00:02:50 -!- lambdabot has quit (Read error: Connection reset by peer).
00:08:25 -!- lambdabot has joined.
00:13:05 -!- Phantom_Hoover has joined.
00:16:50 -!- Sgeo|web has joined.
00:17:04 <Sgeo|web> Is there any reason that ddrescue on a non-damaged HD would be a bad idea?
00:20:24 <int-e> That sounds rather harmless.
00:20:56 <Sgeo|web> Could be a waste of space, I guess
00:21:16 <Sgeo|web> But I think most of the HDs in question are 80GB or so, the 1TB one is questionably broken
00:21:24 <Sgeo|web> (I guess ddrescue will tell me)
00:21:31 <Sgeo|web> So, planning on buying an 8TB NAS
00:24:11 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds).
00:26:29 -!- variable has quit (Ping timeout: 252 seconds).
00:27:47 -!- variable has joined.
00:28:44 -!- yorick_ has changed nick to yorick.
00:30:20 -!- Naviaux has joined.
00:31:33 -!- TodPunk has quit (Read error: Connection reset by peer).
00:31:34 -!- Tod-Autojoined has joined.
00:46:54 <pikhq> Sgeo|web: If there aren't any read errors, ddrescue is basically just dd.
00:50:53 -!- Sgeo|web has quit (Ping timeout: 246 seconds).
00:52:22 <Naviaux> I was wondering if there is anyone that would like to help with making an esolang called Chronos
00:59:08 -!- rdococ has quit (Quit: Page closed).
01:03:31 <HackEgo> [wiki] [[Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42556&oldid=42542 * Esowiki201529A * (+9) It's stub
01:05:24 <HackEgo> [wiki] [[Gibberish/JavaScript/HTML interpreter]] http://esolangs.org/w/index.php?diff=42557&oldid=42538 * Esowiki201529A * (+9)
01:09:08 <HackEgo> [wiki] [[Language list]] http://esolangs.org/w/index.php?diff=42558&oldid=42516 * Esowiki201529A * (+60) /* G */
01:13:24 -!- Sgeo|web has joined.
01:13:44 <Sgeo|web> Can ddrescue be badly screwed up? Other than by forgetting to use a logfile, I mean
01:14:21 <Sgeo|web> Also, webchat sucks, it disconnected and this fact was not obvious until I tried to send a message
01:15:13 <Sgeo|web> Like, am I liable to do the wrong thing and regret ever using ddrescue? And should just send an HD (potentially with personal information) to professional data recovery service?
01:15:20 -!- magician has joined.
01:17:43 -!- Guest3780 has quit (Ping timeout: 245 seconds).
01:17:43 -!- magician has changed nick to Guest3780.
01:27:12 -!- variable has quit (Ping timeout: 250 seconds).
01:28:55 -!- Tod-Autojoined has changed nick to TodPunk.
01:49:50 -!- Naviaux has quit (Quit: Page closed).
01:50:21 <oren> should I try my hand at translating a manga?
02:04:59 <HackEgo> [wiki] [[Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42559&oldid=42556 * Esowiki201529A * (+51)
02:08:20 <HackEgo> [wiki] [[Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42560&oldid=42559 * Esowiki201529A * (+17) /* Escape sequence */
02:20:36 -!- Sprocklem has joined.
02:25:46 <HackEgo> [wiki] [[Special:Log/move]] move * Esowiki201529A * moved [[Gibberish]] to [[Gibberish (programming language)]]: disambiguation?
02:25:46 <HackEgo> [wiki] [[Special:Log/move]] move * Esowiki201529A * moved [[Talk:Gibberish]] to [[Talk:Gibberish (programming language)]]: disambiguation?
02:28:38 -!- bb010g has quit (Quit: Connection closed for inactivity).
02:28:55 <HackEgo> [wiki] [[Gibberish]] http://esolangs.org/w/index.php?diff=42565&oldid=42562 * Esowiki201529A * (-6) Redirected page to [[Gibberish (disambiguation)]]
02:29:39 <izabera> !bf ++++++++++++[>++++++++<-]> > +++++[>+++++<-]>+ [ [->>+<<]>> [-<+<+>>]<- ] <[<]< [->>[+>]<[<]<] >> [.>]
02:29:41 <myndzi> zyxwvutsrqponmlkjihgfedcba
02:29:41 <myndzi> zyxwvutsrqponmlkjihgfedcba
02:29:41 <EgoBot> zyxwvutsrqponmlkjihgfedcba
02:32:30 <HackEgo> [wiki] [[Gibberish (disambiguation)]] N http://esolangs.org/w/index.php?oldid=42566 * Esowiki201529A * (+65) Created page with "* [[Gibberish (programming language)]] * [[Gibberish/JavaScript]]"
02:32:35 <mitchs> !bf <+++++++++++++++++++++++++++++++++.
02:34:38 <mitchs> !bf ++++++++++++[>++++++++++>++<<-]>++>++[<.->-]
02:34:38 <myndzi> zyxwvutsrqponmlkjihgfedcba
02:34:38 <myndzi> zyxwvutsrqponmlkjihgfedcba
02:34:38 <EgoBot> zyxwvutsrqponmlkjihgfedcba
02:35:51 <HackEgo> [wiki] [[Gibberish (programming language)]] http://esolangs.org/w/index.php?diff=42567&oldid=42561 * Esowiki201529A * (+31)
02:37:26 <elliott> myndzi: might want to turn that off in here, or at least make it only run once per line :p
02:39:24 <mitchs> btw the ! wasn't meant to convey anything, it's just the first non-whitespace printable character, and i wanted to check if the bf interpreter lets you go left of starting point
02:39:40 <elliott> wow um that last line broke a bit
02:39:43 <fungot> \o| c.c \o/ ಠ_ಠ \m/ \m/ \o_ c.c _o/ \m/ \m/ ಠ_ಠ \o/ c.c |o/
02:39:44 <myndzi> | c.c.c | ¯|¯⌠ `\o/´ | c.c.c | `\o/´ ¯|¯⌠ | c.c.c |
02:39:44 <myndzi> | c.c.c | ¯|¯⌠ `\o/´ | c.c.c | `\o/´ ¯|¯⌠ ¦ c.c.c |
02:39:44 <myndzi> |\ c.c /^\ /| | | /'\ c.c /| | /< |/< c.c >\
02:39:44 <myndzi> /'\ c.c |\ /| | | /'\ c.c >\ | |\|´¸¨ c.c /|
02:40:01 <elliott> is everything running twice or something
02:40:33 <elliott> it's meant to be a multiocular o
02:40:53 <elliott> it's... getting kind of messed up by everything running twice though
02:41:45 -!- ascl-bot has joined.
02:41:45 -!- ascl-bot has quit (Client Quit).
02:47:16 <izabera> is anyone else disturbed by the small penis in this part of the drawing? /`\
02:47:29 <HackEgo> [wiki] [[Language list]] http://esolangs.org/w/index.php?diff=42568&oldid=42558 * Esowiki201529A * (+33) /* G */
02:48:54 <elliott> would you prefer it be larger? >_>
02:49:06 <HackEgo> [wiki] [[Hello world program in esoteric languages]] http://esolangs.org/w/index.php?diff=42569&oldid=42518 * Esowiki201529A * (+33) /* Gibberish */
02:50:44 <HackEgo> [wiki] [[Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42570&oldid=42560 * Esowiki201529A * (+23) /* See also */
02:53:06 <HackEgo> [wiki] [[EsoInterpreters]] http://esolangs.org/w/index.php?diff=42571&oldid=42121 * Esowiki201529A * (+33) /* Main table */
02:54:14 <HackEgo> [wiki] [[List of quines]] http://esolangs.org/w/index.php?diff=42572&oldid=42076 * Esowiki201529A * (+33) /* Gibberish */
02:55:05 <HackEgo> [wiki] [[User:Marinus/Brainfuck interpreters]] http://esolangs.org/w/index.php?diff=42573&oldid=20310 * Esowiki201529A * (+33) /* Gibberish */
02:57:24 <HackEgo> [wiki] [[Gibberish (disambiguation)]] http://esolangs.org/w/index.php?diff=42574&oldid=42566 * Esowiki201529A * (+41)
02:58:21 <HackEgo> [wiki] [[User:Marinus]] http://esolangs.org/w/index.php?diff=42575&oldid=38074 * Esowiki201529A * (+33)
02:58:41 -!- bb010g has joined.
02:59:43 <HackEgo> [wiki] [[User:Marinus]] http://esolangs.org/w/index.php?diff=42576&oldid=42575 * Esowiki201529A * (+33)
03:00:23 <HackEgo> [wiki] [[User:Javawizard]] http://esolangs.org/w/index.php?diff=42577&oldid=16780 * Esowiki201529A * (+33)
03:01:30 <myndzi> something is very wrong here
03:02:06 <myndzi> why does everything run twice :(
03:02:48 <myndzi> one is my work client and one is my home client
03:03:01 <HackEgo> [wiki] [[User:Quintopia]] http://esolangs.org/w/index.php?diff=42578&oldid=40213 * Esowiki201529A * (+33) /* Interpreters written in Python */
03:03:27 <myndzi> well, i imagine the work one is locked up from the bf hax above
03:03:38 <myndzi> i didn't write it defensively :P
03:12:03 <fungot> \o| c.c \o/ ಠ_ಠ \m/ \m/ \o_ c.c _o/ \m/ \m/ ಠ_ಠ \o/ c.c |o/
03:20:25 <myndzi> well i could load it up again but then i'd just have to turn it off again
03:20:42 <myndzi> 2 clients 1 bnc KeKeKe
03:23:04 <oren> every time I try to use Gimp I just wish I was using MS Pailnt
03:24:00 <hjulle> "Pailnt" is an interesting word.
03:24:35 <oren> I can barely pronounce it as one syllable
03:26:02 <oren> but seriously Gimp has an awful user interface
03:26:49 <hjulle> I'd probably ponounce it like Pail'nt, or pailent with a silent e.
03:27:55 <oren> OH, right. I'll just use Wine to run paint.net
03:28:12 <hjulle> What about KPaint or something?
03:28:42 <oren> I have paint.net already installed on my other partition, so meh
03:29:17 <oren> hmmm... it isn't working
03:30:15 <hjulle> (Assuming that ".net" means that it's a dotnet program.)
03:31:14 <oren> it seems like people have been saying good things about "pinta"
03:32:08 <oren> hell yes this is exactly what I want
03:32:27 <oren> i.e. it is basically the same as MS paint
03:32:46 <oren> (the version before Windows 8 ruined everything forever)
03:34:19 <hjulle> "Pinta is ... modeled after Paint.NET", so, yes it seems like exaktly what you want.
03:34:36 <myndzi> windows 7 ruined paint
03:34:41 <myndzi> you can still get paint xp
03:34:51 <myndzi> http://www.mspaintxp.com/
03:35:44 <oren> Windows 7 paint was bad too? I jumped (was forced to jump) from XP to 8
03:36:03 <myndzi> 7 ~ 8 as far as paint is concerned, i think
03:41:36 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
04:09:44 -!- Froo has quit (Quit: *bubbles away*).
04:18:49 <Sgeo|web> I want to mount, read-only, a raw image file located on the network
04:33:07 <oren> the terriblosity would depend on what kind of network it is
04:37:10 -!- oerjan has joined.
04:40:11 * oerjan spends 2 hours writing an unusually long ghc trac post only to have it start breaking down just before he's ready to submit.
04:41:14 <oren> I never trust in-browser text editing
04:41:48 <oerjan> me neither, i _did_ have enough clue to copy it first
04:42:28 <izabera> with my browser i can launch a terminal with vim by pressing C-e in any editable area
04:42:49 <oerjan> i already have vim open, i just didn't use it to write
04:43:15 <oerjan> trac has the nice feature of automatic preview, as long as it isn't half-dying
04:44:05 <oerjan> also i did this instead of sleeping argh
04:44:28 <oerjan> (figured once i started thinking mathy i wouldn't get any sleep anyway)
04:46:12 <shachaf> oerjan: it's a good thing that after this is fixed Typeable will be bulletproof
04:49:30 <quintopia> i guess this esowiki guy is nothing if not thorough
04:54:05 <oren> i think haskell is a pretty cool guy, eyht makes evrything a manod and doesn't afraid of anything.
04:55:00 <oerjan> WHERE'S YOUR MONAD NOW
04:55:19 <shachaf> oerjan: diagonalization hth
04:55:47 <oerjan> shachaf: doesn't obey the laws hth
04:55:57 <shachaf> but everyone thinks it does hth
04:56:15 <oerjan> no they don't, it's the goto "this cannot be a monad" example
04:57:03 <shachaf> i think Cont is the goto monad example hth
04:57:36 <shachaf> Actually you can probably make a better goto monad than Cont.
04:58:15 <shachaf> You'd want to be able to jump forwards as well as backwards.
04:58:48 <oerjan> no, i'm just wondering whether i should go back to bed
04:59:10 <lambdabot> Local time for oerjan is Thu Apr 16 07:00:50 2015
04:59:40 <oerjan> i did that > 3 hours ago tdnh
05:00:28 <oerjan> but indeed, this would have been the perfect time if i hadn't woken up too early
05:05:36 -!- Sprocklem has quit (Ping timeout: 250 seconds).
05:13:54 -!- Sprocklem has joined.
05:22:26 -!- Sprocklem has quit (Ping timeout: 245 seconds).
05:29:21 -!- variable has joined.
05:37:08 <oerjan> <izabera> well, factorization is np complete... [...] <-- that is neither believed nor proved.
05:38:17 <oerjan> at least not the usual way of thinking of it.
05:39:37 <oerjan> @tell izabera <izabera> well, factorization is np complete... [...] <-- that is neither believed nor proved. at least not the usual way of defining it.
05:50:22 <oerjan> easy things are also np
05:51:11 <izabera> sorry i meant it takes ridiculously long on large enough inputs
05:51:35 * oerjan hides his secret quantum computer
06:12:06 <coppro> izabera: It's not believed to be in P
06:12:13 <coppro> you are confusing "in NP" with "NP-complete"
06:17:29 <oren> I have already forgotten, mostly, what P and NP even mean, but NPC means "at least as hard as everything in NP"
06:17:43 <elliott> npc means non-player character actually
06:18:10 <shachaf> why do you call people nerds?
06:18:44 <elliott> they're not people, they're just NP-complete, obviously
06:18:48 <oren> because he's emulating Piccolo from DBZA?
06:19:12 <elliott> yes. that is what I do all day, 24/7/365
06:19:17 <shachaf> it's probably only faux-rude, but i don't particularly like it
06:20:01 <oren> dragon ballz abridged, the famous comedy youtube redub of DBZ
06:20:53 <elliott> dragon ballz just looks like the plural of dragon ball
06:20:59 <elliott> like, most dragons have 2 dragon ballz
06:21:28 <oren> https://www.youtube.com/watch?v=m0Wc50iEOtc
06:29:57 <b_jonas> (besides, don't judge from just one episode.)
06:30:49 <izabera> i didn't say i'm not going to watch the rest
06:38:01 <lambdabot> CYYZ 160600Z 00000KT 15SM SCT240 05/M02 A3040 RMK CI4 SLP298
06:43:18 -!- Sgeo|web has quit (Quit: Page closed).
07:00:07 <lambdabot> ENVA 160650Z 26020KT 9999 -RASN FEW005 SCT009 BKN011 02/01 Q1002 TEMPO 4000 SHRASNGS BKN010CB RMK WIND 670FT 27021KT
07:01:29 <oerjan> airport weather reports
07:02:58 <b_jonas> forecast says it's not going to rain here though, for today
07:06:38 -!- variable has changed nick to trout.
07:11:57 -!- trout has quit (Ping timeout: 250 seconds).
07:39:35 <Taneb> Possibly relevant/interesting paper: http://arxiv.org/pdf/1203.1895v1.pdf
07:40:35 <oerjan> itym http://arxiv.org/abs/1203.1895v1 hth
07:42:18 <oerjan> from which you can learn that there's a new version
07:51:58 <Taneb> I didn't know you could do that
07:52:05 <Taneb> I was just given the link direct to the PDF
07:52:27 <oerjan> "always link to the abstract" hth
07:53:14 <oerjan> for one thing, there's links only one way, otherwise you need to know the format
07:53:23 <shachaf> oerjan: they should use a better url scheme hth
07:54:11 <oerjan> shachaf: true, it's a bit awkward to guess
07:54:32 <shachaf> in the past i deleted the .pdf, and it put it back in
07:54:43 <shachaf> and i gave up on guessing the right url and just pasted the pdf link in
08:04:52 -!- aretecode has quit (Read error: No route to host).
08:07:56 -!- aretecode has joined.
08:15:43 <Taneb> Does anyone remember when I started to learn Haskell?
08:16:34 <oerjan> wild guess: shortly after you came here
08:16:52 <Deewiant> 2011-07-18 11:26:19( Taneb) Haskell installed
08:17:10 <Taneb> Huh, that feels later than it should be
08:18:24 <Deewiant> One day earlier 2011-07-17 14:55:47( Taneb) Well, I'm not going to learn Haskell just yet
08:23:03 <oerjan> Taneb: that was just the day after the hexham revelation.
08:23:17 <Taneb> I came to this channel later than I though
08:23:45 <Taneb> Jeez, I was already 16 when I got here?
08:24:31 <oerjan> that's about what i thought
08:24:51 <oerjan> i think you _briefly_ visited some time earlier
08:29:22 <Taneb> It feels like this has been part of my life for much longer
08:29:45 -!- Patashu has joined.
08:38:56 -!- Tod-Autojoined has joined.
08:39:32 -!- TodPunk has quit (Read error: Connection reset by peer).
08:41:41 <HackEgo> [wiki] [[Duck Duck Goose]] M http://esolangs.org/w/index.php?diff=42579&oldid=42294 * Vriskanon * (+245) /* Sample program */ Added cat program and multiple sample programs formatting
08:43:58 -!- hjulle has quit (Ping timeout: 265 seconds).
08:57:33 <oerjan> hm wikipedia is renaming accounts to get global logins
08:58:41 <oerjan> mine seems to stay, but i accidentally got to see the talk page message about Oerjan@wiktionary
08:59:12 <oerjan> (not that he would care, hasn't been around since 2007)
08:59:37 <oerjan> i don't _think_ it's me, anyway, i don't recognize those edits.
09:04:18 -!- Phantom_Hoover has joined.
09:12:21 <oren> Hmm... I can tell I'm still young, because none of the languages I know has become obsolete yet
09:13:19 <oren> Whereas my dad knows Perl 3 and Fortran or something
09:13:29 <oren> and therefore is old
09:14:19 * oerjan waves from Pascal-land
09:15:51 <oren> I think Perl as glue is dying in favor of python as glue though... so it won't be long...
09:19:15 <fizzie> oerjan: It's the bizarro-oerjan, your goateed evil counterpart.
09:20:46 <oerjan> fizzie: you've got it backwards hth also HackEgo is damn slow
09:22:40 <fizzie> I assume it's the VPS provider.
09:22:44 <shachaf> I thought it just wasn't working?
09:22:57 <fizzie> The wiki bridge was working less than an hour ago.
09:23:14 <shachaf> Yes, but ` hasn't been working for much longer than that.
09:23:55 <fizzie> That seems plausible, given the logs.
09:24:24 <fizzie> I remember it getting stuck once, due to a thing that I fixed.
09:24:39 <fizzie> It was something related to file permissions, some bit had gone missing.
09:24:51 <fizzie> Of course I didn't write down the details, because where's the fun in that?
09:25:20 <b_jonas> but it worked a few days ago
09:25:39 * oerjan recalls some time when `help worked but not user commands
09:26:07 <coppro> I should probably go to sleep
09:27:36 <fizzie> Well, it's also being slow-ish over SSH, especially if I try to do anything disk-related. But that's probably not the main problem.
09:28:53 <fizzie> It's not on the channel, that's what's up.
09:29:11 <oerjan> um according to my tab completion, it is
09:29:22 <fizzie> According to my /n, it's not.
09:29:30 <fizzie> Perhaps your tab completion is based on incoming messages.
09:30:04 <oerjan> surely it's based on /quits etc.
09:30:09 <fizzie> I think it's a nicely-formatted 'WHO', but it could also be based on 'NAMES'. I'm not entirely sure.
09:30:17 -!- oerjan has left.
09:30:17 -!- oerjan has joined.
09:30:49 -!- HackEgo has joined.
09:31:06 <oerjan> so it was some kind of irc network failure, actually
09:31:13 <HackEgo> Ørjan is oerjan's good twin. He's banned in the IRC RFC for being an invalid character. Sometimes he publishes papers.
09:32:04 <fizzie> (I didn't realize we're so non-'+n' inclusive we allow messages from people not on the channel.)
09:32:47 <oerjan> i think it's irssi getting confused, right
09:34:12 <oerjan> right, HackEgo never was here since i joined, so it must have been the wiki announcement that did it
09:34:34 <oerjan> fizzie: i think it's an experiment?
09:35:27 <oerjan> i think this is the first time a problem has appeared with -n
09:35:53 <oerjan> or well, that i've noticed anyway
09:36:17 <fizzie> Now you'll presumably have that stuck in your tab-completion list, since it won't get any part/quit.
09:36:49 <oerjan> maybe people have messaged from out of channel left and right, and i never noticed they weren't here.
09:38:42 <fizzie> An evil person could make tab-completion arbitrarily difficult by making conflicting prefixes for every nick at every length.
09:42:10 <int-e> But the solution is obvious: only allow single letter nicknames. (Good morning!)
09:43:44 -!- mroman has changed nick to fizzia.
09:44:57 -!- oerjan has left.
09:44:57 -!- oerjan has joined.
09:46:42 <oerjan> int-e: i made a new trac post in the thread
09:46:56 <oerjan> i assume since i didn't get a Cc: you didn't either
09:47:43 <oerjan> now with an actual proof.
09:47:50 <fizzie> "#esoteric: Total of 104 nicks" that's a bit problematic for single-letter (printable ASCII not-reserved-by-IRC) names.
09:49:20 <fizzie> "[0 ops, 0 halfops, 0 voices, 104 normal]" I'm not sure that last word is exactly the right one.
09:53:01 <oerjan> all matter of hairy business, including splitting
09:54:32 <fizzie> In retrospect, I think irssi's tab-complete is based on last-one-who-spoke-first-on-the-list, which would make that less efficient.
10:00:03 <fizzia> irssi has intelligent tab-completion yes.
10:11:04 -!- aretecode has quit (Ping timeout: 246 seconds).
10:14:30 <int-e> fizzie: that can't be, my plan was perfect!
10:17:24 -!- int-e has set topic: Foiled by numbers | Dilapitated depilators | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
10:17:37 <oerjan> int-e: btw if you see any fatal flaw in my proof (including any form of Typeable-capable types that don't fit in either branch of argument) that might mean another exploit that works against spj's fix.
10:18:25 -!- aretecode has joined.
10:19:20 <int-e> oerjan: Maybe later. I'm not sure what you mean by "kind-ambiguous" right now, which means I don't know why the list you gave is exhaustive.
10:24:41 <oerjan> i mean, basically, any two type constructors with the same TypeRep, which afaik now only happens because of the */Constraint ambiguity in (), (,), ..., -> / =>
10:25:48 <oerjan> because Haskell distinguishes them, but core doesn't, and the kind parameters for Typeable are extracted from core.
10:26:36 <oerjan> or well not quite core maybe, but the upshot is this list has no kind parameters that get used for the TypeRep.
10:27:15 -!- boily has joined.
11:30:17 -!- boily has quit (Quit: LIBERATED CHICKEN).
11:58:38 -!- bb010g has quit (Quit: Connection closed for inactivity).
12:37:27 -!- hjulle has joined.
12:50:01 -!- Patashu has quit (Ping timeout: 276 seconds).
12:58:43 -!- ProofTechnique has quit (Ping timeout: 246 seconds).
12:59:39 -!- Tod-Autojoined has changed nick to TodPunk.
13:19:58 -!- dianne has quit (Quit: byeannes).
13:25:58 -!- ProofTechnique has joined.
13:49:41 <b_jonas> `oeis 16271255119687320314
13:49:43 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: oeis: not found
13:52:28 -!- stuntaneous has joined.
13:53:26 -!- stuntaneous has quit (Client Quit).
13:57:06 -!- bb010g has joined.
14:00:08 <int-e> @oeis 16271255119687320314
14:00:21 <Jafet> `factor 16271255119687320314
14:00:21 <HackEgo> 16271255119687320314: 2 19 137 3989 61519 12736309
14:00:33 <lambdabot> Tetranacci numbers: a(n) = a(n-1) + a(n-2) + a(n-3) + a(n-4) with a(0)=a(1)=...
14:01:26 -!- `^_^v has joined.
14:03:28 <b_jonas> int-e: unless I made a mistake, it's the number of sets of squares of a 8x8 chessboard such that each row and each column has some but not all squares in the subset
14:04:05 <Jafet> @google 16271255119687320314
14:04:14 <b_jonas> but someone should check that
14:05:31 <int-e> so relations on {1..8} with full domain and codomain?
14:07:30 <b_jonas> what's the domain of a relation in that sense?
14:07:40 <int-e> domain and codomain are defined by dom(R) = {a | (a,b) in R}; cod(R) = {b | (a,b) in R}.
14:08:11 <b_jonas> in which case there would be (2**8-2)**8 such relations
14:08:16 <b_jonas> but I restrict both rows and columns
14:08:43 <int-e> What? Surely you mean 255^8, not 254^8. But you're wrong.
14:09:09 <b_jonas> the codomain is teh columns
14:09:54 <b_jonas> in that case I want relations such that the domain and codomain is full, and also the domain and codomain of the negation (complement) of that relation are also full
14:10:14 <int-e> Ah, I missed that.
14:10:59 <b_jonas> that's probably why it's not in oeis
14:14:28 <tromp> you want chessboards with no monochromatic ranks or files
14:16:59 <b_jonas> tromp: yes, black and white chessboards with no monochromatic ranks or files
14:17:11 <j-bot> b_jonas: 1.73249e19
14:17:58 <izabera> `` printf '\x03,%dx' {0..33}
14:17:59 <HackEgo> ,0x,1x,2x,3x,4x,5x,6x,7x,8x,9x,10x,11x,12x,13x,14x,15x,16x,17x,18x,19x,20x,21x,22x,23x,24x,25x,26x,27x,28x,29x,30x,31x,32x,33x
14:18:13 <izabera> `` printf '\x03%dx' {0..33}
14:18:14 <HackEgo> xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
14:23:36 <int-e> > let c n 0 = 1; c n k = c (n-1) (k-1); f n = (2^n - 2)^n + sum [(-1)^k * (2^((n-k)*n) * (2^k - 2) + (2^(n-k)-1)^n*2) * c n k | k <- [1..n]] in map f [1..8]
14:23:38 <lambdabot> [0,4,174,34204,22874190,55079812684,496433383427646,17190560520625432252]
14:24:00 <int-e> > let c n 0 = 1; c n k = c (n-1) (k-1) * n `div` k; f n = (2^n - 2)^n + sum [(-1)^k * (2^((n-k)*n) * (2^k - 2) + (2^(n-k)-1)^n*2) * c n k | k <- [1..n]] in map f [1..8]
14:24:01 <lambdabot> [0,2,102,22874,17633670,46959933962,451575174961302,16271255119687320314]
14:24:34 <b_jonas> [ 16271255119687320314=16271255119687320314
14:24:47 <b_jonas> hmm, how does that computation work?
14:24:53 <b_jonas> I used a more complicated computation
14:25:52 <int-e> Inclusion-exclusion principle: Starting from (2^k-2)^n solutions satisfying the column criterion, it computes how many of those have k=1..8 equal rows.
14:26:19 <b_jonas> I wondered if something like that would work
14:26:41 <b_jonas> but eventually wrote a computation that's more complicated, though still fast for 8x8 on a computer
14:27:34 <int-e> so that gives a c n k factor (for selecting the k equal rows), and then (2^k - 2) cases with 2^((n-k)*n) solutions (where the monochromatic rows have mixed colors) and 2 cases with (2^(n-k)-1)^n solutions because the monochromatic rows all have the same color.
14:27:49 <b_jonas> but don't you need to multiply by a binomial somewhere?
14:27:58 <int-e> c n k is a binomial.
14:28:26 <int-e> @oeis 0 2 102 22874
14:30:21 <b_jonas> here's my solution: http://rextester.com/QNQCE30194
14:31:36 <tromp> what is running time as function of boardsize?
14:31:52 <b_jonas> tromp: of my solution? exponential
14:32:20 <tromp> exponential in n or in n^2 for an nxn board?
14:32:21 <b_jonas> O((3+epsilon)**width) I think
14:32:32 <b_jonas> exponential in n for an nxn board
14:32:59 <tromp> can't you compute this with the inclusion-exclusion formula?
14:33:10 <b_jonas> sure, but this occurred to me first
14:33:15 <int-e> tromp: scroll back a bit :)
14:33:29 <b_jonas> the inclusion-exclusion formula int-e shows is a better way
14:33:58 <tromp> ah, you're ahead of me, int-e :)
14:34:04 <b_jonas> in fact, I think mine could be optimized to a somewhat more complicated polynomial time solution
14:34:53 <b_jonas> it would have a somewhat higher exponent but whatever
14:34:57 <b_jonas> I won't implement that though
14:38:15 <int-e> at least https://oeis.org/A048291 exists
14:43:02 <tromp> meanwhile, my 19x19 go computation is still going...
14:44:19 <b_jonas> hehe, this is a strange channel, I give a computation and two people try to implement it rightaway
14:44:27 <b_jonas> tromp: or are you computing something else?
14:44:51 <tromp> i'm computing the number of legal 19x19 Go positions, which is somewhat more involved:)
14:45:20 <tromp> should get finished this year though
14:47:15 <HackEgo> [wiki] [[Lesser known programming languages]] http://esolangs.org/w/index.php?diff=42580&oldid=7995 * 50.23.113.210 * (+51)
14:48:46 <HackEgo> [wiki] [[Lesser known programming languages]] http://esolangs.org/w/index.php?diff=42581&oldid=42580 * B jonas * (-51) revert spam by User:50.23.113.210
14:50:43 -!- Sprocklem has joined.
14:57:53 <int-e> tromp: do you know the answer modulo some primes (2^64-ish, I guess) already?
15:00:03 <HackEgo> [wiki] [[Babble]] http://esolangs.org/w/index.php?diff=42582&oldid=42550 * Rottytooth * (+34) clarity
15:00:40 -!- GeekDude has joined.
15:00:56 <HackEgo> [wiki] [[Babble]] M http://esolangs.org/w/index.php?diff=42583&oldid=42582 * Rottytooth * (+16) /* Concept */
15:01:59 <tromp> no, it will take another 6 weeks or so to get the first congruence (mod 2^64)
15:02:36 <tromp> you can see the progress at http://tromp.github.io/go/legal.html
15:08:39 -!- rdococ has joined.
15:12:43 -!- variable has joined.
15:15:42 <rdococ> variable: since I'm so [not
15:16:09 <rdococ> variable: since I'm so [not] kind, I'll elevate you to a first class object... lol... get it?
15:22:16 <b_jonas> what? you're computing it modulo something?
15:23:14 <int-e> b_jonas: modulo several things, to be combined by the chinese remainder theorem to obtain the full answer
15:23:27 <b_jonas> right, but how does that help in this case?
15:23:42 <int-e> b_jonas: it helps keeping the numbers to machine word size
15:23:51 <b_jonas> sure, but you're traversing a graph
15:24:06 <b_jonas> isn't it better to do all the computations at once?
15:24:09 <b_jonas> I mean, it doesn't have to be radixal
15:24:15 <b_jonas> but all the congruences at the same time
15:24:29 <b_jonas> so you just have to apply chinese remainder once at the very end, which is very quick
15:24:50 <b_jonas> as opposed to computing the result with one modulus first
15:26:36 <tromp> the problem is almost IO bound
15:27:09 <int-e> > 3^20 -- too low because it doesn't take connectivity into account.
15:27:12 <b_jonas> hmm, maybe it's just that you don't know any more efficient way to parallize the problem
15:27:39 <tromp> with filesize proportional to number of state-cnt pairs that can be held in memory
15:28:31 <b_jonas> tromp: sure, but if you eg. computed the values modulo some 128 bit number instead of 64 bit number, how much larger would the io work become? twice larger, or just 1.1 larger?
15:28:39 <FireFly> tromp: will you do 19×17 too, afterwards? To complete all {1..19}×{1..19} combinations
15:28:55 <tromp> so with 566 bit counts, you cldn't hold many states in memory, and get files with few states, which makes IO a much bigger problem
15:28:59 <int-e> > 363324268018 / 3^20
15:29:32 <tromp> i will do all 19xn, n up to 20
15:30:03 <b_jonas> tromp: do you get separate counts for black to go next and white to go next?
15:30:05 <tromp> b_jonas: about 1.9 times larger
15:30:27 <b_jonas> tromp: I see. in that case parallelizing this way _might_ help, if you read the io sequentially enough
15:30:36 <tromp> i'm counting positions. there is no notoin of who is to move
15:30:40 <b_jonas> I can't imagine how this could work
15:31:15 -!- variable has quit (Ping timeout: 240 seconds).
15:31:59 <int-e> "Allowing for some redundancy, we need from 10 to 13 servers, each with at least 32 cores, 512GB RAM, and ample disk space (10-15TB), running for about 3-5 months." -- this disk space seems quite tightly calculated.
15:32:31 <int-e> (going by the 363324268018 number)
15:32:45 <tromp> each "vector" takes about 4.3TB
15:33:21 <int-e> > 363324268018 * 8 -- plus some allowance for efficient access?
15:33:54 <tromp> each modulus needs at least the current and new vector, but preferably also the previous and an older one for error recovery
15:34:41 <tromp> int-e: you need between 1 and 2 bytes on avg for the state-delta
15:35:16 <tromp> and the vectors have states duplicated
15:36:14 <tromp> each state could occur up to 3 times, in different files of a state vector
15:36:55 <tromp> but most duplicates are alrd eliminated by the in-memory sorting
15:36:58 <b_jonas> compared to this, we work with small data and quick computations
15:37:15 <tromp> i dont actually need 512GB that was rather conservative
15:37:25 <tromp> i can manage with 128GB
15:38:00 <tromp> the main requirement is high disk IO bandwidth
15:38:25 -!- ais523 has joined.
15:38:40 <b_jonas> even still, the beefyest machine I use here has 32GB of ram only, and 24 cores, though it does have a RAID with like twenty terabytes of disk space and fast IO bandwidth because we're recording raw image data quickly and compressing it later
15:38:56 <b_jonas> ais523: I was thinking about what you said, and you're right
15:39:31 <b_jonas> you only need to handle the case with identical reduce actions for any possible lookahead token (the reduce action can be a one symbol reduce), not the case of an identical shift
15:40:54 <b_jonas> (also, it's a NUMA machine)
15:41:24 <b_jonas> (and with a crazy fast CPU)
15:47:20 -!- ais523 has quit (Ping timeout: 264 seconds).
15:51:05 -!- nszceta has joined.
15:51:52 -!- nszceta has quit (Max SendQ exceeded).
15:53:57 -!- nszceta has joined.
15:55:08 -!- nszceta has quit (Max SendQ exceeded).
15:55:48 -!- nszceta has joined.
15:58:39 -!- bb010g has quit (Quit: Connection closed for inactivity).
16:05:27 -!- nszceta has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
16:12:31 -!- nszceta has joined.
16:12:31 -!- nszceta has quit (Read error: Connection reset by peer).
16:14:05 -!- MDude has joined.
16:20:24 -!- vodkode has joined.
16:22:31 -!- Sprocklem has quit (Quit: Bye).
16:30:47 <rdococ> is an OISC with a command to copy the bit at one address to a second, then flip the first, turing complete
16:32:14 <int-e> @oeis 1 3 9 31 113
16:32:15 <lambdabot> G.f.: (1-sqrt(1-4x-4x^2))/(2(1+x)).[0,1,1,3,9,31,113,431,1697,6847,28161,117...
16:32:34 <int-e> @oeis 1 3 9 31 113 435
16:36:21 -!- hjulle has quit (Ping timeout: 250 seconds).
16:41:34 <lambdabot> Fibonacci numbers: F(n) = F(n-1) + F(n-2) with F(0) = 0 and F(1) = 1.[0,1,1,...
16:50:16 <rdococ> I didn't even noticed a spambot got hold of my talk page...
16:50:24 -!- rdococ has quit (Quit: Page closed).
16:51:54 -!- rdococ has joined.
17:01:33 <rdococ> a finite state automaton could be a described as a finite set of ordered triples p = {(x,y,z)}.
17:02:18 <rdococ> and if they're numbers, then you can make a raster image with it
17:04:17 <rdococ> assuming you have infinite color depth
17:04:31 <rdococ> or atleast unbounded, so
17:04:39 -!- ProofTechnique has quit (Quit: ZNC - http://znc.in).
17:07:22 -!- ProofTechnique has joined.
17:16:01 -!- variable has joined.
17:16:09 -!- Hijiri has quit (Ping timeout: 256 seconds).
17:22:34 -!- Guest3780 has quit (Read error: Connection reset by peer).
17:24:21 -!- Guest3780 has joined.
17:42:26 -!- CADD has joined.
17:43:10 <rdococ> why do programs have to exist
17:44:43 <rdococ> syntax error: unknown channel
17:45:40 <rdococ> oh and apparently I'm banned from ##philosophy -.- forgot how notorious I look
17:46:00 <rdococ> I feel so lonely... nearly every channel rejects me...
17:46:20 <rdococ> I'm always the lonely one... the weird one...
17:48:26 <rdococ> well atleast this channel still likes me
17:53:44 <rdococ> ...I don't like monologues
17:57:28 -!- variable has quit (Ping timeout: 276 seconds).
17:58:50 <int-e> oerjan: #9858 comment 90 would be more helpful if the required extensions had been listed... I've narrowed it down to KindSignatures + ScopedTypeVariables + TypeFamilies for the first one and PolyKinds + ScopedTypeFamilies for the second one. Interestingly, TypeFamilies seems to be essential for the first one...
18:00:34 <int-e> oh wait, GADTs works as well.
18:07:17 <b_jonas> let me read your sieve formula, I tried to reproduce the sieve but got the wrong result
18:07:27 <b_jonas> I have to multiply by a power of 2
18:08:38 <b_jonas> > let c n 0 = 1; c n k = c (n-1) (k-1) * n `div` k; f n = (2^n - 2)^n + sum [(-1)^k * (2^((n-k)*n) * (2^k - 2) + (2^(n-k)-1)^n*2) * c n k | k <- [1..n]] in map f [1..8]
18:08:40 <lambdabot> [0,2,102,22874,17633670,46959933962,451575174961302,16271255119687320314]
18:09:16 <b_jonas> > let { c n 0 = 1; c n k = c (n-1) (k-1) * n `div` k; } in map (c 8) [0..8]
18:10:59 <b_jonas> that's different from the formula I tried
18:13:04 <b_jonas> ok wait, mine gives the wrong result already for a 2x2 chessboard
18:20:28 -!- TESSSSSSSSSSSSSS has joined.
18:20:54 -!- TESSSSSSSSSSSSSS has changed nick to Sprocklem.
18:29:37 <tromp> first get 1x1 correct (0)
18:30:07 <b_jonas> tromp: that one was correct, and 0x0 too
18:30:16 <tromp> then 1x2 and 2x1 (also 0)
18:32:21 <b_jonas> no, the formula I used is just wrong, and I know why
18:32:28 <b_jonas> I'll just have to try to write the correct formula now
18:32:33 <tromp> answer for 2xn is 2^n-2
18:33:48 <int-e> oerjan: oh, news on #9858, looks like you got what you wanted?
18:40:37 -!- zadock has joined.
18:46:20 -!- zadock has quit (Max SendQ exceeded).
18:47:05 -!- zadock has joined.
18:59:28 -!- bb010g has joined.
19:00:07 -!- zadock has quit (Quit: Leaving).
19:03:03 -!- AnotherTest has joined.
19:07:35 -!- AnotherTest has quit (Ping timeout: 250 seconds).
19:25:44 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
19:27:31 -!- Phantom_Hoover has joined.
19:28:57 <b_jonas> heh, I made another wrong formula
19:30:38 <b_jonas> [ 3 :'+/,(([!y-])*(]!y"_)*(_1^+)*y^~(y>:+)*+/&(0&=)-~2^y-+)"0/~i.>:y'"0 i.9x
19:30:38 <j-bot> b_jonas: 1 0 2 102 22874 17633670 46959933962 451575174961302 16271255119687320314
19:31:19 <b_jonas> it could be turned to a single sum but then there'd be more conditions so this is simpler to write
19:32:16 -!- nszceta has joined.
19:36:45 -!- Hijiri has joined.
19:52:18 -!- Sprocklem has quit (Ping timeout: 255 seconds).
20:00:52 -!- Sprocklem has joined.
20:01:01 -!- Frooxius has joined.
20:15:09 -!- Encapsulation has joined.
20:15:37 <Encapsulation> I'm facing a challenge right now in my program. I need objects running methods and having their attributes change at regular intervals and in reaction to events while simultaneously providing I/O to the user. Which language would be best suited for me?
20:21:11 <izabera> Encapsulation: maybe #esoteric isn't the best place to ask such a question.....
20:23:28 <MDude> Abstract State Machine Language?
20:25:19 -!- variable has joined.
20:30:59 <rdococ> #esoteric is for esoteric programming languages
20:31:08 <rdococ> ie, ones you probably don't want to use, lol
20:31:28 <rdococ> what? I just wanted a laugh... lol
20:31:38 <rdococ> I need to stop using lol, lol
20:38:31 -!- Frooxius has quit (Ping timeout: 244 seconds).
20:42:35 -!- Phantom_Hoover has quit (Ping timeout: 256 seconds).
20:43:58 -!- Hijiri has quit (Quit: WeeChat 1.0.1).
20:44:07 -!- Frooxius has joined.
20:45:16 -!- Froox has joined.
20:47:13 -!- Froo has joined.
20:48:58 -!- Frooxius has quit (Ping timeout: 265 seconds).
20:49:48 -!- Froox has quit (Ping timeout: 256 seconds).
20:53:02 <b_jonas> [ 3 :'(y^~_2+2^y)--/(!&y*(2*y^~_1+2^y&-)+(2^y*y&-)*_2+2&^)>:i.y'"0 i.9x
20:53:03 <j-bot> b_jonas: 1 0 2 102 22874 17633670 46959933962 451575174961302 16271255119687320314
20:53:16 <b_jonas> ^ that one is just a single summation
20:53:33 <b_jonas> [ 3 :'+/,(([!y-])*(]!y"_)*(_1^+)*y^~(y>:+)*+/&(0&=)-~2^y-+)"0/~i.>:y'"0 i.9x
20:53:34 <j-bot> b_jonas: 1 0 2 102 22874 17633670 46959933962 451575174961302 16271255119687320314
20:53:53 <b_jonas> [ rv0=: 3 :'+/,(([!y-])*(]!y"_)*(_1^+)*y^~(y>:+)*+/&(0&=)-~2^y-+)"0/~i.>:y'"0 i.9x
20:54:02 <b_jonas> [ rv2=: 3 :'+/,(([!y-])*(]!y"_)*(_1^+)*y^~(y>:+)*+/&(0&=)-~2^y-+)"0/~i.>:y'"0 i.9x
20:54:10 <j-bot> b_jonas: 1 0 2 102 22874 17633670 46959933962 451575174961302 16271255119687320314
20:54:31 <b_jonas> somehow it even came out shorter, though probably the expression for the double sum could be written more concise too
20:55:11 -!- Phantom_Hoover has joined.
20:55:30 <b_jonas> I'm comparing the wron gthing
20:55:33 <b_jonas> the single summation is longer
20:55:46 <b_jonas> [ ]rv2=: 3 :'(y^~_2+2^y)--/(!&y*(2*y^~_1+2^y&-)+(2^y*y&-)*_2+2&^)>:i.y'"0 i.9x
20:55:47 <j-bot> b_jonas: 1 0 2 102 22874 17633670 46959933962 451575174961302 16271255119687320314
20:56:17 <b_jonas> no, the single summation is shorter afterall
20:56:51 <b_jonas> [ rv0=: 3 :'+/,(([!y-])*(]!y"_)*(_1^+)*y^~(y>:+)*+/&(0&=)-~2^y-+)"0/~i.>:y'"0 i.9x
21:00:10 -!- nszceta has quit (Quit: Textual IRC Client: www.textualapp.com).
21:05:51 -!- Patashu has joined.
21:10:13 -!- ProofTechnique has quit (Ping timeout: 250 seconds).
21:13:25 -!- variable has changed nick to trout.
21:20:23 <rdococ> I didn't know there was a trout class.
21:20:44 <trout> rdococ, you didn't know that trout is directly related to programming?
21:21:08 <rdococ> anyway, I have another boring esolang coming, but atleast it'll have a punny name
21:21:28 <int-e> . o O ( is this an example of trolling fish? )
21:22:49 -!- Sprocklem has quit (Ping timeout: 256 seconds).
21:26:41 -!- Patashu has quit (Ping timeout: 265 seconds).
21:27:11 <olsner> isn't it spelled trawling when you do it to fish?
21:27:24 <olsner> or is that just how I imagine it should be
21:27:28 <oren> Hmm... What if there was a language sufficiently flexible that you could write it to look like C, Fortran, Pascal, or Basic?
21:28:08 <HackEgo> [wiki] [[State of the Art]] N http://esolangs.org/w/index.php?oldid=42584 * Rdococ * (+1182) new esolang
21:28:20 <oren> e.g. int x; and dim x as integer would both be valid
21:28:40 <HackEgo> [wiki] [[State of the Art]] M http://esolangs.org/w/index.php?diff=42585&oldid=42584 * Rdococ * (+9) lil' stubby
21:28:58 <int-e> olsner: "Trolling can be phonetically confused with trawling, a different method of fishing where a net (trawl) is drawn through the water instead of lines." says Wikipedia
21:29:32 <rdococ> it could be trolling because the fish put on the rage face
21:30:13 <int-e> Pouting trouts are not a pretty sight.
21:30:45 <HackEgo> [wiki] [[User:Rdococ]] http://esolangs.org/w/index.php?diff=42586&oldid=42553 * Rdococ * (+23) /* My esoteric programming languages */
21:31:07 <int-e> trout: Apologies for the puns, I hope you do not take them seriously.
21:31:16 <olsner> according to what I hear, these days it's all about the bass
21:31:22 <rdococ> sorry about those puns...
21:31:23 <trout> int-e, puns are fine, that is the whole point of having such a username
21:31:31 -!- sebbu has quit (Ping timeout: 265 seconds).
21:31:40 <rdococ> also, scroll bar wont scroll down
21:31:51 <trout> int-e, I do have a question about your name though
21:32:01 <trout> are you intenctionally casting it?
21:32:25 <olsner> int e is a page fault, maybe that's it
21:32:44 <rdococ> he's more likely to be 3, I think
21:32:48 <int-e> rdococ: that's a better guess than I've seen in a while.
21:33:08 <oren> in C iirc (int) rounds down
21:33:32 -!- sebbu has joined.
21:33:32 <trout> oren, it truncates, yes
21:33:33 <rdococ> don't use C, prefer lua
21:33:56 <rdococ> because low-level stuff derps and costs money and herps.
21:34:10 -!- sebbu has quit (Changing host).
21:34:10 -!- sebbu has joined.
21:35:12 <oren> trout: so it rounds toward zero. I think it's probably a bad idea to rely on that though.
21:40:32 <rdococ> I'd like a programming language based on nothing
21:41:14 <rdococ> should I have just created a bf derivative?
21:44:05 -!- aretecode has quit (Quit: Toodaloo).
21:46:52 -!- `^_^v has quit (Quit: This computer has gone to sleep).
21:54:40 <rdococ> maybe I should just add some hashing requirement
22:00:27 <izabera> unzip unrar unxz... gunzip
22:00:47 <izabera> i type ungzip every fucking time and i keep getting command not found
22:02:55 <oren> put an alias in your bashrc
22:03:24 <oren> I have alias fuckyou as pkill
22:04:36 <izabera> silent () { "$@" > /dev/null 2>&1; }
22:04:38 <izabera> silout () { "$@" > /dev/null; }
22:04:40 <izabera> silerr () { "$@" 2> /dev/null; }
22:06:06 -!- trout has quit (Quit: 1 found in /dev/zero).
22:06:42 <oren> I have a lot of aliases doing that stuff because, after canonical ruined everything with unity, every GUI program outputs:
22:07:08 <oren> Gtk-Message: Failed to load module "overlay-scrollbar"
22:07:15 <oren> every few seconds
22:07:24 -!- Phantom_Hoover has quit (Ping timeout: 244 seconds).
22:07:40 <elliott> izabera: install http://hartlich.com/deco/ and just use "deco" for all of them :p
22:08:06 <elliott> (deco{,-archive} are in the AUR)
22:08:26 <elliott> also protects against tar bombs
22:08:28 -!- Phantom_Hoover has joined.
22:10:05 -!- idris-bot has quit (Quit: Terminated).
22:10:18 <izabera> well honestly i never had problems with tar bombs.......
22:10:39 <izabera> probably gnu tar has an option for it <.<
22:11:21 <izabera> it does, see the overwrite control section in the man page
22:11:42 -!- idris-bot has joined.
22:11:51 <elliott> well, it's mostly .zip bombs
22:12:06 <elliott> a lot of the time I'm like "oh okay this is a .zip not a tar so uh 'unzip foo.zip'... fuck, the files went everywhere"
22:12:14 <elliott> hence deco's consistency is nice :p
22:13:24 <izabera> i believe files in a zip archive can't have hardcoded paths...
22:13:58 <izabera> i mean the worst that can happen is that they clobber the current directory
22:14:12 <elliott> yeah but that's the problem when the current directory is ~ or a downloads dir or something
22:14:16 <int-e> and of course tar files can do that as well.
22:14:20 <elliott> it's just a minor annoyance
22:14:57 <izabera> int-e: yeah but in a tar you can have a file that extracts to /bin/whatever
22:15:41 <int-e> izabera: and GNU tar at least strips leading slashes by default.
22:18:43 <oren> generally I don't just untar things
22:18:45 -!- bb010g has quit (Quit: Connection closed for inactivity).
22:18:56 <oren> I extract the specific files I want
22:20:55 <oren> tar -tzf foo.tar
22:21:17 <int-e> "By default, GNU tar drops a leading `/' on input or output, and complains about file names containing a `..' component."
22:22:04 <int-e> with that behaviour it's no worse than zip
22:22:52 -!- dianne has joined.
22:22:57 <int-e> (well, I suppose you can create interesting symlinks or special files)
22:23:40 <oren> Also if you extract a whole archive, use tar -xkzvf foo.tgz
22:24:03 -!- augur has quit (Remote host closed the connection).
22:24:19 <oren> k makes it not clobber existing files
22:26:12 <oren> I learned about tar after seeing that XKCD comic
22:26:17 <int-e> yeah, well, no. I'm a tar xzf person.
22:27:37 <int-e> (but I have a scratch directory where I unpack stuff, and usually do a tar tzf first when I'm not dealing with a source code distribution of software I'm used to...)
22:29:39 <int-e> (I've lost more data to hitting ^J instead of ^K than I have lost to extracting tar files in a different place than I wanted.)
22:29:57 -!- augur has joined.
22:30:26 -!- FreeFull_ has changed nick to FreeFull.
22:31:20 <oren> I know ^J is the same as enter...
22:31:27 <elliott> alternative explanation: d$
22:32:49 <oren> anyway I don't understand how you can lose data to hitting enter instead of cutting the line?
22:32:55 -!- Froox has joined.
22:33:51 <izabera> don't take it too seriously...
22:33:57 <int-e> by executing a half-typed command instead of deleting it. as a silly example, consider rm -rf ./ [err, no, I don't want to actually delete anything], followed by ^A (beginning of line), ^J.
22:36:27 -!- Froo has quit (Ping timeout: 250 seconds).
22:37:24 <int-e> (Perhaps I should use ^U instead of ^A^K. But U is also next to J.)
22:37:41 <oren> What about just down?
22:37:44 -!- Froox has quit (Ping timeout: 245 seconds).
22:37:59 <izabera> perhaps consider using vi mode <.<
22:38:53 <oren> nonononononononon
22:39:19 <elliott> I use vim but emacs keybindings in everything else
22:39:24 <int-e> oren: down doesn't delete anything for me (bash / readline)
22:39:31 -!- ProofTechnique has joined.
22:39:35 -!- Frooxius has joined.
22:39:58 <int-e> elliott: no more weird than any other hacker I've ever met
22:40:07 -!- Froox has joined.
22:40:18 <oren> Hm? oh. yeah for some reason I usually start every command by hitting up a fe tmes
22:40:23 <elliott> and also sometimes I use emacs but rarely these days
22:40:32 <elliott> but when I do use emacs I often edit ~/.emacs with vim
22:40:43 <oren> I use midnight commander's internal editor
22:40:44 <int-e> there are so many ways of doing stuff that nobody does everything the same way anyway
22:41:25 <oren> which makes even that nano people go "uh. what's that?"
22:41:39 <int-e> I use vi, too, for quick and simple edits. I switch to emacs when it involves moving stuff around (cut & paste)...
22:42:15 <int-e> nano gives me the vi effect (HOW DO I QUIT?)
22:42:38 <rdococ> real programmers use butterflies
22:42:43 <int-e> Err, I use vim, I should say. Cursor keys...
22:43:55 -!- Frooxius has quit (Ping timeout: 246 seconds).
22:44:26 <elliott> I get terrified when I do, like, "sudo -e" on a random system and nano pops up
22:44:41 <elliott> please. what condemned me to this fate. will I succeed in making the simple edit I'm going for here
22:45:04 <oren> nano has internal help tho
22:45:21 <int-e> (Actually quitting nano is not so hard anymore, but I still struggle with saving files. What's wrong with the "Save" verb? "WriteOut" makes no sense. Wordstar has died in the mid 1990s!)
22:45:24 -!- Froox has quit (Ping timeout: 256 seconds).
22:45:33 <oren> as in, the commands are at the bottom of the screen
22:46:05 <int-e> (needless to say the shortcuts are different from the wordstar ones as well...)
22:47:30 <int-e> "[ XOFF ignored, mumble mumble ]" ... cute.
22:48:39 <elliott> oren: it's hard to read when you're panicking though
22:49:39 <oren> I wish vi had like three or four lines of help at the bottom
22:51:09 <int-e> that would waste more than 10% of a 25 line terminal's screen estate...
22:51:47 <oren> it should enable it if it detcts you press escape like 5 times
22:52:22 * int-e wonders whether joe still exists
22:53:05 <int-e> (I don't seem to have it installed anywhere)
22:54:20 <oren> joe seems to exist, as in it prompts me to apt-get it
22:55:52 <izabera> vim has internal help too >.>
22:57:10 -!- vodkode has quit (Quit: Leaving).
22:57:39 <int-e> izabera: but it's hard to find when some other program (like crontab -e) launched it for you to edit a file.
22:59:25 <oren> izabera: where? http://snag.gy/CUxag.jpg
22:59:58 <oren> I don't see a list of commands or a thing saying "insert mode" or anything!
23:00:20 <izabera> how can you code with that
23:01:24 <FireFly> <izabera> i type ungzip every fucking time and i keep getting command not found ← don't forget bunzip2
23:01:39 <izabera> yeah that's another fucker -.-
23:01:42 -!- Hijiri has joined.
23:01:55 <int-e> right. without transparency ;) http://int-e.eu/~bf3/tmp/vi.png
23:02:17 <izabera> <esc>:h topic-you-want-help-about
23:02:21 <int-e> . o O ( Same as gzip, I use bzip2 -d )
23:02:57 <int-e> izabera: You know about vi already. We're discussing a first user experience here. Or so I thought.
23:03:11 <izabera> What is the meaning of life, the universe and everything? *42*
23:03:14 <izabera> Douglas Adams, the only person who knew what this question really was about is
23:03:16 <izabera> now dead, unfortunately. So now you might wonder what the meaning of death
23:04:04 <oren> anyway, mcedit has a list of commands at bottom, a status line at top with the name of the file. http://snag.gy/fBnvC.jpg nano has similar.
23:04:04 <int-e> Death is when you stop worrying about taxes.
23:04:38 <oren> Also you can click on the commands.
23:04:38 * int-e can't see anything in oren's screenshots.
23:04:56 <int-e> (same issue as izabera mentioned)
23:05:40 <oren> http://snag.gy/u2q1D.jpg
23:05:57 <FireFly> oren: when you start vim, you should get a startup message that tells you about :help
23:06:08 <FireFly> unless you explicitly disable it
23:06:44 <FireFly> "type :help<Enter> or <F1> for on-line help"
23:07:03 <int-e> FireFly: that works (for me) if you start 'vi', but not when you use 'vi file', which is what tools like crontab -e will do[D[D[D[D[D[D[D[D[D[D[D.
23:07:14 <int-e> (oops, slow terminal)
23:07:14 <oren> vim fake help file ... great. just great.
23:07:58 <oren> VIM - fake help file for vim-tiny The Vim online help is not installed on this Debian GNU/Linux system.
23:08:07 -!- Sprocklem has joined.
23:08:12 <FireFly> Well, what can I say.. blame Debian :P
23:08:25 <oren> anyway I set EDITOR to nano
23:09:21 -!- ProofTechnique has quit (Ping timeout: 256 seconds).
23:09:44 <FireFly> you could use the STANDARD EDITOR
23:09:47 <izabera> what does :version say in vim-tiny?
23:10:32 <FireFly> I know vim well enough, so the fact that ed ~ ex should help..
23:10:47 <int-e> ooh, nasty thought: set $EDITOR to /bin/rm ...
23:10:53 <elliott> izabera: have you ever used teco?
23:11:00 <elliott> it was nicer than ed when I tried it I think
23:11:11 <oren> ed is good. I can use ed well. but vi REALLY needs a help line at least telling me what mode i'm in. ed doesn'tneed it becuase you can see the commnads.
23:11:17 <elliott> acme is cooler than sam :p
23:11:40 <elliott> sam is the one that's ed + viewer windows, basically
23:11:56 <elliott> acme is the weird futuristic one
23:12:36 <elliott> I like it. it's minimal and flexible
23:12:42 <FireFly> oren: in vim, :set nocompatible and you get a mode thingy in the bottom left
23:13:11 <FireFly> ("nocompatible" meaning "drop strict vi emulation; I want vim powers")
23:13:44 <oren> I see... so now the arrow keys work
23:14:28 <izabera> i mapped the arrows to resize the split windows <.<
23:14:53 <oren> how do you move around the file then?
23:15:13 <elliott> I use hjkl(yubn) in roguelikes but arrow keys in vim.
23:15:34 <FireFly> I use hjklyubn in nethack because it's too messy to remap
23:15:39 <elliott> sometimes I try yubn in vim and am surprised to find they don't work
23:15:40 <FireFly> it's kinda.. weird. with dvorak though
23:15:45 <oren> why doesn't backspace work
23:16:02 <FireFly> Because of stupid defaults
23:16:02 <izabera> oren: set bs=eol,indent,start
23:16:17 <elliott> vim comes with some sample vimrc you can copy that's better than nothing
23:16:22 <elliott> it's probably in /usr/share somewhere
23:16:26 <FireFly> I hope neovim adopts nicer defaults
23:16:49 <FireFly> like, nocompatible and bs=eol,indent,start should definitely be default in 2015..
23:17:18 <FireFly> Yeah, that too, although I think that's less uniform than nocompatible and bs
23:18:01 <FireFly> The only downside I can think of for mapping Y to y$ is that you forget about the original Y when doing vimgolf..
23:18:44 <izabera> nnoremap n nzz <- also this and a couple more...
23:18:53 <oren> augh I think I prefer ed
23:19:04 <oren> ed doesn't have modes
23:19:10 <FireFly> You could use ex for a slightly more user-friendly ed
23:19:20 <izabera> it has normal and insert mode
23:19:30 <oren> (well it does, but they are like heredocs)
23:21:07 <FireFly> Hm, I guess ed is pretty straightforward after all
23:21:19 <oren> Hmm... maybe I can write my own ed with syntax highlighiting
23:22:35 <izabera> i wrote a readline wrapper to add history to ed
23:22:48 <izabera> so that the arrow keys are a little more useful
23:23:36 <FireFly> I was thinking "hey, ed is light enough that you could implement it in an esolang".. then I remembered :s
23:23:38 <oren> I've never written an editor before, but maybe I'll call it OED
23:25:06 <oren> well you can do ed -p:
23:25:19 <oren> then you get a prompt of :
23:25:19 <FireFly> With an esolang implementation of regex matching, I guess the regex engine would be bigger than the rest of the editor..
23:25:45 <izabera> actually a basic regex engine isn't that hard...
23:25:54 <shachaf> they support intersections and all sorts of things
23:26:31 <FireFly> izabera: yes, that is true..
23:28:18 <FireFly> shachaf: you mean using /\= ?
23:29:18 <FireFly> I was thinking of \@= which acts a bit like (?=) in PCRE
23:29:58 <FireFly> Apparently \& is a \@= with implicit parens around everything to the left of it
23:30:17 <shachaf> Well, maybe \@= isn't the same as (?=)
23:30:36 <FireFly> "Note that using "\&" works the same as using "\@=": "foo\&.." is the same as "\(foo\)\@=..". But using "\&" is easier, you don't need the braces."
23:31:01 -!- GeekDude has changed nick to `test.
23:31:33 -!- augur_ has joined.
23:31:43 <FireFly> \@= tries to match the preceding atom zero-width, as far as I know that's the same as Perl's (?=...) (and the help seems to say so, too)
23:31:52 -!- Hijiri has quit (Ping timeout: 256 seconds).
23:32:58 <shachaf> If it does actual regular language intersection, then there's a difference.
23:33:26 <izabera> sorry guys i found this in ltrace's output regcomp(0x7ffeda758840, 0x740470, 1, 0) = 0
23:34:05 <izabera> i found that and regexec(0x7ffeda758840, 0x7403f0, 3, 0x741680) = 1
23:34:09 <izabera> and regfree(0x7ffeda758840, 0x7fb569fddc08, 0x741600, 0x740ec0) = 0
23:35:14 -!- augur has quit (Ping timeout: 245 seconds).
23:35:22 <izabera> ok np i just thought that some of you could know it <.<
23:37:05 -!- Hijiri has joined.
23:37:14 -!- augur_ has quit (Remote host closed the connection).
23:39:29 -!- augur has joined.
23:46:15 -!- `test has changed nick to GeekDude.
23:47:42 <elliott> izabera: I think you can teach strace that the first two parameters are pointers?
23:47:50 <elliott> strace knows types and follows pointers, at least
23:47:59 <elliott> you could just use gdb and breakpoint on regex stuff
23:48:48 <elliott> izabera: http://man7.org/linux/man-pages/man5/ltrace.conf.5.html
23:48:53 <elliott> make one of these for the functions in question
23:50:29 <izabera> i seriously hate debugging -_-
23:59:44 <shachaf> How efficiently can two regular expressions be compared for equivalence?