00:16:45 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
00:35:54 -!- Sprocklem has quit (Ping timeout: 244 seconds).
01:07:27 -!- nooga has joined.
01:12:04 -!- nooga has quit (Ping timeout: 250 seconds).
01:14:41 -!- Dulnes has quit (Quit: Connection closed for inactivity).
01:19:36 <Sgeo> Am I a bad person for recommending Node.JS to someone trying to build a chat? It makes the most sense to me...
01:26:01 <HackEgo> [wiki] [[GridScript]] http://esolangs.org/w/index.php?diff=41267&oldid=41256 * SuperJedi224 * (+64)
01:26:23 <HackEgo> [wiki] [[GridScript]] http://esolangs.org/w/index.php?diff=41268&oldid=41267 * SuperJedi224 * (+3) /* Command Summary */
01:31:29 -!- EisenHerz has changed nick to Sauvin.
01:46:14 <HackEgo> [wiki] [[Popular problem]] http://esolangs.org/w/index.php?diff=41269&oldid=30415 * 122.102.45.251 * (+266)
01:48:25 -!- oren has joined.
01:50:55 <HackEgo> [wiki] [[Popular problem]] http://esolangs.org/w/index.php?diff=41270&oldid=41269 * Orenwatson * (-266) Undo revision 41269 by [[Special:Contributions/122.102.45.251|122.102.45.251]] ([[User talk:122.102.45.251|talk]]) spam
01:53:04 <oren> seo melbourne? why melbourne
01:53:10 <elliott> fizzie: captcha got solved
01:53:22 <elliott> maybe it was just a human though
01:54:00 <oren> is seo even a real thing anymore
01:54:58 <oren> or does it just refer to making the pages bot-readable?
01:58:35 <oren> HOLY SH%$$%@%$
01:58:55 <oren> I just got a email back from the maker of MNNBFSL
01:59:17 <oren> He wrote in English
01:59:23 <oren> "Thank you for writing an article of MNNBFSL."
01:59:37 <oren> "I read http://esolangs.org/wiki/MNNBFSL ."
01:59:47 <oren> "In this artcile, "It appears to have been ... in late August 2014."
01:59:49 <oren> but it is invented in July.
02:00:05 <elliott> japanese people found to know english, world shocked :p
02:00:28 <oren> "MNNBFSL is inspired by Brainfuck and Forth. Forth can manipulate return stack. Forth programmers can write puzzle like program by this feature."
02:01:06 <oren> "I think it is interesting, so I want to create esoteric language that can manipulate return stack."
02:01:22 <lifthrasiir> elliott: they *do* surely know English, they just don't write in it normally :p
02:01:31 <oren> This is very good English for a Japanese person.
02:01:39 <oren> In my experience anyway
02:01:52 <elliott> they get taught english y'know
02:02:08 <lifthrasiir> so do some 2 billion people around the world
02:02:11 <oren> In high school, but that's like I'm supposed to knwo French
02:02:26 <oren> but i can't speak a word
02:02:36 <Bike> japanese people have more incentive to learn english than you do to learn french, if you're not canadian
02:02:39 <elliott> lifthrasiir: yeah, that's my point
02:02:47 <lifthrasiir> at least I can barely remember rendezvous is a correct spelling
02:02:54 <oren> i'm canadian but i'm from Ontario
02:03:25 <Bike> not that canada, the other one
02:03:32 <elliott> oren: well, foreign language classes in the english world are basically hobbies
02:03:57 <lifthrasiir> and foreign language classes in the non-english world are basically for survivals
02:04:02 <elliott> knowing english is more of a necessity...
02:04:15 <elliott> (as in, treated as one by education systems)
02:07:12 <FireFly> Even without school-taught english, plenty of non-english-speaking countries consume lots of english media and pick it up that way
02:07:19 <FireFly> At least for countries that don't dub everything they import
02:08:18 <elliott> oren: btw, you know anarchy golf (where the author of MNNBFSL, yshl, participates) is a japanese site with many japanese players, right? :)
02:08:22 * FireFly 's pronounciation was terrible initially from having picked up english from pokemon and online forums
02:09:02 <oren> japanese people put subtitles in Japanese on everything
02:10:17 * oren , when he was in japan, saw a lot of badly used English
02:11:13 <FireFly> We sub movies/TV series as well, though games remain untranslated
02:11:14 <elliott> it is relevant that avoiding english while being a programmer who knows a lot of languages etc. is incredibly difficult...
02:11:27 <HackEgo> [wiki] [[MNNBFSL]] http://esolangs.org/w/index.php?diff=41271&oldid=41229 * Orenwatson * (+37) updtaed with info from inventor
02:11:44 <elliott> (so you'd expect programmers to generally have a decent grasp of english even regardless of the general statistics)
02:12:21 <oren> yeah becuase programming languages and libraries are based on english keywords
02:12:38 <oren> (with very few exceptions)
02:12:43 <elliott> well, moreso documentation
02:13:11 <elliott> ("def" and "int" aren't english words, so just treating keywords as opaque wouldn't be too much of an obstacle)
02:13:43 <oren> things like that
02:14:40 <oren> but yeah a lot of documentation doesn't bother having any other languages.
02:14:43 <elliott> I like how the full japanese name of the language is light novel title length.
02:15:00 <oren> that's the Japanese way
02:15:11 <oren> poketto monsuta! pokemon!
02:17:24 <Taneb> The full japanese name of the language
02:17:24 <oren> まだ名前のないBF風スタック言語
02:17:40 <Taneb> I assumed you were talking about Japanese for some reason
02:19:34 <oren> nah in Japanese the language is just called 日本語 "nihongo"
02:19:50 <oren> not light novel title length
02:19:52 <elliott> it translates as "My Little Sister Forced Me To Create A Brainfuck Derivative, And Now The Whole School Is Laughing At Me!"
02:20:22 <elliott> (it doesn't but it should)
02:25:23 <oren> 俺の妹はBF元着く言語を無理矢理着かせた！
02:25:42 <oren> My little sister forced me to make a BF derivative
02:27:16 <oren> 今、学校のみんなが俺を笑う！
02:27:33 <oren> now, eveyone at school is laughing at me!
02:27:44 <elliott> please don't make this language
02:28:08 <oren> trolololol.gif
02:28:39 <elliott> you know IRC doesn't support images, right
02:28:48 <Bike> some web clients do
02:29:05 <oren> yeah but when you see the name of the image you know what image it is
02:29:35 <oren> like doge.jpeg
02:29:53 <elliott> you can actually allude to things without adding file extensions
02:30:35 <oren> but its funnier when it's ohmaigaa.webm
02:31:24 <oren> wilhelmscream.mp3
02:31:28 <elliott> my favourite meme is rotten_dog.tiff.shar.bz2
02:32:30 <dianne_> isn't that one a bit old by now
02:32:32 -!- dianne_ has changed nick to dianne.
02:34:43 <shachaf> remember C:\Windows\media?
02:35:11 <oren> dan dadan dan dan daaaaan
02:36:27 <oren> 2girls1cup.mp4
02:37:00 <oren> trolololol.mp3
02:37:41 <oren> yerawizardharry.avi
02:38:45 <elliott> I'm glad this channel is getting even worse :p
02:39:43 <FireFly> I'm sad this channel is getting even worse :(
02:40:22 <oren> allthewhosdowninwhovillecryboohoo.webm
02:41:22 <elliott> okay this got old like five minutes ago
02:41:22 <Taneb> I'm mad this channel is getting even worse >:(
02:41:40 <oren> tomeitwastuesday.wmv
02:43:17 <Taneb> Sometimes I'm also Nathan
02:43:24 <Taneb> Although not on IRC
02:56:21 -!- nooga has joined.
03:00:00 -!- CADD has quit (Read error: Connection reset by peer).
03:00:47 -!- nooga has quit (Ping timeout: 255 seconds).
03:34:36 -!- CrazyM4n has joined.
03:34:37 -!- AndoDaan has joined.
03:45:02 -!- Sprocklem has joined.
03:47:13 <zzo38> Can a Pokemon card puzzle possibly be made up that the goal is for the players to cooperate to end the game as quickly as possible (it doesn't matter who wins or if it would end in a draw)? There may be many other conditions to be possible too
03:51:02 -!- shikhout has quit (Ping timeout: 258 seconds).
03:55:40 <J_Arcane> https://github.com/charcole/Z3
04:11:59 <zzo38> Conceding isn't allowed of course
04:16:09 <shachaf> zzo38: Yes, such a puzzle could possibly be made up.
04:21:46 -!- GeekDude has changed nick to GeekZzZzZ.
04:21:49 -!- GeekZzZzZ has changed nick to GeekZzZzZz.
04:24:28 <zzo38> Do you know how to make it good though?
04:31:37 -!- Dulnes has joined.
04:31:51 <Dulnes> Mmm not much discussed
04:40:34 -!- MoALTz_ has joined.
04:40:53 -!- GeekZzZzZz has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
04:43:43 -!- MoALTz has quit (Ping timeout: 265 seconds).
04:45:17 -!- nooga has joined.
04:46:37 -!- AndoDaan has quit (Quit: Quit).
04:48:21 -!- CrazyM4n has quit (Quit: Lost terminal).
04:49:37 -!- nooga has quit (Ping timeout: 240 seconds).
04:52:28 -!- dianne has left.
05:23:29 <Sgeo> `slist supdate
05:23:31 <HackEgo> slist supdate: Taneb atriq Ngevd Fiora Sgeo ThatOtherPerson alot
05:29:01 -!- ZombieAlive has quit (Remote host closed the connection).
05:37:45 -!- cluid has joined.
05:39:46 <newsham> whoa https://github.com/charcole/Z3
05:40:58 <shachaf> #haskell is such a disaster
05:42:55 <cluid> Interesting newsham
05:43:37 <cluid> @newsham I was reading alangsec article an dyou were mentioned
05:45:09 <cluid> I was thinking about making an esolang page on ROP, this is one thing i read http://langsec.org/papers/Bratus.pdf
05:45:16 <cluid> what do you think about such an article?
05:46:11 -!- ZombieAlive has joined.
05:47:17 <newsham> i think the topic is interesting, though i'm not sure if looking at them through the lense of "weird machiens" is all that enlightening...
05:47:26 <newsham> i guess if thats your passion, its a valid way of looking at it..
05:47:50 <cluid> The best article is geometry of flesh on the bone
05:47:51 <newsham> thats a pretty flattering mention of my name.. i like it..
05:48:33 <newsham> everyone seems to be trying to be really deep and philosophical and clever about it, but its just understanding the computer and making it do stuff...
05:49:06 <newsham> i dont think its that great of a leap of faith to say that if you control the stack frame of a program you basically control the control flow and the program
05:49:33 <newsham> running a program using page faults is a little more clever..
05:50:41 <shachaf> i,i nop-oriented programming
05:52:53 <cluid> newsham, yeah some of the langsec stuff I see is slightly askew, but seeing security people finally say that you shosuld be recognizing languages rather than regex blacklighting 'bad' stuff makes me really happy
05:54:04 <newsham> security professionals should be able to tell you that whitelisting is better than blacklisting anyway :)
05:54:10 <newsham> and regexps arent often the best tool
05:55:12 <cluid> but I think ROP is a cool esolang
05:55:16 <cluid> and deserves a place on the wiki, no?
05:55:43 <newsham> yah, rop is kinda a neat esolang
05:55:53 <newsham> well, family of esolangs really
05:57:05 <newsham> ever see massalin's superoptimizer paper?
05:57:28 <cluid> no, ill have a look now
05:57:33 <newsham> http://web.stanford.edu/class/cs343/resources/cs343-annot-superopt.pdf
05:58:31 <cluid> he doesn't qualify "finds shorted program that computes a given function" with any caveats lol
05:58:47 <newsham> would be neat "growing" programs organically (ie. search/annealing/genetic prog) and then trying to analyze them
05:59:23 <newsham> i bet sooner or later you'd find much cooler things than rop going on
06:05:31 <cluid> http://web.stanford.edu/class/cs343/
06:05:56 <cluid> is that a 'trusting trust' compiler?
06:06:52 <cluid> il have to look around his course notes, I've wanted to write a compiler but found it very difficult
06:07:11 <newsham> http://www.dwheeler.com/trusting-trust/dissertation/wheeler-trusting-trust-ddc.pdf
06:19:11 -!- oerjan has joined.
06:23:24 <int-e> All Dominosa secrets have been revealed. Fizzie used trits.
06:23:49 <oerjan> i see you've adapted some of mine. now to look at actual code...
06:23:59 <int-e> And tails kindly taught me a dc trick (obvious in retrospect, why didn't I think of that...)
06:25:59 <int-e> I should've found the gcd thing
06:25:59 <int-e> the >, that's clever :)
06:26:04 <oerjan> ah you missed both of those? yeah i did feel clever about those
06:26:27 <int-e> but I got some value out of filter instead
06:28:15 <int-e> oerjan: I also liked this adjacency test: j-i==6^(div j 6-div i 6) but it's too long.
06:28:26 -!- Sprocklem has quit (Ping timeout: 256 seconds).
06:30:30 <oerjan> i got a bit annoyed about the defined precedence of `elem` vs. >, could have saved the parentheses if they were just a little different
06:31:13 -!- shikhin has joined.
06:31:41 -!- shikhin has quit (Read error: Connection reset by peer).
06:32:11 <int-e> now I'm hoping for fizzie to translate the binary encoding into burlesque
06:33:17 <int-e> http://sprunge.us/GJcW?hs
06:33:41 <oerjan> ...why would he use trits anyway?
06:34:08 -!- nooga has joined.
06:34:38 <int-e> I really expected the decoding function to grow more when I changed the encoding to binary.
06:36:13 -!- shikhin has joined.
06:37:48 <int-e> (it actually shrunk by 2 characters)
06:38:10 <oerjan> oh i wondered how to shorten concat.lines / filter(>' ') but if you are using filter everywhere that's not a problem
06:38:17 -!- shikhin has quit (Read error: Connection reset by peer).
06:38:33 -!- nooga has quit (Ping timeout: 240 seconds).
06:38:41 <int-e> though that's a bit hard to measure; is the [0..29] part of the decoding function or not?
06:40:17 <oerjan> ooh the g(<'@') is also clever, i tried to somehow use show(i,j) or the like but kept fooling around with init and tail to remove the parens
06:40:37 <oerjan> _i_ should have found that, i guess :P
06:41:12 -!- shikhin has joined.
06:41:19 <int-e> it's actually nice to finally have independent tricks that could be combined :)
06:42:34 <oerjan> (btw henkma hasn't showed up for this one...)
06:43:06 <int-e> He didn't attempt Make 24 either.
06:43:32 <int-e> maybe he prefers the non-compression tasks, or perhaps he just didn't catch up to me.
06:44:41 -!- Dulnes has quit (Quit: Connection closed for inactivity).
06:45:18 <int-e> (henkma has a 9986 average, I'm pretty sure he never submits solutions that score fewer than 10k points upon submission.)
06:46:09 <oerjan> you found several of the other tricks i thought you might miss, though. i really felt so clever about this one overall.
06:46:24 <int-e> the zip[0..] one was hard.
06:46:39 <oerjan> well yeah that's like the basis for the algorithm
06:46:47 -!- shikhin has quit (Read error: Connection reset by peer).
06:48:10 -!- shikhin has joined.
06:48:16 <int-e> well, I was playing with this version at one point: http://sprunge.us/jEBB?hs
06:50:16 -!- shikhin has quit (Read error: Connection reset by peer).
06:53:10 -!- shikhin has joined.
06:53:28 <oerjan> i tried several ways of constructing a representation of an unordered pair before i realized it was shorter to just stick both versions in literally
06:53:32 -!- shikhin has quit (Read error: Connection reset by peer).
06:54:14 <int-e> anyway, I had trouble making the transition from "filter index j from list" to "filter index j *and* its associated character, which we happen to know already, from a list."
06:55:30 <int-e> with a *really* short fromEnum, I'd have liked to use fromEnum u*fromEnum v as an unordered pair.
06:56:03 <oerjan> i am pretty sure i had some versions that didn't but they were early and unsubmitted and i didn't save them
06:56:31 <oerjan> actually the first submitted did zip[0..29] :P
06:57:16 <int-e> I wrote that, and then ... hmm ... oh...
06:57:41 <int-e> But I didn't submit before I reached 175.
06:58:10 -!- shikhin has joined.
06:58:21 <int-e> I still find it amazing how much the knowledge that things can still be improved helps (me, at least).
06:58:45 <oerjan> i've found it amazing how much that helps you too ;)
06:59:14 <int-e> I tend to try crazier ideas.
06:59:36 <int-e> Because, you know, one of them *has* to work.
07:00:38 <oerjan> btw would fromEnum u*fromEnum v have worked? it's not entirely correct
07:00:55 <oerjan> or maybe it is, for the character ranges that show up
07:01:10 <int-e> for '0'..'4', right.
07:01:42 -!- shikhin has quit (Read error: Connection reset by peer).
07:02:01 <int-e> 7:49 5:50 17:51 13:52 <-- except for '0', each has a unique prime factor.
07:02:17 <oerjan> 48 49 50 51 ... hey i was about to deduce that :P
07:02:50 <oerjan> um i think you're off by 1
07:02:55 <int-e> > sort [fromEnum a*fromEnum b | a<-['0'..'4'], b<-[a..'4']] -- what I actually did.
07:02:56 <lambdabot> [2304,2352,2400,2401,2448,2450,2496,2499,2500,2548,2550,2600,2601,2652,2704]
07:03:10 -!- shikhin has joined.
07:04:15 <oerjan> hm it would work a bit longer than '0'..'4'
07:05:14 <oerjan> does it work for all of '0'..'9'
07:05:47 <int-e> > nub $ map length $ group $ sort [fromEnum a*fromEnum b | a<-['0'..'9'], b<-[a..'9']]
07:06:46 * oerjan should learn not to start typing when he's already told you the idea
07:06:48 <int-e> the argument gets a bit messy though; 49 and 56 rely on the former being a square.
07:08:14 <int-e> well, the prime factor argument works by not considering prime factors 2 and 3 at all. so 56*56 and 49*48 end up in the same category. (and 49 being a square does not help...)
07:08:31 <oerjan> next question: does this work for arbitrary multisets :P
07:08:51 <int-e> after adding some offset, yes.
07:08:57 <int-e> the offset may be large, though.
07:08:58 -!- shikhin has quit (Read error: Connection reset by peer).
07:09:37 <oerjan> (probably not, but what's a counterexample)
07:09:45 <oerjan> primes > 7 always work
07:09:56 <oerjan> because they are only in one of the numbers
07:10:15 <oerjan> and then maybe some others get excluded
07:10:38 -!- shikhin has joined.
07:11:04 <oerjan> 48,49,50,51,52,53,54,55,56,57 -> 51,52,53,54,55 are out
07:12:40 <oerjan> lifthrasiir: at the moment, wondering if products of 48..57 (i.e. ASCII values of digits) can be uniquely factored back
07:13:24 <int-e> so 48, 49, 54 and 56 use only 3 primes total, so no.
07:13:25 <lifthrasiir> do you have to accept two digits (for the purpose of codegolf) and nothing else?
07:14:38 <oerjan> lifthrasiir: it's about comparing unordered pair of digits, although this is actually a suboptimal solution to the golfing problem, but it's a bit mathematically interesting
07:14:55 <oerjan> it works for pairs, btw.
07:15:23 <oerjan> so now it was about products of lists
07:15:53 <lifthrasiir> oerjan: so if the input is "13247" and "74312" it should say "YES" while "12345" vs. "123456" should say NO etc?
07:16:22 <oerjan> lifthrasiir: well yeah...
07:18:45 -!- shikhin has quit (Read error: Connection reset by peer).
07:19:31 -!- shikhin has joined.
07:19:52 <oerjan> 48 = 2^4*3, 49 = 7^2, 50 = 2*5^2, 54 = 2*3^3, 56 = 2^3*7
07:20:05 <lifthrasiir> oerjan: wouldn't the arithmetic progression work?
07:20:20 <int-e> > 48^18 * 49^11 == 54^6 * 56^22
07:21:09 <oerjan> lifthrasiir: i'm not actually asking for variations of the problem btw
07:21:49 <oerjan> but yeah, there should be some offset that works, as int-e already mentioned
07:21:56 <int-e> oerjan: I think that's the smallest counterexample, but I may have missed something. integer lattices are hard.
07:22:27 <oerjan> probably NP-hard, i should imagine
07:23:43 <oerjan> (4,1,0,0),(0,0,0,2),(1,0,2,0),(1,3,0,0),(3,0,0,1)
07:24:13 -!- shikhin has quit (Read error: Connection reset by peer).
07:24:31 <oerjan> oh right 5 is actually excluded at that point
07:25:12 <oerjan> *(4,1,0),(0,0,2),(1,3,0),(3,0,1)
07:25:19 -!- shikhin has joined.
07:26:15 <fizzie> Yes, I went with trits to have a simple "skip/+1/+6" ternary-choice for each digit. But it's very possible the thing you did in dc would be shorter.
07:26:47 -!- Patashu has joined.
07:27:36 <fizzie> For the curious, http://sprunge.us/hOjN is my log, though it's pretty much just about the non-cheating version.
07:28:41 <fizzie> Oh, the commented version is not up-to-date. Oh well.
07:29:04 <int-e> Yeah, I did some serious number crunching for the dc version (just a couple of hours of CPU time, but still).
07:30:38 -!- shikhin has quit (Read error: Connection reset by peer).
07:31:07 -!- MDude has changed nick to MDream.
07:31:37 -!- shikhin has joined.
07:31:59 -!- shikhin has quit (Read error: Connection reset by peer).
07:32:54 <int-e> all to get from FB88?1+3%d5408*3095-*- (manual polynomial interpolation) to ?B74+I22094|3908- and then ?988-2 72889|21+
07:35:32 <fizzie> I did wonder where you snatched ?988-2 72889|21+ from.
07:36:21 <int-e> I got extremely lucky there :)
07:36:42 -!- shikhin has joined.
07:36:54 <int-e> even so, I wouldn't be surprised if there were some shorter code for the initialisation. it's just hard to search for.
07:38:52 -!- shikhin has quit (Read error: Connection reset by peer).
07:40:44 <oerjan> int-e: i think your counterexample is correct (and unique up to a constant multiple in the powers)
07:47:06 <oerjan> that new brainfuck problem is _definitely_ a compression problem.
07:48:13 -!- MoALTz_ has quit (Quit: Leaving).
07:49:48 <int-e> not a nice one either.
07:50:20 <oerjan> looks a bit two-pronged
07:50:57 <int-e> why +++[>+++++<-]> and not >+[>-[-<]>>]>-
07:51:48 <int-e> (and how does that +[>-[-<]>>] loop work anyway?)
07:51:49 -!- shikhin has joined.
07:52:37 <int-e> ok, with some luck each is the lexicographically smallest program that works.
07:53:39 <int-e> (modulo the usualy problems when one solves the halting problem)
07:54:33 -!- shikhin has quit (Read error: Connection reset by peer).
07:54:39 <int-e> 6^16 is a bit too large though :)
07:55:21 <oerjan> i think that +[>-[-<]>>] loop goes to the left of the tape end, no
07:55:53 <int-e> [->+++++<] is lexicographically smaller than [>+++++<-], so there goes that theory.
07:56:26 <oerjan> they linked to the wiki article so probably just snatched them from there
07:57:04 <int-e> yeah, that skips over the left end.
07:57:39 <int-e> so let's assume there's a 0 there... then...
07:59:12 <oerjan> <int-e> why +++[>+++++<-]> and not >+[>-[-<]>>]>- <-- i think "because the wiki article misses the latter" is a pretty obvious answer, there
07:59:30 <oerjan> and the reason it misses it is because the first is older and the latter isn't shorter
08:00:35 <oerjan> (not that i trust the wiki article to be comprehensive in any way beyond the old "shortest 2-cell wrapping" entries)
08:00:59 <oerjan> which i heard were found by exhaustive search
08:01:14 <int-e> Oh, I copied it wrong. There's a > in front of the loop.
08:01:23 <b_jonas> What are the brainfuck derivatives that are so old they are expemt from the guideline against creating brainfuck derivatives just because of their age?
08:01:45 <oerjan> b_jonas: Ook! Ook? Ook.
08:01:53 <oerjan> and some say, that's all.
08:02:11 <b_jonas> Ook! definitely, and brainfuck definitely as well
08:02:29 <oerjan> brainfuck isn't a brainfuck derivative hth
08:02:47 <Bike> it's the trivial derivative!
08:03:02 <b_jonas> When was bitfuck invented?
08:04:13 <oerjan> well boolfuck is exempt because it has actual theoretical interest.
08:04:22 <oerjan> whether or not it is old.
08:05:38 <int-e> isn't the trivial derivative zero...
08:06:23 <Bike> aren't you zero
08:06:33 <oerjan> int-e: https://esolangs.org/wiki/Zero isn't among the most trivial :P
08:06:44 <oerjan> but it _is_ a brainfuck derivative
08:07:30 -!- shikhin has joined.
08:08:05 <oerjan> it seems to have the Turing-complete category, although afair that's still unproved
08:08:28 <Bike> a list of the first hundred thousand brainfuck programs. nifty
08:08:30 <oerjan> maybe it can be considered poved
08:08:42 <Bike> non-halting ones. w/e
08:09:07 <oerjan> yeah 100000 _should_ be plenty enough to write a bf interpreter in
08:11:00 <oerjan> actually it fails to be turing-complete in the sense that it's not interpretable by a TM
08:11:20 <Bike> dbfi is like 1300 bits ish sooooo yes.
08:11:21 <oerjan> but people seem to disagree on whether that's a requirement
08:11:50 <int-e> oh so it's merely Turing-hard?
08:12:15 <oerjan> (by the analogy with NP-complete / NP-hard it should be, but people seem to disagree on whether that analogy holds)
08:12:53 <int-e> it's a rather impractical notion
08:17:26 <Jafet> Zero⊕, where bit n of the sequence is the parity of the number of n-1-bit programs that halt
08:19:05 <int-e> oerjan: also, does "Turing-hardness" require that we can construct a universal machine, or merely that it exists, mathematically speaking?
08:19:25 <Bike> probably depends on your intuitionism
08:19:39 <int-e> It's very weak. I like classicaly logic.
08:19:52 <int-e> Oh, another stray "y", where do they come from?
08:19:52 -!- Dulnes has joined.
08:20:00 <oerjan> int-e: i'd think it requires constructability
08:21:00 <oerjan> it requires that we can translate any turing machine computation to it, constructively - but the constructive construction doesn't itself have to be known hth
08:21:58 <oerjan> (the unknown constructive construction needs to handle _all_ computations for a universal TM.)
08:21:59 <shachaf> oerjan: i don't think that's covered in 1984 hth
08:22:01 <int-e> well, we can translate the TM into a universal machine (which we know exists even if we can't write it down) and some suitable input.
08:22:46 <int-e> so even if we could not calculate a single bit of the "padding sequence" we'd still be in the Turing-hard domain.
08:23:00 -!- nooga has joined.
08:23:08 <Dulnes> What does it mean when lambdabot says mueval-core: time limit exceeded
08:23:24 <int-e> Dulnes: it means that mueval's time limit was exceeded.
08:23:43 <int-e> Your computation took too long!
08:24:09 <int-e> This may happen due to circumstances unrelated to your code, like the server starting to swap heavily.
08:24:45 <HackEgo> [wiki] [[Zero]] http://esolangs.org/w/index.php?diff=41272&oldid=41010 * Oerjan * (+420) /* Computational class */
08:25:00 <Jafet> So, there's this tacit assumption that if a computation model is at least Turing-complete, you can construct a machine that comes preloaded with the input to any other machine
08:25:02 <Dulnes> I need to make a language that is like the alphabet but uses ~=_<>{ charactors as letters and numberz
08:25:24 <Dulnes> ~=_<>{}[]|$€£¥¢￦§^`°¿¡\«»®©
08:25:27 <Jafet> But if you have a language that explicitly requires its own halting sequence to work, that may not be possible
08:26:16 <int-e> Jafet: stop diagonalizing, it makes brains explode.
08:26:56 <int-e> though things may be fine for languages that accept the empty program
08:27:26 <Dulnes> Mmm that would be hard if i had to use {} [] as letter\number sequence what would i use as a bracket?
08:27:26 -!- nooga has quit (Ping timeout: 256 seconds).
08:27:53 <Jafet> In other words, if your definition of Turing-completeness requires the translation itself to be computable
08:28:03 <Jafet> (by a Turing machine, I suppose)
08:28:08 <oerjan> Jafet: it's not that tacit, you can prove it i should think
08:28:35 <oerjan> Jafet: my definition does
08:28:45 <Dulnes> Well if i compare ~ to q its the same in my perspective
08:28:52 <Jafet> You might run into a bit in the Zero sequence that corresponds to an unprovable theorem
08:29:07 <oerjan> Jafet: however my definition includes the input in what needs to be translated
08:29:32 <oerjan> Jafet: and _also_ allows using the input of the target, which is essential for Zero.
08:30:18 <oerjan> (my definition also allows a final translation of output at the end)
08:31:25 <Bike> something like zero, but oriented so that you can program around unprovable bits, to some extent. an encoding of something more robust than brainfuck probably.
08:31:48 <Dulnes> L = Bracket, [ = letter
08:31:57 <Jafet> That definition makes sense (eg. for fractran) but I'm not sure if just using it to translate every TM to one universal interpreter in the target language is in the right spirit
08:32:03 <oerjan> Dulnes: btw you realize that any language which is _solely_ about using cryptic syntax is uninteresting to nearly every regular in this channel?
08:32:39 <Bike> something where even though the sequence is computable the language is less ambiguously TC
08:32:52 <Dulnes> Im trying to occupy space
08:34:32 <Jafet> re. space, IRC is like being on a bus with three hundred people
08:34:33 <Dulnes> {]} £[<>{ |_<€]® = you dirty person
08:35:02 <Dulnes> I liked cramped quarters
08:35:36 <Dulnes> Sooooooon ill just be typing in wingdings
08:35:58 <Dulnes> Or is that wingdings 2 - 3 idk
08:36:03 <oerjan> Jafet: we have invented the terminology "BF-complete" for languages that need to be able to deal with somewhat reasonable I/O
08:37:31 <Dulnes> I think im going insane
08:37:49 <oerjan> maybe you should go lie down a bit hth
08:38:16 <Dulnes> Ive been scrolling to long
08:38:42 <Dulnes> If i teach my stuff the stuff i can make the stuff
08:38:59 -!- shikhin has quit (Ping timeout: 244 seconds).
08:42:04 <Dulnes> I can just imagine what it will look like when im done :0
08:43:13 * Dulnes dies of cafine over dose
08:45:06 <HackEgo> [wiki] [[Pikalang]] N http://esolangs.org/w/index.php?oldid=41273 * Grotr * (+1743) Initial commit
08:46:00 <HackEgo> [wiki] [[Joke language list]] http://esolangs.org/w/index.php?diff=41274&oldid=41171 * Grotr * (+15) /* Brainfuck derivatives */ add Pikalang
08:56:24 <cluid> was Alan himself Turing complete?
08:57:50 <oerjan> cluid: i've been wondering before whether there are rule 110 metacells of some kind
08:58:19 <cluid> I wrote a CA simulator to prepare to search for things that might help finding metacells
08:59:02 <int-e> oh, GG got published early today
08:59:14 <int-e> priorities, priorities, gotta have them straight.
08:59:23 <oerjan> there's complications, though, because the max speeds of transmission leftwards and rightwards are different in "ether" than in a zero background
08:59:38 <cluid> so rule 110 does not move to the right?
08:59:55 <oerjan> but through ether it does
09:00:10 <oerjan> leftwards, it moves faster through zeros
09:00:35 <oerjan> no, ether is that special pattern that tends to quickly show up from random starting conditions
09:01:02 <cluid> http://upload.wikimedia.org/wikipedia/en/b/bd/CA_rule110.png would that be the majority of stuff here
09:01:13 <oerjan> it's probably a big part of what makes rule 110 so intricate
09:01:52 <oerjan> cluid: yes, the "background" so to speak
09:02:19 <cluid> that makes this metacell stuff seem a lot harder
09:02:53 <oerjan> but you might be able to make a metacell that moves at a different speed
09:03:21 <cluid> so the input pattern would have to be encoded at a slope?
09:03:27 <oerjan> or, as i've mentioned before, with "relativistic" speed - where you don't translate time to time
09:03:45 <oerjan> yes, and possibly even with time sloped
09:09:32 <cluid> ## ### # in rule 30 this repeats
09:09:41 <cluid> i got it from wiki
09:09:46 <oerjan> cluid: in fact that png shows the phenomenon where any disturbance moving leftward through zeros turns into a pattern of ether possibly with "gliders" in it
09:10:17 <cluid> and a spaceship too?
09:10:20 <cluid> that thing going down
09:10:21 <oerjan> rule 30 is that other complicated rule iirc, but much harder to do anything with because it's reversible
09:10:39 <oerjan> cluid: well for rule 110 "glider" is the term for spaceship
09:10:41 <cluid> rule30 would have to encode other reversible rules only
09:10:47 <oerjan> been used in papers and stuff
09:10:48 <cluid> limiting search that way might be helpful
09:11:10 <cluid> i havent look at any research on 1D CAs
09:11:18 <cluid> i wonder if its worth it to do that
09:11:24 <oerjan> @google rule 110 gliders
09:11:39 <lambdabot> Plugin `search' failed with: <<timeout>>
09:11:43 <oerjan> @google rule 110 gliders
09:11:58 <lambdabot> Plugin `search' failed with: <<timeout>>
09:12:09 <elliott> oklopol does research on 1D CAs I think
09:12:32 <oerjan> cluid: the top google hit for that gives a paper with lots of rule 110 glider and repeating patterns
09:12:40 <oerjan> (at least it's top for me)
09:12:53 <cluid> ok but im not especally interested in rule 110
09:12:58 <cluid> but i wil have a look at this
09:13:19 <oerjan> cluid: ok, rule 110 is like the simplest rule that has gliders though.
09:13:34 <oerjan> well useful ones, anyway
09:17:22 <cluid> Rule 110 - International Center of Unconventional Computing
09:18:43 <cluid> http://en.wikipedia.org/wiki/Andrew_Adamatzky
09:19:15 <cluid> He has used slime moulds to plan potential routes for roadway systems
09:19:42 <cluid> He has also shown that the billiard balls in billiard-ball computers may be replaced by soldier crabs
09:19:52 <cluid> http://en.wikipedia.org/wiki/Mictyris_guinotae#Crab_computing
09:21:56 -!- dts has joined.
09:27:55 <fizzie> oerjan: We'd better wear some armor, then.
09:31:43 * oerjan vaguely assumes he's missing a pun
09:32:13 <cluid> http://en.wikipedia.org/wiki/Q_code these codes look similar to an instruction set
09:34:06 <fizzie> oerjan: Well, I mean, you want to be a bit wary if someone's after your heart. You will need it to live, after all.
09:42:11 <int-e> oerjan: I don't know.
09:44:29 <lambdabot> http://www.youtube.com/watch?v=b-3BI9AspYc
09:44:37 <oerjan> seems to have fixed itself
09:49:25 <oerjan> <elliott> okay this got old like five minutes ago <-- it got old even while browsing it in the logs hth
10:04:14 <oerjan> fwiw http://oerjan.nvg.org/r110.txt is an old file i made. a bit down in it is a list of all the (6, iiuc my own notes) limiting ways for a disturbance to travel leftward into zeros.
10:05:51 <oerjan> some of them have the additional gliders moving at -1/2, the same as the maximal speed in ether, which means you can treat them as essentially growing a leftward infinite ether in a way that nothing can catch up with.
10:08:02 -!- nooga has joined.
10:08:49 <int-e> is rule 110 known to be TC against a zero background?
10:09:14 <oerjan> which is part of what made me do this classification
10:09:20 <int-e> (I don't really like that ether business very much)
10:09:38 <oerjan> of course i don't think it's known to be TC against an ether background either
10:09:55 <oerjan> (you need infinite setup both ways.)
10:10:07 <oerjan> the ether is cool. embrace the ether.
10:11:22 <cluid> I feel like this stuff is awfully difficult
10:12:13 <oerjan> you have to search through a large number of patterns.
10:12:28 <int-e> oh. the proof has periodic control structures to both sides... it's worse than I thought.
10:12:34 -!- nooga has quit (Ping timeout: 264 seconds).
10:13:02 <cluid> is there reason to believe the simple metacell approach metaPattern = concatMap encodeCell is not working? where encodeCell :: Bool -> Vector n Bool
10:13:22 <oerjan> to get it working in a zero background you'd need to invent some glider guns, i think
10:13:47 <cluid> we can use an ether background by having encodeCell False = ether?
10:14:02 <int-e> oerjan: ok, with the given state of affairs, having something work against a plain ether background would be a great improvement.
10:14:38 <oerjan> mind you, i don't know if there have been recent improvements.
10:15:02 <oerjan> i assume it's not a very hot research area :P
10:16:17 <int-e> right. no grant money in it
10:16:47 <oerjan> cluid: as long as you don't require metacells to move at full speed, i don't see a real reason why that couldn't possibly work.
10:17:25 <oerjan> but then, the life metacell doesn't move at full speed either.
10:19:22 <cluid> http://lpaste.net/115417 I've put a clearer explanation
10:19:24 <oerjan> because the main speed obstacles i see have to do with large fields of zeros - you cannot get anything to move rightwards into them, and you cannot _stop_ anything from moving leftwards into them.
10:19:55 <oerjan> and also i _suspect_ zeros are the only thing that allow maximal leftward speed in them
10:20:53 <cluid> maybe simulation would have to allow for drift
10:20:56 <oerjan> cluid: but i'd say what you're suggesting is the "obvious" thing to try first.
10:21:25 <cluid> simulations may be have to occur in a wrapping world
10:22:03 <oerjan> well if you want not to run out of memory...
10:23:34 <oerjan> cluid: you can probably do compression and hash-life like things to get simulation of metacells to have only a constant overhead.
10:24:13 <cluid> that's a good idea to speed up simulation and search
10:25:27 <cluid> http://lpaste.net/115417
10:25:34 <cluid> I gues I got a lead for what to search for
10:25:48 <cluid> 1. find repeating patterns of the same length
10:25:58 <cluid> 2. try putting them together and search for pyramids
10:26:20 <cluid> im sure that if I try this I will fail and not find anything but if it worked that would be nice
10:37:44 -!- cluid has quit (Quit: Leaving).
10:38:55 <int-e> @tell cluid "im sure that if I try this I will fail and not find anything but if it worked that would be nice" <-- this is called "research", hth.
10:42:15 -!- shikhin has joined.
10:44:41 -!- nooga has joined.
11:04:41 -!- Dulnes has quit (Quit: Connection closed for inactivity).
11:15:45 -!- boily has joined.
11:15:48 -!- nooga has quit (Ping timeout: 272 seconds).
11:22:38 <boily> Good morning, shachaf.
11:23:02 <boily> @localtime shachaf
11:23:02 <lambdabot> Local time for shachaf is Mon Dec 1 03:22:21 2014
11:23:12 <lambdabot> Local time for boily is Mon, 01 Dec 2014 06:22:31 -0500
11:24:20 -!- AndoDaan has joined.
11:25:18 -!- oerjan has quit (Quit: Later).
11:55:34 -!- Patashu has quit (Ping timeout: 258 seconds).
12:23:16 -!- boily has quit (Quit: RESERVED CHICKEN).
12:56:20 -!- AndoDaan_ has joined.
12:58:24 -!- AndoDaan has quit (Ping timeout: 250 seconds).
13:12:19 -!- Phantom_Hoover has joined.
13:34:33 -!- _AndoDaan_ has joined.
13:34:43 -!- AndoDaan_ has quit (Ping timeout: 245 seconds).
13:44:57 <oren> good morning everyone
13:45:09 <oren> i have just finished breakfast
13:48:58 <J_Arcane> oren: I have just write one-half of a struct system.
13:50:12 <oren> for what platform?
13:51:05 <J_Arcane> In Heresy, my BASIC/Lisp thing.
13:51:37 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
13:53:30 <HackEgo> [wiki] [[GridScript]] http://esolangs.org/w/index.php?diff=41275&oldid=41268 * SuperJedi224 * (+160)
13:56:24 <oren> basic mixed with lisp... yeah 'Heresy' is a great name for something like that!
13:56:58 <oren> sounds like someything i'd like though. I started programming with Visual Basic
13:57:05 <J_Arcane> https://github.com/jarcane/heresy
13:57:39 <J_Arcane> Yeah. It started as a joke, but in the future I might like to actually use it as a sort of 'gateway LISP' for guys like me who started out in BASIC and wanna learn LISP and FP.
14:05:52 <oren> ah, it's implemented in racket!
14:06:30 <oren> i used racket for a few of my courses
14:12:28 -!- _AndoDaan_ has quit (Quit: Quit).
14:21:33 <oren> oh god. pi pi pika ka chu
14:26:44 <oren> well mightas well then
14:34:01 -!- `^_^v has joined.
14:39:04 <HackEgo> [wiki] [[User:Orenwatson]] http://esolangs.org/w/index.php?diff=41276&oldid=41257 * Orenwatson * (+258) i hope your happy elliot
14:40:02 -!- shikhin has quit (Ping timeout: 258 seconds).
14:42:52 <HackEgo> [wiki] [[User:Orenwatson]] M http://esolangs.org/w/index.php?diff=41277&oldid=41276 * Orenwatson * (+0)
14:47:19 <oren> i hope yor happy elliott
14:49:23 <HackEgo> [wiki] [[User:Orenwatson]] http://esolangs.org/w/index.php?diff=41278&oldid=41277 * Orenwatson * (+101) added translation hth
14:50:24 <oren> yor idia yor falt elliott
14:51:32 <oren> 「俺の妹はＢＦ元着く言語を無理矢理作らせたから、学校のみんなが俺を笑う！」
14:53:33 -!- oren has quit (Quit: Lost terminal).
15:12:55 <HackEgo> [wiki] [[GridScript]] http://esolangs.org/w/index.php?diff=41279&oldid=41275 * SuperJedi224 * (+17) /* Command Summary */
15:14:57 -!- shikhin has joined.
15:22:35 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
15:22:41 -!- Phantom__Hoover has joined.
15:29:25 -!- drdanmaku has joined.
15:48:08 -!- oren has joined.
16:04:38 -!- oren has quit (Quit: Lost terminal).
16:05:14 -!- Sprocklem has joined.
16:07:54 -!- oren has joined.
16:20:26 -!- oren has quit (Read error: Connection reset by peer).
16:44:27 -!- mihow has joined.
16:56:32 -!- Sprocklem has quit (Ping timeout: 252 seconds).
16:58:17 -!- Sprocklem has joined.
16:59:35 -!- oren has joined.
16:59:41 <J_Arcane> ow. I think my brain just went into an infinite loop ...
17:00:06 <oren> why did firefox hang my whole computer just now?
17:00:33 <oren> I had to restart my whole system
17:00:49 -!- GeekDude has joined.
17:01:45 <oren> is there a command line tool that will run a program but limit it to 1GB of memory total?
17:03:07 <J_Arcane> But it might be more extreme than what you want: http://www.sandboxie.com/
17:03:27 <oren> in C there is setrlimit.
17:03:30 <Melvar> You can run ulimit first,
17:03:38 -!- Sprocklem has quit (Ping timeout: 240 seconds).
17:05:35 <oren> what the hell FF, why do you just segfault when you can't hog any more memory???
17:05:59 <oren> it really wants ALLLLL my memory
17:07:05 <Melvar> A segfault from a memory limit?
17:07:37 <oren> if you ulimit 1000000000 firefox just segfaults as soon as it starts
17:08:08 <oren> it needs 1400000000 just to start up
17:08:30 <oren> I remember when XP needed 64 meg and we called it a hog
17:09:36 <oren> i am using oftlimit
17:09:41 <oren> softlimit command
17:09:49 <oren> which is in bytes
17:11:38 <oren> if you do softlimit -a 1000000000 firefox it segfaults
17:11:55 -!- b_jonas has quit (Ping timeout: 250 seconds).
17:12:28 -!- b_jonas has joined.
17:14:52 <oren> ok so to prevent firfox form hanging and eating my whole system
17:15:12 <oren> i'm gonna compromise and use softlimit -a 1500000000
17:15:28 <oren> limit it to 1.5 GB.
17:15:52 <oren> that allows it to start up but stops it if JS starts to hog everything
17:16:11 -!- nortti has changed nick to lawspeaker.
17:17:33 <Melvar> Funny, I don’t even know that problem. Must be NoScript.
17:17:45 -!- lawspeaker has changed nick to nortti.
17:17:50 -!- bb010g has quit (Quit: Connection closed for inactivity).
17:18:00 <Melvar> Or possibly it merely gets lost in the underflow.
17:18:12 <oren> I use noscript too but if you allow too many tabs to run
17:18:26 <oren> it can stillhappen asit did 5 minutes ago
17:18:36 <oren> i had to hold down power button
17:18:53 <newsham> eight megs and constantly swapping
17:19:01 <newsham> now emacs is a model of leanness
17:19:26 <oren> FF needs 1.5 GB! GB!
17:19:43 <oren> unbelievable to the me of 2000
17:20:01 <oren> (yes I knew what a GB was when I was 7)
17:20:11 <Melvar> Mine normally eats about 6GB …
17:20:42 <Melvar> That’s what I get for over 9000 tabs open.
17:21:19 <oren> yeah i only have a few GB of memory total
17:22:13 * Melvar has a look on his netbook.
17:22:37 <oren> this laptop cost me 260 dollars
17:22:57 <oren> it is very very cheap
17:24:25 -!- nortti has changed nick to lawspeaker.
17:24:56 <newsham> my pet peeve is when things that used to work on an old laptop stop working...
17:25:08 <newsham> like recently my chromebook has been having cpu load issues with netflix and youtube
17:25:12 <newsham> which worked perfectly fine for years
17:25:32 <newsham> so either the software got more bloated on my chromebook or the codec choices on the server got more demanding
17:26:20 <oren> there should be a button on firefox that stops all JS on the page
17:26:41 <oren> so i can hit it when the memory spikes
17:26:55 <oren> like ctrl-fuckjs
17:27:40 -!- cluid has joined.
17:27:42 -!- lawspeaker has changed nick to nortti.
17:28:12 <oren> we were talking about how FF is a memory hog
17:29:13 <oren> X by itself used to hog memory, now FF uses 4 times as much
17:30:57 <oren> and everyone is for some reason fine with that
17:31:27 <oren> when moores law hits a fundamental limit, everyone will be in for
17:31:28 <cluid> I am interested in Gopher
17:31:31 <newsham> 1) browsers turn out to be really useful these days, 2) in dollar amounts, its not using that much memory
17:31:40 <cluid> Gopher is extremely light weight
17:31:47 <cluid> You can implement it with low memory in a page of cod
17:31:55 <newsham> honestly i'm more concerned about browsers security issues than its memory consumption issues
17:32:14 <newsham> the http protocol itself is not elegant but is not that complicated
17:32:31 <cluid> web standards have grown ridiculously out of proportion
17:32:33 <newsham> TLS is a bit ... but its also very useful...
17:32:36 <oren> when a page of JS can hang my system it is a giant security issue
17:32:57 <newsham> maybe you should be using chrome and not ffox ;-)
17:33:01 <cluid> JS should be implemented ina resource bounded way, not using crazy JIT stuff to run it as fast as possibl
17:33:24 <oren> does chrome put a limit on the amount of memory one tab uses
17:33:41 <oren> firefox hung my system with 1 tab open. 1.
17:33:44 <cluid> I don't like chrome, it's annoying that i can't use proxies
17:34:04 <Melvar> oren: What was in the tab?
17:34:20 <oren> discourse forum on dailywtf
17:34:54 <newsham> you can use proxies with chrome. i use proxies all the time with chrome
17:35:05 <oren> i am now running FF with a resource limit of 1.5 G
17:35:09 <Melvar> Okay, the firefox on my netbook starts out at 180M, and it is by no means empty.
17:35:13 <cluid> it does not let me
17:35:35 <newsham> maybe you dont have as much malware, melvar.
17:36:02 <oren> i am using a whole bunch of plugins
17:36:21 <oren> hold on lemme disable them and we will see if they are the problwm
17:37:00 -!- MDream has changed nick to MDude.
17:38:10 <Melvar> Oh yeah, I don’t think there’s flash on my netbook …
17:38:20 <cluid> javascript is a bad language
17:38:32 <oren> flash is in a separete process so it isn't being counter
17:38:55 -!- S1 has joined.
17:39:06 <oren> 1.1 GB at startup, one tab, google with most of my addons disabled
17:39:08 <Melvar> Oright, but then the same counts for other plugins. Unless you actually meant addons.
17:39:36 <oren> i dont recognize that there is a difference
17:39:51 <oren> its the same thing anyway
17:40:13 <J_Arcane> Phew. Done: http://pasterack.org/pastes/73293
17:41:22 <oren> sisabled adblock plus and we are down to 0.9 gb
17:41:35 <oren> actually more like 1 gb
17:42:16 <Melvar> IIRC, Plugins are out-of-process programs using NPAPI, addons are JS+(HTML/XUL)+CSS packages inside the browser.
17:42:35 <oren> addons are in JS?!?!? FUUUUUUUUUU
17:42:51 <cluid> the architecture of browser seems to encourge plugins being js
17:43:00 <oren> thats why they eat memory like my uncle at a bbq
17:43:05 <Melvar> With regards to firefox, “plugin” and “addon” are technical terms.
17:43:29 <oren> actually he's lost weight these days
17:44:01 <cluid> Gopher clients dont generally eat memor
17:44:03 <Melvar> Nah, it exists to keep my life-state in.
17:44:15 <J_Arcane> The other day I found a rant from 2006 complaining about it consuming 1GB of memory.
17:44:17 <cluid> you can interact with gopher using a simple command line client
17:44:33 <J_Arcane> I laughed and laughed ,because it's still as bad.
17:44:48 <oren> well as i said XP needed 64 megs and we called it an abomination
17:45:09 <cluid> what OS uses low memory?
17:45:45 <HackEgo> [wiki] [[GridScript]] http://esolangs.org/w/index.php?diff=41280&oldid=41279 * SuperJedi224 * (+45) /* Ackermann Function */
17:46:27 <oren> my aunt still runs DOS with WordPerfect to do her work
17:46:34 <oren> as a theologian
17:46:42 <J_Arcane> cluid: FreeBSD is pretty nice, yes.
17:47:15 <oren> floppy disks 4eva
17:47:22 <J_Arcane> FF literally eats more memory than my entire OS, Xorg, window manager, and emacs put together. By multiples.
17:48:00 <Melvar> The only time my firefox eats more than expected is after I reconnect to the network when the router faceplanted. It seems proportional to how long it has been running, but I have yet to figure out what it actually tries to do then.
17:48:57 <J_Arcane> Melvar: Runtime affects it, yes. It leaks memory. It literally has since beta, and it's never been fixed. There are joke websites about 'how long until the FF memory leak will be fixed'. It's a well known issue.
17:49:23 <cluid> web browsers are the new OS
17:49:44 <J_Arcane> In the early days it was a running joke that the new patch notes would announce they finally fixed the memory hole. And then so would the next, and the next, and ...
17:49:51 <HackEgo> [wiki] [[POGAACK]] M http://esolangs.org/w/index.php?diff=41281&oldid=41261 * BCompton * (+2) Fixed dead link template
17:49:56 <Melvar> J_Arcane: I mean, this only happens when I reconnect to the internet after falling off, and why that is is what I can’t figure out.
17:50:28 <J_Arcane> Melvar: Yeah, that is an odd case. Maybe something in it starts duplicating processes attempting to reconnect?
17:50:37 <Melvar> Also, I expect they keep fixing memory holes, but new ones get introduced at about the same rate.
17:51:33 <oren> ugh. from now on i'mma use a site to download youtube videos instead of using the stupid youtube player
17:51:59 <cluid> youtube-dl command line program
17:52:10 <oren> YES someone make that
17:52:11 <Melvar> Well, no, it starts *after I reconnect*. And the growth is proportional to the running time, not the previous held memory.
17:53:01 <Melvar> oren: It’s there, I’ve been using it for ages for videos I want to sit back and enjoy, ’cause my pipe isn’t fat enough to stream even 720p reliably.
17:53:09 <oren> Melvar that sounds like a bug from terrible design... why would firefox even need to know how you are connected.
17:53:34 <Melvar> I actually select the video url and press super+y to start downloading the video.
17:54:08 <Melvar> oren: Outside of actually trying to load something, yeah.
17:54:56 <Melvar> This involves a WM keybind and a script that uses xclip and youtube-dl.
17:55:20 <oren> meh, i'll just CC CV into the command line
17:56:04 <cluid> it would b egood to use more technologes that are not web based
17:56:10 <oren> with this i have no more reason to use FF instead of lynx
17:56:20 <oren> goodbyefirefox
17:58:08 <oren> seriously this is the best
18:01:23 -!- FreeFull has quit (Ping timeout: 245 seconds).
18:02:08 -!- FreeFull has joined.
18:02:08 -!- FreeFull has quit (Changing host).
18:02:08 -!- FreeFull has joined.
18:05:11 -!- cluid has quit (Quit: Leaving).
18:14:38 <J_Arcane> Holy crap this is cool. Gambit Scheme inside Emacs running inside a browser tab: https://feeley.github.io/gambit-in-emacs-in-the-browser/
18:16:10 <oren> I use a lot of terminal stuff mostly because my first computer was Redhat Linux without X
18:16:25 <oren> my second computer was WinXP
18:16:57 <oren> I do not have a grudge against either and cannot understand why anyone would
18:18:03 <oren> but things inside browsers really make me annoyed
18:18:18 <oren> like soundcloud whyyyyy
18:18:38 <oren> what was wrong with downloading music before playing it
18:18:50 <oren> or while playing it
18:19:49 <oren> but yes browsers are recapitulating the history of operating systems
18:20:41 <J_Arcane> oren: It's basically the thin-client race all over again, only they're starting to win.
18:20:54 <oren> I blame google
18:21:27 <J_Arcane> If Windows apps weren't shit, people wouldn't have started this crap in the first place.
18:21:33 -!- nooga has joined.
18:22:33 <oren> firefox and chrome is like windows 95 vs unix in terms of multitasking
18:23:07 <J_Arcane> Chrome has issues, but the sandboxing and process management is way better most of the time.
18:26:05 <oren> well for playing music i'm not about to switch from vlc -Incurses
18:26:49 <oren> to soundcloud inside chrome
18:27:56 <oren> hmmm i should make an ncurses module for scrip7
18:28:18 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
18:28:46 <oren> or maybe an ncurses module for befunge
18:29:04 <oren> (though i dunno if befunge is easily extensible)
18:32:04 <fizzie> There is already a ncurses fingerprint for Funge-98.
18:33:23 <fizzie> Implemented at least by cfunge and Rc/Funge-98, IIRC.
18:34:40 <J_Arcane> I do wish Racket had PDCurses support; or at least better libtcod bindings that actually work.
19:01:26 -!- S1 has changed nick to S0.
19:05:40 -!- GeekDude has joined.
19:23:56 -!- Sauvin has quit (Remote host closed the connection).
19:24:32 <J_Arcane> http://jarcane.github.io/blog/2014/12/01/inventing-a-thing.html
19:27:19 -!- S0 has changed nick to S1.
19:29:11 <oren> J_Arcane: cool! yeah, the racket structs are a pain
19:33:47 -!- G33kDude has joined.
19:34:59 -!- GeekDude has quit (Ping timeout: 256 seconds).
19:35:05 -!- G33kDude has changed nick to GeekDude.
19:36:17 -!- myndzi has left.
19:36:27 -!- myndzi has joined.
19:41:38 <HackEgo> [wiki] [[Pikalang]] http://esolangs.org/w/index.php?diff=41282&oldid=41273 * Zzo38 * (-37)
19:44:34 <elliott> oren: that 1.5 GB may not be measuring what you think it is.
19:45:52 <elliott> oren: if you're using adblock plus don't blame firefox for memory usage :) it's a massive memory hog
19:47:04 <oren> i disbled it which reduced usage to 1.1 GB. that's still a hog
19:47:16 <elliott> meh, was the entire backlog today nerds complaining about boring stuff?
19:47:51 <oren> yah i think so... J-Arcane wrote a cool BASIC-Lisp hybrid
19:48:15 <elliott> oren: and are you sure that 1.5 GB is measuring what you think it is? http://virtualthreads.blogspot.co.uk/2006/02/understanding-memory-usage-on-linux.html https://stackoverflow.com/questions/7880784/what-is-rss-and-vsz-in-linux-memory-management
19:49:44 <oren> its the VIRT column on top?
19:49:56 <oren> which i think also includes swap usage
19:50:28 <elliott> you should probably at least read the SO answer. measuring memory usage is non-trivial and a column saying "1.5 GB" doesn't really correspond to 1.5 GB closer to running out of RAM.
19:51:37 <Bike> a lot of times processes on my netbook have more in the VIRT column than i have ram chips. weird, right
19:51:40 <elliott> anyway, firefox uses a lot of RAM but most of the absurd figures are down to bad measurement or addons
19:51:58 <Bike> actually doesn't firefox have an internal memory usage thingie
19:52:01 <Bike> about:memory or something
19:52:06 <elliott> chrome does, at least. maybe ff too
19:52:29 <oren> well firefox segfaults if i run it with a ulimit of 1 GB
19:52:31 <elliott> a lot of people like to say chrome uses tons of RAM because it uses multiple processes and it looks like every process is using tons of memory because it's sharing stuff with the others
19:52:46 <Bike> i googled a big before realizing that i was googling using firefox and could just check.
19:52:46 <elliott> oren: and what is that ulimit limiting? do you know what that ulimit means?
19:52:59 <oren> whoops not ulimit the other one
19:53:45 <oren> softlimit -a 1000000000 firefox
19:53:53 <Bike> "19.73 MB (01.60%) ++ top(https://www.google.com/search?q=firefox+about+pages" how ironic
19:53:55 <oren> limits all memory
19:54:03 <Bike> "all memory" huh
19:54:19 <oren> like all the memory put together
19:54:49 <Bike> That doesn't mean much of anything.
19:54:51 <oren> the total of all segment sizes
19:55:28 <elliott> okay, I'll stop linking you helpful information to try and clear up any misconceptions you have, since it seems you're certain you don't have any.
19:55:46 <oren> well how come it segfaults?
19:55:54 <Bike> read the fucking links then ask
19:55:55 <coppro> elliott: htop reports that chrome causes about 1GB more memory to be used, or thereabouts
19:56:03 <coppro> a bit less when I first spin it up
19:56:19 <Bike> COOL FACT: the .bss segment is reserved for memory that starts off as all zeroes. WEIRDLY, programs do not just reserve the size of the bss on physical ram chips and zero it out. WHO KNOWS WHY THOUGH
19:56:26 <coppro> it also uses a lot of CPU
19:56:36 <oren> doesn't that mean they tried to allocate something with malloc, it returned null and they gnored that?
19:56:50 <Bike> no. what the fuck?
19:56:56 <oren> that's the only legitimate reason for a segfault
19:57:09 <oren> in a prperly made app anyway
19:57:30 <elliott> ah yes, properly made apps don't even bother checking malloc's return value
19:57:37 <Bike> You don't know how memory works, shut up about "properly made" if you're not only going to not know things but refuse to learn things.
19:57:49 <elliott> coppro: sure. I can believe that. I have my doubts about the measuring process being used for this firefox figure, though.
19:58:09 <oren> if malloc returns null what areyou going to do? keep in mind you might need memory to do it.
19:58:10 <elliott> I do think it's more or less irrelevant though, the far more relevant question is how the memory usage grows with additional tabs
19:58:25 <elliott> oren: you don't need any RAM to check for NULL and print out an error message and abort
19:58:42 <elliott> please stop assuming I'm more ignorant than you, it's really condescending.
19:59:11 <oren> if you're going to print an error then abort,that's ok for console. Firefox is a X program
19:59:15 <elliott> some programs even gracefully recover from malloc failure! though it's tricky to trigger one on linux by default
19:59:34 <elliott> oren: malloc failing doesn't take away any of your previous memory. you can reserve some at any point previously
19:59:35 <oren> setrlimit to some low value
19:59:59 <oren> will cause malloc to fail in many programs
20:00:11 <elliott> I can practically guarantee you firefox checks the return value of malloc, if only to abort()...
20:00:26 <oren> i don't understand how it can segfault then
20:00:26 <elliott> (well, okay; I suppose it probably mostly uses C++ new or such)
20:00:43 <oren> just from limiting its memory
20:00:52 <elliott> well, I'm not saying it doesn't have any bugs. anyway, do you know about overcommit and the OOM killer?
20:01:08 <elliott> if you want to be testing malloc failures the default linux configuration is a bad one, since it's unlikely to produce any.
20:01:25 -!- ais523 has joined.
20:02:42 <oren> like i said, if you setrlimit malloc will fail right?
20:02:45 -!- perrier has quit (Remote host closed the connection).
20:03:00 <elliott> I don't know how rlimits interact with overcommit/OOM killer behaviour.
20:05:43 <oren> brk and sbrk fail when the rlimit has been set. src:man setrlimit
20:06:38 <oren> btw i'm not assuming you don't know anything
20:06:47 <elliott> but memory isn't only allocated through brk and sbrk. firefox uses jemalloc; that uses mmap too
20:06:48 <oren> i just am describing my understanding
20:06:59 <ais523> elliott: the rlimit affects how much can be overcommitted
20:07:08 <ais523> as in, any memory malloc claims to have given you counts against the rlimit
20:07:20 <ais523> once the limit is reached, malloc returns null
20:07:31 <elliott> ais523: firefox doesn't use libc malloc, though, for one
20:07:39 <elliott> the kernel doesn't know about malloc, really...
20:07:41 -!- Patashu has joined.
20:07:44 <Bike> i should write my own malloc.
20:07:47 <oren> RLIMIT_AS limits brk, mmap and mreamap
20:07:50 <ais523> elliott: to be precise, sbrk and mmap return null
20:07:54 <Bike> i guess doing it in terms of mmap is the thing.
20:09:11 <zzo38> TeX implements its own memory allocation. SQLite can use its own memory allocation too. Lemon and bzip2 both allow you to specify your own alternate memory allocation instead of using malloc/free.
20:09:33 -!- augur has quit (Remote host closed the connection).
20:13:09 <fizzie> And RLIMIT_AS also limits things that are not really "using" memory, like mmap'ing actual files. And will count things like shared memory blocks and libraries as if the process in question was their only user.
20:13:12 <fizzie> I vaguely recall we had that discussion about the address space use of a "hello, world" Java program on the Sun JVM being <some big number>, due to <some particular thing the JVM did>.
20:13:50 <elliott> "Related fun fact: the x86 architectures have a direction flag that can be set to cause the processor to run backwards. This is how the backwards copying overlapping memmove is implemented." meh, I was hoping this meant the ip actually moves backwards, but it's hyperbole
20:14:10 -!- augur has joined.
20:14:32 <fizzie> Just the *si/*di register postincrement/decrement, sadly.
20:14:53 <Bike> still waiting for x86 increment immediate
20:15:41 <olsner> istr JVM needing something like 2GB virtual memory to start at all (after running into a badly set ulimit), but that was virtual memory
20:16:06 <fizzie> fis 19104 2.0 0.1 6555876 28496 pts/1 Sl+ 22:14 0:00 java -cp . tmp
20:16:20 <fizzie> Six and a half gigs of address space, ~28 megs resident.
20:16:28 <oren> so is or is not RLIMIT_AS a good way of estimating how much memory firefox is actually using?
20:17:47 <oren> or which RLIMIT is the right one?
20:18:41 <elliott> Bike: like, it would self-modify the instruction?
20:18:52 <fizzie> http://sprunge.us/QMWR <- not unless you think that meant it takes more than 4 gigabytes to run a trivial Java program.
20:18:56 <Bike> yes. i've heard vax had it but that might be wrong.
20:19:06 <fizzie> There is no "right one".
20:19:13 -!- Phantom__Hoover has quit (Ping timeout: 272 seconds).
20:19:30 <Bike> firefox gives you a thing to analyze its memory usage. i think it is a good place to start.
20:20:01 <oren> i can't analyze it after my system hangs
20:20:48 <oren> from huge memory usage on certain websites by firefox
20:20:52 <Bike> for example, it has "heap-allocated" of about 772 MB here. top shows 2600 MBish VIRT. so, i mean.
20:21:53 <elliott> oren: you can easily measure memory usage at startup with it, though.
20:21:56 <Bike> https://developer.mozilla.org/en-US/docs/Mozilla/Projects/Emscripten/Techniques/Out_of_memory_error_reports here's something
20:23:12 <Bike> hm i think that's for js though
20:23:50 <oren> this morning i went on dailywtf forums, and my whole system froze and i had to hold down power button
20:24:17 <oren> that is how i came to this problem
20:24:34 <elliott> that was your browser protecting you from that site
20:24:55 -!- ais523 has quit (Read error: Connection reset by peer).
20:25:02 <oren> it can also happen on games written in JS
20:25:02 -!- ais523 has joined.
20:25:38 -!- Patashu has quit (Ping timeout: 240 seconds).
20:25:52 <oren> i want firefox to die of a segfault instead of hanging
20:26:15 <oren> that is how i found out that setting the limit to 1GB or less causes immediate segfault
20:26:23 <Bike> How do you know it's memory usage?
20:26:35 <oren> it is RLIMIT_AS
20:26:48 <Bike> No, I mean how do you know the hangs are caused by memory usage, yes.
20:27:42 <oren> i know because the hard disk light turns on and doesn't turn off, indicating swapping constantly
20:28:00 <oren> also because i once watched it happen on top
20:28:14 <oren> when i went to ctrlalt f1
20:28:21 <oren> and opened top
20:28:30 <oren> it had 100% memory used
20:29:04 <elliott> how much RAM do you have, by the way?
20:29:19 <oren> that should be enough for anyon
20:29:29 <elliott> as your hours of complaining demonstrate, it clearly isn't. :p
20:29:39 <elliott> are you using a 64-bit OS?
20:29:48 <Bike> that should be enough for anyone -- bill gates, never
20:29:56 -!- bb010g has joined.
20:29:58 <elliott> you only have 2 gigabytes of RAM and 64-bit OSes use more memory!
20:30:24 <Bike> my 2 gb netbook uses 64 bit, the old model was 32 bit
20:30:25 <elliott> there's a reason that the x32 ABI (x86-64 + 32-bit pointers) is a thing.
20:30:26 <oren> bah, that means i effectivly have 1GB that is still a lot
20:30:36 <Bike> alas, i liked finding books on the obscure "x86" architecture"
20:30:40 <Bike> bugs, not books
20:30:43 <Bike> though i also found books.
20:30:44 <elliott> you're just looking for an argument, not help, right?
20:31:15 <oren> well what do i do if i want firefox to segfault instead of freezing the whole ubunut?
20:31:26 <oren> is or is not RLIMIT_AS the one i want?
20:31:39 <elliott> fizzie: you wanna take over?
20:31:40 <zzo38> I believe VAX has a "increment immediate" instruction; x86 doesn't have any such thing.
20:31:49 <Bike> do you want it to segfault if it mmaps a big file? no? well, there you go then
20:31:59 <oren> why would it do that?
20:32:25 <oren> its a web browser not a database
20:32:29 <fizzie> oren: If you can manage to set a RLIMIT_AS that's both large enough for Firefox to run and small enough so that you can fit the limit plus the rest of your system in your physical memory at the same time, maybe that could be a practical solution.
20:32:44 <fizzie> oren: It doesn't sound terribly likely, though.
20:33:12 <fizzie> elliott: I doubt I'll have anything more constructive to offer than "things don't look good" + "look for lightweight browsers".
20:33:23 <fizzie> (They'll all suck, of course.)
20:33:34 <oren> well i guess i can use lynx for almost everything
20:34:06 <oren> except some of my school websites
20:34:37 <elliott> you could just install a 32-bit OS, which would have a measurable concrete benefit and not involve guessing what random rlimits mean.
20:34:44 -!- callforjudgement has joined.
20:35:06 <oren> ormaybe ill just buy a new laptop
20:35:08 <elliott> btw, I bought a really cheap laptop recently, and it still had 4 gigabytes of RAM (but a 1.5 ghz processor).
20:35:13 <elliott> you picked cheep laptop unwisely :p
20:35:33 <oren> this one is already 1.5 years old
20:35:57 <oren> i tend to replace them every 2 years anyway
20:36:09 <elliott> you could buy a more expensive laptop that would last 4 years
20:36:11 <Taneb> This one has 4GB and it's second hand and I believe the previous owner had it for years
20:36:28 <oren> this one cost 260 bucks
20:37:00 <fizzie> (Speaking of browsers and their memory use, Chrome's own task manager says the GPU process is using 800 megs of it.)
20:37:17 <Taneb> oren, this cost me £50
20:37:38 <oren> 50 punds = how many canadian dollars
20:37:53 <elliott> oh, canadian dollars? no wonder it was bad
20:38:04 -!- ais523 has quit (Ping timeout: 258 seconds).
20:38:04 -!- shikhout has joined.
20:38:10 -!- callforjudgement has changed nick to ais523.
20:38:17 <elliott> the one I bought cost more like 400 CAD, but everything is expensive in europe
20:38:24 -!- shikhin has quit (Disconnected by services).
20:38:32 -!- shikhout has changed nick to shikhin.
20:38:48 <oren> canadian dollars ar not worth much
20:38:58 <oren> thats why we got rid of our pennies
20:39:20 <fizzie> elliott: Speaking of which, your capital city seems stupidly expensive.
20:39:51 <fizzie> elliott: I don't understand why my unnamed future employer must be in there of all places. Except maybe it's "hip"?
20:39:51 <elliott> fizzie: you can probably make a trade-off with commutes, but...
20:40:10 <elliott> haven't you named that employer in here already?
20:40:17 -!- Sprocklem has joined.
20:40:37 <fizzie> elliott: Only over the speech recognition, so I have plausible deniability. (I don't know why I did that, it's not a secret.)
20:40:56 <elliott> everyone will know you're going to work for microsoft
20:41:17 <oren> i worked for Bering MEdia
20:41:38 <fizzie> I remember Astalavista, but not what it actually searched.
20:42:10 <fizzie> Sorry, "computer security information" place.
20:42:22 <elliott> is this bering media company entirely dedicated to creepy ad targetting
20:43:20 <oren> i, an intern had access to the names, phone numbers, mailing addresses of many many people in theUS
20:43:37 <oren> (my NDA is over so nyah nyah)
20:43:37 <elliott> so much for "doubleblind privacy(TM)"
20:43:57 <oren> well i didn't have their IP addresses
20:43:58 <elliott> Privacy is a popular topic today – the online advertising debate of privacy versus functionality is a big conversation in the media. Too often privacy is viewed at odds with the goals of industry, but we think that’s the easy way out.
20:44:06 <oren> thats what it actually means
20:44:14 <fizzie> What does "zip+4" mean?
20:44:23 <elliott> fizzie: those extended US zips, I think?
20:44:25 <oren> it is a mailing address
20:44:29 <elliott> like 123456-1942 or something
20:44:34 <elliott> I think they target a residence directly
20:44:48 <oren> i could send anyone a letter
20:44:52 <elliott> An extended ZIP+4 code, introduced in 1983, includes the five digits of the ZIP code, a hyphen, and four more digits that determine a more specific location within a given ZIP code.
20:45:09 <oren> it usually at least targets the specific building
20:45:40 <oren> anyway the idea was you could target an area by centre and radius
20:46:22 <fizzie> Re Bering Media, I was at least amused by that blog post title on the front page: "The EU ‘cookie law’: what has it done for us?"
20:46:26 <fizzie> Sadly, it was not a link to a Monty Python parody.
20:46:55 <oren> when i was there they only had the USA
20:47:05 <oren> even though it is a canadian company
20:47:37 <oren> actually we only had names for some of the database
20:47:41 -!- dario has joined.
20:47:59 <oren> i think it was compiled from public information
20:48:15 <oren> but having all of it in one place is kind of...
20:49:36 <oren> yes now i remember, they got most of the info from the US postal service somehow
20:50:32 <oren> and then the ISPs cooperated to associate ip's with locations
20:50:54 <oren> they had deals with several medium size ISP's
20:51:39 <oren> although all of that is public if you read the blogs of the company
20:52:15 <oren> to be fair however almost 100% of clicks on ads come from bots
20:52:27 <oren> that is another thing i learned there
20:52:53 <oren> so the websites with highest click counts are websites almost no-one uses
20:53:51 <HackEgo> dario: 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.)
20:54:13 <oren> and the people who click a lot on ads are people with virus-infected machines
20:54:31 <oren> that is, Russian-controlled botnets
20:56:17 <oren> the other reason i can talk about bering media is because my boss at the time quit the company
20:57:52 <oren> becuase he did'nt like what they were doing
20:59:29 <oren> the problem is that no-one wants to pay for bot-clicks, but websites want bots to click
21:00:01 <oren> the web ad industry is under attack from both ends
21:03:28 <oren> so they write code to try to identify the user so they can 'ban' robots
21:04:06 <oren> but it doesn't really work becuase a large botnet can issue one click for each of its nodes
21:04:36 <oren> and that adds up to a fe dollars worth of clicks
21:05:23 <oren> which the ad network has to pay for and charge the client
21:05:37 <ais523> oren: almost 100% of ad clicks coming from bots doesn't surprise me at all
21:05:44 <ais523> most people don't follow ads at all
21:06:28 <ais523> actually, many advertisers are in favour of adblock because of this, it means they aren't charged for views that probably wouldn't lead to a follow anyway
21:06:52 -!- Bicyclidine has joined.
21:06:59 <oren> but the people who use advblock are least likely to be infected
21:07:54 <oren> i dunno the whole thing seemed unsolvable to me
21:09:27 <oren> which is why i didn't think what they were doing was wrong, just pointless
21:09:58 <ais523> oren: I also concluded that internet adverts are probably pointless
21:10:18 <oren> distinguishing a bot from a human when neither is cooperating with you is probably impossible
21:11:00 <oren> over a network, i mean
21:11:07 <ais523> oren: right, that seems like an even harder problem than CAPTCHAs
21:11:32 <ais523> because for CAPTCHAs, at least the humans are cooperating
21:11:35 <oren> they had an idea involving checking how fast the mouse moved over the ads
21:11:49 <ais523> easily fakable once the botnets determine that that's what's happening
21:12:02 <ais523> which wouldn't take them long if your ad network was at all large
21:12:13 <fizzie> Don't be so discouraging, internet ads pay my (future) salary. :/
21:12:17 <oren> the bots move at human speeds already
21:12:48 <oren> hey the industry is thriving even if everyone in it knows it's kind of a sham
21:13:29 <oren> because people see the ads even if they don't click
21:13:40 <oren> and advertisers think that's enough
21:14:05 <ais523> oren: well, for TV ads
21:14:07 <ais523> you /can't/ click on them
21:14:12 <ais523> and yet they're still pretty popular
21:14:48 <oren> and the endless hordes of untechnical users make up for nerds using adblock
21:15:35 <ais523> adverts probably aren't the right way to contact the nerds anyway
21:16:29 <oren> interestingly there is a browser that comes with adblock: maxthon
21:16:49 <oren> but only even bigger nerdsthan me use that
21:16:51 <ais523> most email clients have adblock nowadays
21:18:13 <ais523> actually, interesting that web ads are so much easier to block
21:18:24 <ais523> because all the ad companies use complex third-party hosting methods and the like
21:19:36 <zzo38> You can also use filters with whatever email client you are using; or define a filter that comes before the client is another way
21:19:58 <ais523> rather than just getting the sites displaying the ads to also serve the ads, which would be the sensible way (also, a rather less evil way because then the ad company doesn't get tracking data)
21:20:08 <ais523> actually that's probably why they don't do things sensibly
21:20:23 -!- Froox has joined.
21:22:38 <oren> tracking is important because you don't want 50 clicks from one computer
21:25:03 -!- Phantom__Hoover has joined.
21:25:21 <ais523> oren: you see the clicks, just not views
21:25:32 <ais523> because you still control the destination of the link, just not the source any more
21:25:38 -!- nooga_ has joined.
21:26:08 -!- yorick_ has joined.
21:28:20 -!- puritania has joined.
21:28:54 -!- aloril_ has joined.
21:33:46 <oren> heyeverybody.avi
21:34:50 -!- S1 has quit (Disconnected by services).
21:34:52 -!- puritania has changed nick to S1.
21:35:49 -!- nooga has quit (*.net *.split).
21:35:51 -!- Frooxius has quit (*.net *.split).
21:35:55 -!- yorick has quit (*.net *.split).
21:35:55 -!- aloril has quit (*.net *.split).
21:40:57 <HackEgo> olist 969: shachaf oerjan Sgeo FireFly boily nortti
21:43:42 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
21:44:03 <oren> what does olist do
21:44:48 <HackEgo> bin/danddreclist \ bin/dontaskdonttelllist \ bin/don'taskdon'ttelllist \ bin/emptylist \ bin/erflist \ bin/list \ bin/listen \ bin/llist \ bin/makelist \ bin/mlist \ bin/olist \ bin/pbflist \ bin/pbflistdeluxe \ bin/slist \ bin/smlist \ bin/testlist
21:47:43 <HackEgo> /hackenv/bin/!: 4: exec: ibin/echo: not found
21:48:22 <shachaf> FireFly: but i have /hilight on the rot13 of my name :'(
21:48:43 <FireFly> Good thing you weren't in the testlist, then
21:49:08 -!- callforjudgement has joined.
21:49:13 -!- ais523 has quit (Disconnected by services).
21:49:15 -!- callforjudgement has changed nick to ais523.
21:49:22 <FireFly> Also who doesn't enjoy fun puns
21:49:23 <ais523> I'm disappointed that `list stopped working :-(
21:50:16 <zzo38> The danddreclist is my own one and there are many others too
21:50:41 <HackEgo> echo -n "$(basename "$0")${@:+ }$@: "; tail -n+2 "$0" | xargs; exit
21:51:29 <zzo38> You can be notified of these list in many ways: [1] Notify when your name is mentioned, if it is on the list. [2] Notify when the list command is entered (for example, if a message starts with "`danddreclist"). [3] Access the logs and find a message where the command is entered.
21:51:43 <zzo38> I don't know if these programs can access the internet; that would be another way though.
21:52:59 <shachaf> olist is for Order of the Stick updates
21:53:38 -!- J_Arcane has quit (Ping timeout: 240 seconds).
21:54:08 <ais523> zzo38: HackEgo is at least capable of downloading from the Internet, but I think possibly it can only do that on specific commands
21:54:10 <zzo38> See if the wisdom directory contains information about it? Perhaps that can help too
21:54:31 <shachaf> ais523: Is it still? I thought that stopped working.
21:54:42 <fizzie> zzo38: They used to have access via a proxy server that only allowed certain whitelisted pages.
21:54:59 <fizzie> But that at least may have stopped working.
21:55:18 <fizzie> I think `fetch still works? Not sure.
21:55:25 <ais523> fizzie: I was thinking of `fetch
21:55:34 <ais523> `fetch http://nethack4.org/pastebin/nonexistent-page
21:55:39 <HackEgo> http://nethack4.org/pastebin/nonexistent-page: \ 2014-12-01 21:54:55 ERROR 404: Not Found.
21:56:19 <ais523> not sure which of these are HackEgo
21:56:28 <ais523> someone here is running a link-looking script
21:56:51 <zzo38> Send the message privately to HackEgo then.
21:56:55 <shachaf> HackEgo is 162.248.166.242
21:57:00 <fizzie> Do you log user-agents?
21:57:39 <HackEgo> [wiki] [[Mmmm()]] N http://esolangs.org/w/index.php?oldid=41283 * SuperJedi224 * (+1464) Created page with "Mmmm() is an esoteric "microlanguage" by SuperJedi224, inspired in part by [[Brainf***]]. The only characters the interpreter actually pays any attention to are "M", "m", "("..."
21:58:40 <shachaf> Someone should make a language called Brainf***.
21:59:07 <Taneb> Hmm, could I treat modulo arithmetic as the motivating example for quotient groups?
21:59:09 <ais523> shachaf: we have it mentioned as one of the bowlderlizations of brainfuck
21:59:15 <ais523> together with B****fuck
21:59:45 <zzo38> And then you should also need B****f*** and b******** and if you *really* want to be really confusing, even *********
22:00:19 <HackEgo> [wiki] [[Mmmm()]] http://esolangs.org/w/index.php?diff=41284&oldid=41283 * SuperJedi224 * (+12)
22:00:41 <shachaf> zzo38: What word is "*really*" supposed to be?
22:01:36 <zzo38> The asterisks are supposed to be unmentionable characters so you cannot actually type the proper one
22:01:37 <HackEgo> [wiki] [[Mmmm()]] http://esolangs.org/w/index.php?diff=41286&oldid=41284 * SuperJedi224 * (+11)
22:02:11 <fizzie> What's the lowest N in widespread aNb's? There's l10n with 10. Maybe 7 is too low.
22:02:28 <ais523> shachaf: there are no words that match that in my /usr/share/dict/words
22:02:44 <ais523> although if you treat the * as an actual *, not ?
22:02:57 <ais523> you have "ethereally" and "funereally" as well as the obvious "really"
22:03:18 <shachaf> There is also "preallying"
22:03:31 <shachaf> Which has letters on both sides.
22:03:43 <fizzie> "cereally". In a cereal manner.
22:03:49 -!- dario has quit (Quit: Ex-Chat).
22:04:01 <shachaf> I use /usr/share/dict/words/american-english-insane
22:04:42 <ais523> presumably that contains more words than the regular dictionary?
22:04:55 <Taneb> No, it only contains insane words
22:05:19 <fizzie> Didn't realize it was an actual adjective.
22:05:22 <fizzie> The adj cereal has 1 sense (no senses from tagged texts)
22:05:23 <fizzie> 1. cereal -- (made of grain or relating to grain or the plants that produce it; "a cereal beverage"; "cereal grasses")
22:05:38 <HackEgo> [wiki] [[Language list]] http://esolangs.org/w/index.php?diff=41287&oldid=41210 * SuperJedi224 * (+13)
22:07:27 -!- Sprocklem has quit (Ping timeout: 264 seconds).
22:07:29 <ais523> fizzie: I don't consider that an adjective, it's a noun used as a modifier
22:07:44 <ais523> you can swap "cereal" out for pretty much any other concrete noun and the examples still make sense, that's how you can tell
22:08:32 -!- Bicyclidine has quit (Ping timeout: 258 seconds).
22:08:53 -!- Sprocklem has joined.
22:10:15 -!- Bicyclidine has joined.
22:12:35 <ais523> elliott: a dog that's good at sniffing out cereal, I guess
22:12:45 <ais523> it's a meaningful combination, just not one you'd be likely to use
22:13:10 <HackEgo> [wiki] [[Mmmm()]] http://esolangs.org/w/index.php?diff=41288&oldid=41286 * SuperJedi224 * (+190)
22:13:50 <HackEgo> [wiki] [[Mmmm()]] http://esolangs.org/w/index.php?diff=41289&oldid=41288 * SuperJedi224 * (-87)
22:15:51 -!- Bicyclidine has quit (Ping timeout: 264 seconds).
22:16:54 -!- Bicyclidine has joined.
22:18:37 <oren> hmm it appears I can in fact edit the wiki using lynx
22:18:59 <oren> yeah its official i'm ditching firefox
22:19:43 <ais523> I ended up blanking Talk:Main Page on Wikipedia once by editing it through Mozilla 1 for SunOS
22:20:02 <elliott> you could at least use links2 or elinks or something.
22:20:19 <oren> are those better?
22:20:43 <Bicyclidine> if the pope shat in the woods would he be catholic
22:20:56 <oren> i'm open to suggestions aslong as they don't support JS
22:20:59 <elliott> depends on your definition of better, but I would pick w3m or links2 over lynx any day.
22:21:07 <elliott> you know you can just turn off javascript, right
22:21:17 <oren> and don't hog memory
22:21:35 -!- Sprocklem has quit (Ping timeout: 264 seconds).
22:21:36 <oren> CSS can hog memory too
22:22:07 <oren> once a page with complex CSS hogged my cpu and overheated it on my last computer
22:22:46 <oren> too many levels of CSS divs to figure out i guess
22:23:37 <oren> maybe i just won't install x next time
22:25:41 <oren> or maybe i won't buy another craptop
22:34:05 -!- oren has quit (Quit: Lost terminal).
23:02:28 -!- Bicyclidine has quit (Ping timeout: 252 seconds).
23:02:38 -!- mihow has quit (Quit: mihow).
23:09:18 -!- Bicyclidine has joined.
23:13:26 -!- Bicyclid1ne has joined.
23:14:51 -!- Bicyclidine has quit (Ping timeout: 258 seconds).
23:28:24 -!- `^_^v has quit (Ping timeout: 255 seconds).
23:28:35 -!- pikhq_ has quit (Ping timeout: 272 seconds).
23:31:22 <Sgeo> Have you ever just wanted to facepalm someone? http://stackoverflow.com/questions/27222103/the-order-of-the-keys-in-jsonobject
23:32:16 <ais523> I didn't think "facepalm" was a transitive verb
23:33:01 * Sgeo transitives 'facepalm' and verbs 'transitive' and 'verb'
23:33:27 <Bicyclid1ne> isn't palming someone else's face just a "slap"
23:33:34 <Bicyclid1ne> in which case, yes. i have wanted to slap people
23:34:12 <Bicyclid1ne> this person i don't want to slap. i do want to groan obnoxiously at them, however.
23:34:18 -!- Bicyclid1ne has changed nick to Bicyclidine.
23:35:41 <Bicyclidine> Of course, the non-simulated version would use sound, rather than text, among other differences.
23:36:15 <Bicyclidine> Also. What does "verb" mean used as intransitive verb.
23:36:41 -!- pikhq has joined.
23:37:31 -!- Dulnes has joined.
23:37:32 <Sgeo> I never said I intransitived 'verb'
23:37:49 -!- bb010g has quit (Quit: Connection closed for inactivity).
23:38:13 <Bicyclidine> no, but you transitived 'facepalm', which was previously an intransitive verb, so by analogy, to transitive something means to make an intransitive verb transitive
23:38:18 <Dulnes> Does lambdabot keep you from trying to do @tell lambdabot ( eg )
23:38:42 <Sgeo> Yes, but I didn't transitive transitive and verb
23:39:00 <Dulnes> What are you two up to
23:41:08 -!- boily has joined.
23:42:45 <Dulnes> What was i doing yesterday?
23:47:18 <Dulnes> If you have a 60 GB compressed tar ball how do you uncompress it without lagging your computer
23:48:42 <Dulnes> Guys i have something to say
23:51:37 <boily> what is the something that you have to say about?
23:52:47 -!- FreeFull has quit (Ping timeout: 264 seconds).
23:53:31 <boily> nutella is good. I approve of nutella.
23:54:17 -!- FreeFull has joined.
23:59:18 -!- nooga_ has quit (Ping timeout: 240 seconds).