00:09:29 -!- S1 has quit (Quit: S1).
00:10:48 -!- qwertyo has quit (Ping timeout: 240 seconds).
00:11:23 -!- boily has quit (Quit: EXQUISITE CHICKEN).
00:24:11 <int-e> oerjan: Hint, he's saying Perl because nobody would admit to writing mission critical software in PHP.
00:24:32 <CrazyM4n> As if the universe is mission critical
00:25:23 <int-e> Good point, we're in no position to tell.
00:26:10 <CrazyM4n> Hell, it's probably slapped together in BF or something as an example of cellular automata
00:40:15 <oerjan> if the gods are so alien they think BF is a good language for slapping things together we're all doomed anyway
00:43:02 -!- AndoDaan has joined.
00:43:09 <MDude> Universes as a target for code golfing.
00:43:55 <CrazyM4n> I got this life sustaining universe down to 1,300,000 chars!
00:44:22 <oerjan> i think that's a bit bloated
00:45:52 <oerjan> see, if you use complex numbers instead of probabilities you can cut it down to just 14,000 and the inhabitants will never notice unless they start probing ridiculously small distances
00:46:04 <int-e> the trick is to keep your code simple and plant all future behaviour in the ether.
00:46:20 -!- mihow has quit (Quit: mihow).
00:48:09 <int-e> oerjan: did you see the third Haskell entry to Euclidean Norm? It should make you feel better ;-)
00:48:27 <olsner> fungot: how many characters are you?
00:48:27 <fungot> olsner: finns are generally very suspicious about all new ideas, have to admit that i didn't need to see them
00:48:32 <CrazyM4n> You really have to exploit the emergent behavior of some simple quantum rules, let it run long enough and you can get a good seed
00:53:31 <oren> hey, did you know that matlab doesn't have a good way to programmatically save a diagram as an image? I sure love matlab!
00:54:01 <oerjan> today's NSFL: wikipedia main page
00:54:49 <olsner> "Gross pathology ..." indeed
00:58:47 <oren> so in other words i have to draw rectangles pixel by pixel.
00:59:11 <oren> might as well write a brainfuck program to do it
01:07:24 <int-e> oerjan: is that L = life?
01:07:53 <oerjan> it's not like i invented the acronym...
01:08:25 <int-e> I'm extrapolating from NSFW, and you're quicker than google.
01:08:25 -!- shikhout has joined.
01:11:40 -!- shikhin has quit (Ping timeout: 260 seconds).
01:16:41 -!- dts|pokeball has quit (Read error: Connection reset by peer).
01:29:40 -!- Lilin has joined.
01:30:25 <oerjan> not reddit, that's for sure
01:30:39 <int-e> Lots of stars weighing heavy on our souls, whose only saving grace is the inverse square law.
01:31:03 <oren> wow you're a poet!
01:32:11 <oren> Engelina - Walking in the Sky
01:35:38 <Lilin> Makin' my way down town
01:36:55 -!- oerjan has quit (Quit: zzleepy).
01:37:06 <int-e> oren: If I were I'd probably not mistreat the language so much. In this case, I used an adjective instead of an adverb.
01:39:57 -!- TodPunk has quit (Read error: Connection reset by peer).
01:47:00 -!- TodPunk has joined.
01:49:52 <CrazyM4n> So I'm going to try to golf down a befunge interpreter
01:50:41 <CrazyM4n> Ah screw you I already wrote like half of that
01:51:06 <CrazyM4n> Gotta stay on top of the things I start
01:51:30 <oren> sounds good. yeah i have that problem too, i mash out a prototype and never finalize it
01:54:21 <CrazyM4n> http://i.imgur.com/PQZEVYa.png Perl and your weird function names
01:55:24 <oren> yeah... the keywords are unClike too: last? next?
01:55:43 <Lilin> open without death
01:55:59 <oren> i have a cheat sheet somewhere with the differences.
01:56:56 <oren> yeah open or die; appears in a lot of my scripts
01:58:07 <oren> but in C i prefer if(!F=fopen())goto hell;
01:58:15 <CrazyM4n> It's actually a death threat against the file
01:58:55 <Lilin> I find languages to be more evil nowadays
02:01:17 <oren> i also sometimes do things like for(){if()goto heaven;}
02:02:23 <oren> actually imperative languages rarely make it easy to specify a loop with two separate ending conditions leading to different results
02:02:48 <oren> with tail recursion it is easy
02:04:52 <CrazyM4n> It's all easy in functional languages until you realize you're neck deep in code that's neck deep in recursions and category theory that you don't understand why the code uses it
02:06:40 <oren> that is why instead of using functional languages, we steal features and smuggle them into our imperative kingdom in the dead of night.
02:07:01 <zzo38> oren: Well, usually in such cases I don't need to use goto anyways, although sometimes it can help a bit
02:07:15 <zzo38> Especially in C; in BASIC programs I will write GOTO a bit more often than in a C program.
02:08:01 <oren> in C i often use return instead of goto, but it comes to the same thing
02:08:29 <oren> you have control flow that forks in the middle of a loop
02:09:15 -!- CrazyM4n has quit (Quit: brb).
02:12:38 -!- oren has quit (Quit: batteries out bleh!).
02:12:48 -!- olsner has quit (Ping timeout: 240 seconds).
02:53:23 -!- MoALTz__ has joined.
02:57:15 -!- MoALTz_ has quit (Ping timeout: 272 seconds).
03:31:52 -!- dts|pokeball has joined.
03:48:45 -!- mitchs_ has joined.
03:51:55 -!- mitchs has quit (Ping timeout: 258 seconds).
03:59:05 -!- CrazyM4n has joined.
04:06:40 -!- shikhout has quit (Ping timeout: 260 seconds).
04:09:00 -!- adu has quit (Quit: adu).
04:12:14 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
04:14:57 -!- tswett has joined.
04:24:26 -!- MDude has changed nick to MDream.
04:27:13 -!- augur has joined.
04:30:57 -!- olsner has joined.
04:31:51 -!- augur has quit (Remote host closed the connection).
04:32:45 -!- augur has joined.
05:05:07 -!- tswett has quit (Ping timeout: 250 seconds).
05:11:23 -!- Lilin has changed nick to Decensum.
05:13:14 <shachaf> Decensum: Why do you keep changing your nick?
05:13:18 <shachaf> I thought you were going to stop.
05:18:33 -!- qwertyo has joined.
05:20:56 -!- CrazyM4n has quit (Remote host closed the connection).
05:26:10 -!- oren has joined.
05:37:44 -!- qwertyo has quit (Ping timeout: 244 seconds).
05:38:46 -!- nisstyre has quit (Ping timeout: 244 seconds).
05:47:36 -!- b_jonas has quit (Ping timeout: 256 seconds).
05:54:53 -!- b_jonas has joined.
06:08:13 -!- nisstyre has joined.
06:16:10 -!- MoALTz__ has quit (Quit: Leaving).
06:23:24 -!- dts|pokeball has changed nick to illuminaughty.
06:23:38 -!- illuminaughty has changed nick to dts|pokeball.
06:26:04 -!- dts|pokeball has changed nick to marketbot.
06:26:31 -!- marketbot has changed nick to dts|pokeball.
06:28:28 -!- Dr_pattabhi has joined.
06:29:36 -!- ZombieAlive has quit (Remote host closed the connection).
06:29:42 -!- Dr_pattabhi has left.
06:32:24 * oren casts a curse of cursed cursing on matlab and whatever scientist decided that everything had to eb wirtten in it
06:32:47 -!- shikhin has joined.
06:35:11 -!- AndoDaan has quit (Ping timeout: 250 seconds).
06:57:39 <shachaf> Decensum: Please don't do that.
07:05:30 -!- augur_ has joined.
07:06:32 -!- augur has quit (Read error: Connection reset by peer).
07:08:09 -!- shikhin has quit (Write error: Connection reset by peer).
07:09:16 -!- shikhin has joined.
07:23:32 <HackEgo> käämmiksenne kourisimmillasi lämmetsompaamilta reltäni neuvomissani karistäviänne nielillamme kehomollasi näkyvälistä mummanne
07:24:49 <fizzie> Just realized there's a Finnish word that's not *too* colloquial, yet breaks vowel harmony: "tällainen".
07:24:58 <fizzie> (It's a contraction from the compound "tämänlainen".)
07:26:10 -!- Patashu has joined.
07:26:25 <fizzie> http://en.wiktionary.org/wiki/-lainen#Etymology_2 "From compounds with laji (“kind, sort”) + -inen. The origin as a separate word can still be seen in the imperfect accommodation to vowel harmony."
07:57:02 -!- Decensum has quit.
08:05:28 -!- AndoDaan has joined.
08:09:02 -!- atehwa has quit (Remote host closed the connection).
08:10:46 -!- Frooxius has joined.
08:28:59 -!- weissschloss has quit (Max SendQ exceeded).
08:31:17 -!- weissschloss has joined.
08:31:50 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
08:44:45 -!- idris-bot has quit (Ping timeout: 250 seconds).
08:44:57 -!- Melvar has quit (Ping timeout: 264 seconds).
10:04:25 -!- nisstyre has quit (Changing host).
10:04:25 -!- nisstyre has joined.
10:13:53 -!- cluid has joined.
10:30:17 -!- AndoDaan has quit (Ping timeout: 265 seconds).
11:00:45 <HackEgo> [wiki] [[Special:Log/newusers]] create * Domi382 * New user account
11:15:02 -!- boily has joined.
11:15:12 -!- Melvar has joined.
11:20:01 <HackEgo> [wiki] [[GERMAN]] N http://esolangs.org/w/index.php?oldid=41386 * Domi382 * (+289) GERMAN
11:22:52 <int-e> Or should I say ORIGINELL.
11:24:39 <fizzie> Who's going to slap a [[Category:Brainfuck equivalents]] on it, though?
11:25:54 -!- oerjan has joined.
11:26:48 <HackEgo> [wiki] [[GERMAN]] http://esolangs.org/w/index.php?diff=41387&oldid=41386 * 213.162.68.138 * (+36) "Who's going to slap a [[Category:Brainfuck equivalents]] on it, though?" -- Well, who?
11:28:44 <boily> sometimes, there are cues that we are being watched by otherwordly powers who listen to everything we say...
11:30:00 <HackEgo> [wiki] [[قلب]] M http://esolangs.org/w/index.php?diff=41388&oldid=41384 * Ehird * (+10) clarify
11:30:29 <oerjan> boily: why are you mispasting from the other esoteric and why haven't you told us where it is tdnh
11:31:51 <oerjan> boily: also, have you heard about the codu log stealth mode
11:32:15 <boily> the stealth mode. I forgot about it.
11:32:29 <int-e> . o O ( Sorry, Randall, but university homepages are more about securing budgets than about helping prospective students, and even less about helping current students, invited speakers and the like. )
11:32:31 <oerjan> admittedly there is no proof that ip isn't someone on channel.
11:32:55 <oerjan> int-e: wait, there's a new xkcd today?
11:33:16 <fizzie> Stalker mode, I think.
11:33:19 <int-e> No. It's yesterday's.
11:33:53 <oerjan> int-e: i don't think yesterday's xkcd is relevant to that statement...
11:33:58 <int-e> Even worse, I hit the "random" button and didn't realize.
11:34:19 <fizzie> What's up with what-if, anyway?
11:34:20 <int-e> https://xkcd.com/773/ -- sorry.
11:34:54 <oerjan> fizzie: is there a new one now? it's been hiated for a couple weeks afaihn
11:35:45 * int-e is using an advanced stealth mode where IRC host != web browsing host. (Also known as "irssi inside screen on VPS")
11:37:26 <oerjan> wait, someone made GERMAN when there's already German?
11:37:51 <boily> mayber GERMAN is germanier?
11:38:24 <boily> (does German have comparative and superlative like in English?)
11:38:32 <int-e> I don't like the German esolang, but it is more original than simply stating the names of the actions.
11:39:02 <oerjan> did you know that if you visit a page in IE that is just a different capitalization of the one you were coming from, it doesn't add it to history tdnh
11:40:09 <int-e> boily: yes, sure. Unnötig, unnötiger, am unnötigsten.
11:40:44 <boily> oerjan: that's terrifying tdnhaa
11:41:35 <oerjan> boily: in fact german does them in essentially the same way as english, except that it isn't afraid of long words so it rarely does the equivalent of "more/most"
11:41:36 <int-e> (unnecessary, more unnecessary, most unnecessary.)
11:42:26 <oerjan> i think norwegian is somewhere in between
11:43:09 <int-e> `learn int-e is not Swedish.
11:43:11 <HackEgo> Learned 'int-e': int-e is not Swedish.
11:43:26 <int-e> There, that'll help.
11:43:47 <cluid> I had an idea for branfuck
11:44:11 <oerjan> int-e: what have you done neither that nor my previous pun is showing up in the repository
11:44:23 <cluid> we could maybe figure out in a lot of programs which registers are used
11:44:37 <cluid> I imagine most brainfuck programs used registers in a linear repeating pattern
11:45:09 <int-e> oerjan: Nothing. Hackego is not committing anything, apparently.
11:45:15 <cluid> I know that my most complex programs only use that format, some constant initial segment & nx+b & nx+c & nx+d
11:45:20 <int-e> fizzie: I'm sure this is your fault ;-)
11:45:39 <oerjan> hm maybe it didn't yesterday either
11:45:58 <int-e> maybe a full partition?
11:46:07 <HackEgo> df: cannot read table of mounted file systems: No such file or directory
11:46:48 <int-e> right. even `learn would've complained. but is the repo on the same partition?
11:47:30 <oerjan> `run echo hi >test; cat test
11:47:48 <boily> fizzie: /^[AÄEËIÏOÖUÜYŸ]{5,}$/
11:52:51 <oerjan> cluid: all programs with <> balanced loops, in particular; that's a known optimization.
11:53:32 -!- lambdabot has quit (Quit: tweak).
11:54:51 * oerjan hates when his touchpad accidentally starts dragging a window
11:56:04 <cluid> I only really have one program that doesn't have balanced loops
11:56:49 <oerjan> it's easier to program that way when you really _do_ have a fixed number of cells.
11:56:59 -!- Patashu has quit (Ping timeout: 245 seconds).
11:57:03 <oerjan> but it won't do for doing things like arrays.
11:58:00 -!- lambdabot has joined.
11:58:08 <fizzie> oerjan: There is none, but it still says "What If will return next week. Happy Thanksgiving!" on top.
11:59:17 <oerjan> cluid: i vaguely recall people talking about trying to deduce what cell you are on even when not all loops are balanced, but it's far harder because then the compiler has to actually guess what memory layout the program uses and check that it's consistent.
11:59:25 <fizzie> Re HackEgo, I'll have a look, but I should be doing useful things at the moment.
12:00:08 <fizzie> Okay, "useful" is rather arguable.
12:00:55 <fizzie> It's not just a full disk, at least.
12:01:30 <oerjan> there's nothing about my last commit that looks weird, at least.
12:01:56 <int-e> that's what you think!
12:02:08 <fizzie> `run echo "let's see if the log gets any messages" > useless_file.txt
12:02:17 * oerjan is mentioning that because we _have_ manage to lock up HackEgo's repository mechanism with commits in the past
12:02:36 <fizzie> Nothing in the stdout log, at least.
12:02:53 <cluid> oerjan, I was thinking generalize the assumption to linear equations and you should be able to handle some unbalanced loops?
12:02:54 <int-e> run hg verify on the repo?
12:03:59 <fizzie> If I can remember where it is.
12:04:34 <fizzie> Oh, right, it's the unassuming "env" directory.
12:05:04 <oerjan> this is when we discover the repository has been broken since 2011
12:05:19 <int-e> Also we didn't notice this for more than a day, you can investigate later :P (this is what I always tell myself when lambdabot is acting up.)
12:06:23 <oerjan> int-e: THAT EXPLAINS SO MUCH
12:08:03 <lambdabot> No instance for (Data.Typeable.Internal.Typeable a0)
12:08:03 <lambdabot> arising from a use of ‘M339982716941561429716085.show_M3399827169415614297...
12:08:03 <lambdabot> The type variable ‘a0’ is ambiguous
12:09:25 <int-e> At some point there was a Show instance for functions ... not sure what happened to it.
12:09:31 <boily> > id :: Char -> Char
12:09:41 <int-e> thanks, right, that was stupid
12:09:43 <oerjan> int-e: i think at some point there were _two_
12:10:11 <int-e> there *used* to be an instance for a -> b that just printed <function>
12:10:16 <int-e> and > id worked just fine with that.
12:11:18 <fizzie> "abort: no username supplied (see "hg help config")"
12:11:36 <oerjan> it's because ExtendedDefaultRules still requires one of the classes to be in a limited list, iirc
12:12:05 <int-e> Hmm, so I could try to adding Typeable to that list.
12:12:18 <oerjan> i'm not sure it is customizable
12:12:54 <oerjan> it includes the numeric standard classes and Show, at least
12:13:38 <fizzie> `run ls useless_file.txt
12:13:54 <HackEgo> ls: cannot access useless_file.txt: No such file or directory
12:14:40 <fizzie> It works now, but I'm not sure what I did. I mean, there was that uncommitted file "hmm\n? hello" in the directory that I cleaned up, but that's been there for a while.
12:15:33 <int-e> oerjan: right. there's a feature request about this here: https://ghc.haskell.org/trac/ghc/ticket/8171
12:15:45 <int-e> I *can* patch GHC the next time I upgrade though ;-)
12:16:15 <oerjan> `learn int-e är inte svensk
12:16:17 <HackEgo> Learned 'int-e': int-e är inte svensk
12:16:26 <int-e> (Without looking, shouldn't be *too* hard because at least the compiler already knows about the type class.)
12:17:21 <int-e> (who cares why it works if it works)
12:17:31 <int-e> (who, except hackers... err, never mind.)
12:18:45 <int-e> And I mean that term in the traditional sense.
12:20:00 <oerjan> `learn int-e är inte svensk.
12:20:02 <HackEgo> Learned 'int-e': int-e är inte svensk.
12:20:27 <oerjan> with `learn_append around, not ending wisdoms with punctuation gets awkward.
12:24:34 <int-e> oerjan: would `learn-append int-e . have done the trick?
12:24:54 <int-e> `cat bin/learn-append
12:24:54 <HackEgo> cat: bin/learn-append: No such file or directory
12:25:19 <int-e> `cat bin/learn_append
12:25:20 <HackEgo> #!/bin/bash \ topic=$(echo "$1" | lowercase | sed 's/^\(an\?\|the\) //;s/s\? .*//') \ stuff=$(echo "$1" | cut -d' ' -f2-) \ perl -i -p -e 's/\n/ /' "wisdom/$topic" \ echo "$stuff" >>"wisdom/$topic" \ echo -n "Learned '$topic': " \ cat "wisdom/$topic"
12:26:28 <int-e> oh. perl -i -p -e 's/\n/ /'
12:26:48 <int-e> (perl -pie anyone?)
12:27:18 <oerjan> i dunno, i cannot read proto-indoeuropean very well
12:27:39 -!- boily has quit (Quit: TEMPEST CHICKEN).
12:28:39 <int-e> oerjan: http://www.vexxarr.com/archive.php?seldate=112714 hth
12:29:50 <oerjan> i think that film's a turkey
12:31:24 <oerjan> does this mean there are Gobbeldammerung 1-5
12:32:37 <oerjan> looks like it http://www.vexxarr.com/archive.php?seldate=112813
12:35:30 <oerjan> http://www.vexxarr.com/archive.php?seldate=112212
12:36:10 <int-e> oerjan: http://sprunge.us/JHeL hth
12:36:15 <oerjan> http://www.vexxarr.com/archive.php?seldate=112411
12:37:00 <int-e> oerjan: sorry, I was faster this time.
12:37:03 -!- GeekDude has joined.
12:37:21 <oerjan> int-e: you have committed a gravy mistake
12:42:05 <HackEgo> [wiki] [[RingCode]] N http://esolangs.org/w/index.php?oldid=41389 * TomPN * (+1711) Created page with "'''RingCode''' is an esoteric programming language invented by Tom Price-Nicholson in 2014. RingCode is based on the Tengwar alphabet from Lord of the Rings. The Tengwar are d..."
12:43:01 <int-e> oerjan: That's not punny, it needs another helping to meat minimum requirements.
12:44:28 <oerjan> this is getting out of hand, i cranberry anymore
12:45:37 <oerjan> (this about exhausts my knowledge of thanksgiving)
12:53:38 <int-e> Right. Let's berry the subject.
12:54:20 <HackEgo> [wiki] [[قلب]] M http://esolangs.org/w/index.php?diff=41390&oldid=41388 * Ehird * (+12) /* Examples */ rtl
12:54:47 <fizzie> I forgot to apply an abs to this vector before plotting, but on the other hand I think it looks better this way: https://dl.dropboxusercontent.com/u/113389132/Misc/20141209-plot.png
12:58:09 <elliott> what's it supposed to look like?
12:58:37 <oerjan> it's supposed to gave abs hth
13:02:55 <fizzie> It's supposed to look like https://dl.dropboxusercontent.com/u/113389132/Misc/20141209-plot-1.png which is, like, so square.
13:03:29 <fizzie> Maybe I could use the non-absified version in the cover or something.
13:04:21 <oerjan> a thesis with a blooper appendix, now that's something
13:04:40 <int-e> oerjan: btw, I'm curious. have you tried the Polyominoes golf problem?
13:06:13 <fizzie> I think what the original reminded me of was one of these http://cdn.arstechnica.net/Science/August10/lhc.jpg things.
13:07:00 <fizzie> Except maybe the 90-degree turn there at around (0.5, 0) is not so likely in a particle accelerator.
13:08:58 <int-e> must be a collision :)
13:09:07 -!- shikhout has joined.
13:09:25 <int-e> that's a fancy magnetic field you have there though, does it vary over time?
13:10:05 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41391&oldid=41389 * TomPN * (+513)
13:10:50 -!- ais523 has joined.
13:12:26 -!- shikhin has quit (Ping timeout: 256 seconds).
13:13:26 -!- ais523 has quit (Client Quit).
13:13:34 -!- ais523 has joined.
13:14:45 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41392&oldid=41391 * TomPN * (+189)
13:18:59 <int-e> ah, I wish the article would focus on semantics instead of syntax.
13:19:49 <int-e> The way it looks now, it could still be a Brainfuck clone.
13:20:21 <oerjan> hm a brainfuck equivalent that is so obfuscated we never notice
13:29:37 <ais523> OK, what about a esolanguage that is normally interesting in an eso sort of way
13:29:47 <ais523> however, it has a 0.1% chance of interpreting any input program as brainfuck instead
13:30:11 <int-e> oerjan: "This is my new universal language which I developed specificially for writing a Brainfuck interpreter in!"
13:30:54 <int-e> oerjan: Would omitting that sentence count as obfuscation?
13:31:31 <ais523> why would you develop a language for writing a BF interpreter in? BF's too simple
13:31:50 <int-e> ais523: I'm speculating about the meaning of "obfuscation".
13:31:54 <ais523> I mean, I developed a language for writing INTERCAL optimizers in, but that's complex enough that it could do with a language of its own
13:32:14 <ais523> how do we define brainfuck equivalency, btw? at the command level?
13:32:30 <int-e> I was thinking "Turing completeness."
13:32:44 <ais523> well, yes, I'm trying to find the definition we /actually/ mean
13:33:02 <int-e> Ok, so we've identified the same problem :)
13:33:03 <ais523> something like "programs in this language can be translated to BF in linear time" are defeated by the bundle-an-interpreter trick
13:34:08 -!- _AndoDaan has joined.
13:34:27 <oerjan> something that's to brainfuck like zzo38's 1d chess is to chess
13:36:35 <b_jonas> fizzie: here's what that graph reminds me to http://www.math.bme.hu/~ambrus/pu/missing-abs
13:36:41 <ais523> I don't know of zzo38's 1D chess
13:36:45 <ais523> is it anything like my 1D sokoban?
13:37:50 <oerjan> a most intriguing game
13:38:24 <int-e> I'm tring to figure out whether it could be made interesting. Say, by adding the ability to jump over a box.
13:38:51 <ais523> each square allowed you to teleport to a specific other square
13:39:11 <ais523> other than that, it was identical to regular sokoban
13:39:26 <oerjan> did the blocks teleport too?
13:39:27 <b_jonas> ais523: does it teleport boxes too?
13:39:36 <ais523> no, the boxes just move normally
13:39:54 <b_jonas> ais523: can you teleport on a box?
13:40:42 <b_jonas> do you get teleported at most once after each move?
13:41:04 <ais523> you can choose when to teleport, and can teleport multiple times in a row if you like
13:41:08 <ais523> you can't teleport under a box
13:41:14 <ais523> and there are no walls except one at each end
13:41:29 <ais523> (and the teleport destinations are all on the map, obviously)
13:41:34 -!- shikhout has changed nick to shikhin.
13:41:48 * int-e is reminded of the cyberbox level subset in xsok
13:42:24 <int-e> though teleportation there just picks the nearest free teleporter, so it's not the same.
13:43:39 <b_jonas> so there's no mandatory teleport
13:43:55 <b_jonas> I guess walls aren't really necessary, you can emulate them with boxes anyway
13:48:01 <oerjan> i can find logs where zzo38 talks about it but no actual link
13:48:42 <oerjan> point was, it was 1d yet entirely equivalent to chess
13:50:25 <ais523> presumably the board was 64 by 1, and each piece had a set of distances it could move forwards or backwards
13:50:33 <ais523> main problem there is preventing the sides of the board wrapping
13:51:29 <shachaf> <zzo38> Vorpal: Yes. It is not quite like category theory, though. I do not understand category theory completely, so it is not category theory.
13:52:24 <shachaf> oerjan: http://www.chessvariants.org/index/msdisplay.php?itemid=MSeeeeeeeeeeeeee hth
13:52:37 <fizzie> b_jonas: Oh, it's a common theme?
13:53:01 <ais523> I'd be very surprised if 1D Sokoban were equivalent to Sokoban
13:53:20 <ais523> if it is, maybe there's a rule that "everything is PSPACE-complete unless it obviously isn't"
13:53:36 <oerjan> ais523: sounds like a corollary to that wolfram thing
13:53:44 <oerjan> or the similar TC thing
13:54:02 <oerjan> and of course there has be one for NP
13:54:40 <ais523> oerjan: I don't see any reason why turing-completeness being common would cause pspace-completeness to be common
13:54:56 <ais523> (especially as np-completeness seems more common in practice, and I don't think anyone really believes that pspace=np)
13:55:29 <oerjan> ais523: i think it's basically about the kind of resources you have available
13:55:55 <oerjan> if your game is about finding a fixed solution setup, it's NP-complete
13:56:14 <ais523> right; the paper about computational complexity of computer games mentioned, for most of those games, that they weren't above NP because all moves were irreversible
13:56:21 <oerjan> if it's about finding a solution that moves around inside a setup, it's PSPACE-complete
13:56:35 <oerjan> and if it actually has unbounded state, it's TC
13:57:09 <ais523> now I want to think up a counterexample to that
13:57:19 <oerjan> of course there will be some in theory
13:57:20 <ais523> although, by that heuristic, 1D sokoban is PSPACE-complete
13:58:05 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41393&oldid=41392 * TomPN * (+2108)
13:58:15 <oerjan> there are hierarchies of complexity classes, after all
13:59:02 <HackEgo> [wiki] [[Musical notes]] http://esolangs.org/w/index.php?diff=41394&oldid=41336 * TomPN * (+70) /* See also */
13:59:02 <shachaf> and below them there are loerarchies
13:59:27 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41395&oldid=41393 * TomPN * (-3) /* See also */
13:59:40 <HackEgo> [wiki] [[Dimensions]] http://esolangs.org/w/index.php?diff=41396&oldid=41328 * TomPN * (+70) /* See also */
13:59:50 <oerjan> ais523: of course _all_ of them, in addition to resources, require your game or the like to have enough local logical primitives
13:59:56 <HackEgo> [wiki] [[Quantum Dimensions]] http://esolangs.org/w/index.php?diff=41397&oldid=41099 * TomPN * (+70) /* See also */
14:00:45 <ais523> oerjan: right, but a good esoprogrammer can craft logical primitives out of almost anything
14:00:49 <HackEgo> [wiki] [[Language list]] http://esolangs.org/w/index.php?diff=41398&oldid=41368 * TomPN * (+15) /* R */
14:01:06 <ais523> that said, it still frustrates me that some of the constructions in the computer games complexity paper are wrong
14:01:26 <ais523> e.g. they claim the Metroid construction also works for Super Metroid but it doesn't due to the existence of the mid-air morph
14:02:43 <HackEgo> [wiki] [[User talk:TomPN]] http://esolangs.org/w/index.php?diff=41399&oldid=40923 * TomPN * (+14) /* Other esolangs */
14:02:59 <b_jonas> ais523: I hope you wrote to the authors to note these mistakes
14:03:06 <b_jonas> maybe they just don't play those games enough
14:03:40 <ais523> interesting idea, though
14:04:27 <oerjan> ais523: that's almost the _definition_ of a good esoprogrammer there
14:04:43 <oerjan> it's certainly the way i work, at any rate
14:04:58 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41400&oldid=41395 * TomPN * (+0) /* Adding tehtar to form instructions */
14:05:02 <b_jonas> ais523: tell them. try to do a quick search first for later articles of theirs correcting the mistakes of course.
14:05:05 <elliott> oerjan: btw you should work on clue (keymaker)'s tcness
14:05:27 -!- nortti has changed nick to lawspeaker.
14:06:05 -!- lawspeaker has changed nick to nortti.
14:06:14 <ais523> elliott: is clue (oklopol)'s obviously TC, or obviously sub-TC? I have a feeling it's one or the other
14:06:24 <elliott> obviously TC, there's an ski implementation
14:09:01 <oerjan> ais523: although my best work also involves being able not to be overwhelmed by the complexity of putting it all together.
14:09:51 <ais523> that can be a hard part
14:24:05 -!- ZombieAlive has joined.
14:25:13 -!- drdanmaku has joined.
14:30:24 -!- ais523 has quit (Read error: No route to host).
14:30:34 -!- ais523 has joined.
14:57:41 -!- _AndoDaan has quit (Remote host closed the connection).
15:07:05 -!- spiette has joined.
15:15:02 -!- `^_^v has joined.
15:29:08 <oerjan> elliott: clue seems to fall into the same class as self-bct and dupdog: too complicated to prove non-tc, too unruly to see any obvious way of controlling the complexity
15:35:57 <b_jonas> oerjan: does deciding whether an input to McCulloch's second machine is immortal fall into the same class? or is that obviously controllable?
15:36:18 -!- Sprocklem has joined.
15:36:43 <ais523> mcculloch has far too many machines
15:36:49 <ais523> you'll have to list the rules, rather than just name it
15:37:00 <b_jonas> ais523: I have them listed on the esolang wiki
15:37:14 <b_jonas> ais523: http://esolangs.org/wiki/McCulloch%27s_second_machine
15:39:28 -!- Decensum has joined.
15:39:38 <oerjan> hm the digits 6-9 are redundant with each other, it seems
15:40:09 -!- ChanServ has set channel mode: +o oerjan.
15:40:20 -!- oerjan has kicked Decensum You are getting annoying.
15:40:30 <ais523> oerjan: oh, it's dulnes?
15:40:35 <b_jonas> oerjan: that's irrelevant if you ask only about immortal numbers, becaues in this machine any number containing a digit 1 or 6 or 7 or 8 or 9 is mortal
15:40:48 <ais523> also, first -n message!
15:40:59 -!- oerjan has set channel mode: -o oerjan.
15:41:09 <ais523> has dulnes been kicked before?
15:41:21 <ais523> and if so, what was the offence?
15:41:25 -!- Decensum has joined.
15:42:20 <oren> what does -n mean
15:42:54 <b_jonas> ais523: as in channel mode -n , right?
15:42:57 <oren> oh i guess i know nothing about irc
15:43:29 <oren> add it to the list of things i know nothing about
15:43:48 <oerjan> b_jonas: oh so you cannot avoid a digit getting to the start
15:43:57 <ais523> hmm, now the list of things that oren knows nothing about has been removed from the list of things that I know nothing about
15:44:11 -!- Decensum has changed nick to Decensum|Away.
15:44:17 <b_jonas> oerjan: yes. I can tell you the proof but it might be more interesting if you try to figure it out yourself
15:45:32 <shachaf> Any ideas for clever ways to take the tail of an FMList? http://stackoverflow.com/a/27382987
15:47:13 * oerjan decides his brain doesn't have room for both FMLists and machines at the moment
15:47:31 <shachaf> whoa, you're doing things with -- oh, not that machines
15:48:34 <b_jonas> oerjan: isn't it the same thing? they both want to access a list from both ends quickly
15:48:44 <b_jonas> (don't take that seriously(
15:49:17 <elliott> I've kicked dulnes I think
15:49:57 <oerjan> elliott: are you sure you've not been replaced with facekicker
15:50:17 <elliott> anyway I'm going to take his advice and go to bed.
15:50:17 <oren> is that a regex
15:50:18 <oerjan> cluid: a regular expression hth
15:50:19 <ais523> cluid: a regex, by the look of it
15:50:23 <J_Arcane> Heh heh. http://en.wikipedia.org/wiki/Heisenbug
15:50:56 <ais523> aha, a spam I just got: "You have been selected as a winner for using Google services"
15:51:04 <j-bot> b_jonas: |limit error
15:51:04 <j-bot> b_jonas: | +/ i.1000000000
15:51:07 <j-bot> b_jonas: 499999500000
15:51:11 <j-bot> b_jonas: |limit error
15:51:12 <j-bot> b_jonas: | +/ i.100000000
15:51:14 <j-bot> b_jonas: |limit error
15:51:14 <j-bot> b_jonas: | +/ i.10000000
15:51:19 <ais523> when I'm in the minority of the Internet-using population that doesn't (at least, not those that require login, and I normally use a separate browser even for the ones that don't)
15:51:35 <shachaf> ais523: Is it signed by the "CEO of Google UK"?
15:51:37 <b_jonas> j-bot: 3 :'+/i.1e6'"0]i.1e5
15:51:40 <shachaf> I got spam like that recently.
15:51:52 <cluid> what is difficult about the mcculloch machine thing
15:52:07 <cluid> have you tried checking each rule against the "regex"
15:52:14 <ais523> ©2014 Google Terms & Privacy
15:52:17 <b_jonas> j-bot: 3 :'+/i.1e6'"0]i.1e4
15:52:32 <shachaf> I pasted the whole thing into an IRC channel.
15:52:32 <ais523> that's quite some bizarre copyright notice
15:52:36 <b_jonas> j-bot: +/3 :'+/i.1e6'"0]i.1e3
15:52:39 <j-bot> b_jonas: 499999500000000
15:52:49 <ais523> is Matt Brittin the actual CEO of Google, I wonder?
15:53:25 <b_jonas> cluid: I don't claim it's difficult, but I couldn't solve the problem I gave
15:53:34 <ais523> oh, it says "find attached email with more details" and has an attached jpeg
15:53:36 <shachaf> Larry Page is the CEO of Google.1
15:53:40 <ais523> which I'm not going to lok at
15:53:53 <shachaf> According to Wikipedia Matt Brittin is a VP.
15:54:17 <oren> ais523: are you paranoid?
15:54:25 <cluid> The machine is defined for exactly the digit strings matching the following regular expression: /^[345]*2[1-9][1-9]*$/
15:54:30 <cluid> is this derived from the rules?
15:54:34 <ais523> oren: to some extent, yes
15:54:34 <cluid> or is this a definition
15:54:39 <b_jonas> it's derived from the rules
15:54:50 <cluid> so this needs proof
15:54:59 <cluid> it would be a shame if there was a mistake in that
15:55:47 <ais523> cluid: any string with no 2 is mortal
15:55:54 <ais523> because that's the base case of the induction
15:57:02 <b_jonas> shachaf: not since http://dilbert.com/fast/2014-11-15/ I think
15:57:40 <cluid> R = /^[345]*2[1-9][1-9]*$/
15:57:40 <cluid> dR/d2 = /^[1-9][1-9]*$/
15:59:21 <cluid> this is extremely difficult
16:01:12 -!- MDream has changed nick to MDude.
16:02:34 -!- tswett has joined.
16:02:40 <oerjan> b_jonas: that loads so fast i cannot see it
16:02:51 <cluid> b_jonas, why not [345]*2[1-9]* ?
16:03:19 <cluid> I tried to derive that regex myself
16:03:24 <cluid> but I got a different answer
16:03:28 <b_jonas> cluid: because the rules specifically forbid giving an empty string
16:03:39 <cluid> this string always has 2 in it though
16:04:05 <cluid> so [345]*2[1-9]+ ,which is what you have
16:04:27 -!- Sprocklem has quit (Quit: [).
16:04:32 -!- Sprocklem_ has joined.
16:04:52 -!- Sprocklem_ has changed nick to Sprocklem.
16:07:00 <cluid> http://lpaste.net/116084
16:07:21 <cluid> I think this characterizes the strings which are mortal, so this might be a step in the right direction?
16:08:09 <cluid> oh it would be !R union those (whic lets us delete the last one)
16:09:57 <cluid> the negation of the regex R
16:10:15 <oerjan> that's not what mortal means
16:10:21 <cluid> I don't think regex is powerful enough for this problem though, since you have m x ++ 2 ++ m x
16:10:29 <cluid> the same string occurs twice
16:10:38 <oerjan> it can take any number of iterations to die
16:10:46 <cluid> oh I see what you mean, thanks
16:12:06 <cluid> maybe you could express mortal strings as a CFG?
16:12:08 <oerjan> b_jonas: this does resemble dupdog...
16:13:17 <cluid> yeah i gues thats hard
16:13:38 <cluid> well another approach might be: If you can decide if a string diverges then you wi
16:13:57 <cluid> that also looks very hard to find though
16:15:24 <oerjan> well you can rephrase the question but you have to have some reason if that rephrasing is going to be easier to solve...
16:15:49 <cluid> I think this problem is too hard to solve, but it would be nice to solve it
16:15:50 -!- ais523 has quit.
16:15:55 * oerjan realizes he just rephrased the problem of finding a solution
16:16:57 <int-e> It has worked before...
16:17:18 <cluid> M xs when not (P xs) or (P xs and M (m xs))
16:17:25 <cluid> this is theform of the mortal predicate
16:17:29 <oerjan> int-e: now that's crazy talk
16:17:32 <cluid> where P tells you if a string is in the machine
16:20:08 <cluid> are there strings which grow large then get smaller?
16:20:13 <cluid> if not its probably decidable
16:20:31 <cluid> well ther are strings like 5:5:5:5:something
16:20:38 <cluid> which gets really big them the 'somethings' might get smaller
16:20:45 <cluid> 555555555444 for example
16:21:04 <cluid> that would need pathced I guess
16:21:05 <int-e> wait, can some op add +n to the channel modes, please?
16:21:19 <int-e> seeing messages from kicked people is ... strange.
16:21:32 <b_jonas> int-e: ais deliberately has it at -n until it becomes a problem
16:22:16 <cluid> this problem is very hard :(
16:22:18 <int-e> well, consider this as one vote for it being a nuisance then.
16:22:43 <cluid> i dont think its -n that is a nuisance
16:23:16 <cluid> anyway, can you give ma hint for solving the McCulloch's second machine
16:23:29 <int-e> it is to me. it breaks my mental model of IRC. ;-)
16:23:31 <oerjan> int-e: note that iirc an actual ban still prevents talking
16:23:34 -!- hjulle has quit (Ping timeout: 264 seconds).
16:23:43 <cluid> int-e, I was joking that the real problem is Dulnes
16:24:12 * oerjan swats cluid for explaining jokes -----###
16:24:19 <b_jonas> oerjan: the ban doesn't apply for people who aren't joined,
16:24:21 <b_jonas> for implementation reasons
16:24:37 <cluid> oerjan, any insights into this
16:24:38 <int-e> cluid: the -----### depicts a paddle, btw.
16:24:49 <cluid> i thought it was a swatter
16:24:55 <oerjan> int-e: no, it's a flyswatter hth
16:25:21 <int-e> oerjan: oh. how did you depict mapoles?
16:25:37 <b_jonas> oerjan: basically the server caches which joined users can talk on a channel, so it doesn't have to recompute it from the channel modes at each message, but it doesn't want to cache anything for non-joined users, so it just allows everyone
16:25:47 <cluid> so help me with math please :)
16:25:49 <oerjan> <cluid> anyway, can you give ma hint for solving the McCulloch's second machine <-- this is a problem that _we_ don't know how to solve yet, so unlikely.
16:26:16 <b_jonas> oerjan: besides, you can see messages from recently kicked users because of multiple server stuff
16:26:20 <cluid> I thought of something
16:26:22 <int-e> Anyway, my impression of the McCulloch machine was that it's a hard problem, and that there may just be enough rope to hang it on the TC nail.
16:26:45 <cluid> all infinite loops occur in the following way: x is in R, and m x is in R
16:26:59 <b_jonas> int-e: if it matters, my guess is that there is a relatively short algorithm to decide mortality, and has a human-understandible length of proof.
16:27:22 -!- ZombieAlive has quit (Remote host closed the connection).
16:27:24 <cluid> oh wait that's wrong
16:27:38 <oerjan> int-e: the mapoles aren't mine so i don't generally depict them
16:27:40 -!- ZombieAlive has joined.
16:27:43 <cluid> you need a language R' such that: forall x in R', m x is in R'
16:28:07 <cluid> the loops live in R'
16:28:48 <cluid> maybe you could define R as a CFG?
16:29:54 <int-e> there, for future reference.
16:30:30 <cluid> http://lpaste.net/116088
16:31:17 <cluid> R' -> 2 R' | 3 R' 2 R' | 4 Z | 5 R' R'
16:31:17 <cluid> Z -> Z 2 | Z 2 Z 3 | R' 4 | Z Z 5
16:31:50 <cluid> this language is empty :/
16:31:58 <cluid> this approach doesn't work
16:33:06 <oren> perhaps you could write a prover in prolog?
16:33:18 <oerjan> cluid: note that you are going to have a _very_ hard time finding a "general" method to solve it, because only takes a bit different rules to get a "tag system" for which the problem _is_ unsolvable.
16:33:43 <oerjan> so if it is solvable it needs looking at specific properties of the rules used
16:35:35 <cluid> I don't really have any other ways I could approach the problem though
16:36:32 <int-e> In fact, I'd say that tag systems look simpler!
16:36:43 <cluid> if S is the language of looping machines, S -> 2 S | 3 K | ...
16:36:58 <cluid> then K is the language of machines such that m x ++ 2 ++ m x is a looping machine
16:38:21 <cluid> http://lpaste.net/116090 it seems unlikey to transform this program into something which can be computed
16:38:31 <oerjan> int-e: the only thing that might make this simpler is that you don't have many constant strings.
16:38:44 <cluid> it needs "base cases"
16:40:11 <cluid> SopaXorzTaker, what is it?
16:40:47 <b_jonas> [ ];._1' can I talk after parted?'[+/([:+/@i.1e6"_)"0 i.6e3
16:40:51 -!- j-bot has left.
16:41:04 <int-e> oerjan: And of course it may be possible that mortal strings are easy to characterize and you can *still* embed a TC language in the immortal one (say, some CA).
16:41:10 <oren> do you mean actually functional, or just procedural?
16:41:23 <int-e> oerjan: But that's just crazy talk at this point.
16:42:09 <oerjan> int-e: your attempt to solve identity matrix just triggered a henkma incident hth
16:42:52 <b_jonas> is that like a collapsing hrung disaster?
16:43:07 <oerjan> b_jonas: much worse hth
16:44:12 <shachaf> what would you say is the state of matter of the identity matrix twh
16:44:22 <int-e> 5 characters shorter and completely different statistics. Sigh.
16:44:42 <shachaf> perhaps i got that backwards
16:45:03 <int-e> oerjan: but at least it's an endless problem, plenty of time to catch up ;-)
16:46:15 <b_jonas> SopaXorzTaker: you know it's generally not allowed to create brainfuck variants these days, unless you have a very good excuse, right?
16:47:30 <int-e> oerjan: I have a post-mortem toy problem, http://golf.shinh.org/p.rb?postfix+to+infix
16:47:41 <cluid> could you encode collatz type problems into this machine?
16:48:39 <b_jonas> why are there no active problems?
16:49:36 <oerjan> because no one made any
16:51:43 -!- nortti has changed nick to lawspeaker.
16:52:25 -!- lawspeaker has changed nick to nortti.
16:54:04 <shachaf> you know the thing where you're writing a recursive function and you accidentally unfold it one step for no good reason?
16:55:50 <shachaf> now that you're not thinking about machines you should tell me whether this fmlist business is possible twh
16:56:11 <shachaf> i'm actually not completely sure whether it's possible even though i have a working implementation
16:56:48 -!- oerjan has quit (Quit: leaving).
17:05:03 -!- Sprocklem has quit (Ping timeout: 258 seconds).
17:12:35 -!- MDream has joined.
17:14:35 <HackEgo> [wiki] [[KittyKittyMewMew]] http://esolangs.org/w/index.php?diff=41401&oldid=38827 * 67.78.57.11 * (+141) /* Interpreters for KittyKittyMewMew */
17:14:38 -!- MDude has quit (Ping timeout: 258 seconds).
17:16:13 -!- SopaXorzTaker has changed nick to ummjackson.
17:16:20 -!- ummjackson has changed nick to SopaXorzTaker.
17:17:33 -!- scounder has quit (Read error: Connection reset by peer).
17:30:52 -!- scounder has joined.
17:37:06 -!- AndoDaan has joined.
17:53:41 -!- mihow has joined.
18:03:45 <int-e> shachaf: sorry, what fmlist business, didn't the stackoverflow thread already answer all questions?
18:04:10 <shachaf> int-e: I linked to my answer in the thread. I want a non-cheating way to do what I did there.
18:04:56 -!- Decensum|Away has quit (Quit: Connection closed for inactivity).
18:05:16 <shachaf> Yes. Without using any types that break the monoid laws.
18:08:07 <FireFly> shachaf: that did not help but was a pretty what? rude reply?
18:08:46 <shachaf> perfectly reasonable response
18:20:20 <int-e> shachaf: http://sprunge.us/WJWB?hs doesn't break any rules, I think.
18:21:36 <HackEgo> [wiki] [[RingCode]] http://esolangs.org/w/index.php?diff=41402&oldid=41400 * BCompton * (+94)
18:22:54 <shachaf> int-e: See also http://lpaste.net/116099 from glguy in #haskell-lens
18:23:02 <int-e> shachaf: using the basic property that uncons ((a:b) ++ c) = Just (a, b++c), so UAux stors the head, the tail, and a reconstruction of the original list.
18:23:30 <int-e> right, that's the same idea.
18:24:23 <oren> i find Haskell a lot harder to understand than lisp; is that normal
18:25:37 -!- MDream has changed nick to MDude.
18:28:26 <int-e> oren: only if you know Lisp.
18:29:52 <oren> what if you don't relly know either?
18:32:43 -!- spiette has quit (Remote host closed the connection).
18:32:57 <cluid> lisp has a lower barrier for entry
18:33:07 <cluid> haskell is more complex and unusual
18:34:23 <oren> i think it's that haskell is somewhat an ALGOL dialect
18:34:55 <cluid> in my picture of things scheme (a lisp) descends from ALGOL
18:35:06 <cluid> haskell comes from miranda and gopher and lazy ml
18:35:33 <oren> well you probably know more than me so ok
18:36:08 <shachaf> int-e: I think your solution comes up with the same result as glguy's.
18:36:13 <int-e> shachaf: Now I'm wondering whether the (a,b) <> (_,d) = (a <> d, b <> d) semigroup has a name.
18:36:16 <shachaf> (And mine, after I changed it not to unfold the recursion.)
18:39:10 <shachaf> I encourage y'all to submit answers with your solutions.
18:39:56 <int-e> What is a solution, if it's not an answer?
18:40:28 <shachaf> To submit answers to the stackoverflow question that contain the solutions.
18:40:42 <int-e> Ah. Not going to happen.
18:41:59 <int-e> shachaf: I actually parsed the sentence wrong: (I encourage y'all (to submit answers) with your solutions.)
18:42:14 <int-e> rather than (answers with your solutions)
18:42:25 <shachaf> I don't follow your parentheses.
18:42:43 <shachaf> Use your solutions to encourage you to submit answers?
18:43:01 <int-e> Well, it didn't make sense to me either.
18:43:35 <int-e> It sounded almost like something a teacher would tell her students, but not quite.
18:49:23 <shachaf> I still like this free magma thing.
18:49:33 <shachaf> lambdabot should have it so we can inspect monoid trees.
18:50:47 <shachaf> I have a 9-line @let definition suitable for pasting, but only I can use that easily.
18:54:41 <shachaf> > foldMap N (S.fromList [1..6])
18:54:42 <lambdabot> (N 1 ◇ (N 2 ◇ N 3)) ◇ (N 4 ◇ (ε ◇ (N 5 ◇ N 6)))
19:09:04 -!- shikhout has joined.
19:12:09 -!- shikhin has quit (Disconnected by services).
19:12:12 -!- shikhout has changed nick to shikhin.
19:22:10 <shachaf> 11:20 <glguy> shachaf: int-e's solution is a little too strict
19:22:10 <shachaf> 11:20 <glguy> minefield = (singleton 0 <> singleton 1 <> undefined) <> (singleton 2)
19:22:48 <shachaf> int-e: You should be in #-lens or something, it's the future.
19:25:07 -!- MoALTz has joined.
19:28:59 <int-e> Ah yes, nasty strictness. So using tuples for separating out the full part is essential.
19:34:45 -!- S1 has joined.
19:38:56 <int-e> actually no, the type is fine... but I need the selectors.
19:41:59 <int-e> so that leads to http://sprunge.us/acbH?hs
19:42:59 <shachaf> I'm going to stop conveying messages between you two now, because it's silly.
19:43:30 <shachaf> Either you join #-lens or glguy joins this channel. Or something.
19:43:42 -!- glguy has joined.
19:44:24 <glguy> shachaf: I'm still OK with you relaying messages for me
19:44:30 <glguy> this way I can make sure they are relayed correctly
19:45:22 <HackEgo> glguy: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
19:45:40 <int-e> (just following standard operating procedure)
19:46:15 <int-e> so did shachaf relay the http://sprunge.us/acbH?hs link?
19:46:57 <int-e> glguy: in any case, thanks for pointing out the strictness bug in my code.
19:47:05 <glguy> Yeah, he pasted that in another channel
19:47:28 <glguy> I didn't know where the discussion had originated but we were discussing it there, too
19:48:27 <int-e> shachaf: no. I'm aiming for maximum confusion. :P
19:48:58 <glguy> this is what I had come up with : http://lpaste.net/116099 (I don't know if shachaf managed to share that over here)
19:50:57 -!- relrod has joined.
19:51:28 <HackEgo> relrod: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
19:51:53 -!- Sprocklem has joined.
19:55:13 <HackEgo> bin/bienvenido \ bin/emoclew \ bin/ozcome \ bin/relcome \ bin/rwelcome \ bin/tervetuloa \ bin/wehlcohme \ bin/welcome \ bin/welcome \ bin/WeLcOmE \ bin/WELCOME \ bin/welcome13 \ bin/wlcm \ bin/wow \ bin/zalgreet
19:55:44 <shachaf> whoa, I didn't know about all these.
19:55:54 <HackEgo> shachaf: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
19:56:24 <HackEgo> welcome "$@" | sed 's/aeiou//'
19:56:41 <shachaf> That program has a high bug density.
19:57:17 <FireFly> `` sed -ir 's/.$/g&/' bin/wlcm; cat bin/wlcm
19:57:20 <HackEgo> welcome "$@" | sed 's/aeiou//g'
19:57:28 <HackEgo> FireFly: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
19:57:34 <shachaf> That's one bug, but there's a more important one.
19:57:39 <FireFly> I didn't spot the other obvious bug
19:57:46 <FireFly> (but I did now, of course)
19:58:09 <HackEgo> F͓̈́i̻͓r͈ͨe̵ͫF̲͂ĺ͚y̖̕:͓ͮ ͨͤW͇̐e͎̣lͯ̓c̳̊o̯̰m̶͏ĕ̟ ̣̀t̻͕o̢͆ ͖̓t̗̝h͍͑eͪ̉ ̨ͫi͇͒n̢̼t̥̘eͬ͟ṛ̀n̾̈́a̫͏t̸͒i͉ͬŏͦn̳͗a̫ͮl̿ͤ ̓ͭh͗͊ų͎b̢ͪ ͯ̒f̹̀ơ͉r̨͗ ̋ͬeͮͭs͝ͅo̝͛t̹ͮe̎͋r̵̪i͎ͥc̿͞ ̱͗pͥ͏r̓͞o̅ͥg̡͑rͭͫa͉͡m̜̑mͭ็iͩ̓n̨̕g̈́͒ ͣ͜l̺͖ã̲n͠
19:58:30 <FireFly> `` sed -ir 's/sed.*/tr -d aeoiu/' bin/wlcm; cat bin/wlcm
19:59:07 <FireFly> zalgreet might've been too much for HackEgo
19:59:26 <HackEgo> yourself: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
19:59:27 <HackEgo> T͈͑hͩͭaͫ̉n̍̾k̪ͬs͈̐,̘̌ ̯́s̾͊ḧ̜a̛̚c̙ͦh̸̃a͔̽f͎̰.̙̅ ͗̄T̷͇h́͊ä̇c̘ͨh̛̻å̎f̂ͨ.̱̊
19:59:28 <HackEgo> welcome "$@" | tr -d aeoiu
19:59:52 <HackEgo> :-( \ 113500 \ bdsmreclist \ bin \ canary \ cat \ complaints \ :-D \ dc \ dir \ dog \ etc \ factor \ faith \ head \ hello \ hello.c \ ibin \ index.html?dl=1812 \ interps \ lib \ paste \ pref \ prefs \ py.py \ quines \ quotes \ script.py \ share \ src \ test.c \ Wierd \ wisdom \ wisdom.pdf
20:00:10 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
20:00:33 <HackEgo> HackEgo, also known as HackBot, is a bot that runs arbitrary commands on Unix. See `help for info on using it. You should totally try to hax0r it! Make sure you imagine it's running as root with no sandboxing.
20:02:16 <HackEgo> h: Wlcm t th ntrntnl hb fr strc prgrmmng lngg dsgn nd dplymnt! Fr mr nfrmtn, chck t r wk: <http://slngs.rg/>. (Fr th thr knd f strc, try #strc n rc.dl.nt.)
20:02:20 <int-e> `cat cat dog canary
20:02:21 <HackEgo> cat: cat dog canary: No such file or directory
20:02:25 <int-e> `` cat cat dog canary
20:03:12 -!- Patashu has joined.
20:24:02 -!- Sprocklem has quit (Ping timeout: 245 seconds).
20:25:42 -!- Patashu has quit (Ping timeout: 245 seconds).
20:59:04 -!- Sprocklem has joined.
21:07:21 <oren> how come so many languages have REPLs but don't porvide the ability to save functions/variables you have defined on the REPL?
21:10:00 <int-e> because they are read-eval-print-loops, not editors.
21:10:38 <oren> but BASIC had that ability in the 1980s
21:11:08 <oren> and matlab has it
21:11:23 <oren> why can't GOOD languages do it?
21:11:50 <cluid> i dont get it either
21:11:51 <int-e> Not trying to be cocky. It's a mindset that comes from editing code in editors, then loading it to run it. Once you reach that point it's easier to modify the code in the editor and reload it.
21:12:23 <int-e> The strange thing is that nevertheless, simple REPL loops often grow and acquire the ability to define new functions.
21:13:40 <oren> i just had to scroll up in my terminal, and copypaste it and edit the prompts out
21:13:50 <oren> that should not be necessary
21:16:21 <int-e> (The REPLs I can think of don't even keep the entered source code around.)
21:17:12 -!- Vorpal has quit (Ping timeout: 255 seconds).
21:17:31 <int-e> It's noteworthy BASIC comes from a completely different era when running two programs simultaneously was generally not done. So of course you needed an IDE, rather than just a REPL, and an IDE that can't save code is worthless.
21:18:05 <glguy> With a repl you can define definitions in terms of previous effects: x = sizeOfFile "somefile"; f y = y + x, so it might not be enough to just save a function definition. You'd probably need to save the whole transcript
21:18:11 -!- Vorpal has joined.
21:18:14 -!- Vorpal has quit (Changing host).
21:18:14 -!- Vorpal has joined.
21:30:10 <Jafet> You can save the current state of the REPL
21:31:26 <oren> yeah that's what I mean: you should be able to turn the state where there is a variable x and it contains 19, into a file containing "x = 19"
21:31:36 -!- ocharles_ has joined.
21:32:05 <Jafet> Some REPLs store the transcript in a list which you can write out or replay manually (ipython, mathematica)
21:32:37 <glguy> Others let you save a snapshot of the heap to reload later
21:33:18 <Jafet> oren: what about the state where x is a file descriptor?
21:33:50 -!- tswett has quit (Ping timeout: 272 seconds).
21:34:03 <oren> in that case it should either fail to save x or turn it into fopen...fseek...
21:34:27 <oren> not that the language would necessarily use C's functions
21:35:20 -!- Melvar has quit (Ping timeout: 245 seconds).
21:35:31 <Jafet> What if the file is an open TCP socket
21:35:59 <glguy> obviously you'd just also save a version of the remote program alongside your REPL state
21:36:12 <oren> then it should definitely fail (with an error like "cannot save this type of value: TCP-socket")
21:36:13 <glguy> and bring both back up when you were ready
21:36:27 <int-e> I would think that saving transcripts would be useful. Going beyond that in languages that don't have native code inspection is probably not going to pay off.
21:37:13 <glguy> Isn't this basically how smaltalk environments work?
21:37:19 <int-e> So ... not Lisp, Scheme or TCL, did I miss any? ;-)
21:37:37 <int-e> The smalltalk environments that I've seen (very few) use Heap images.
21:38:13 <glguy> which presumably do something to abstract out these rough edges when resuming from a saved heap
21:38:18 <glguy> (where they can be smoothed)
21:38:36 -!- Melvar has joined.
21:38:37 <oren> why can't it save as source code?
21:38:48 <oren> like BASIC did?
21:39:27 <cluid> you should be able to
21:39:30 <oren> that way i can develop a function in the repl and then compile it
21:39:46 <cluid> you could make a better REPL with this feature
21:40:37 <oren> i am working on one for scrip7...
21:40:55 <int-e> But the approach to IDEs has been turned upside down. Rather than extending a REPL with editing functionality, we have editors that integrate the REPL.
21:41:20 <int-e> If you want code editing, the state of the art is to use an editor, not a REPL.
21:41:26 <cluid> scrip7 is really cool
21:41:54 <oren> it will be cooler... current version kinda sucks
21:42:24 <oren> it needs better repl, which i am making in ncurses
21:46:04 <oren> the registers will have context-relevant display above the normal scrip7 command line
21:47:51 <oren> and of course the new version of langugae will have better IO commands providing functionality as described above
21:48:28 <oren> and the context-relevant display is fully editable of course
21:49:04 <oren> e.g. point register 0 to a 64-bit value, then edit the displayed value directly
21:50:21 <oren> a drop-in debug/programming console for any C program is what it ends up being
21:52:21 -!- nys has joined.
21:54:25 -!- idris-bot has joined.
21:59:14 -!- Frooxius has quit (Quit: *bubbles away*).
22:00:17 -!- cluid has quit (Remote host closed the connection).
22:04:27 -!- Frooxius has joined.
22:21:35 -!- Sprocklem has quit (Ping timeout: 245 seconds).
22:33:31 <Taneb> Heh, Agda's source has its own definition of lenses
22:33:55 <Taneb> It's essentially a veeery stripped down version of lens
22:49:32 -!- Froox has joined.
22:49:33 -!- Frooxius has quit (Read error: Connection reset by peer).
22:56:50 -!- oerjan has joined.
22:58:03 -!- Decensum|Away has joined.
22:58:15 <oerjan> goddammit, the asus smarttouch (or something like that) driver has died again
22:58:52 <oerjan> the only way i know to fix it is rebooting windows
23:04:24 <oerjan> of course it has to happen when i have 24 tabs open
23:05:02 <oerjan> Decensum|Away: what are you comenting on?
23:06:32 -!- CrazyM4n has joined.
23:06:59 -!- AndoDaan has quit (Read error: Connection reset by peer).
23:07:08 <oren> or C++ without boost
23:07:24 -!- Decensum|Away has changed nick to Decensum|Zzz.
23:07:55 -!- AndoDaan has joined.
23:08:00 <zzo38> But what program is it anyways?
23:08:04 <Decensum|Zzz> Client stays on for about 5 hours so im gonna go sleep
23:08:30 -!- Vorpal has quit (Ping timeout: 240 seconds).
23:08:47 <oren> if he's asking "C or C++" he must have a need for speed
23:09:04 <oren> and possibly a need for low-level hardware access
23:10:14 -!- Vorpal has joined.
23:10:25 <oren> rust doesn't look stable enough...
23:10:44 <oren> look at what happened to the python community
23:11:26 -!- `^_^v has quit (Ping timeout: 250 seconds).
23:13:02 <myname> 1.0 beta should be out at the beginning of next year
23:13:28 <oren> The foolish man builds his house upon sand.
23:13:38 -!- boily has joined.
23:14:15 <oren> The wise man builds his house upon rock: C90, no gnu extensions.
23:16:26 <oren> myname: it's a biblical reference. the point is you should build important things upon a grounding of things that won't fail.
23:17:08 <myname> oren: i don't build anything
23:17:43 <myname> at this way i am actually totally okay with unstable languages
23:19:05 <oren> well, ok. but until told otherwise i'mma assume that whatever Decensum is building is something he'd rather not have to constantly revise.
23:27:58 -!- dts|pokeball has quit (Ping timeout: 244 seconds).
23:34:03 -!- dts|pokeball has joined.
23:43:13 -!- dts|pokeball has quit (Read error: Connection reset by peer).
23:47:19 -!- Frooxius has joined.
23:50:42 -!- Froox has quit (Ping timeout: 244 seconds).
23:50:43 -!- Froo has joined.
23:54:18 -!- Frooxius has quit (Ping timeout: 252 seconds).