00:17:13 -!- hppavilion1 has joined.
00:17:40 -!- Phantom_Hoover has joined.
00:17:59 <Sgeo_> https://www.youtube.com/watch?v=GFLkou8NvJo
00:18:53 -!- aretecode has quit (Quit: Toodaloo).
00:24:30 <Sgeo_> nortti, I understand the end of the world a bit better now https://www.youtube.com/watch?v=jWeAsCzHvTo
00:26:49 <nortti> thanks. do you know if there exists a list of all songs released up to date, so I can check what I've missed
00:27:10 -!- variable has joined.
00:29:32 <Sgeo_> http://theevilliouschronicles.wikia.com/wiki/Category:Songs might be your best bet
00:30:17 <nortti> :/ would have liked a chronologically ordered one, but eeh, good enough
00:30:20 -!- tromp__ has joined.
00:30:58 <Sgeo_> chronologically as in in-world time, or as in release order?
00:33:29 <Sgeo_> http://theevilliouschronicles.wikia.com/wiki/Evillious_Chronicles#Song_Listing
00:33:39 <Sgeo_> Not certain if release order, but at least a few songs look right
00:35:18 <Sgeo_> I should make a playlist at some point
00:35:24 <Sgeo_> I don't think I've seen every song
00:52:19 <Sgeo_> nortti, hmm, that translation of Boy of the End seemed to imply to me that ONE of the endings is the true one, but wiki seems to suggest all of them will come true
00:57:17 <Sgeo_> THE WORLD ENDED ALREADY?!?
00:57:17 <Sgeo_> http://theevilliouschronicles.wikia.com/wiki/Punishment#History
00:57:37 <Sgeo_> I thought the songs were building up to revealing what the end would be, with the birth of Irregular etc.
00:59:19 -!- tromp__ has quit (Remote host closed the connection).
00:59:24 -!- variable has changed nick to constant.
01:04:29 -!- hppavilion1 has quit (Ping timeout: 246 seconds).
01:04:29 -!- hppavilion[1] has quit (Ping timeout: 246 seconds).
01:04:41 -!- fungot has quit (Ping timeout: 260 seconds).
01:05:33 <oerjan> the world _must_ be ending, fungot has left
01:06:27 -!- mihow has quit (Quit: mihow).
01:30:38 -!- mihow has joined.
01:30:44 -!- mihow has quit (Remote host closed the connection).
01:31:04 -!- mihow has joined.
01:33:49 -!- Phantom_Hoover has quit (Remote host closed the connection).
01:43:50 <zzo38> Is this a proper way of to make man page? http://sprunge.us/HBIM It is work for me, but is the stuff written there how it should be written man page?
01:49:06 -!- mihow has quit (Quit: mihow).
01:49:30 -!- mihow has joined.
01:56:34 -!- Herbalist has joined.
01:56:44 -!- Herbalist has left.
02:01:49 -!- mihow has quit (Quit: mihow).
02:14:19 -!- oerjan has quit (Quit: Gooed knight).
02:24:40 -!- mihow has joined.
02:35:19 -!- zzo38 has quit (Remote host closed the connection).
02:47:49 -!- tromp__ has joined.
02:52:10 -!- tromp__ has quit (Ping timeout: 252 seconds).
02:52:46 -!- mihow has quit (Quit: mihow).
03:45:02 -!- zzo38 has joined.
03:49:13 -!- tromp__ has joined.
03:53:02 <zzo38> Do you believe my guess of how the example codes in the [[Tangle bracket language]] is working, is correct?
03:54:10 -!- tromp__ has quit (Ping timeout: 264 seconds).
04:05:16 -!- Wright_ has joined.
04:05:16 -!- Wright has quit (Read error: Connection reset by peer).
04:09:17 <zzo38> Putting a bent paperclip in the power supply worked!
04:14:59 <zzo38> Can I get Windows-like function key operation for bash?
04:24:24 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
04:24:48 -!- tromp__ has joined.
04:39:51 -!- password2 has joined.
04:47:12 <zzo38> Such as F3=repeat last command (but keep what you have currently typed and delete those first few characters from the added text), F2=type another character and repeat previous like F3 but stop just before that character, F5=save last command but don't execute it, F8=search in history for command starting with what you have typed, F1=repeat one character from last command.
04:49:21 <zzo38> But it would help if F5 function it change text into different color so that you can know it didn't execute
04:57:29 <oren> I recommend learning the default keys for bash, that way you can use them on every linux box
04:58:14 <zzo38> Yes, but are any of these functions available?
04:58:43 <oren> ctrl-r is search history
04:58:46 <oren> http://www.bigsmoke.us/readline/shortcuts
04:59:25 -!- tromp__ has quit (Remote host closed the connection).
05:06:43 -!- llue has joined.
05:06:43 -!- llue has quit (Changing host).
05:06:43 -!- llue has joined.
05:10:01 -!- lleu has quit (Ping timeout: 256 seconds).
05:20:43 -!- mihow has joined.
05:32:43 -!- password2 has quit (Remote host closed the connection).
05:58:28 <zzo38> Some terminal emulator implement you can't read back the window title, for security purpose. But I have another idea, to have "secure read title" option, which when the program request the window title it will remember the current window title and send the index into its array of storage, so that it can still put back the window title how it was before, even without knowing what it is.
05:59:55 -!- tromp__ has joined.
06:04:44 -!- tromp__ has quit (Ping timeout: 264 seconds).
06:10:57 -!- MDream has joined.
06:15:32 -!- hppavilion[1][42 has joined.
06:15:45 -!- hppavilion[1][42 has changed nick to hppavilion[1][0].
06:21:01 <zzo38> OK, what kind of esoteric assembly language?
06:24:09 <myname> i do think most oisc languages cannot be put together
06:24:24 <myname> since you don't actually mention the instruction
06:24:41 <myname> you'd have to make the instruction depending on the value
06:24:55 <myname> it may be hard to do this without losing tc
06:25:18 -!- radza has joined.
06:25:40 <hppavilion[1][0]> What I meant was taking the _possible instructions for use_ in a OISC
06:25:43 <myname> that would be pretty hard for e.g. subleq
06:26:51 -!- radza has left.
06:29:53 -!- mihow has quit (Quit: mihow).
06:40:05 -!- _256Q has quit (Read error: Connection reset by peer).
06:53:36 -!- mihow has joined.
07:01:21 -!- tromp__ has joined.
07:05:52 -!- tromp__ has quit (Ping timeout: 260 seconds).
07:22:49 -!- hppavilion[1][0] has quit (Ping timeout: 246 seconds).
08:04:57 -!- mihow has quit (Quit: mihow).
08:06:30 -!- white_bear has joined.
08:27:38 -!- Patashu has joined.
08:50:45 <coppro> ok so CPT-invariant thermodynamics is trippy
08:50:50 <coppro> but CP-invariant thermodynamics is bizarre
08:57:55 <b_jonas> which type of CPT-invariant termodynamics? the one that blames the edge conditions for low entropy?
09:11:27 -!- x10A94 has joined.
09:20:39 <zzo38> Can GCC optimize a multiplication of a boolean by a number?
09:21:12 -!- hjulle has joined.
09:21:24 <Taneb> zzo38, that should either be 0 or an unspecified number?
09:21:33 <Taneb> I am not sure if such an optimization is valid
09:21:53 <Taneb> Say, if you're working with unsigned 16-bit ints
09:22:17 <zzo38> Most boolean operations are 0 or 1
09:22:33 <Taneb> zzo38, it's actually unspecified, I think
09:22:35 <zzo38> (Some might not be though)
09:24:46 <zzo38> As far as I know it is 0 or 1 if the built-in operators (not functions) are used, it seem to be. But in BASIC and Forth it is 0 and -1 instead.
09:27:39 <zzo38> And even in case where it is unspecified, you can guarantee to be zero if the second number is also zero. Possibly in a few cases this might be useful I don't quite know
09:27:54 <coppro> b_jonas: no, the general consequence of what happens to themodynamics under the T symmetry
09:27:58 <coppro> does it flip, or does it not?
09:29:09 <coppro> if it flips, that provides a potential explanation of the abundance of matter (it's thermodynamically favourable for antimatter to turn into matter, but not vice versa)
09:29:12 -!- hjulle has quit (Quit: ZNC - http://znc.in).
09:29:19 <coppro> and it implies that in general, antimatter is not entropic but syntropic
09:29:53 <Jafet> `` cd /tmp && echo 'unsigned f(unsigned a, unsigned b) { return (a==b)*b; }' > bmul.c && gcc bmul.c -c -g -O3 -march=native && objdump -d bmul.o
09:30:08 <coppro> if thermodynamics does not flip, then antimatter is entropic, which means that the universe as a whole does not obey CPT symmetry, raising two questions: what will replace QFT for the theory of everything, since QFT can't violate CPT symmetry, and why does the arrow of time exist?
09:30:27 <HackEgo> \ bmul.o: file format elf64-x86-64 \ \ \ Disassembly of section .text: \ \ 0000000000000000 <f>: \ 0:31 c0 xor %eax,%eax \ 2:39 f7 cmp %esi,%edi \ 4:0f 94 c0 sete %al \ 7:0f af c6 imul %esi,%eax \ a:c3 retq
09:30:54 <Jafet> So, this gcc does not.
09:31:05 <coppro> syntropic antimatter also raises the question of whether or not you can convey information backwards in time using it
09:31:16 <Jafet> `` cd /tmp && echo 'unsigned f(unsigned a, unsigned b) { return a==b? b : 0; }' > bmul.c && gcc bmul.c -c -g -O3 -march=native && objdump -d bmul.o
09:31:19 <coppro> since it would obey retrocausality, not causality
09:31:23 <HackEgo> \ bmul.o: file format elf64-x86-64 \ \ \ Disassembly of section .text: \ \ 0000000000000000 <f>: \ 0:89 f0 mov %esi,%eax \ 2:39 f7 cmp %esi,%edi \ 4:ba 00 00 00 00 mov $0x0,%edx \ 9:0f 45 c2 cmovne %edx,%eax \ c:c3 retq
09:38:56 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: clang: not found
09:39:07 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: tcc: not found
09:41:41 -!- AnotherTest has joined.
09:48:52 -!- tromp__ has joined.
09:50:34 -!- llue has quit (Read error: Connection reset by peer).
09:50:57 -!- llue has joined.
09:50:58 -!- llue has quit (Changing host).
09:50:58 -!- llue has joined.
09:53:26 -!- tromp__ has quit (Ping timeout: 244 seconds).
10:00:03 -!- llue has quit (Quit: That's what she said).
10:00:30 -!- lleu has joined.
10:00:30 -!- lleu has quit (Changing host).
10:00:30 -!- lleu has joined.
10:01:08 <zzo38> Can you access memory usage data including buffer/cache in Linux without parsing the /proc/meminfo file?
10:14:06 -!- Wright_ has quit (Remote host closed the connection).
10:15:11 <Deewiant> Yes, you can be root and parse all the /proc/$pid/smaps files instead
10:16:31 -!- Patashu has quit (Remote host closed the connection).
10:16:48 -!- Patashu has joined.
10:18:22 <zzo38> I mean don't parsing any file
10:18:32 <zzo38> (nor the output of the free command)
10:19:51 <zzo38> I did look at the source-codes of how it is implemented in /proc/meminfo it says cached = global_page_state(NR_FILE_PAGES) - total_swapcache_pages() - i.bufferram;
10:20:37 <int-e> zzo38: does sysinfo() miss anything you need?
10:20:52 <zzo38> Yes, it misses the cache memory amount
10:21:47 -!- boily has joined.
10:33:11 <b_jonas> zzo38: apart from directly accessing the memory usage data, you can try running the "free" or "top" programs.
10:34:42 <b_jonas> but those programs probably actually get most of their data from /proc files on Linux
10:35:11 <zzo38> Is it possible to tell Firefox to display the percent-encodings of Unicode characters in the location bar rather than the characters themself?
10:35:27 <b_jonas> zzo38: apart from that, you can try using the sysinfo() system call, which gives some memory usage statistics data
10:36:04 <zzo38> b_jonas: Yes but sysinfo() doesn't have the cache amount
10:37:59 <b_jonas> zzo38: as for that, I think if you copy the url to the clipboard (eg. by pressing control-L control-C ), it copies the original version with percent encodes.
10:38:36 <zzo38> Yes I know, but I want it to display the percent encodings all the time
10:39:19 <Deewiant> zzo38: For details like that I think parsing /proc is the only way to go in user space. Alternatively you can write a kernel module to get a more machine-readable form
10:41:40 <b_jonas> zzo38: I don't know how to do that then. Maybe there's some well-hidden setting or a firefox extension that does that.
10:46:07 <zzo38> I want the status bar also to display percent encodings
11:04:53 <zzo38> Actually I think I found the problem, it is a function called "losslessDecodeURI" in browser.js, but I am not sure how to override it (in userChrome.js)
11:05:50 <Taneb> Is there a proof that brainfuck is NOT turing complete with only two (unbounded) cells?
11:15:38 -!- Phantom_Hoover has joined.
11:19:57 -!- boily has quit (Quit: KERNED CHICKEN).
11:33:01 <zzo38> I managed to fix that function, it is not so difficult to fix. I did: window.losslessDecodeURI=function(aURI) { return aURI.spec; };
11:33:18 -!- SopaXT has joined.
11:38:20 -!- tromp__ has joined.
11:39:14 <Jafet> It's probably true, but no one's made a proof yet
11:42:43 <zzo38> Then you should try to prove it by yourself.
11:43:10 -!- tromp__ has quit (Ping timeout: 265 seconds).
11:45:26 <Jafet> I'm not sure what kind of reduction would be used to prove it, since most lower computational classes are finite in some sense
12:24:01 -!- oerjan has joined.
12:35:20 -!- Patashu has quit (Ping timeout: 264 seconds).
12:39:06 -!- tromp__ has joined.
12:43:35 -!- tromp__ has quit (Ping timeout: 250 seconds).
13:18:59 -!- ais523 has joined.
13:33:58 -!- SopaXT has quit (Read error: Connection reset by peer).
13:40:11 -!- `^_^v has joined.
13:43:55 -!- ais523 has quit.
13:44:50 -!- ais523 has joined.
13:45:29 -!- oerjan has quit (Quit: leaving).
13:57:01 -!- SopaXT has joined.
14:02:58 -!- ais523 has quit.
14:03:56 -!- ais523 has joined.
14:12:12 -!- hilquias has joined.
14:14:31 -!- ais523 has quit.
14:15:49 -!- ais523 has joined.
14:23:47 -!- ais523 has quit.
14:25:03 -!- ais523 has joined.
14:36:20 -!- fungot has joined.
14:37:29 <fungot> fizzie: chipmunk says: nice job on ooe on c.l.s. right now i'm using cvs, but it's a pretty lowlevel basic right?
14:37:45 <fizzie> fungot: CVS, very retro.
14:37:45 <fungot> fizzie: it is quite unreadable...)) expression)
14:56:12 -!- ais523 has quit.
14:56:27 -!- ais523 has joined.
14:59:18 -!- GeekDude has joined.
14:59:33 -!- nycs has joined.
15:02:23 -!- `^_^v has quit (Ping timeout: 244 seconds).
15:09:59 -!- constant has quit (Ping timeout: 256 seconds).
15:14:57 <tswett> Wisdom from the neural net:
15:15:00 <tswett> "It's not much for a turing complete IP to use somewhat bad parts, and then you will confuse every cpu call"
15:18:07 <ais523> what input did you give it to produce that?
15:18:20 -!- ais523 has quit.
15:18:35 -!- ais523 has joined.
15:18:36 <ais523> I mean, it's not actually a meaningful sentence, but it does look meaningful until you read it multiple times
15:18:37 <ais523> I mean, it's not actually a meaningful sentence, but it does look meaningful until you read it multiple times
15:18:52 <tswett> The #esoteric IRC logs.
15:19:03 <tswett> I think the ones for April and part of May, maybe?
15:20:01 <tswett> Taneb: a finite state machine with access to a two-unbounded-cell brainfuck tape can simulate Fractran, so that would be Turing-complete.
15:20:21 <tswett> So if brainfuck with two unbounded cells is *not* Turing-complete, it must be because brainfuck's flow control isn't powerful enough.
15:21:27 -!- SopaXT has quit (Remote host closed the connection).
15:21:28 <shachaf> `` ln -s universe wisdom/the\ universe
15:21:32 <Taneb> tswett, oerjan proved that three cells was enough
15:21:47 <Taneb> So I was wondering if it's possible to go deeper, so to speak
15:22:02 <shachaf> `` sed -i 's/Go,/Go, the universe,/' wisdom/tanebvention
15:23:01 <ais523> wait, you can't just arbitrarily declare Taneb to have invented arbitrary things
15:24:33 <tswett> `le/rn fundamental theorem of Taneb/The Fundamental Theorem of Taneb states that for all strings S, if S describes a thing, then it is provable that Taneb invented the thing described by S; and, furthermore, that it is provable that there exists a string T that describes a thing that Taneb did not invent.
15:24:35 <HackEgo> Learned «fundamental theorem of taneb»
15:24:41 <tswett> `? fundamental theorem of Taneb
15:24:42 <HackEgo> The Fundamental Theorem of Taneb states that for all strings S, if S describes a thing, then it is provable that Taneb invented the thing described by S; and, furthermore, that it is provable that there exists a string T that describes a thing that Taneb did not invent.
15:28:30 <ais523> is that the Taneb version of omega-incompleteness?
15:28:33 -!- tromp__ has joined.
15:28:37 <ais523> (if I have an omega on my compose key, I don't know where it is)
15:29:39 -!- comodvs has joined.
15:29:45 <tswett> What's omega-incompleteness?
15:29:59 <comodvs> does anyone here have The Book of Nymphs, Sylphs, Pygmies, and Salamanders by Paracelsus
15:31:06 <HackEgo> comodvs: 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 EFnet or DALnet.)
15:31:50 <comodvs> i'm looking for this book all over the internet, but i can't find it in any place :/
15:32:56 -!- tromp__ has quit (Ping timeout: 244 seconds).
15:36:17 <ais523> possibly the wrong #esoteric, hard to tell though
15:36:29 <ais523> tswett: oh, I was thinking of omega-inconsistency
15:36:57 <ais523> it's where a proof system allows you to prove that a property is true for some number, but also allows you to prove it to be false for any specific number
15:48:08 <Jafet> Inaccessible fungot.
15:48:08 <fungot> Jafet: can i help you?. if you have " all permissions"? implementing them for some time, could you just send a /msg if you want something like this
16:05:29 -!- nycs has quit (Quit: This computer has gone to sleep).
16:09:26 <tswett> Doink. In my client, ais523 and fungot's nicks have the same length and color, so I thought that ais523 said what fungot said.
16:09:26 <fungot> tswett: my own test cases with m-o c to send to a scheme simulator: i was reading it and pondering about iguanas...)
16:09:53 <tswett> But it looked pretty fungotty, so I figured maybe ais523 invoked fungot privately and then just pasted fungot's response to the channel.
16:09:53 <fungot> tswett: stop being an associative list. but that's just to annoy sonera disconnects me?
16:11:04 <shachaf> is tswett being an associative list again
16:11:21 <Jafet> fungot: you're in the UK now, Sonera can't disconnect you.
16:11:23 <tswett> I'm: sorry; I: can't help it;
16:12:13 <ais523> `addquote <fungot> tswett: stop being an associative list. [...] <shachaf> is tswett being an associative list again <tswett> I'm: sorry; I: can't help it;
16:12:14 <fungot> ais523: i'm just a mere functionary who knows how that happens, matters, unless humans start pursuing and succeed in the turing test...)
16:12:16 <HackEgo> 1249) <fungot> tswett: stop being an associative list. [...] <shachaf> is tswett being an associative list again <tswett> I'm: sorry; I: can't help it;
16:13:59 <Jafet> fungotionary, the herald of the turing test
16:13:59 <fungot> Jafet: take it up the ass. boxes are as bad as last time we did this using the above formula as 0 1 2 3))
16:15:10 <Jafet> I don't think that aspect of human behaviour is exercised by the test.
16:15:40 <tswett> Lessee if I can find that list of bot prefixes.
16:15:57 <fungot> Bot prefixes: fungot ^, HackEgo `, EgoBot !, lambdabot @ or ?, thutubot +, metasepia ~, idris-bot ( , jconn ) , blsqbot !
16:16:43 <tswett> I wonder if lambdabot also responds to its name.
16:18:41 -!- juzo has joined.
16:20:07 <HackEgo> ¡Bienvenido al centro internacional para el diseño y despliegue de lenguajes de programación esotéricos! Por desgracia, la mayoría de nosotros no hablamos español. Para obtener más información, echa un vistazo a nuestro wiki: http://esolangs.org/. (Para el otro tipo de esoterismo, prueba #esoteric en EFnet o DALnet.)
16:20:21 <juzo> como estan todos
16:20:24 <tswett> Pero yo hablo español un poco.
16:20:47 <x10A94> Holy hell, why do you need so many bots?
16:21:04 <ais523> we don't need this many, but there are a bunch of programmers here
16:21:07 <ais523> so it just tends to happen
16:21:15 <ais523> often we just write the bots for the sake of something to write
16:21:21 <ais523> (and most of them aren't connected at any given moment in time)
16:21:29 <tswett> No los *necesitamos*, pero nos gustan.
16:21:34 -!- `^_^v has joined.
16:21:46 <Taneb> x10A94, we don't have that many, just fungot EgoBot glogbot HackEgo lambdabot idris-bot j-bot zemhill__ and half of myndzi
16:21:46 <fungot> Taneb: " fnord" either, but i
16:21:48 -!- juzo has left.
16:22:01 <ais523> Taneb: you forgot about clog and glogbot
16:22:05 <ais523> although that's understandable because they don't talk
16:22:07 <Taneb> ais523, I got glogbot
16:22:22 <tswett> Pues, voy a ducharme. Hasta pronto.
16:22:23 <Taneb> Still, we're at less than 10% bot
16:22:25 <ais523> I assumed that forgetting one would make you forget the other
16:22:31 <ais523> when was cmeme last here, btw?
16:22:52 <Taneb> ais523, I tend to use glogbot's logs and not clog's logs
16:23:14 <tswett> Man, I can't wait to get autochat9000 in here.
16:23:22 <tswett> One sec, lemme register that nick.
16:23:27 <Taneb> I should resurrect pietbot
16:23:35 <ais523> did pietbot ever end up in-channel?
16:23:37 -!- tswett has quit (Changing host).
16:23:37 -!- tswett has joined.
16:23:41 -!- tswett has changed nick to autochat9000.
16:23:50 <ais523> I don't have a huge urge to resurrect thutubot, but it was fun for a while
16:23:53 -!- autochat9000 has changed nick to tswett.
16:23:59 <Taneb> Never actually said anything
16:23:59 <ais523> anyway, in terms of bots we actually /need
16:24:15 <Taneb> It could connect, register with nickserv, join the channel, idle until commanded to do something, then crash
16:24:23 <Taneb> Or hang, I can't remember
16:24:30 <ais523> probably HackEgo + one logbot would be enough by itself (also, lambdabot is useful at the times this channel turns into a second #haskell rather than doing its day job)
16:24:37 <ais523> and to be fair, like three quarters of HackEgo is useless
16:24:47 <ais523> but a bunch of useful stuff got copied there too
16:24:51 <ais523> the useless stuff is used more often though
16:25:48 <Jafet> `` echo 'bienvenido | rainwords' > bin/arienvenido && chmod +x bin/arienvenido
16:26:45 <ais523> how often do we need a rainbow-coloured welcome in Spanish? answer: even less often than we need a rainbow-coloured welcome in English, which is never
16:26:50 <Jafet> `` echo 'bienvenido "$@" | rainwords' > bin/arienvenido && chmod +x bin/arienvenido
16:27:17 <Taneb> I have a cinema to get to
16:27:25 <ais523> I remember when the topic was the "The international hub…" one for months or years at a time
16:27:35 <ais523> and the ontopic discussion happened more often and was more interesting
16:27:42 <ais523> and the offtopic discussion was terrible and made you not want to be here at all
16:27:52 <fizzie> tswett: FWIW, Freenode recommends you register bots with a separate NickServ account: https://freenode.net/faq.shtml#nicksetup step #6: "The exception to this is where you might want to run a bot. You should register a separate account for your bot."
16:27:56 <ais523> nowadays the offtopic discussion is mostly inoffensive and the ontopic discussion is much less of a draw
16:28:57 <Jafet> Well, I tried proving that two-cell brainfuck is decidable but lost interest
16:29:19 <ais523> two-cell bignum, I take it
16:29:39 <tswett> fizzie: *nod* Do you know why that is?
16:29:54 <ais523> is there ever a benefit to an unbalanced loop in that language?
16:29:57 <Jafet> The idea I had was that each loop leaves one cell zero, so one could try to squeeze this into a pushdown automaton stack
16:30:13 <ais523> tswett: I imagine so that you don't get spammed with "someone is connecting from your account" messages when you're online at the same time as the bot
16:30:27 <ais523> and so that a malfunctioning bot can be banned without affecting its owner
16:30:29 <fizzie> I think I saw some rationale somewhere, but it was more about if your bot is misbehaving it being not tied to your account so much.
16:31:19 <fizzie> And maybe also to not make your channel-operatory privileges available to someone who can take control of the bot.
16:31:27 <Jafet> It's also not clear whether it could be more powerful if <> didn't wrap around or if - on a zero cell has no effect
16:32:44 <ais523> with 2-cell BF, you can do a "multiply by constant" operation
16:32:54 <ais523> on one of the cells, while the other is zero
16:32:59 <ais523> this is a known path to infinite storage
16:33:11 <ais523> the problem is if you can do the corresponding "check for divisibility
16:33:19 <ais523> that's needed to make use of the info
16:33:21 <Jafet> Two-counter minsky machines have a conditional branch though
16:33:31 <ais523> yes, the problem here is that the loops are nested
16:33:33 <Jafet> That's why this problem is interesting, I suppose
16:33:49 <ais523> actually I'm reminded of that BF variant I made with an ambiguously spelled name
16:34:03 <ais523> that's also almost certainly TC if it has goto
16:34:11 <ais523> but the fact that loops need to nest makes things interesting
16:34:25 <ais523> (the two might be equivalent, actually)
16:35:11 -!- ais523 has quit.
16:36:46 -!- ais523 has joined.
16:44:28 -!- ais523 has quit.
16:48:33 -!- ais523 has joined.
17:04:24 -!- white_bear has quit (Quit: leaving).
17:05:52 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
17:17:57 -!- tromp__ has joined.
17:18:28 -!- ais523 has quit.
17:19:23 -!- ais523 has joined.
17:22:28 -!- tromp__ has quit (Ping timeout: 244 seconds).
17:29:52 -!- TieSoul has joined.
17:34:59 -!- mihow has joined.
17:36:01 <ais523> global notices have typos in nowadays? :-D
17:38:49 <tswett> I've noticed that Brits can say "in" where we Americans would have to say "in them" or "in it".
17:39:27 <ais523> it's an elision, which were always /technically/ legal but mostly only used in poetry
17:39:32 <ais523> that one's been catching on though
17:39:49 <tswett> I think for me, "in" implies that it's the sort of thing which is inserted and removed on a regular basis.
17:40:06 -!- ais523 has quit.
17:40:25 <tswett> ais523 saw what I said and immediately ragequit.
17:40:58 <tswett> You can say "a VCR with a tape in", or "a gun with bullets in", but not "a notice with typos in".
17:41:51 -!- ais523 has joined.
17:41:52 <tswett> Come to think of it, you can't say "a glass with water in" either.
17:42:23 <tswett> Unless it has some special compartment for water which is separate from the main part of the glass.
17:43:36 -!- GeekDude has joined.
17:48:08 -!- mihow has quit (Quit: mihow).
17:48:49 -!- mihow has joined.
18:08:39 -!- atrapado has joined.
18:20:37 -!- ais523 has quit (Remote host closed the connection).
18:20:55 -!- ais523 has joined.
18:36:37 -!- hppavilion1 has joined.
18:40:48 -!- ais523 has quit.
18:41:01 -!- ais523 has joined.
18:48:12 <HackEgo> [wiki] [[MOISC]] N http://esolangs.org/w/index.php?oldid=43568 * Hppavilion1 * (+890) Created Page
18:51:34 <lambdabot> LOWI 221820Z 27014KT 9999 FEW070CB SCT085 BKN140 21/18 Q1015 NOSIG
18:55:07 <hppavilion1> So in other news, I'm making a simple assembly language for use by educational institutions teaching Compiler Design
19:02:00 <coppro> hppavilion1: because there aren't enough of those already
19:02:30 <hppavilion1> I've never heard of one, but now that you mention it there are probably thousands
19:13:36 -!- ais523 has quit.
19:13:53 -!- ais523 has joined.
19:14:31 -!- x10A94 has quit (Read error: Connection reset by peer).
19:21:58 -!- oerjan has joined.
19:23:17 <oerjan> wut, the logs have a reasonably correct clock
19:25:59 -!- MDream has quit (Ping timeout: 264 seconds).
19:28:23 -!- ais523 has quit.
19:28:42 -!- ais523 has joined.
19:35:05 -!- TieSoul has quit (Read error: Connection reset by peer).
19:37:11 -!- ais523 has quit.
19:37:28 -!- ais523 has joined.
19:43:01 <zzo38> Is there program it can helping you to make command-line version of many websites?
19:55:47 <oerjan> well, the codu logs. i haven't checked tunes.
19:59:55 -!- hilquias has quit (Read error: Connection reset by peer).
20:00:07 -!- hilquias has joined.
20:17:40 -!- ais523 has quit.
20:24:37 -!- pseudomyne has joined.
20:26:33 -!- hppavilion1 has quit (Ping timeout: 246 seconds).
20:29:59 -!- mauris has joined.
20:58:51 -!- tromp__ has joined.
21:03:42 -!- tromp__ has quit (Ping timeout: 255 seconds).
21:07:47 -!- Patashu has joined.
21:08:42 -!- _256Q has joined.
21:12:06 -!- pseudomyne has quit (Remote host closed the connection).
21:26:45 -!- mauris_ has joined.
21:28:56 -!- mauris has quit (Ping timeout: 246 seconds).
21:36:56 -!- bb010g has quit (Quit: Connection closed for inactivity).
21:37:33 <oerjan> Hm so here is my "ideal" made-up-on-the-spot system. It somewhat combines all of 2, 3 and 4 from https://ghc.haskell.org/trac/ghc/wiki/Commentary/Compiler/GenericDeriving#InteractionwithGeneralizedNewtypeDeriving. It is intended to be backwards-compatible, except for one added warning.
21:37:37 <oerjan> * By 3, classes may be annotated (pragma?) to say they prefer GND or DAC deriving. Builtin-derived classes count as annotated for their own style of deriving. To actually derive a class in a module, any extension for the derivation style still needs to be enabled as well.
21:37:42 <oerjan> * By 4, if `GeneralizedNewtypeDeriving` is enabled, the `newtype` keyword may be used to signify that an instance for a newtype should be GND derived, even if this is against the annotated behavior for the class. This might even include builtin-derived classes like `Show`. (Obviously not `Typeable`, though.)
21:37:46 <oerjan> * Also by 4, if both `GeneralizedNewtypeDeriving` and `DeriveAnyClass` are enabled (or for least surprise, maybe even with just the latter), the `default` keyword may be used to signify that an instance for a newtype should ''not'' be GND-derived, even if this is the annotated behavior for the class.
21:37:51 <oerjan> * By 2, if ''neither'' the newtype deriving nor the class is annotated, then the behavior depends on which of `GeneralizedNewtypeDeriving` and `DeriveAnyClass` is enabled. If ''both'' are enabled, then a ''warning'' should be given. (This prevents surprises when a user adds both extensions for unrelated instances.) Then it defaults to DAC as today.
21:37:56 <oerjan> * Although the proper extensions need to be enabled for whichever annotations/derivation styles end up being used, the ''only'' case where simply changing the extensions enabled will change code from one legal style of derivation to another should be the one in the previous point. (And thus the warning.)
21:38:02 <oerjan> As of now, I don't remember any classes with builtin-derivations that also are useful with `DeriveAnyClass`. So I think there isn't much need to be able to distinguish those two cases. Which also means that none of this matters to `data` declarations, only `newtype`.
21:38:06 <oerjan> i guess you all got to read me latest ghc trac post hth
21:38:32 * oerjan belatedly shortens clipboard
21:39:17 <oerjan> well, anyone awake, anyway.
21:39:45 <Taneb> Ant-Man was good on a second watch
21:40:13 <HackEgo> Thanks, watcher. Thatcher.
21:40:23 <oerjan> Taneb: i was just thinking i shouldn't do that
21:40:31 <int-e> `le/rn watch Too late!
21:40:58 <oerjan> i though `le/rn had got that bug fixed...
21:41:06 <HackEgo> rm: cannot remove `/home/hackbot/hackbot.hg/multibot_cmds/env/.hg/store/data/canary.orig': Is a directory \ Done.
21:41:10 <int-e> `le/rn watch/Too late!
21:41:53 <int-e> `rm wisdom/watch Too late!
21:41:54 <HackEgo> rm: cannot remove `wisdom/watch Too late!': No such file or directory
21:41:55 -!- Patashu has quit (Ping timeout: 244 seconds).
21:42:15 <int-e> `rm wisdom/watch too late!
21:43:12 <nortti> Sgeo_: I think I might get how the four ends work. in the song, it's mentioned that depending on who you are, hänsel appears holding either a knife or a bottle, and either as an angel or a mad familiar. maybe which of the four ends happens depends on the person?
21:43:41 <oerjan> `` echo 'Spjætt!' >canary
21:44:30 <oerjan> those were interjections i remember from a norwegian translation of bloom county (or outland) hth
21:46:32 -!- AnotherTest has quit (Ping timeout: 272 seconds).
21:48:06 -!- `^_^v has quit (Ping timeout: 240 seconds).
21:48:39 -!- mauris_ has changed nick to mauris.
21:50:58 -!- bb010g has joined.
21:53:00 <int-e> `` sed -i 1a'[[ "$1" = */* ]] || exit' bin/slashlearn
21:53:43 <int-e> `le/rn no slash test
21:54:28 <int-e> `le/rn /no keyword test
21:54:55 <int-e> there's no test for lack of content
21:55:07 -!- boily has joined.
21:55:12 <int-e> shachaf: in your dreams!
21:55:21 <shachaf> Though mk also has this issue.
21:55:48 <HackEgo> [[ "$1" == *//* ]] || exit 1; key="${1%%//*}"; value="${1#*//}"; echo "$value" > "$key" && echo "$key"
21:56:16 <int-e> (why did I read the bash manpage to find [[ ]] when I could've looked there?
21:56:36 <HackEgo> /hackenv/bin/mk: line 1: : No such file or directory
21:58:22 <int-e> presumably you could test for ?*//*
21:59:35 <shachaf> `` sed -i 's/\*/?*/' bin/mk
22:03:48 <oerjan> <Taneb> Is there a proof that brainfuck is NOT turing complete with only two (unbounded) cells? <-- hm it _should_ be possibly to deduce from the presumably known fact that a minsky inc/dec machine with 1 cell is not TC
22:04:19 <Taneb> oerjan, isn't that a push-down automaton
22:04:26 <Taneb> known to be sub-TC
22:04:47 <oerjan> Taneb: um i'm not sure, i just assume it's known and stuff
22:05:05 <oerjan> since otherwise, why would they keep blathering about the 2-cell one.
22:05:29 <int-e> that's unclear... a 2 counter minsky machine is TC, and the proof I know works by simultating a 3 counter machine by dealing with numbers 2^a 3^b 5^c
22:05:48 <int-e> (and 3 counters are good for two stacks, hence a tape)
22:06:06 <oerjan> int-e: i'm talking about _1_-cell without multiplication or division
22:06:27 <int-e> but 1 cell isn't even a push-down automaton
22:06:35 <Taneb> oerjan, can you @tell me the answer? I'm going to bed now
22:06:51 <int-e> (well... it is a degenerated one, I guess that's what Taneb meant...)
22:07:19 <oerjan> well my point is that i think 2-cell bf is reducible to 1-cell minsky after a terminating starting phase.
22:07:47 <oerjan> namely, run it until you leave a loop for the first time
22:07:48 <int-e> oerjan: I know. For brainfuck, you need to be very precise about the rules of stepping over the tape boundaries, I think
22:08:17 <oerjan> (alternatively until you get to an infinite innermost loop)
22:08:50 <oerjan> once you've left a loop, you're always in a state where one cell is _known_ to have a finite set of possible values.
22:09:00 <int-e> because if you can stop off the tape, reading zeros, but not being allowed to modify the cells, then you can encode some finite state information in the pointer.
22:09:10 <HackEgo> [wiki] [[Special:Log/newusers]] create * Giargiano * New user account
22:09:41 <oerjan> int-e: um, 2 cells. you can assuming the tape wraps if you want.
22:10:46 <oerjan> int-e: i think that's rather useless since the fs info has to be lost before you can change a cell again...
22:11:02 <oerjan> also, you cannot loop in an all-zero region
22:11:03 <int-e> that would be precise enough. and in that case I think it's strictly more powerful than a minsky machine with a single counter (for example, you can multiply a given input by 2), but I do believe it's not TC.
22:11:42 -!- mihow has quit (Quit: mihow).
22:12:00 <HackEgo> [wiki] [[Randwork]] http://esolangs.org/w/index.php?diff=43569&oldid=25072 * Giargiano * (+88) +Perl interpreter
22:16:01 <oerjan> oh hm that step-off-the tape thing is subtle, because you can use it to escape loops.
22:16:49 <oerjan> i don't think that's very in the spirit of what Taneb is asking, though.
22:17:12 <HackEgo> [wiki] [[Randwork]] M http://esolangs.org/w/index.php?diff=43570&oldid=43569 * Giargiano * (+22) /* External resources */
22:41:46 -!- newbie|2 has joined.
22:44:43 -!- comodvs has quit (Ping timeout: 265 seconds).
22:48:36 <zzo38> I thought of a new kind of package manager idea, where each package except anonymous packages are identified by URI (used only for identification and does not necessarily ahve to point to anything), categories of packages are themself virtual packages, there is a local root package whose specification is specifying which packages are installed, and that you can also load packages from stdin and emit packages to stdout, so it can act as a filter too.
22:54:22 <zzo38> Relation of packages might be: "X requires Y", "X suggests Y", "X is a version of Y", "X is a substitute for Y". Packages can also have custom properties that any package that substitutes for it or is a version of it must override. Virtual packages are allowed to be named or anonymous, and you can't really install/uninstall it rather it determine automatically if it is considered installed or not by its specification.
22:57:58 <zzo38> In order to avoid entering the entire URI for each package you can define a prefix by the package manager's "add-prefix" command.
22:58:00 -!- mauris has quit (Ping timeout: 255 seconds).
22:58:22 -!- atrapado has quit (Quit: Leaving).
23:03:08 -!- _256Q has quit (Read error: Connection reset by peer).
23:05:10 <zzo38> Other relations can be defined using the existing ones. The virtual package specifications might include: any of X, all of X, none of X. Therefore a "X conflicts with Y" relation might be noted by: X :requires [ :not Y ]
23:06:29 -!- mihow has joined.
23:11:41 <zzo38> Actually the way I mentioned custom property isn't quite best way, I think; a better way is: A package can specify which custom properties must be overridden by packages with the specified relation to this package. For example if the root package has a custom property for a configuration path and it specifies that any package that depends on it must override it, then the root package depends on that package and the root package will specify the conf
23:12:27 <zzo38> If a package is installed only due to requirement from some other package, then the root package does not depend on it and therefore it can know to uninstall when not needed.
23:15:24 <zzo38> Do other package managers use a "root package" like this?
23:16:00 <zzo38> This would also easily allow you to make packages that can be installed for individual users because each user can also have their own root package too.
23:19:58 <zzo38> What is your opinions of all of these things?
23:21:41 -!- mihow has quit (Quit: mihow).
23:23:36 -!- mihow has joined.
23:25:59 -!- mihow has quit (Client Quit).
23:26:26 -!- Herbalist has joined.
23:26:26 -!- Herbalist has quit (Changing host).
23:26:26 -!- Herbalist has joined.
23:27:23 -!- mihow has joined.
23:29:40 -!- mihow has quit (Client Quit).
23:33:51 -!- mihow has joined.
23:37:38 -!- Wright has joined.
23:45:16 <zzo38> Ordering of versions numbers, including possible of breaking compatibility, can be specified using the stuff I have mentioned so far! (Do you see how?)
23:48:26 -!- tromp__ has joined.
23:53:13 -!- tromp__ has quit (Ping timeout: 244 seconds).
23:54:57 <zzo38> But the thing still not mentioned yet is for a virtual package's criteria to check for values of custom-properties, for example you might have a "FamicomVM" virtual package; packages that substitute for it (i.e. FamicomVM implementations, or NES/Famicom emulators) must override custom-properties to indicate mapper and input device implementations, and a game might specify a dependency on a virtual package which requires iNES mapper #0 and keyboard e
23:59:09 <zzo38> And I believe that is all of the stuff you will need, isn't it?