesthiswork1 has quit (Read error: Connection reset by peer). 18:06:25 -!- doesthiswork has joined. 18:06:44 (Implementing SSL on the next proxy in line instead) 18:06:58 mmm 18:08:27 -!- doesthiswork1 has joined. 18:08:27 -!- doesthiswork has quit (Read error: Connection reset by peer). 18:17:03 Bah, the libreoffice 5.something I have here at home does a much worse job of editing-and-saving this .doc file than the libreoffice 4.something I had at work. 18:20:34 Then install the other version as well (if it is possible to install both) 18:31:48 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 18:33:07 -!- AnotherTest has joined. 18:43:06 fizzie, ouch 18:44:18 hizzie 18:44:21 Horpal 18:50:22 <\oren\> gVd aftRnUn evrEbudE! 18:51:08 This sending of pieces of paper around is totally silly. 18:51:31 They want two printed copies of the thesis manuscript along with the application form. 18:51:44 fizzie: "they" = ? 18:51:44 best idea ever: an OS that boots two kernels, and every time you start a process, it randomly chooses one of the kernels to run it on 18:51:46 \oren\: Id spel it "evribudi" 18:52:02 <\oren\> university policies are made by the oldest, most tenuruous professors 18:52:10 also does it have to be properly bound? or is a temporary binding enough? 18:52:11 ais523: Part of the same conspiracy which is holding back the publication of your thesis, presumably. 18:52:46 shachaf: at least my university now has a (very sensible) rule that if the thesis examiner and student agree, you can send the corrected version of a thesis by email rather than needing to send a printed copy 18:52:50 and then not get it printed until it's been approved 18:52:56 <\oren\> nortti: then how would you spell 'it'? 18:53:00 ais523: Well, the university. And it's supposed to be perforated and in a plastic folder, but you're expressly forbidden to go through the actual get-a-printed-copy process at this stage. 18:53:04 \oren\: "it" 18:53:07 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 18:53:07 But your thesis has been approved, right? 18:53:11 shachaf: yes 18:53:13 ages ago 18:53:14 But we still can't read a copy. 18:53:16 right 18:53:22 I can link you to the page that says you can't read a copy, if you like 18:53:26 Sure. 18:53:46 <\oren\> nortti: that is bad. the vowel in 'it' and the vowels at the end of 'every' are very sifferent 18:53:48 ais523: Come to think of it, I assume you could get it properly printed, if you want to do it at your own expense. 18:53:55 shachaf: here: http://etheses.bham.ac.uk/6120/ 18:54:20 \oren\: I hear them as the same, while the one in "see" and end of "every" as very different :/ 18:54:31 -!- AnotherTest has joined. 18:54:42 fizzie: I could; I've already done that for the "official" versions, and both the University and bookbinder were quite clear that I was allowed to print as many copies as I like so long as I paid for them 18:54:52 I had to pay for the two official copies too 18:54:56 <\oren\> nortti: well the good thing sabout my system is you can spell wrods according to your own accent 18:55:14 indeed 18:55:33 that's what I was meaning 18:55:59 ais523: Ah. At our university, the university pays for up to 500 EUR of printing costs, as long as the dissertation is published in the university's series. 18:56:38 And the minimum number of copies is 30. 18:56:44 <\oren\> fizzie: at UofT I got 1000 free printing pages every year as part of my tuition 18:57:06 <\oren\> so i guess not 'free' 18:57:13 hmm, 1000 pages isn't so much 18:57:41 <\oren\> there were no rule as to "no printing pages entirely black[C" 18:57:42 30 copies of my thesis would currently require 6420 pages. 18:57:47 \oren\: I personally think you should have a way to mark scwas, like in the beginning of "about". in a spelling system me and heddwch developed, we used 'y' for iy 18:57:50 *it 18:58:04 why not use ə? 18:58:18 I've seen phonetic spellings that used English letters entirely except they used ə for schwa 18:58:19 \oren\: Is this, like, properly-bound-in-a-book pages, or just, you know, printer printing? 18:58:36 <\oren\> nortti: I just use whatever the vowle would be if it weren't reduced 18:58:48 <\oren\> or 'u' 18:58:53 \oren\: that's hard to tell phonetically, though 18:59:12 \oren\: but don't you have special forms for [ər], 'R'? 18:59:16 <\oren\> to me the first vowel in 'about' is u as in 'but' 18:59:20 <\oren\> nortti: yes 18:59:25 ais523: Well, the university. And it's supposed to be perforated and in a plastic folder, but you're expressly forbidden to go through the actual get-a-printed-copy process at this stage. 18:59:33 fizzie, this sounds stupidly complicated? 18:59:42 <\oren\> because those are reduced in my accent to sylabic R 18:59:56 fizzie, when I did my bachelor thesis it was all digital. 19:00:01 <\oren\> as in 'butter' -> butR 19:00:05 fizzie, no paper copy anywhere along the line 19:00:11 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 19:00:43 <\oren\> fizzie: it was printing on up to a page size of twice a letter-page 19:00:49 Maybe it is different for master theses? 19:01:03 Vorpal: From what I've heard informally, at least some of the Doctoral Programme Committees no longer even look at the paper copies. I think the two print copies are those that will be sent to the pre-examiners, though. Not that I'm so sure they want the paper copies either. 19:01:22 fizzie: the originally marked (i.e. before editing) copies of my PhD thesis had to be paper, and they were bound in a special "temporary binding" way designed for the process which is pretty cheap 19:01:29 fizzie, heh 19:01:47 ais523: I think that's approximately equivalent to our "put it in a plastic folder" rule. 19:02:01 -!- AnotherTest has joined. 19:02:10 fizzie: well temporary bindings are quite strong, you could probably rip the pages out if you tried hard but they're unlikely to fall out by accident 19:02:11 ais523: If I email them to ask for an early copy do you think they would oblige? 19:02:12 unlike plastic folders 19:02:28 shachaf: I'm guessing no, but I doubt it would hurt 19:02:47 fizzie, also I thought you were already working at google in UK? 19:02:53 On android stuff iirc 19:02:57 <\oren\> ais523> why not use ə? <- because ə isn't on a normal keyboard 19:03:06 it's on mine! :-( 19:03:11 ais523, compose? 19:03:14 are you calling my keyboard abnormal? 19:03:16 Vorpal: of course 19:03:19 <\oren\> very 19:03:19 compose e e 19:03:24 ais523-lang: Impossible to use 19:03:32 (unless you're ais523) 19:03:35 what's "ais523-lang"? 19:03:38 (or stole his keyboard) 19:03:39 Vorpal: I didn't do a bachelor's (I squeaked by just before they modernized our programs to involve one; before, it was straight-to-master) but I think my master's thesis involved a printout for examination. And I got 20 copies bound; there's a particular binding specifically for this kind of theses ("diplomityö"). 19:03:40 I've made tons of languages 19:03:50 hmm, I haven't made any natural languages yet, though 19:03:53 Vorpal: And yes, you thought correctly. 19:03:54 <\oren\> one that uses all these crazy letter you can apparently type 19:03:56 OK, what about this: esoteric conlang, that has zero words 19:04:01 yes 19:04:06 this is obviously a good idea 19:04:09 fizzie, so why messing around with plastic folders now then? 19:04:27 Vorpal: Well, I mean, as opposed to what? Leaving my PhD unfinished? 19:04:33 \oren\: it's a completely stock UK 101 (103?) key laptop keyboard, the only change I made was to rebind one of the keys 19:04:38 fizzie, ah 19:05:05 UK keyboard? Hmm... 19:05:14 shachaf: I guess you at least get to read the abstract 19:05:26 ais523, I doubt any laptop keyboard is 101 or 103 keys, there are always strange keys not found on classic keyboards 19:05:37 I thought Fn was the main one 19:06:00 ais523: By the way, did you see my question about substructural types the other day? 19:06:02 ais523, Mine has browser back/fwd for example, and fn. Also no right windows key 19:06:08 19:03 it's interesting that in BCKW, CKW correspond to the three types of structure people talk about in substructural type systems 19:06:10 the only weird key placement I can see here is that for some reason, pause/sysrq is fn-shift 19:06:11 I did get some ex-colleagues at the university to do the physical printing, at least, so I don't have to ship the printouts from UK to Finland. 19:06:14 19:04 what sort of substructural type system has reordering, discarding, and duplication, but not composition? 19:06:15 wait, no, just pause 19:06:17 19:04 is it really boring? 19:06:18 sysrq is on prtsc as usual 19:06:22 shachaf: I didn't see it first time 19:06:38 Since you're an expert in substructural types and all. 19:06:43 shachaf: composition corresponds to the Application rule of first-order logic, right? 19:06:52 ais523, my desktop keyboard also lacks a right windows key. Even though it is a microsoft keyboard. heh 19:06:52 <\oren\> my prtsc is between alt and ctrl 19:07:03 so there's no way to eliminate lambdas at this point, so Abstraction is pointless 19:07:07 ais523, it has a fn too, between altgr and right ctrl 19:07:12 <\oren\> where the right windows key should be 19:07:17 I'm not sure. 19:07:21 and there's no point in forming tuples because there's no way to destructure them 19:07:22 Although it seems that I might have to send the application form, because apparently it loses the magical powers of binding if it goes through a scan-and-print step. 19:07:25 ais523, and "menu" is on fn-right shift 19:07:26 At least it's not signed in blood. 19:07:29 thus everything is happening on the left of the turnstile 19:07:30 ais523, which is super-weird 19:07:49 What's the Application rule? 19:07:53 what you end up with is a definition of a set, I think 19:08:04 I haven't asked if they'd accept a faxed copy, because (AIUI) fax machines often preserve the magic. 19:08:15 fizzie, AIUI? 19:08:21 As I understand it. 19:08:21 <\oren\> My keyboard does have an excellent "lock" button (trust a business laptop to have a builtin boss key) 19:08:26 right 19:08:36 fizzie, go ask them that ;P 19:08:43 fizzie, then find a fax machine 19:08:46 Application rule derives, from (G |- M : t' -> t) and (D |- N : t'), (G,D |- MN : t) 19:08:48 <\oren\> it has also a ""new web browser window key 19:09:04 it normally ends up corresponding to categorical composition, if you use a category for your semantics (most people do) 19:09:22 Vorpal: We've got a fax-sending service in the intranet, but maybe that would make the spirits angry. 19:09:30 but you have to use the closed structure to make t' point in the opposite direction 19:09:33 I guess you represent t' as 1 -> t' or something like that? 19:09:44 shachaf: not in first-order logic, you don't 19:09:55 In the category you use for your semantics. 19:09:59 fizzie, you what? 19:10:03 fizzie, heh 19:10:04 the normal categorical method is to represent the entire judgements as G -> (t' -> t), D -> t' 19:10:12 and (G,D) -> t 19:10:16 Ah. 19:10:31 bleh, I wish I could link to my thesis where all this is explained 19:10:36 I guess I can read http://etheses.bham.ac.uk/5741/1/Smith15PhD.pdf until your thesis comes out. 19:10:38 (except with G and D in Greek) 19:10:58 haha, that's a thesis by someone else called Smith, I take it? 19:11:04 also it has way too many pages 19:11:08 Vorpal: I haven't looked at it, I just saw some mention somewhere that you can just "go/fax" if you have to communicate to someone stuck in the past. 19:11:24 There was also something a bit more elaborate if you need to "regularly receive faxes". 19:11:32 hah 19:11:35 fizzie: clearly I have Feather on my mind, I thought that communicating with someone stuck in the past might change history 19:12:54 (although in Feather, changing history can change the present, and in fact probably /should/ or else you'll get into a time loop) 19:13:20 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 19:13:48 -!- TieSoul has quit (Read error: Connection reset by peer). 19:14:21 -!- AnotherTest has joined. 19:14:22 Is Feather any closer to actually existing than it has used to be? 19:14:32 no 19:14:35 well slightly 19:14:48 I've realised that The Underlambda Project might inadvertently remove one of the many obstacles to its existence 19:15:00 Tens of billions of people are stuck in the past. 19:15:36 whoever put that {{stub}} on the Esolang article about Feather is a genius, btw 19:15:37 Sir Thomas More, for example. 19:15:57 He is currently in the past, with no way of getting to the present. 19:16:24 I think it's up to a hundred billion. 19:16:27 We should help him. 19:18:59 Let's help him! Quick, to the fax machine! 19:21:31 -!- Alcest has quit (Remote host closed the connection). 19:26:25 ais523: I sent them an email. 19:26:52 I'm curious to know what their answer (if any) will be 19:27:05 although given that it's Saturday, we'll have to wait a few days because I seriously doubt they reply to emails on weekends 19:28:21 whose answer? 19:28:22 Is there a reason you can't publish a copy directly, by the way? 19:28:24 what? 19:28:48 shachaf: I'm not sure 19:28:53 wait, you're sending messages to the past? 19:28:55 and so far, haven't done so in case I get into trouble 19:29:37 can you send one to Ben Franklin please/ 19:29:48 the paper copies don't seem to be publicly available either (although the reason for that might be that /both/ of them are currently undergoing refurbishment) 19:30:23 b_jonas: the hard part is, that to prevent terribly bad things happening a message sent to the past has to have no visible effects until to just before you choose to send the message, at which point something happens to stop you sending the message 19:30:42 this is quite hard to achieve, and one of the currently unsolved problems blocking Feather from existing 19:33:38 ais523, at that point, it appears you can't use the "sending message to past" for anything useful? 19:34:18 Vorpal: well, you can, say, check to see if an object has a method, if it doesn't, you can retroactively add it 19:34:35 it won't make any visible effects to the past because as the object didn't have the method, clearly nothing called it 19:34:48 but your check will see that the method is already there 19:35:13 ais523, heh 19:35:18 well, bbl 19:37:08 * shachaf . o O ("What's expected of us" by Ted Chiang in 2005-07-07 at ) 19:37:23 ski: did i do it right twh 19:43:35 ais523, b_jonas what were the rules to that tiny Magic variant you were talking about the other day? 19:43:53 Taneb: there were actually two such variants 19:44:00 which are similar 19:44:25 ais523: I see 19:44:26 the one I suggested was: you have three cards in your library (Vintage banlist, no restricted list), you don't lose for drawing from an empty library, otherwise normal Magic rules 19:44:55 b_jonass was: you have five cards in your library, Legacy banlist, starting hand size is 0, during mulligans you can reorder your library, otherwise normal Magic rules 19:44:58 *b_jonas's 19:45:16 ais523: yes, and default maximum hand size is still 7. 19:46:30 right, this is starting hand size as in Vanguard 19:46:48 b_jonas: come to think of it, we could call your format Doomsday 19:47:12 because it's almost literally the situation you're in after resolving Doomsday (five cards of your choice in any order) 19:48:31 -!- [1]blurelIse has joined. 19:48:43 There's also a well-known tiny _limited_ magic variant, where in the first duels, each player starts with a single sealed booster pack that he has to shuffle to his library with ten basic lands, without looking at the cards first, then it's a single elimination tournament and the winner of the match in the first round gets to use a deck from both boosters in that match. 19:48:54 I think that one is called "duel masters", but I'm not sure. 19:49:19 Hmm, I only ever played a one-game variant of that. 19:49:30 b_jonas: no, duel masters is a different game Wizards makes 19:49:35 ais523: that may have been what gave the original idea, I don't know 19:49:43 there is a name for that format but I can't remember what it is 19:49:46 oh right, Pack Wars 19:49:51 ok 19:50:14 shachaf: that predictor is a devious device 19:50:48 -!- blurelIse has quit (Ping timeout: 264 seconds). 19:50:48 -!- [1]blurelIse has changed nick to blurelIse. 19:52:23 I also make up a variant called the "Standard Solitaire" variant; it is a solitaire variant where the only change to the standard rules is that rule 104.2a is not applicable. Try to make up a very strange deck in order to make a more interesting kind of game with this variant; add cards that damage yourself, cards that have unusual requirements, as well as cards with effects that just let you to win the game. 19:52:32 We would call them "pack battles" which I guess explains why there was only one game. 19:52:50 Also eventually we replaced most words in the game with the word "jam". 19:52:54 !104.2a 19:53:06 `comprules 104.2a 19:53:07 ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: comprules: not found 19:53:32 "jam" originally (well, in the context of Magic: The Gathering) meant "+1/+1 counter" 19:53:40 Later on it meant any kind of counter. 19:53:44 "A player still in the game wins the game if all of that player<92>s opponents have left the game. This happens immediately and overrides all effects that would prevent that player from winning the game." 19:53:49 that's what 104.2a says 19:53:53 It also grew to mean "card" and "booster pack" 19:54:01 Giant Growth would jam up a card until end of turn. 19:54:08 Playing Magic: The Gathering was referred to as "jamming" 19:54:13 do we need a comprehensive rules bot? 19:54:26 Wouldn't hurt. 19:54:30 ais523: there is one on the efnet/#mtgrules channel 19:54:41 ais523: but we could easily teach the rules to HackEgo 19:54:41 I used to be in that channel. 19:54:49 I have recently been having problems connecting to EFnet 19:54:54 "Your Rules Advisor Membership is about to expire on 2015-10-12" 19:54:56 Should I renew it? 19:55:02 ais523: there's just one problem, some rules are too long to fit in an irc line 19:55:10 -!- ais523 has quit (Quit: dinner). 19:55:25 ais523 always leaves very abruptly. 19:55:35 but we could add a `more command to HackEgo to list continuations with a second command I think 19:55:47 lambdabot has that 19:56:06 But adding it to HackEgo in the obvious way would involve making hg commits for read-only commands. 19:56:12 Which would be kind of annoying. 19:56:16 shachaf: it's easy to add to HackEgo too, with a short script to bin 19:56:28 shachaf: doesn't it have some non-version-controlled temp directory? 19:56:37 That survives across reboots? 19:56:42 no 19:56:44 dunno 19:56:45 whatever 19:56:49 -!- function has joined. 19:57:00 Maybe we can use one of the bugs we know of that causes that behavior. 19:57:01 shachaf: that's how buubot worked too by the way, it could only store versioned data 19:57:22 I still stored state in versioned data though 19:57:29 with macros 19:57:47 It would be nice to have a non-versioned directory for things like this. 19:57:56 Though people would probably abuse it. 19:58:31 why would they? 19:58:57 `culprits wisdom/culprit 19:58:59 int-e ais523 badger badger badger badger badger badger badger badger mushroom mushroom 19:59:01 ais523.................. 19:59:13 I mean, how would they abuse it more than they already abuse the versioned one? 19:59:31 int-e int-e int-e int-e int-e int-e int-e int-e oerjan oerjan 19:59:45 um, the non-versioned dir wouldn't be in the path, so you could see if something points there 19:59:51 I mean in the PATH 20:00:00 shachaf: yes yes yes yes ...?! 20:00:19 int-e: just fitting the rhythm hth 20:00:24 `? culprit 20:00:25 ​`culprits` is a program that lists the lists the nicks responsible for a wisdom entry. Usage: `culprits wisdom/ENTRY 20:00:43 `culprits bin/culprits 20:00:44 Jafet Jafet tswett tswett shachaf shachaf shachaf FireFly FireFly FireFly FireFly FireFly FireFly shachaf 20:01:20 `culprits wisdom/culprits 20:01:22 No output. 20:01:29 `culprits wisdom/culprit 20:01:31 int-e ais523 badger badger badger badger badger badger badger badger mushroom mushroom 20:02:01 cute 20:02:07 `? culprit 20:02:08 ​`culprits` is a program that lists the lists the nicks responsible for a wisdom entry. Usage: `culprits wisdom/ENTRY 20:02:24 `? firefly 20:02:25 FireFly was a short-running but well-loved sci-fi TV series released in 2003, starring Nathan Fillion and directed and written by Joss Whedon. 20:02:27 lists the lists the lists the lists the lists the lists the lists the lists the wisdom wisdom 20:02:35 ... wow, a fact 20:02:59 `culprits wisdom/firefly 20:03:01 int-e ais523 oerjan elliott Bike FreeFull Tanea 20:03:37 `? ais523 20:03:38 Agent “Iä” Smith is an alien with a strange allergy to avian body covering, which he is trying to retroactively prevent from ever evolving. On the 3rd of March, he's lawful good. 20:04:09 `culprits wisdom/ais523 20:04:11 int-e ais523 oerjan elliott Bike FreeFull elliott Sgeo Bike oerjan Taneb ais523 ais523 elliott oerjan elliott FreeFull oerjan oerjan oerjan oerjan FreeFull shachaf shachaf nitia 20:04:21 `? I 20:04:22 i love monoids 20:04:23 shachaf: The existing bin/paste already kind of makes commits for "read-only" commands. 20:04:30 `? prim 20:04:31 prim? ¯\(°​_o)/¯ 20:04:32 `? prime 20:04:33 fizzie: Yes. 20:04:33 prime? ¯\(°​_o)/¯ 20:04:35 `? factor 20:04:36 factor? ¯\(°​_o)/¯ 20:04:37 Is the part about retroactively preventing it from evolving a reference to that one esolang that shouldn't be named? 20:04:38 `? binomial 20:04:39 binomial? ¯\(°​_o)/¯ 20:04:45 fizzie: But at least when you use it you're expecting that to happen. 20:05:02 fizzie: When I upload some text to a paste site, I also expect to be modifying that site. 20:05:02 `wisdom 20:05:04 ​⌨/You are probably using one right now! 20:05:10 FireFly, you mean Feather? Ais talked about it himself today 20:05:17 Vorpal: oh. 20:05:20 `unidecode ⌨ 20:05:21 ​[U+2328 KEYBOARD] 20:05:30 that is why he left abruptly 20:05:36 Possibly 20:05:57 isn't it Brainfuck that we want to retroactively cause not to be evolved? 20:06:12 no 20:06:17 just its derivatives 20:06:18 b_jonas, I don't mind brainfuck as much as all the clones 20:06:32 Most of them are just boring 20:06:39 ook. 20:06:47 That is the classical pointless one 20:07:08 Vorpal: well sure, and I don't mind the _first_ stem of weed in my garden as much as all the rest it spawns, but that's where you have to start extermination 20:07:19 bitfuck looks kind of interesting at first, until you realise there is a simple recipe for translating any brainfuck program into a bitfuck one 20:07:28 of course ook's creator has redeemed himself 20:07:33 what? no! Ook! is the best BF clone, one of the very few non-pointless ones 20:07:39 Oook! Oook! 20:07:40 int-e, oh? 20:07:50 Ook. Ook? 20:07:51 Vorpal: DMM is also the creator of Piet 20:07:54 int-e, ah 20:08:01 int-e: and of Chef, more usefully 20:08:20 b_jonas, really? It is just plain boring. And no, the Discworld reference doesn't make it any better 20:09:26 Ok, points... It does feature periods and exclamation marks. 20:09:39 And question marks I guess. 20:20:38 -!- sc00fy has quit (Ping timeout: 246 seconds). 20:37:29 -!- ais523 has joined. 20:38:31 -!- JesseH has joined. 20:39:57 -!- FreeFull has quit (Ping timeout: 256 seconds). 20:40:27 -!- FreeFull has joined. 20:50:24 -!- Thisbe has joined. 20:50:26 fizzie, there? 20:50:45 I'm currently trying to understand this program the fuzzer generated: &ITR"\ <80>$59kbfkkk?^@kkk^MQ( o 20:50:59 It sometimes crashes, sometimes hangs 20:51:16 Note that <80> and so on are non-printable 20:51:21 And so is ^@ 20:51:25 and ^M 20:51:32 (this is how less displays it) 20:52:23 I'm utterly confused how it even managed to get into the k code. It looks to me as if & and o should both just reverse... (Since o is disabled for safety reasons) 20:54:11 Oh wait that is ^M, of course... 20:56:40 Well that program is pure evil... 20:59:57 Vorpal: run it through cat -v 21:00:02 and fuzzing programs tend to be 21:00:09 I'm amused that that fuzzer uses so many k's 21:00:21 ais523, yes 21:01:04 ais523, I'm ... not sure I will even try to fix that. The k code is way too complicated... Especially the handling of nested k 21:01:54 ais523, I don't even understand the specifics of how it crashes, but it seems related to ? generating a specific set of < and > but never any ^ and v presumably 21:02:23 Vorpal: what about using a determinstic RNG seed? 21:02:27 I should maybe patch ? to be deterministic yeah 21:02:32 also, that program seems to be a one-liner 21:02:39 so ^ and v would just loop back round to the ? 21:02:42 ais523, no it isn't because of the embedded ^M 21:02:47 ah right 21:02:48 ais523, which is a CR iirc 21:02:52 there's still nothing in the column containing the ? 21:03:06 ais523, also k on a movement doesn't apply that on itself possibly 21:03:19 due to f*** k 21:03:57 -!- gamemanj has quit (Ping timeout: 265 seconds). 21:05:20 ais523, I think it possibly applies at some nesting of k or other, but not at the top k nor ? itself 21:05:42 ais523, still I apparently got an stack overflow of nested k calls 21:05:51 Somehow 21:10:13 Vorpal: what happens with a program that's just a single k? 21:10:28 hm, 21:10:36 AFAICT the /intended/ behaviour there is a stack overflow 21:11:08 -!- Thisbe has quit (Quit: Thisbe). 21:11:13 ais523, nothing 21:11:24 ais523, since that pops 0 iterations from the stack and runs k zero times 21:11:53 Vorpal: oh, I see 21:12:06 you'd somehow need an infinite number of non-zeroes on the stack 21:12:09 first 21:12:13 yes 21:12:23 with nested k it is a bit easier 21:12:26 what about ykk? 21:12:39 Well y will put *something* on the stack certainly 21:12:51 But I think it will just end up running y a lot 21:12:53 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 21:13:10 I'm trying to work out if you get a self-sustaining loop of nested k 21:13:21 `! funge98 ykk 21:13:22 ​/hackenv/bin/!: 4: exec: ibin/funge98: not found 21:13:34 `! befunge ykk 21:13:35 Unsupported instruction 'y' (0x79) (maybe not Befunge-93?) \ Unsupported instruction 'k' (0x6b) (maybe not Befunge-93?) \ Unsupported instruction 'k' (0x6b) (maybe not Befunge-93?) \ Unsupported instruction '' (0xffffffff) (maybe not Befunge-93?) \ Unsupported instruction 'y' (0x79) (maybe not Befunge-93?) \ Unsupported instruction 'k' (0x6b) (may 21:13:45 huh, it put an EOF on the playfield 21:13:45 ais523, I think it involves ? to move the IP around between runs alas 21:14:08 Vorpal: the ? is redirecting the k into itself, I bet 21:14:17 hm 21:14:20 think about a program along the lines of k^ (in a oneliner) 21:14:35 (I don't bet very much because randomly generated Befunge is hard to mentally parse) 21:15:07 ais523, you need to get ? to move it to a k for the next loop one level up 21:15:11 I think 21:15:28 So you need at least 3 k to start with 21:15:34 I *think* 21:15:40 this code is super gnarly 21:15:57 Hm that is an English word isn't it? The spell check disagrees 21:16:08 "gnarly" that is 21:16:32 Vorpal: it's pretty colloquial 21:16:56 it wouldn't surprise me if it didn't turn up in the spellchecker's data sources 21:17:02 which are maybe more formal 21:17:08 right 21:18:07 ais523, anyway I think only cfunge implements nested k at all. Deewiant basically said "**** this" and ignored it, since the standard says nothing about it, and there is no obvious "good" way to handle it 21:18:08 iirc 21:18:35 well yes, it's clearly ridiculous :-) 21:18:50 Underload has something similar, but it's uncontroversial that if you chain it indefinitely you get a stack overflow 21:19:00 -!- Phantom_Hoover has quit (Ping timeout: 250 seconds). 21:19:06 ^ul (:*:^):^ 21:19:06 ...too much stack! 21:19:12 wow, that was /quick/ 21:19:52 `! ul (:*:^):^ 21:19:53 ​/hackenv/bin/!: 4: exec: ibin/ul: not found 21:19:56 `! underload (:*:^):^ 21:20:05 that was less quick 21:20:09 maybe derlo has higher limits 21:20:21 or maybe it just doesn't print error messages, that seems more likely 21:20:57 right 21:21:01 ugh, this file has troll indentation 21:22:15 `` echo (:*:^):^ | ibin/underload -a -d3 21:22:16 bash: -c: line 0: syntax error near unexpected token `:*:^' \ bash: -c: line 0: `echo (:*:^):^ | ibin/underload -a -d3 ' 21:22:23 `` echo '(:*:^):^' | ibin/underload -a -d3 21:22:24 Attempt to execute unknown command 45 21:22:28 hmm 21:22:37 `` echo '(:*:^):^' | ibin/underload -a -d3 /dev/stdin 21:22:38 Attempt to execute unknown command 45 21:22:42 oh right 21:22:45 `` echo '(:*:^):^' | ibin/underload -o -d3 /dev/stdin 21:22:46 Attempt to execute unknown command 45 21:22:50 hmm again 21:23:08 `` echo '(Hello, world!)S' | ibin/underload -o -d3 /dev/stdin 21:23:09 Attempt to execute unknown command 45 21:23:13 `unicode 45 21:23:14 E 21:23:25 but there isn't a capital E in the program 21:25:42 ais523, hm even with a fixed srandom() it appears there is still some variability in the program 21:25:54 Since the fuzzer reports it 21:26:01 (it is an fuzzer that instruments the program) 21:26:09 (for more educated guesses) 21:26:24 american fuzzy lop, presumably? 21:26:27 Yes 21:26:27 no wonder there are so many k 21:26:31 ais523, hah :P 21:26:40 ais523, you used it? 21:26:53 no, but I've read about it 21:27:00 including the mathematical basis behind how it works 21:27:07 ais523, it actually hasn't found any crashes yet, I only got that crash when investigating a hang and it crashed one of the times I ran it 21:28:13 -!- Phantom_Hoover has joined. 21:28:41 Oh I think I see 21:28:54 Confusing #ifdef for arc4random overriding it 21:33:28 @wn gnarly 21:33:29 *** "gnarly" wn "WordNet (r) 3.0 (2006)" 21:33:29 gnarly 21:33:29 adj 1: used of old persons or old trees; covered with knobs or 21:33:29 knots; "gnarled and knotted hands"; "a knobbed stick" 21:33:29 [syn: {gnarled}, {gnarly}, {knotted}, {knotty}, 21:33:31 {knobbed}] 21:34:07 yes 21:34:13 that is what the k code is 21:34:25 A person and/or a tree. 21:34:43 I guess most code is a (parse) tree. 21:34:45 it's normally used as a metaphor, I think 21:35:00 Really? Still variable? How? 21:35:11 Well, less than before at least 21:35:22 I guess there could be some time thingy 21:36:27 "sH{wt 21:36:30 this apparently hangs 21:36:36 hm 21:37:06 ais523: The fungot ^ul time limits are in terms of Underload "operations", so you can have some pretty slow programs, e.g. if you push two strings so that the stack is almost to the limit, and then loop ~. 21:37:06 fizzie: just a minor reference here. on the flip side of the pond fnord hence the name monk fish) than the professional records? fnord fnord 03:38, 18 may 2006 ( utc 21:37:20 -!- Patashu has joined. 21:37:34 ^style 21:37:35 Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp* youtube 21:37:51 You can tell from the timestamp. 21:38:05 Oh yes, t of course 21:41:12 ais523, afl is particularly fond of loading HRTI for some odd reason 21:41:59 (I gave all the fingerprints loading as input, since finding valid fingerprints might be hard) 21:42:20 "ITRH"4(!tNNyyyyyy 21:42:25 hm not sure how that hangs 21:42:33 Oh wait, t againb 21:42:38 and N will of course reflect 21:42:45 I was going to ask, what does N do? 21:42:47 I should probably disable t 21:42:55 ais523, nothing in HRTI it appears 21:44:04 my head is hurting trying to figure out what yyyyyy does 21:44:09 apart from push a bunch of garbage to the stack 21:49:00 ais523, well that is what it does, though not all of it, since y is also pick 21:49:35 Vorpal: I know 21:49:44 Yeah I'm running non-concurrent now, since I got way too many "unique" t related hangs that weren't really unique from my point of view 21:50:52 ais523, I patched in a instruction count limit, and "find next instruction" cycle limit (code doing p at super large coordinates then cfunge iterating forever to get to that point) and so on 21:51:02 since those hangs are just the program doing stupid stuff 21:51:07 -!- sc00fy has joined. 21:51:19 I guess that is the downside, finding *interesting* hangs is hard when fuzzing an interpreter 21:51:23 yes 21:52:33 ais523, it is not like I haven't fuzz tested before though, so some of that stuff existed, but then I did it with alarm(), but that seems less compatible with afl to some extent 21:53:08 And that fuzzing was just generating random programs and feeding it into cfunge 21:54:22 ais523, any idea how to tunnel SSH through a proxy that fakes SSL cert btw? 21:54:33 as in, it doesn't just pass HTTPS through 21:54:38 it MITMs 21:54:41 Vorpal: what sort of proxy? http only? 21:54:47 ais523, http and https yes 21:54:52 but the HTTPS one MITMs 21:55:06 I guess you use a standard ssh-over-http, then (not sure where you get those from) 21:55:18 the s in the https will just make things awkward because the mitmness will mean the certs won't verify 21:55:35 indeed 21:56:12 ais523, I used to ssh over 443 because it didn't MITM in the past and thus anything on 443 just went through 21:56:29 But now I need to change tactic 21:56:59 I have an acquaintance in another channel who keeps telling people to tunnel over DNS 21:57:04 but while possible, that's really slow 21:57:31 dear god 21:57:44 ais523, also wouldn't it saturate the DNS caches? 21:58:06 I assume you have a special authorative DNS server that listens for requests and returns other ones 21:58:10 I have also had idea before to tunnel stuff over DNS (just data though, not interactive sessions) in order to be able to use it from hotels and stuff where you can access DNS without needing an account or registration or whatever there 21:58:11 Probably with short TTL 21:58:47 Vorpal: I think you communicate one way using the exact URL being requested (that's a subdomain of a domain whose authoratitive DNS server you control) 21:58:51 ais523, since I can't make direct DNS requests there, only through the DNS server DHCP provides. I could possibly tunnel over ping because that goes through unhindered 21:58:56 and the other way using the IP addresses, TXT fields etc. being returned 21:59:03 ais523, my idea was tunneling SSH over websockets somehow 21:59:13 also tunnel over ping has almost certainly been done already 21:59:24 ais523, seems even worse 22:00:31 ais523, anyway, figuring out a way to tunnel over http seems easier, the question is if it is compatible with running nginx 22:00:47 I guess I could do it with a virtual host or something 22:00:53 websockets is probably the sanest method, to be fair 22:01:21 See if the server supports using CONNECT, just make if it receive CONNECT first then it can be used like SSH after that for the current session, might be one way? 22:01:38 ais523, can't find any ready-made software for it though 22:02:07 zzo38, pretty sure the new one doesn't, though I could test a bit more 22:02:10 -!- hppavilion[1] has joined. 22:02:57 zzo38, I know you can only connect to port 80 and 443 with HTTP and HTTPS respectively. You can't reach, say. a server on port 8080 on the internet 22:06:01 "PSD3"4(??????kkkkkkkkkkkkkkkkkkkkkkkk???????ӈ^@^H . 22:06:04 ais523, ^ 22:07:03 ӈ isn't even a real command 22:07:16 and what does 3DSP do anyway? 22:07:28 ais523, vector maths iirc 22:08:12 it is an RCFunge one 22:08:23 vector and matrix operations 22:09:12 "ESAB"4( N 22:09:16 It's a pretty strange one. 22:09:17 I wonder why N from BASE hangs 22:09:23 fizzie, 3DSP? 22:09:27 Well yeha 22:09:29 yeah* 22:09:30 Yes. Sort of very arbitrary. 22:09:36 what's N from BASE? 22:09:42 ais523, print in base 22:09:52 maybe it's to do with how I/O is connected? 22:09:57 or maybe it's trying to print infinitely many zeroes? 22:10:12 Hm 22:10:23 Oh 22:10:29 ais523, it is printing in base 1 22:10:34 of course 22:10:39 since ( leaves the loaded fingerprint 22:10:44 as 1 22:10:47 on the stack 22:10:57 ah, so it's printing a very large numer in unary 22:11:03 if (base == 1) { 22:11:03 while (val--) 22:11:03 cf_putchar_unlocked('0'); 22:11:03 cf_putchar_unlocked(' '); 22:11:05 yep 22:11:15 and that is something that's impossible to do quickly 22:11:36 Well, you could perhaps use vmsplice()? 22:11:41 But no not really 22:11:43 gk{ 22:11:47 Hm 22:12:00 No I don't see why that would hang 22:12:04 huh, I learned a new syscall today 22:12:09 Though the idea of k on { is annoying 22:12:11 but it only works on pipes 22:12:14 ais523, which one? 22:12:17 vmsplice 22:12:20 ah 22:12:25 ais523, well there is splice too 22:12:53 ais523, it is all going for that fabled zero-copy-IO 22:12:53 I knew that one, though 22:13:33 SPLICE_F_GIFT looks abusable 22:14:01 really it should probably be specified to munmap the given pages… 22:14:22 ais523, wasn't there a root exploit using vmsplice soon after it was introduced? 22:14:28 I don't know 22:14:35 Either vmsplice or splice 22:15:56 ais523, here are the hangs so far http://sprunge.us/bQJU 22:16:13 We can assume that ^ is from cat -e I think 22:16:45 And some of them I don't know why they hang, possibly my iteration limit is just too close to the actual time limit 22:17:53 -!- Sprocklem has quit (Quit: [). 22:17:54 -!- atrapado has quit (Quit: Leaving). 22:19:23 that double M-2 is confusing 22:19:35 and that ! on its own even more so 22:19:59 -!- Sprocklem has joined. 22:20:09 ais523, yes. I think it is just that the execution count is too close to the 20 ms limit in some cases 22:20:18 execution count limit* 22:20:29 My λ-calculator is coming along nicely 22:20:39 Currently setting up the AST evaluation functions 22:20:43 *methods 22:25:43 night 22:25:55 It's interesting that whenever I start talking, everyone goes silent and/or leaves xd 22:25:57 *xD 22:26:16 hppavilion[1]: could just be timezones 22:26:23 Fair enough 22:26:27 OR no one likes me xD 22:26:47 it could be just that this channel isn't very noisy anyway 22:27:02 It isn't, that's true 22:27:16 Should I publish what little I have for LIME so far? 22:27:27 maybe we prefer concise esolangs (like burlesque) over verbose ones (like Shakespear) 22:27:35 Or should I wait until I have at least one working thing before I publish to GitHub? 22:33:43 oh, this reminds me 22:33:51 I should push what I've done on The Underlambda Project so far somewhere public 22:33:55 I was going to hours ago but forgot 22:34:29 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds). 22:35:53 here: darcs clone http://nethack4.org/esolangs/the-underlambda-project 22:36:01 for anyone who wants to see how it's progressing so far 22:44:09 hppavilion[1]: I think you can publish what you have so far if you want to do so 22:44:20 zzo38: OK 22:44:23 I'm adding: 22:45:16 * λ-expressions 22:45:16 * Named Expressions 22:45:16 * Operators 22:45:16 * Anything else? 22:45:33 I don't know, that may be good enough so far I suppose 22:51:26 ais523: fwiw, linux doesn't currently implement SPLICE_F_GIFT (this is documented for the page-stealing half of SPLICE_F_MOVE of the splice system call) 22:51:36 int-e: oh good 22:51:43 then it doesn't matter that its semantics are insane 22:51:58 they can always be fixed when it's implemented 22:54:02 hppavilion[1]: a let or letrec form, some sort of numbers with built-in arithmetic and comparison primitives, possibly mutable cells or even mutable bindings (variables) created by the lambdas, some basic IO operations, a repl 22:55:23 b_jonas: It's a lambda calculus playground. You kind of have to define your own numbers and arithmetic 22:56:24 hppavilion[1]: you still can do that if you want 22:56:37 but it's useful to have them built-in 22:56:38 I can, but I kind of want to make users define their own things xD 22:57:28 do you want people to even have to transform let and letrec to lambda expressions by hand? 22:57:33 or will you at least add that 23:01:40 hppavilion[1]: ok, let me ask differently. if I want a lambda calculus calculator, what extra would this give compared to any other interpreter that supports lambda expressions, such as a perl/ruby/scheme etc interpreter? 23:02:03 (assume a not very old version of perl; old versions have a cripping bug if you try to use them as a lamdba calculus interpreter) 23:02:34 ais523, hi 23:02:37 hppavilion[1]: Nothing at all. It's just a simple program for people to do lambda calculus in. 23:02:40 went back to computer for a moment 23:02:45 it had found one crash 23:02:46 http://sprunge.us/FRGO 23:02:50 hppavilion[1]: ok. 23:02:54 after minimizing 23:03:12 ais523, doesn't crash the non-fuzz build though :/ 23:03:31 hppavilion[1]: does it do some sort of reference counting or garbage collection to reclaim memory consumed by functions and bindings? 23:04:14 ais523, clearly related to this though: 23:04:16 * In k: iteration: 1128597731 instruction: y (121) 23:04:16 * In k: iteration: 1128597730 instruction: y (121) 23:04:16 * In k: iteration: 1128597729 instruction: y (121) 23:04:20 b_jonas: I might make it do that later, but it's a python-based system. So mostly no, unless python has some pretty advanced built-in garbage collection 23:04:52 um, python does have reference counting and even advanced garbage collection 23:04:58 OK 23:05:01 so that's probably a yes 23:05:08 Then yes, it does that 23:05:13 though you can write the interpreter in a bad way that doesn't let it work of course 23:05:20 True 23:05:50 will look at that tomorrow or next weekend (rather busy with real life tomorrow) 23:05:51 it's tricky, you have to make sure to hold references to only those bindings that are really needed 23:06:41 hppavilion[1]: no, it's called true, that's a built-in expression; True is an old symbol kept from compatibility back from python 1 that you shouldn't use. 23:07:05 true is \x.\y.x :-P 23:07:20 Wait, what? 23:07:31 ais523: correct 23:07:34 Vorpal: now I'm trying to figure out how to get onto the second line 23:07:51 presumably you need the [ to run where the y is currently (using offset k( 23:07:53 *offset k 23:07:53 ais523, well, [ 23:08:04 the [ can send you past the second line 23:08:17 but you can't run along it unless it runs on the second line somehow 23:08:20 but perhaps the second line's a red herring 23:08:38 ais523, I will need to make a debug build with fuzz testing enabled to debug this tomorrow 23:09:03 -!- oerjan has joined. 23:09:34 ais523, well it was still there when minimizing, and the minimizing turned it from this: http://sprunge.us/PHTZ 23:09:36 -!- evalj has quit (Remote host closed the connection). 23:10:44 ais523, so clearly the minimising did *something* 23:12:04 going to bed for real now 23:13:41 -!- Phantom_Hoover has joined. 23:15:06 ais523, this is an ingenious case of variable behaviour though: "ITRH"4(G( 23:15:15 Now going to bed for really real 23:15:18 night 23:15:28 also that program is beautiful :-) 23:15:36 all it's missing is another capital letter 23:15:43 ais523, why 23:16:02 ais523, it uses the computed timer resolution to determine how to load the next fingerprint 23:16:15 All the other variable cases also use HRTI 23:16:19 But much messier 23:16:33 Vorpal: I mean, so that the newly loaded fingerprint actually runs 23:17:02 http://sprunge.us/OESc 23:17:12 ah 23:18:33 ais523, it probably evaluated to 1 or 2, which trigger different code paths in cfunge 23:20:25 anyway I thought you were going to bed :-P 23:25:41 -!- function has changed nick to trout. 23:44:11 -!- boily has joined. 23:52:27 Right now I'm learning how to program from John Von Neumann and it is exhilarating 23:56:43 watch out for brain explosions 23:57:35 goerjanod evening. 23:58:25 bood evenily 23:59:12 -!- sc00fy has quit (Ping timeout: 255 seconds).