←2015-09-30 2015-10-01 2015-10-02→ ↑2015 ↑all
00:12:47 <\oren\> imback
00:13:52 <\oren\> b_jonas: Hmm, the kanji are showing up fine on my computer at home and the one at work
00:14:40 -!- Frooxius has quit (Quit: *bubbles away*).
00:16:22 <\oren\> god damn caching!
00:17:23 <\oren\> probably the cause of all this crap is that my.css is being cached
00:18:46 <\oren\> I need a way to solve this caching problems for good!
00:26:47 -!- doesthiswork has joined.
00:27:02 <HackEgo> [wiki] [[English]] http://esolangs.org/w/index.php?diff=44564&oldid=44056 * 71.178.9.97 * (+167) /* Compilers */ adding an example of computer programs that can compile English.
00:27:38 <\oren\> hmmm... I'll make a cgi script that can read a filename and some random numbers, and return just the file. Then I can change the random numbers to defeat caching
00:29:20 <doesthiswork> You know how complex numbers are equal if their magnitudes match and their angles match modulo 2pi? If we take out the modulo 2pi part then every number has a single square root
00:30:29 <doesthiswork> what we lose is the ability to add numbers with different angles
00:36:37 <Phantom___Hoover> a very substantial loss
00:37:39 <\oren\> doesthiswork: why do we lose that
00:40:28 -!- GoToTell has joined.
00:42:46 <GoToTell> @tell mroman_ http://mathworld.wolfram.com/UniformSumDistribution.html (7) (8) (9)
00:42:46 <lambdabot> Consider it noted.
00:42:52 <doesthiswork> because multiplying just adds the angles, but I'm not sure how to combine them when I add two numbers, so i just leave them separate
01:05:13 -!- Phantom___Hoover has quit (Read error: Connection reset by peer).
01:18:10 -!- boily has joined.
02:00:30 <boily> Oh dear fungot in the sky. Please drive me away from temptation, namely binging on webcomics before going to sleep.
02:00:30 <fungot> boily: to the room of a dwarven assassin who happens. that includes any " wizard formerly known as " eagle-eyed pete" before bozzok caught, we would like my mode of transit returned, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i n
02:00:44 <boily> right, need to check if there were any oots updates lately...
02:02:46 <shachaf> only one hth
02:03:00 <shachaf> ^style
02:03:00 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots* pa qwantz sms speeches ss wp youtube
02:03:06 <shachaf> whoa whoa whoa
02:03:09 <shachaf> is that new?
02:03:21 <shachaf> fungot: what's with durkon
02:03:21 <fungot> shachaf: we need to get the hell, i rolled a 1 gp fee, we'll get you thinking like a rogue," as " par for my people, i would get the message would actually put us in mortal danger is infinitesimally low.
02:04:31 <boily> hellochaf. there was only one.
02:05:34 <boily> I think I'll succumb to the siren calls of my mattress and be comatose for the next few hours, responsibly like the responsible adult that I am. or something like that to that generic effect.
02:05:40 -!- boily has quit (Quit: PANCREATIC CHICKEN).
02:50:11 -!- danofthedeep has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
03:10:25 -!- GoToTell has quit (Ping timeout: 250 seconds).
03:48:23 -!- Frooxius has joined.
04:01:42 -!- MDude has joined.
04:07:17 <newsham> ^style c64
04:07:18 <fungot> Selected style: c64 (C64 programming material)
04:07:54 <newsham> ^show
04:07:55 <fungot> echo reverb rev rot13 rev2 fib wc ul cho choo pow2 source help hw srmlebac uenlsbcmra scramble unscramble asc ord prefixes tmp test celebrate wiki chr ha rainbow rainbow2 welcome me tell eval elikoski list ping def a thanks tmp2 8ball rreree rerere botsnack bf
04:09:25 <newsham> ^botsnack woot
04:09:25 <fungot> Oh nom nom nom!
04:09:38 <newsham> ^bf +>+>+
04:12:33 <newsham> what does the style get used for?
04:13:18 <newsham> fungot: irq
04:13:18 <fungot> newsham: stack requirements: 2 2 zero page addressing instructions of the programmer's reference guide at your local dealer, is used in the character shape that you need to test more than one machine language
04:13:38 <newsham> fungot: sys
04:13:38 <fungot> newsham: load"",1,1 ( looks for the target line number value written to tape buffer pointer to the end of each harmonic present is proportional to the screen is divided into two nybbles of each sprite has a keyboard buffer
04:33:06 <HackEgo> [wiki] [[Wordy]] M http://esolangs.org/w/index.php?diff=44565&oldid=42292 * Iamevn * (+680) BNFish thing added
04:47:08 -!- hppavilion[1] has quit (Ping timeout: 272 seconds).
05:03:23 -!- mauris has joined.
05:14:11 -!- mauris has quit (Ping timeout: 244 seconds).
05:15:32 -!- doesthiswork has quit (Read error: Connection reset by peer).
05:24:07 -!- hppavilion[1] has joined.
05:28:56 -!- hppavilion[1] has quit (Ping timeout: 272 seconds).
05:37:19 -!- J_Arcane_ has joined.
05:41:03 -!- J_Arcane has quit (Ping timeout: 244 seconds).
05:41:14 -!- J_Arcane_ has changed nick to J_Arcane.
05:46:46 <izabera> followed that link posted by hppavilion[1] and i ended up here http://www.jayconrod.com/posts/35/parsing-keyvalue-pairs-in-bash
05:47:06 <izabera> the rest of the site looks cute, and they have a bash section, and it's pure crap
05:47:17 -!- J_Arcane has quit (Ping timeout: 265 seconds).
05:59:33 -!- MDude has changed nick to MDream.
06:05:33 <Jafet> fungot: elite funge98 demos
06:05:33 <fungot> Jafet: for best results, the basic rom, or modify program lines give each function name.
06:07:44 <b_jonas> \oren\: put in some HTTP headers. there's like three different ones, and clients and proxies look at different sets
06:11:28 -!- hppavilion[1] has joined.
06:22:46 <Sgeo> https://twitter.com/badedgecases
06:23:46 -!- Frooxius has quit (Quit: *bubbles away*).
06:24:37 -!- sc00fy has joined.
06:27:50 -!- hppavilion[1] has quit (Ping timeout: 272 seconds).
06:43:42 <Jafet> Huh, Kiribati and Kiritimati (pronounced "Kiribas" and "Kirismas", naturally) are at UTC+14:00
06:49:52 -!- sc00fy has quit (Ping timeout: 246 seconds).
06:55:11 -!- sc00fy has joined.
07:11:01 -!- oerjan has joined.
07:13:10 -!- ^v has quit (Ping timeout: 246 seconds).
07:13:49 -!- sc00fy has quit (Ping timeout: 268 seconds).
07:14:32 <oerjan> @tell izabera *MWAHAHAHA*
07:14:32 <lambdabot> Consider it noted.
07:15:12 <oerjan> and that was even a guest comic
07:25:37 <oerjan> @tell Gregor <gamemanj> `echo `whoami` is on <-- you know it would be brilliant if that (well, corrected) actually gave the nick
07:25:37 <lambdabot> Consider it noted.
07:26:47 <oerjan> might not always fit lexically
07:28:13 <oerjan> `` ls wisdom/c*
07:28:15 <HackEgo> wisdom/c \ wisdom/c# \ wisdom/c++ \ wisdom/cake \ wisdom/cakeprophet \ wisdom/california \ wisdom/can \ wisdom/canada \ wisdom/canary \ wisdom/caps lock \ wisdom/cat \ wisdom/catamorphism \ wisdom/categorical product \ wisdom/category \ wisdom/category-helpdesk \ wisdom/category theory \ wisdom/cello \ wisdom/certainly \ wisdom/certainty \ wisdom/c
07:28:36 <oerjan> hm forgot what it was
07:28:49 <oerjan> `ls /proc
07:28:49 <HackEgo> 1 \ 10 \ 2 \ 281 \ 285 \ 286 \ 287 \ 288 \ 289 \ 290 \ 291 \ 292 \ 3 \ 4 \ 47 \ 49 \ 5 \ 51 \ 6 \ 68 \ 7 \ 76 \ 77 \ 8 \ 9 \ buddyinfo \ bus \ cgroups \ cmdline \ config.gz \ consoles \ cpuinfo \ crypto \ devices \ diskstats \ driver \ execdomains \ exitcode \ filesystems \ fs \ interrupts \ iomem \ ioports \ irq \ kallsyms \ kcore \ kmsg \ kpageco
07:29:10 <oerjan> `cat /proc/cmdline
07:29:11 <HackEgo> initrd=/usr/bin/../lib/umlbox/umlbox-initrd.gz ubda=/tmp/16088.conf mem=256M con1=null,fd:3 con2=fd:5,fd:8 con=null,null root=98:0
07:29:22 <oerjan> hmph it's not in there
07:29:53 <oerjan> `cat /proc/buddyinfo
07:29:53 <HackEgo> Node 0, zone Normal 1 3 1 2 2 4 4 3 2 2 57
07:30:15 <oerjan> `` ls /proc | tail
07:30:16 <HackEgo> sysemu \ sysrq-trigger \ sysvipc \ timer_list \ tty \ uptime \ version \ vmallocinfo \ vmstat \ zoneinfo
07:30:38 <oerjan> `` ls /proc | tail -30
07:30:38 <HackEgo> ioports \ irq \ kallsyms \ kcore \ kmsg \ kpagecount \ kpageflags \ loadavg \ locks \ meminfo \ misc \ mounts \ net \ pagetypeinfo \ partitions \ self \ slabinfo \ softirqs \ stat \ sys \ sysemu \ sysrq-trigger \ sysvipc \ timer_list \ tty \ uptime \ version \ vmallocinfo \ vmstat \ zoneinfo
07:30:55 <oerjan> `cat /proc/seld
07:30:56 <HackEgo> cat: /proc/seld: No such file or directory
07:30:58 <oerjan> `cat /proc/self
07:30:59 <HackEgo> cat: /proc/self: Is a directory
07:31:15 <oerjan> `ls /proc/self
07:31:16 <HackEgo> auxv \ cgroup \ clear_refs \ cmdline \ comm \ coredump_filter \ cpuset \ cwd \ environ \ exe \ fd \ fdinfo \ limits \ maps \ mem \ mountinfo \ mounts \ mountstats \ net \ ns \ oom_adj \ oom_score \ oom_score_adj \ pagemap \ personality \ root \ smaps \ stat \ statm \ status \ task \ wchan
07:31:56 <oerjan> `cat /proc/self/personality
07:31:57 <HackEgo> 00000000
07:35:27 -!- mauris has joined.
07:35:39 <oerjan> `` which wh
07:35:40 <HackEgo> No output.
07:35:42 <oerjan> `` which who
07:35:43 <HackEgo> ​/usr/bin/who
07:36:05 <oerjan> `` who are you
07:36:06 <HackEgo> No output.
07:41:52 <oerjan> ] x
07:42:01 <oerjan> j-bot: x
07:42:01 <j-bot> oerjan: x
07:42:11 -!- mauris has quit (Ping timeout: 264 seconds).
07:42:12 <oerjan> j-bot: help
07:42:13 <j-bot> oerjan: |value error: help
07:42:28 <oerjan> j-bot: 1+1
07:42:28 <j-bot> oerjan: 2
07:43:16 <oerjan> oh
07:43:21 <oerjan> [ 2^10
07:43:22 <j-bot> oerjan: 1024
07:45:11 -!- hppavilion[1] has joined.
07:50:06 -!- hppavilion[1] has quit (Read error: Connection reset by peer).
07:50:58 * oerjan swats hppavilion[1] in absentia for quitting _just_ as he was finishing his message -----###
07:51:15 <b_jonas> oerjan: ohai
07:51:20 * oerjan swats that "he" for ambiguity -----###
07:51:25 <oerjan> ohayo
07:54:56 -!- mauris has joined.
07:56:51 -!- neo1-9-7-S has joined.
07:57:38 <shachaf> `? swat
07:57:39 <HackEgo> swat? ¯\(°​_o)/¯
07:59:10 -!- mauris has quit (Ping timeout: 246 seconds).
07:59:23 -!- mauris has joined.
08:06:12 -!- mauris_ has joined.
08:08:42 -!- mauris has quit (Ping timeout: 268 seconds).
08:15:06 -!- mauris_ has quit (Ping timeout: 255 seconds).
08:18:39 <oerjan> @tell boily <boily> Oh dear fungot in the sky. Please drive me away from temptation, namely binging on webcomics before going to sleep. <-- most inefficient prayer ever tdnh
08:18:39 <lambdabot> Consider it noted.
08:18:39 <fungot> oerjan: changing banks. once the channel is left out the computer responded with bit 0 for external audio signals, the graphics display is no line x, y is a mistake in your programmable characters in length, and
08:20:56 -!- mauris has joined.
08:22:56 <FireFly> fungot: so y is the debugging channel?
08:22:57 <fungot> FireFly: there is a good description of the first 8 pixels is erased: the test is accomplished for voice 1 at location 53276 ( d01c). it is recommended that a high-pass filter reduces the volume of the basic interpreter
08:25:21 -!- mauris has quit (Ping timeout: 244 seconds).
08:26:53 <izabera> @messages
08:27:04 <izabera> why did i get a mwahahaha
08:27:29 <izabera> does lambdabot accept @messages in pm?
08:27:56 <izabera> looks like it does
08:30:27 <oerjan> i did that in public anyway
08:32:02 <oerjan> and it was in response to your own
08:38:30 -!- mauris has joined.
08:39:54 <FireFly> @massage
08:39:54 <lambdabot> You don't have any messages
08:42:30 -!- mauris has quit (Ping timeout: 240 seconds).
08:59:26 -!- sammwch has joined.
09:01:31 -!- sammwch has quit (Excess Flood).
09:01:39 -!- sammwch has joined.
09:01:46 -!- heddwch has quit (Read error: Connection reset by peer).
09:02:05 -!- sammwch has changed nick to heddwch.
09:04:34 -!- mroman_ has joined.
09:04:54 <mroman_> https://esolangs.org/wiki/Category:Golfing_language still only shows GolfScript but there are other pages in that category as well o_O?
09:08:20 <oerjan> i see 5 languages there
09:08:28 -!- Patashu has joined.
09:08:48 <oerjan> try reloading
09:09:47 <oerjan> mroman_: iirc there's a bug with the category pages so that only the actual text counts for last modified date, or something
09:11:19 -!- FreeFull has quit.
09:21:03 -!- FreeFull has joined.
09:21:26 -!- gniourf has quit (Quit: Leaving).
09:21:59 -!- gniourf has joined.
09:23:24 <HackEgo> [wiki] [[Category:Golfing language]] http://esolangs.org/w/index.php?diff=44566&oldid=44533 * 160.85.232.152 * (+12) * programming languages
09:24:28 <mroman_> yeah now I see 5 too
09:44:06 <mroman_> Do any of you guys have eso programs that produce fibonacci numbers that you'd release into public domain
09:44:13 <mroman_> or were already released into public domain
09:46:20 -!- Phantom_Hoover has joined.
09:47:44 -!- ais523 has joined.
09:54:55 -!- oerjan has quit (Quit: leaving).
09:59:10 -!- neo1-9-7-S has quit (Quit: Leaving).
10:08:27 <fizzie> <fungot> [...] it is recommended that a high-pass filter reduces the volume of the basic interpreter
10:08:27 <fungot> fizzie: 2) test for the considerations involved), and not in the
10:08:31 <fizzie> I liked that one.
10:08:53 <fizzie> fungot: Do you mean the basic interpreter makes a lot of low-frequency noise?
10:08:53 <fungot> fizzie: bits 0-3 listed in the directory entry will read the paddle fire buttons. see entries for the operator to type the following
10:09:03 -!- Frooxius has joined.
10:23:57 <mroman_> "creatonism motion"
10:24:02 <mroman_> what the fridge is that
10:26:09 <mroman_> and scottish.parliament.uk has a nice "Server Error in '/' Application." error message
10:26:21 <mroman_> looks .NET-like
10:26:47 <mroman_> ah fuck those fucking javascript links
10:26:59 <mroman_> you can't fucking open them in a new tab
10:28:44 <ais523> ^style
10:28:44 <fungot> Available: agora alice c64* ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp youtube
10:28:52 <ais523> I'm not used to this fungot style
10:28:53 <fungot> ais523: there are eight sprites, numbered from 0 again, i.e., but doesn't affect the number of times the one's place, then that sprite is 24 by 21 dots high.
10:29:04 <ais523> ^style c64
10:29:04 <fungot> Selected style: c64 (C64 programming material)
10:34:09 <fizzie> One problem with it is that I didn't properly fill paragraphs.
10:34:25 <fizzie> So it's prone to stopping in the middle of a
10:34:58 <fizzie> fungot: What's the address to poke for the border color?
10:34:59 <fungot> fizzie: 70 goto 20 get1, a zero, then that character. if the device-number is left off.
10:35:20 <fizzie> That doesn't seem right.
10:35:42 <b_jonas> ais523: I looked at ayacc, and I like how the new style C api can be used re-entrantly without globals. I'm still sort of missing multiple entry points (top level symbols), but I can work that around.
10:35:53 <b_jonas> I'll definitely try to test it.
10:36:01 <ais523> b_jonas: it has multiple entry points
10:36:08 <ais523> just give more than one %start declaration
10:36:13 <b_jonas> ais523: oh, it can do that?
10:36:15 <b_jonas> great
10:36:16 <ais523> you'll get a warning saying that POSIX disallows it, but it'll work
10:36:23 <b_jonas> that wasn't clear from the documentation
10:36:25 <ais523> (the new-style API has the start symbol name in it for a reason)
10:36:29 <ais523> I'm not sure if I've documented that yet
10:36:41 <b_jonas> ah! so "_start" isn't literal, but it's supposed to mean an underscore followed by the start symbol name?
10:36:48 <b_jonas> in yyparse_start that is
10:36:57 <ais523> right
10:37:07 <b_jonas> also, another thing I wanted to ask is,
10:37:47 <b_jonas> you have the -p option to prefix all the exported symbols, or make it static. is there a corresponding declaration for this that you can put right inside the yacc input file?
10:38:23 <ais523> b_jonas: currently no; -p is specified by POSIX, -p from inside the file isn't
10:38:54 -!- boily has joined.
10:39:02 <ais523> what's your use case for that? running "make" to run yacc, without an explicit makefile?
10:39:14 <ais523> or using aimake without an explicit rule to add -p?
10:39:32 <ais523> normally the assumption is that people who write programs have control over the build instructions
10:39:49 <b_jonas> and is there a way you can tell it to put all the symbols inside (a) a C++ namespace or set of nested namespaces, (b) as extern "C"? both of these three is easy, you just have to emit a line of code before your declarations, but after any standard #include lines, plus some closing braces at the end.
10:40:12 <b_jonas> ais523: no, I'm just wondering, the -p option is fine. it's inside a namespace that I'm really wondering about.
10:40:27 <ais523> b_jonas: there isn't currently, although some day there will be
10:40:48 <b_jonas> it just seems strange how some of yacc is controlled from command line and some from the yacc file, and I don't see the logic on what is where.
10:41:54 <ais523> AFAIU, the split should be: the yacc file controls what the parsing algorithm is and the mathematical description of what is being parsed; the command line controls the API from which the parser is called from outside
10:41:57 <b_jonas> So for the re-entrant api, the cookie pointer is passed to the yylex and the yyerror functions that I supply. But how do I access the cookie from braced grammar rules?
10:43:23 <b_jonas> ais523: that could make sense in theory, but in practice I won't be using just a simple mathematically nice grammar, but a grammar that only ayacc thinks is LALR(1), but actually it interacts with the lexer input and the braced blocks in strange ways.
10:43:44 <ais523> b_jonas: well it still lets you know where the API split is
10:44:06 <b_jonas> And in that case, some of the magic is in external source files, some in the grammar rules, and some is in the C code at the beginning of the yacc file.
10:44:12 <b_jonas> (C or C++ code)
10:44:21 <ais523> also I don't think brace rules have access to the extra void * at the moment, at least not officially
10:44:30 <ais523> it'd be easy enough to give them access, though
10:44:40 <b_jonas> Please give them access then, I think that would be useful.
10:44:50 <b_jonas> It could be worked around, because the lexer can pass that pointer to it in every symbol,
10:44:59 <b_jonas> but I think the parser already holds that pointer so it's pointless.
10:45:12 <ais523> it actually has a name right now, it's just "internal" and isn't meant to be user-visible
10:46:24 <ais523> "yys.yyp.yylexarg" apparently
10:47:15 <b_jonas> I'd like to use the extra argument because I want to start multiple nested yyparse instances.
10:48:17 <ais523> I guess the correct thing to do here is to add a macro to expose the parameter under a more sensible name
10:48:19 <ais523> do you have one in mind?
10:48:32 <b_jonas> no
10:48:45 <b_jonas> well
10:48:51 <b_jonas> it could have "cookie" or "baton" in it
10:49:10 <ais523> these are weird names to use for the spare void * parameter
10:49:10 <b_jonas> pointers like this are usually called one of "cookie", "baton", "token", or "data", but "data" is stupid, and "token" is confusing in a parser generator
10:49:17 <ais523> normally it's called something along the lines of "arg"
10:49:18 <b_jonas> I prefer "cookie"
10:49:37 <b_jonas> because "cookie" means it's uninterpreted by you
10:49:48 <b_jonas> "baton" is used because it's something the library "passes on" to you.
10:50:11 <b_jonas> yeah, "arg" is almost as bad as "data"
10:50:17 <b_jonas> there's also "environment" or similar
10:50:17 <ais523> my understanding of a "cookie" or "magic cookie" is a special meaningless value that you use as a confirmation
10:50:33 <ais523> sort-of like paranoid_quit but for programs
10:50:45 <b_jonas> um, I think that's "magic value"
10:50:51 <b_jonas> I didn't know it was called for cookie
10:51:30 <ais523> sort-of like the way that reboot(2) requires its first two arguments to be 4276215469 and 677274793, in an attempt to reduce the potential impact of undefined behaviour
10:51:56 <ais523> (a few other options are now available for the second argument; they're used to commemorate important events in Linux's history)
10:52:00 <b_jonas> ais523: exactly
10:52:15 <b_jonas> ais523: there's such things used in data structures to check for accidental memory corruption or invalid pointers
10:52:32 <b_jonas> plus there's also "canary" or "sentinel" bytes used for detecting overwriting past the end of a buffer
10:52:38 <b_jonas> "canary" to detect corruption early
10:52:53 <ais523> https://en.wikipedia.org/wiki/Magic_cookie
10:53:01 -!- ais523 has quit.
10:53:14 -!- ais523 has joined.
10:53:28 <ais523> right, a canary is a value whose purpose is to detect being overwritten
10:53:49 <b_jonas> ais523: a HTTP cookie doesn't mean this though, it just contains a value that isn't interpreted by the HTTP client, but is passed back to the HTTP server
10:53:55 <ais523> yes
10:54:10 <ais523> hmm, Wikipedia's definition of a magic cookie is more general than mine and less general than yours
10:54:23 <b_jonas> is any of these apart from "token" used in some meaning related to parsers? if a word is, we should avoid it
10:54:27 <ais523> it's basically a value that's treated like a handle: the value is used to distinguish between different callers
10:54:44 <b_jonas> oh yes, "opaque handle"
10:55:37 <ais523> Windows uses handles a lot
10:57:03 <b_jonas> let me check some libraries for what they use
10:58:22 <ais523> qsort_r uses "arg"
10:58:38 <ais523> (that was the first function with a callback-style void * that came to mind)
10:59:15 <b_jonas> it seems lua doesn't have a C cookie associated with C functions (that is, lua functions implemented in C), but it lets the function have lua upvalues, and some of those lua upvalues can be "light userdata", that is, C pointers. however, the memory allocator function does have a cookie argument, and it's called variously "ud" and "ptr".
10:59:35 <b_jonas> I'll have to check sqlite3, it has some cookies
11:01:05 <int-e> `? cookie
11:01:06 <HackEgo> cookie? ¯\(°​_o)/¯
11:01:25 <b_jonas> yum
11:01:33 <boily> ^botsnack
11:01:34 <fungot> Oh nom nom nom!
11:01:42 <boily> @massages-loud
11:01:42 <lambdabot> oerjan said 2h 43m 3s ago: <boily> Oh dear fungot in the sky. Please drive me away from temptation, namely binging on webcomics before going to sleep. <-- most inefficient prayer ever tdnh
11:02:18 <int-e> ithm effective
11:02:30 <boily> @tell oerjan fungot answers me. fungot is great. fungot fungot fungot ♪
11:02:30 <lambdabot> Consider it noted.
11:02:30 <fungot> boily: os pointer to point to the 6510 at addresses 53248-57343 ( d000-dfff). specialized control circuitry minimizes software overhead, facilitating use in the sprite shape data in files, so that break points may be an input/ output operations. the
11:03:31 <int-e> fungot: stop almost making sense already
11:03:32 <fungot> int-e: examples: 100*2 50*0 a*x1 r%*14 print 43648 program text. once read, and
11:04:18 <b_jonas> http://vignette2.wikia.nocookie.net/mlp/images/8/86/Cookie_Crumbles_ID_S4E23.png/revision/latest?cb=20140614061241 cookies
11:04:31 <b_jonas> sorry, that's not relevant to the discussion
11:04:45 <int-e> somewhat disturbingly, it got the decimal/hex conversion correct (probably unique tokens, but still).
11:05:45 <int-e> . o O ( Cookie grammar: Cookies go well *with* tea, not *in* tea. )
11:05:56 <boily> b_jhellonas. it's a poney cookie. or a cookie poney.
11:06:10 <b_jonas> fungot, what's 380°F in celsius scale?
11:06:10 <fungot> b_jonas: 1) can be held to 0 of the commodore 64 program cartridge. you must first choose a location as a single value which the
11:06:29 <int-e> your poney sounds phoney
11:06:53 <boily> the fungot hasn't loaded the proper cartridge to convert temperatures.
11:06:53 <fungot> boily: last key pressed. if future changes are made available by altering the contents of the 1k chunk of memory
11:06:58 <int-e> @google 380°F in °C
11:07:00 <lambdabot> http://checkpriceonline.tk/termometro-infrarossi-puntatore-laser-50-380-digitale-modello-1-koka69946/
11:07:00 <lambdabot> Title: Buy Cheap Best TERMOMETRO INFRAROSSI PUNTATORE LASER -50°/380°DIGITALE ...
11:07:08 <int-e> @google 380F in C
11:07:09 <lambdabot> 193.333333 degrees Celsius
11:07:09 <lambdabot> https://www.evi.com/q/380_fahrenheit_to_celsius
11:07:16 <boily> oh the unicodity...
11:07:45 <int-e> the first hit is disturbing
11:08:09 <int-e> @google 451F in C
11:08:11 <lambdabot> 232.777778 degrees Celsius
11:08:11 <lambdabot> http://www.slate.com/articles/health_and_science/explainer/2012/06/ray_bradbury_death_does_paper_really_burn_at_451_degrees_fahrenheit_.html
11:08:11 <lambdabot> Title: Ray Bradbury Death: Does paper really burn at 451 degrees Fahrenheit?
11:08:14 <b_jonas> sqlite3 calls it at least user_data (as in sqlite3_user_data to query it) and pApp (but it uses crazy prefixed names like that for most non-public api symbols), pArg, and sometimes simply unnamed because it appears only as a function parameter
11:09:05 <b_jonas> libev specifically calls it "data"
11:09:28 <b_jonas> sqlite3 also calls it pAppData in a public (but not mandatory) api
11:12:18 <b_jonas> and also pClientData
11:12:26 <b_jonas> ok, so in sqlite3 they just use random names for this
11:13:00 <b_jonas> and even random interfaces: some functions (deallocators of structures you allocate yourself) don't get a cookie at all, some get a cookie pointer passed to it, and some have to call a function to get their cookie pointer.
11:13:04 <b_jonas> crazy
11:13:21 -!- nooga has joined.
11:13:40 <b_jonas> if "cookie" is ambiguous, how about "baton"? does that have other meanings?
11:15:24 <nooga> I just had this idea for a language, which compiler would assume that every programmers statement is a lie, and there would be a keyword "srsly" which would mark statement as not-lie
11:15:34 <nooga> but the compiler wouldn't trust the keyword
11:15:46 <nooga> so it would have to solve liars paradox in compile time
11:15:53 <nooga> and would ask questions to the programmer
11:24:27 -!- AnotherTest has joined.
11:25:12 -!- bender| has joined.
11:27:57 -!- ais523 has quit (Quit: lunch).
11:33:02 -!- boily has quit (Quit: FLUVIAL CHICKEN).
11:35:02 -!- sc00fy has joined.
11:38:30 -!- AnotherTest has quit (Ping timeout: 260 seconds).
11:48:06 -!- sc00fy has quit (Ping timeout: 240 seconds).
12:03:52 -!- J_Arcane has joined.
12:12:23 <mroman_> anybody experience with mod_rewrite
12:12:32 <mroman_> it dosen't accept my (.*)$ rule apparantely
12:14:11 <mroman_> but you can match ^api/(.*)$ hm...
12:15:29 -!- ais523 has joined.
12:19:55 <b_jonas> ais523: ok, I'd say yybaton then, unless baton also has some other meaning
12:19:59 -!- ais523 has quit (Client Quit).
12:20:16 -!- ais523 has joined.
12:20:27 <ais523> b_jonas: the problem is that "baton" isn't commonly used for this sort of argument
12:20:34 <ais523> I'd probably just call it "arg", maybe with a qualifier
12:20:40 <ais523> because that seems to be what everyone else calls it
12:21:44 <b_jonas> ais523: I think it is used somewhere, let me check
12:27:10 <b_jonas> ais523: I think libsvn (subversion C api) uses "baton", but I don't know if that appears in the api names, or only the docs or function argument names
12:27:27 <ais523> maybe even call it yyvoidstar
12:27:33 <b_jonas> ais523: uh no
12:27:34 <ais523> because "the void * argument" is the standard name for it in conversation
12:27:41 <b_jonas> ais523: you can call it yyuserdata but I don't much like that
12:27:56 <ais523> hmm, I like that better than "baton" at least
12:28:14 <b_jonas> or yycontext, yyupvalue, yycontinuation, yyenvironment
12:29:03 <b_jonas> "userdata" is used, but not quite in this context, but in C structures that you may want to extend with your own data, but they're mallocated by the library so you can't just inherit from them
12:29:24 <b_jonas> used this way by the gnome people that is
12:30:02 <b_jonas> gnome as in the group of libraries, not as a race
12:30:17 <ais523> "continuation" is completely wrong, a continuation is basically the functional version of a goto
12:30:39 <ais523> b_jonas: I don't think anyone would have interpreted that as being about gnomes the mythological creatures
12:30:49 -!- AnotherTest has joined.
12:30:53 * ais523 wonders if GNOME is at all related to ELF and DWARF, or if it's just a coincidence
12:31:12 <b_jonas> I think DWARF is related to ELF
12:31:21 <b_jonas> I don't know about GNOME, but I think it's related to GNU
12:31:46 <b_jonas> wait wait
12:32:02 <b_jonas> isn't gnome very old, as in, it might predate the use of ELF on linux?
12:32:08 <b_jonas> how old is gnome?
12:32:23 <ais523> Linux used something before ELF?
12:32:27 <ais523> oh right, a.out
12:32:34 <b_jonas> ais523: yes, but that was way before I started with linux
12:33:02 <b_jonas> when I started with linux, it already used only ELF, and the a.out stuff is for compatibility in the kernel and binutils only. nobody even has a.out libc so you can't actually run a.out programs.
12:33:03 <ais523> you see occasional references to the a.out format in manpages, that's how I know about it
12:33:15 <ais523> (it is quite confusing that it has the same name as the default output file from cc)
12:34:03 <b_jonas> basically the linux libc started in a.out, then it jumped to ELF, at which point it got renamed to gnu libc AND glibc AND libc 2.*, and they used it to break binary compatibility, which is generally a useful thing to do sometimes with libc to get rid of badly designed crap, and should be done more often
12:41:50 -!- Frooxius has quit (Quit: *bubbles away*).
12:45:17 -!- ais523 has quit.
12:45:43 -!- ais523 has joined.
12:48:58 -!- doesthiswork has joined.
12:55:51 <fizzie> I remember Slackware doing the libc5 to glibc switch, and it was somewhat messy.
12:56:49 <mroman_> http://mroman.ch/testmoo/api/H/default meep :)
12:57:16 <fizzie> As for GNOME/ELF, I don't know if they're related or whether GNOME predates widespread ELF-on-Linux, but it certainly doesn't predate ELF itself.
12:58:07 <b_jonas> fizzie: I see
12:58:16 <fizzie> Also I didn't know it no longer stands for "GNU Network Object Model Environment" at all, like it used to.
12:58:42 <b_jonas> I'm young enough that I don't remember that, I'm quite sure that the first linux I installed already used elf, by far.
12:59:01 <b_jonas> fizzie: GCC has been reacronymmed too
12:59:16 <b_jonas> it's now "GNU compiler collection" instead of "GNU C compiler"
12:59:37 <fizzie> That I did know about. Makes sense.
12:59:51 <fizzie> As far as I can tell, GNOME no longer stands for anything.
12:59:52 <ais523> because it compiles other languages too
13:00:14 <b_jonas> ais523: yes, and it's even written in C++ now
13:00:23 <b_jonas> or a mix of C and C++
13:00:35 <b_jonas> but it was renamed before
13:00:46 <b_jonas> it's been compiling C++ and fortran and objective C and ada for quite a long time
13:01:13 <b_jonas> though I always installed it without most of the extra languages, enabling only C, C++, and sometimes fortran
13:01:27 <b_jonas> I don't need to compile the rest
13:05:43 <FreeFull> Linux is still capable of running a.out programs with an extra module loaded
13:06:57 <ais523> and, presumably, libc…3? installed
13:07:02 <b_jonas> FreeFull: yes, the kernel is, but you need an a.out libc and possibly other a.out libraries
13:07:05 <ais523> I know we're on libc6 atm
13:07:05 <b_jonas> which you don't usually have
13:07:52 <FreeFull> Yeah, you'd need those
13:10:47 <mroman_> ``curl -i -v -X POST -d 'code=iooo' http://mroman.ch/testmoo/api/deadfish/default
13:10:48 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `curl: not found
13:10:52 <mroman_> pff
13:11:02 <mroman_> `run curl -i -v -X POST -d 'code=iooo' http://mroman.ch/testmoo/api/deadfish/default
13:11:03 <HackEgo> ​* About to connect() to proxy 127.0.0.1 port 3128 (#0) \ * Trying 127.0.0.1... \ Failed to connect to socket 2. \ % Total % Received % Xferd Average Speed Time Time Time Current \ Dload Upload Total Spent Left Speed \
13:11:31 <mroman_> ``history | grep wget
13:11:32 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `history: not found
13:12:02 <mroman_> `run curl -X POST -d 'code=iooo' http://mroman.ch/testmoo/api/deadfish/default
13:12:03 <HackEgo> Failed to connect to socket 2. \ \ curl: (52) Empty reply from server
13:12:11 <mroman_> :(
13:14:31 -!- Patashu has quit (Ping timeout: 246 seconds).
13:14:35 <b_jonas> ais523: I think http://www.mezzacotta.net/magic/goldfish/ is somewhat similar to those strange small M:tG formats we were discussing.
13:14:52 <ais523> that link goes to mezzacotta, and so is probably worth clicking on
13:15:25 <ais523> b_jonas: this reminds me, there's a thread on Wizard's forums somewhere about the largest finite combo in magic
13:15:30 <ais523> and those forums are going to shut down soon
13:15:39 <b_jonas> ais523: yes, it's linked from the IWC news snippet right now
13:16:06 <ais523> it uses Knuth up-arrow notation
13:16:10 <b_jonas> ais523: yes, the forum will be shut down. that's a bit sad. apparently they're going to delete the whole thing, rather than make it unwritable except for admins.
13:16:22 <ais523> http://community.wizards.com/content/forum-topic/2820281
13:16:47 <b_jonas> ais523: that's probably big, yes
13:17:47 <mroman_> wget doesn't work either on hackego
13:18:12 <ais523> wait, no
13:18:19 <ais523> that thread has smaller numbers than I remember
13:18:25 <b_jonas> though I think you'll need a very specific definition for "infinite combo" for this, and that definition would be difficult because of underspecified rules and computability bounds
13:18:56 <b_jonas> ais523: I'm not saying because of the threads, it's just that I think a large finite combo in M:tG can probably be very large.
13:19:20 <b_jonas> it's just that you rarely need to set up such a thing, because "small" finite combos or infinite combos are usually sufficient
13:20:41 <ais523> aha: this is what really sparked it off: http://www.soniccenter.org/sm/mtg/megacombo.html
13:20:50 <ais523> and this is the resulting forum thread: http://community.wizards.com/forum/cards-and-combos/threads/2230096
13:21:43 <b_jonas> I mean, attacking with 1000 4/5 green elf warrior tokens is enough, I don't generally need more. Or spending exactly 100 mana for Helix Pinnacle, and then surviving till next upkeep with like a dozen expensive defensive spells.
13:22:09 <b_jonas> And infinite combos are generally easier to set up than such big finite combos.
13:22:31 <ais523> oh good, they migrated to mtgsalvation: http://www.mtgsalvation.com/forums/magic-fundamentals/magic-general/615089-most-turn-1-damage-in-a-deck-with-no-infinite
13:22:41 <ais523> (mostly relevant because that's a forum that /isn't/ going to go down in a few weeks)
13:23:41 <b_jonas> ais523: hmm, can you use a combo for that that creates infinite mana, if you're very careful that the cards can't convert it to infinite damage to the opponent?
13:24:22 <HackEgo> [wiki] [[Special:Log/newusers]] create * DewiIoan * New user account
13:24:28 <ais523> b_jonas: you probably /could/ but it'd be less efficient because it reduces the number of things that you can use to add on exponents
13:24:37 <ais523> I do not have a good feeling about that username
13:24:41 <ais523> but let's see what it does first
13:24:51 <ais523> (if anything)
13:25:23 <b_jonas> ais523: no, I mean, you might need lots of mana (at least lots of colorless mana, or mana of some color(s)), and it can be easier to get infinite mana than to get a thousand mana sometimes;
13:25:36 <b_jonas> though with some combos, you can regain more and more mana as you continue, so you might not need it.
13:25:52 -!- ais523 has quit.
13:26:06 <HackEgo> [wiki] [[Talk:True]] http://esolangs.org/w/index.php?diff=44567&oldid=30804 * DewiIoan * (+213)
13:28:52 -!- ais523 has joined.
13:28:58 <b_jonas> ais523: hmm wait
13:29:06 <b_jonas> I don't see how this goldfish draft can work:
13:29:15 <b_jonas> it points out that "Note also that many times your deck will be unimaginably large, containing many powers of ten worth of basic lands to support engines that draw ridiculous amounts of cards."
13:29:41 <b_jonas> can't you sometimes set up a simple combo that deals one damage for each basic land in your deck, minus a few?
13:30:20 <b_jonas> oh, maybe that would count as an infinite combo
13:31:02 <ais523> b_jonas: I think it counts as an infinite combo, yes
13:31:11 <ais523> "could have performed", you could have put more lands in
13:33:54 <b_jonas> yes, but that's an action you perform during deck building, not during the game. you also could have drafted other cards.
13:34:29 <b_jonas> So I'm not sure it's in scope for that rule.
13:40:19 <ais523> b_jonas: hmm, many of these maximum non-infinite combos use different /colours/ of mana to give themselves higher numbers
13:40:25 <ais523> e.g. can convert red mana to green hyperexponentially
13:40:46 <b_jonas> ais523: nice
13:43:10 <b_jonas> if I played my elf deck against a goldfish with the deck ordered, then it could go singly exponential, though I'd probably have to modify it if I wanted it to be more effective in this kind of thing
13:44:41 <b_jonas> hmm, ok, maybe it couldn't go exponential
13:44:46 <b_jonas> I'd have to modify it a bit for that
13:46:50 <ais523> hmm, one of the largest three-card combos available here is apparently casting twinflame on as many targets as possible, then redirecting all the targets onto doubling season
13:47:05 <ais523> that's… scary
13:47:25 <b_jonas> yes it is
13:47:56 <b_jonas> why do you need to redirect? isn't there something that copies the spell for each available target?
13:48:05 <b_jonas> oh, maybe that's what you mean
13:59:56 <b_jonas> hmm...
14:01:04 <b_jonas> oh that would be infinite
14:04:09 <b_jonas> and that other combo is only exponential
14:04:18 <b_jonas> hmm wait
14:04:27 <b_jonas> I should try to make something faster growing than just exponential
14:04:40 <b_jonas> oh, it would be double exponential
14:04:43 <b_jonas> but not better than that
14:08:26 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
14:12:14 -!- nooga_ has joined.
14:13:22 -!- Phantom_Hoover has joined.
14:14:10 -!- nooga has quit (Ping timeout: 240 seconds).
14:14:52 <ais523> b_jonas: haha, I've now reached the stage of the chronology of this thing where they'd run out of colours and were using /snow/ mana
14:15:42 <ais523> and they're now using right-arrow notation rather than up-arrow notation
14:16:20 <mroman_> what's \A in regexes?
14:16:34 -!- ais523 has quit.
14:16:39 <b_jonas> ais523: is it possible to use copies of Rings of Brighthearth rather than copies of Doubling Season?
14:16:45 -!- ais523 has joined.
14:16:47 <b_jonas> mroman_: anchor to beginning of input string
14:16:48 <b_jonas> ais523: is it possible to use copies of Rings of Brighthearth rather than copies of Doubling Season?
14:16:59 <ais523> mroman: beginning of the string, even if ^ is redefined
14:17:11 <ais523> see "man perlrebackslash
14:17:12 <mroman_> you can redefine ^ o_O
14:17:13 <ais523> "
14:17:17 <ais523> mroman_: to match newlines and the like
14:17:40 <mroman_> ah. right.
14:17:41 <mroman_> thx
14:23:34 <b_jonas> ais523: and yes, up arrow notation can describe "small" combos, eg. you'd use N up arrows to describe a combo using N+O(1) resources such that each of them can be be spent on doubling your amount of the next resouce; for things like doubling the same resource with itself such as with Doubling Season, you need right arrow notation or some other Ackerman function.
14:23:44 <b_jonas> hmm no wait
14:23:46 <b_jonas> I'm getting that wrong
14:23:55 <b_jonas> up arrow notation is enough for such a big combo
14:24:01 <b_jonas> argh these big numbers always confuse me
14:25:57 -!- danofthedeep has joined.
14:29:12 -!- TieSoul has joined.
14:30:35 -!- doesthiswork has quit (Quit: Leaving.).
14:38:19 -!- `^_^v has joined.
14:45:29 <Vorpal> Does anyone know if there is a way in gdb to set the ulimits for the process you debug? I can't set it before starting gdb, because then gdb fails
14:45:54 <Vorpal> I guess I could patch in a call to setrlimit() or whatever in the debugged program, but surely there has to be a better way?
15:04:58 -!- nooga_ has quit (Quit: Lost terminal).
15:10:08 <Taneb> I REALLY need to become less absent-minded
15:10:14 <Taneb> Very nearly lost my new laptop
15:10:20 <Taneb> Because I left it at a bus stop
15:10:30 <Taneb> On the way to collect my glasses, which I had left in a cafe yesterday
15:10:53 <coppro> lol
15:11:05 <coppro> Vorpal: print setrlimit(...)
15:15:27 <Taneb> Does anyone have any advice for being less absent-minded
15:15:43 <ais523> Taneb: I suggest minimizing the consequences of being absent-minded, instead
15:15:53 <ais523> for example you could physically tie the laptop bag to your body (e.g. with a piece of string)
15:16:06 <Taneb> That could work
15:16:14 <Taneb> Good idea
15:16:28 <Taneb> Now I just need to remember where I put my string!
15:17:56 <mroman_> I regularly forget to turn my water heater on
15:18:01 <mroman_> and then I try to make tea with cold water
15:19:20 <fizzie> While 'print' probably works, gdb also has a 'call' command that I think is more semantically appropriate.
15:20:02 <ais523> I thought "print" was standard for this sort of thing
15:21:48 <fizzie> I've always used "call".
15:22:17 <fizzie> It also prints the result, so there's probably not that much difference.
15:22:27 <fizzie> Won't take all those print options, though.
15:26:01 <izabera> oh it's october 1st
15:26:07 <izabera> this means it's chick flick week
15:26:20 <Vorpal> coppro, ooh, good idea
15:27:14 <izabera> let's watch devil wears prada together
15:40:39 -!- mroman_ has quit (Quit: Lost terminal).
15:41:37 -!- danofthedeep has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
15:42:11 <ais523> "(This notation can also be extended to expressions that use 3 or more right arrows, but those are beyond the scope of this page, and I doubt the system of Magic will ever be expressive enough to include the types of constructs that would call for their use, outside of cases that truly do go infinite.)"
15:42:14 <ais523> oh, if only you knew
15:43:44 <b_jonas> ais523: hehehe
15:44:20 <ais523> the current research is trying to maximize the number of right arrows that can be used
15:44:49 <b_jonas> ais523: well, we can probably simulate any computable algorithm, and make it output damage, though it may require more than 60 cards, more than one player, or more than a few turns to set up
15:46:11 <ais523> b_jonas: assuming the bug in the stackflow construction is fixed, you can, and you can even set it up on turn 1 of a 2p game; the problem is that you could also set up an infinite loop the same way
15:46:26 <ais523> sure, you can write a busy beaver program, but nothing's preventing you writing an infinite loop instead
15:46:32 <b_jonas> ais523: would that work against a goldfish opponent?
15:47:01 <b_jonas> yes, I guess you probably don't need more than one turn
15:47:02 <ais523> it's set up so that it also works even against a non-goldfish opponent (assuming you go first and they don't have force of will or equivalent)
15:47:19 <ais523> via the standard turn 1 combo techniques (black lotus into show and tell into omniscience into enter the infinite, etc.)
15:47:21 <b_jonas> ais523: sure, but I mean, it doesn't require specific cards in the opponent's deck or something
15:47:27 <ais523> right
15:47:41 <ais523> it needs the opponent to control certain cards, but there's always Donate
15:48:01 <b_jonas> if they only have to control permanents or tokens, yes
15:49:38 <ais523> <spambot> This week offer: Only 17.99 usd to rank your site in top 999.999 Alexa
15:49:44 <ais523> is that a decimal point or a comma?
15:49:51 <ais523> if a decimal point, I don't believe it
15:50:36 <ais523> (also clearly such spambots work, if they work at all, by manipulating alexa toolbar submissions rather than actually driving viewers to your site)
15:51:50 <b_jonas> wants to maximize the number of right arrows? that sounds scary
15:52:18 <b_jonas> wanting to maximize the number of up arrows, I can see that
15:52:40 <ais523> b_jonas: it's much harder to get a right arrow
15:52:46 <b_jonas> yes, definitely
15:53:01 <ais523> they mostly involve 3- or 4-card combos that trigger off a particular basic landcycling effect
15:53:45 <b_jonas> getting three right arrows already sounds impossible (unless the numbers are trivially small so they simplify to two arrows with reasonably non-big numbers)
15:54:04 <b_jonas> I'm not even sure how this right arrow notation works
15:54:19 <b_jonas> though I at least know where to find descriptions of these strange notations
15:55:05 <b_jonas> oh, that reminds me
15:56:34 -!- `^_^v has quit (Ping timeout: 260 seconds).
15:56:51 <b_jonas> there's a nice answer on MathOverflow that's impressed me. someone asked a slightly underspecified but reasonable question, so I answered one interpretation, because you can obviously modify that answer for other reasonable similar interpretations. but then someone gave an algebraist answer, generalizing the problem a lot with higher-order functions and explicitly listed conditions so that his proof covers probably all reasonable interpretations.
15:57:19 <b_jonas> the underspecified part was "Ackermann function", of which there's multiple definitions
15:57:31 <b_jonas> so he proved it for all "Ackermann-like functions" generally
15:58:38 <ais523> b_jonas: here's the basis behind the Ackermann combo (the particular deck provided is potentially broken, but the fundamental behind the combo is sound): http://www.mtgsalvation.com/forums/magic-fundamentals/magic-general/615089-most-turn-1-damage-in-a-deck-with-no-infinite?page=2#c41
16:00:21 <ais523> the current record is based on trying to find a separate Ackermann combo for each colour of mana (including snow)
16:00:40 <b_jonas> nice
16:01:22 <b_jonas> can you use mana with strange restrictions, especially from evolved Sliver Hive?
16:01:33 <ais523> well they're also running out of deck space as it is
16:01:45 <b_jonas> yes, I guess that would be a problem
16:01:47 -!- AnotherTest has quit (Ping timeout: 256 seconds).
16:02:05 <b_jonas> but I'm asking about Sliver Hive because maybe you can Artificially Evolve it to multiple creature types
16:02:18 <ais523> the ackermann combos are a lot less efficient than the merely exponential combos, in terms of deck space
16:02:23 <b_jonas> though it might be hard to make it non-infinite if you can copy stuff that much
16:02:24 <ais523> (although who cares, they have a /way/ higher complexity)
16:02:39 <b_jonas> well sure
16:02:41 <ais523> also Artifical Evolution is much too easy to create loops with
16:03:07 <Phantom_Hoover> someone linked these on another channel http://parsleysagerosemarytimemachine.tumblr.com/post/130181748287/so-i-might-have-stayed-up-late-tonight-creating
16:03:17 <Phantom_Hoover> but i have no understanding of mtg
16:04:10 -!- oerjan has joined.
16:04:18 <ais523> that's just a text post
16:04:41 <b_jonas> ais523: yes, and part of the problem is that you may need a lot of mana to start some of these loops, but you probably don't want infinite mana, or at least not of all colors
16:04:44 <Phantom_Hoover> no it's not? the card images are loading for me at least
16:04:49 <ais523> and really, M:tG is a cross between a card game and a programming language, only the card game is incredibly expensive (because it's only provided as randomized subsets)
16:05:12 <ais523> b_jonas: I believe the standard is to allow unlimited colorless mana, and restrict the usage of each individual colour
16:05:29 <ais523> (the colorless isn't technically "infinite", but rather generated in quantities greater than you need during the combo using Mana Echoes)
16:05:42 -!- J_Arcane has quit (Ping timeout: 244 seconds).
16:05:54 <b_jonas> ais523: sure, but then you may still need like a hundred colored mana just to play fourty cards to set up multiple combos
16:06:08 <b_jonas> ok, probably not a hundred
16:06:47 <ais523> b_jonas: omniscience
16:06:59 <b_jonas> oh! good idea
16:06:59 <ais523> it works out really neatly, because it lets you play any spells you like, but doesn't pay for abilities
16:07:21 <b_jonas> of course you have to be careful not to allow playing spells infinitely then
16:07:22 <b_jonas> but still
16:08:08 <b_jonas> ais523: what's the rules for this again? vintage legal deck of 60 cards, against goldfish, and any number of turns?
16:08:40 <b_jonas> oh, and rules about the opponent not losing from the life loss probably.
16:08:54 <b_jonas> do you have to deal the damage to the opponent?
16:08:54 <ais523> b_jonas: 60 cards, no sideboard (or other cards outside the game), no bans, 4-of limit, do as much damage as you can turn 1 against a goldfish, the goldfish /does/ lose from life loss so you want to do it all as one chunk
16:09:06 <b_jonas> turn 1, ok
16:09:11 <b_jonas> ah
16:09:12 <b_jonas> I see
16:09:14 <ais523> also the goldfish's deck consists of 60 basic lands (this is actually relevant sometimes)
16:09:26 <b_jonas> sure, if you mill him
16:09:42 <Phantom_Hoover> so has mtg been proven TC or are you working on that
16:09:51 -!- bender| has quit (Quit: Ping Pong Fuckout).
16:09:55 <b_jonas> well, I guess the losing from life loss isn't that much of a limitation
16:09:58 <ais523> Phantom_Hoover: it was proven TC ages ago
16:10:06 <ais523> the new problem has been trying to prove it TC with no choices made by the players
16:10:17 <b_jonas> because you can make him unable to lose, temporarily, with that demon thing
16:10:25 <ais523> (i.e. there is no way to stop the game calculating arbitrary Turing-complete calculations no matter what the players do, other than conceding)
16:10:32 <b_jonas> with Abyssal Persecutor
16:10:41 <b_jonas> and then killing that
16:10:49 <ais523> b_jonas: that'd work too but it costs a card
16:11:06 <b_jonas> or two cards, yes
16:11:09 <Taneb> ais523: would a helplessly TC MtG imply a situation when the game is proven to have ended but the winner is undecidable?
16:11:10 <b_jonas> depending on what you pack
16:11:29 <b_jonas> ais523: though it can also help if you cause the opponent to draw repeatedly, and use that to get a number 60
16:11:34 <ais523> Taneb: yes
16:11:56 <ais523> or, hmm
16:11:59 <ais523> unknown, certainly
16:12:04 <ais523> is it possible to make it formally undecidable?
16:12:27 <ais523> I guess not, you can't simulate the rules of M:tG in M:tG itself because of the rule that says infinite loop = draw requires a halting oracle
16:12:34 <b_jonas> ais523: I think it can be undecidable between you winning and a draw
16:12:55 <ais523> unless there's some way to use Shaharazad in order to make the result of a subgame observable and construct a halting oracle that way
16:12:56 <ais523> ooh
16:12:57 <b_jonas> ais523: you don't have to simulate the rules of M:tG, you only have to simulate some known undecidable problem
16:13:24 <ais523> b_jonas: there are no known undecidable problems that don't rely on the details of the axioms used to decide them, though
16:13:53 <b_jonas> ais523: and I think because of the infinite loop rule, if you make it so that you're allowed to make the choice of where to stop, or allowed to choose the amount of fuel before the start, you can make it so that it's definitely not a draw, but undecidable between you winning and the other player winning.
16:14:11 <b_jonas> ais523: yes, but the axioms are the ZFC axioms, not related to M:tG
16:14:22 <b_jonas> or even some axioms weaker than ZFC
16:14:28 <b_jonas> oh yes, it has to be ZFC
16:14:54 <b_jonas> so you simulate a program that searches for a ZFC proof for a statement that says "I do not have a proof in ZFC"
16:15:12 <ais523> b_jonas: yes, but there is no such statement
16:15:13 <b_jonas> then this program won't halt, but you can't prove that in ZFC
16:15:18 <b_jonas> why wouldn't there be?
16:15:27 <b_jonas> sure there is
16:15:31 <b_jonas> you can construct one with quining
16:15:38 <ais523> err, I meant no such proof
16:15:45 <ais523> ZFC is believed to be consistent, right?
16:15:53 <b_jonas> yes, we have to assume it's consistent
16:15:59 <Taneb> Believed to be consistent does not mean that it is consistent
16:16:01 <ais523> if such a proof existed, then it would prove a false statement
16:16:08 <ais523> well, let me put it this way
16:16:33 <b_jonas> ais523: yes, but you can't prove in ZFC that it doesn't have a proof, which means ZFC can't prove that the loop will never terminate,
16:16:38 <ais523> if we construct an M:tG game that bruteforces ZFC, looking for a proof of "this statement has no proof"
16:16:40 <b_jonas> so in the end you get a match for which you can't prove who wins
16:16:49 <ais523> it won't find one
16:16:57 <ais523> it's impossilbe to prove /in ZFC/ that it won't find one
16:17:03 <ais523> however, Magic judges are not bound by the rules of ZFC
16:17:07 -!- AnotherTest has joined.
16:17:24 <ais523> and thus, being human, they can do the reasoning we've done above and determine that no proof will be found, thus the program is a draw
16:17:52 <b_jonas> sure, the Magic judges are bound by the tournament rules, and they'll kick you out of the tournament when you're less than one percent into explaining all the details of how your complicated combo works, or before that if your deck is so large you can't shuffle it.
16:18:17 <b_jonas> you probably can't even set up all your combo in a tournament, because you'll be punished for stalling the time unnecessarily, when you could just win the game instead.
16:18:29 <ais523> b_jonas: anyway, I'm thinking about it like this
16:18:45 <shachaf> ais523: what if someone encodes the statement "Magic: the Gathering judges cannot consistently affirm this statement"?
16:18:47 <ais523> there is currently a halting oracle in the rules that says that nonterminating game with no ability to disrupt it = draw, right?
16:18:48 <shachaf> checkmate
16:18:57 <ais523> shachaf: that needs a model of human behaviour
16:19:03 <ais523> however, I think you can get a contradiction here
16:19:05 <ais523> using subgames
16:19:16 <ais523> subgames allow a win/draw/loss to be observable
16:19:32 <ais523> we already know that if we reach a situation that a) does not allow the stack to clear, and b) does not give anyone any choices until the stack clears
16:19:36 <b_jonas> shachaf: no, chess is way more finite on a finite board. infinite board has some interesting research though, involving large countable cardinal numbers and such.
16:19:39 <ais523> then the game is necessarily an instant win, loss, or draw
16:20:00 <Vorpal> ais523, ooh, is MtG TC?
16:20:03 <ais523> with a subgame, we may be able to determine which (it'll take some care to figure out how much life players have but that shouldn't be impossible)
16:20:05 <ais523> Vorpal: …yes
16:20:08 <Vorpal> neat
16:20:11 <ais523> actually we think it might be higher
16:20:11 <b_jonas> ais523: there's no way you can force the subgame to behave exactly the way you want with a setup before it
16:20:16 <Vorpal> ais523, higher?
16:20:18 <Vorpal> How
16:20:18 <b_jonas> not without allowing decisions
16:20:25 <ais523> b_jonas: oh bleh
16:20:30 <b_jonas> sorry
16:20:34 <ais523> I'm not 100% convinced, although it might require more than a 60-card deck
16:20:50 <b_jonas> ais523: higher than 60 card isn't much of a problem
16:20:50 <ais523> ooh, what about using Lich's Mirror instead? although that doesn't trigger on draws
16:20:57 <Vorpal> ais523, how can it be higher than TC?
16:21:03 <ais523> Vorpal: there's a halting oracle
16:21:12 <b_jonas> a very few actual reasonable to play builds require more than 60 cards
16:21:16 <Vorpal> well... good luck with that
16:21:18 <ais523> we're trying to figure out a way to see if that halting oracle is accessible from within the game itself
16:21:25 <b_jonas> so more than 60 card isn't a problem for a tournament itself
16:21:25 <Vorpal> ah neat
16:21:29 <ais523> the ideal would be to set up a contradiction in which the game halts if and only if it doesn't halt
16:21:48 <b_jonas> ais523: the problem about that is that the halting oracle is underspecified in two ways
16:22:02 <ais523> which may /finally/ persuade the M:tG rules people to fix their rules on infinite loops
16:22:12 <b_jonas> ais523: I don't think it can be fixed completely
16:22:17 <b_jonas> some parts of it can be fixed, sure
16:22:24 <ais523> b_jonas: what do you think are the underspecifications in the halting oracle?
16:22:37 <b_jonas> ais523: the lesser problem is not in the halting oracle, but how it's used,
16:23:08 <b_jonas> wait, let me read the rules, I don't remember exactly
16:23:26 <ais523> b_jonas: here's a potential fix: "if an apparent infinite loop occurs, each player may either offer a draw or name a number of iterations; if both players offer a draw, the game is drawn; otherwise, if the loop continues for more than the stated number of iterations, the player who offered that number of iterations loses"
16:23:43 <Vorpal> <ais523> which may /finally/ persuade the M:tG rules people to fix their rules on infinite loops <-- they don't do it like chess? IIRC that is like "if the same position is repeated more than 3 times, the game is a draw"?
16:23:51 <ais523> that gives you a primitive recursive oracle
16:24:02 <ais523> Vorpal: Magic has an unbounded number of possible positions
16:24:06 <Taneb> ais523: that allows both players to lose
16:24:25 <ais523> Taneb: that can already happen in normal gameplay (e.g. rakdos charm)
16:24:30 <Taneb> Oh, OK
16:24:41 <Vorpal> ais523, hm
16:24:47 <b_jonas> ais523: I think the underspecification is about what actions you must perform to break a loop
16:25:07 <ais523> b_jonas: ah right; the rules do require you to break a loop if you can, but leave it unclear as to precisely what that means
16:25:24 <b_jonas> ais523: eg. I think the rules say you don't have to cast Terror just to break an infinite loop involving creates an opponent sets up
16:25:26 <ais523> I think the conventional wisdom is that they require you to not play abilities/spells that are keeping the loop going
16:25:27 <Vorpal> ais523, what happens when your deck runs out?
16:25:44 <b_jonas> the rule is 716.5 in some version
16:25:45 <ais523> but don't require you to play abilities/spells that aren't part of the loop just for the purpose of disrupting it
16:25:53 <ais523> Vorpal: you lose if you try to draw a card from an empty deck
16:25:56 <b_jonas> "No player can be forced to perform an action that would end a loop other than actions called for by objects involved in the loop."
16:25:59 <Vorpal> b_jonas, that is a lot of rules
16:26:03 -!- newsham has quit (Ping timeout: 255 seconds).
16:26:06 <ais523> Vorpal: they aren't numbered consecutively
16:26:14 <ais523> every now and then it skips to the next multiple of 100
16:26:15 <Vorpal> ais523, so, won't that limit the game length?
16:26:24 <ais523> Vorpal: you can recycle cards back into the deck
16:26:28 <Vorpal> ah
16:26:29 -!- newsham has joined.
16:26:35 <ais523> also, the setups we're talking about typically all happen on turn 1
16:27:03 <Vorpal> ais523, I never played MtG, but I believe heartstone for example will definitely be limited due to this
16:27:15 <Vorpal> ais523, so it is infinite in a single turn?
16:27:18 <b_jonas> ais523: hmm, can you set up a loop where only the opponent can make (meaningful) choices, and to break the loop, he is required to perform some very hard cryptographic NP problem, like factoring a huge composite number or finding a string with a given sha-1 sum?
16:27:28 <ais523> Vorpal: hearthstone is very different than M:tG in this case, because hearthstone doesn't have the instants/responding mechanic
16:27:41 <Vorpal> Ah
16:27:44 <ais523> which is heavily involved in my combo
16:27:57 <ais523> b_jonas: oh, beautiful
16:28:10 <ais523> the answer is probably yes, I can envisage some way to add input to StackFlow
16:28:20 <b_jonas> and I mean using repeatedly such kinds of decisions where the opponent is required to choose to break the loop, such as a "an opponent sacrifices a creature" where he has two creatures he could choose
16:28:35 <Vorpal> ais523, oh that is right... Doesn't MtG have ways to interrupt what the opponent is doing? This is what you refer to?
16:28:45 <ais523> Vorpal: yes
16:28:53 <ais523> also yourself, which is less useful except when setting up stupid combos
16:29:01 <Vorpal> ais523, and they don't limit the "recursion depth" of that?
16:29:19 <ais523> the http://esolangs.org/wiki/StackFlow combo ends up setting up a self-sustaining loop that repeatedly interrupts itself
16:29:24 <ais523> without any human interaction
16:29:31 <b_jonas> Vorpal: sure, but you'd set up the loop in such a way that he doesn't have the resources to interrupt the game in any way, such as make sure he has no cards in the hand, no abilities he can activate, doesn't ever get priority or turn start during the loop, etc
16:29:54 <Vorpal> b_jonas, ouch
16:30:00 <b_jonas> Vorpal: the only thing they are still allowed to do is to concede the game, which any player can do any time, but it immediately causes them to lose the game in an unreversable manner
16:30:19 <b_jonas> ais523: I wonder if such a crypto challenge would force the opponent to concede the game
16:30:29 <b_jonas> no wait
16:30:39 <b_jonas> maybe they could take a shortcut if they can prove there is a solution?
16:30:53 <ais523> I thought such controversial shortcuts had to be agreed by both players
16:30:59 <b_jonas> like, if they prove the number is composite, they can just describe how they make the right actions
16:31:22 <b_jonas> ais523: I don't think they have to. they have to be agreed in that he can't tell what decisions _you_ will make, if you're allowed choices
16:31:29 <ais523> also, even proving a large composite number composite is hard if you don't happen to hit it with one of the probabilistic prime tests, which are hard to do in your test
16:31:32 <b_jonas> he just have to prove the judge that wit works
16:31:34 <ais523> *in your head
16:31:41 <ais523> what's the largest known composite strong pseudoprime?
16:31:54 <b_jonas> ais523: we have deterministic (but slow) prime tests now,
16:31:56 <b_jonas> and you can use paper.
16:31:59 <b_jonas> but yes, even then
16:32:13 <ais523> b_jonas: we had deterministic but slow prime tests even beforehand
16:32:13 <b_jonas> you can't run a prime test on a big number in your head
16:32:26 <ais523> the difference is that they're now polynomial (specifically sextic) rather than exponential
16:32:29 <b_jonas> yep
16:32:41 <Vorpal> ais523, presumably this will require a lot of luck for you to get the right cards in the hand at the same time?
16:32:53 <b_jonas> still, you can probably set up a problem where he can't even determine polynomiallly whether there's a solution, and there's half a chance that there's a solution
16:32:57 <b_jonas> maybe
16:33:30 <ais523> Vorpal: there are "universal combo setups" that let you start off any combo you like with maybe about 5-10% probability, even against an opponent who is trying to stop you
16:33:43 <Vorpal> ais523, heh
16:34:01 <Vorpal> I think MtG has some issues as a game that needs to be ironed out
16:34:04 -!- mauris has joined.
16:34:20 <b_jonas> ais523: start when, your first turn, or the upkeep of an opponent that goes first?
16:34:38 <b_jonas> ais523: and are these Vintage-legal, or do they instead involve four Black lotuses?
16:34:41 -!- gamemanj has joined.
16:34:43 <ais523> b_jonas: the ones I know are on your own turn, and Legacy-legal
16:34:51 <ais523> although Vintage makes it easier because of moxen
16:34:58 <ais523> actually I don't really know any of them
16:35:01 <b_jonas> legacy-legal and 5%? wow
16:35:03 <ais523> I believe it's also doable in Modern, although harder still of course
16:35:15 <ais523> b_jonas: also it's not necessarily turn 1
16:35:22 <b_jonas> that sounds hard because of how many things an opponent can do, with their Simian Spirit Guides
16:35:33 <ais523> oh, I meant against typical opponents
16:35:38 <ais523> not ones which are specifically anti-metagaming your deck
16:35:41 <b_jonas> oh, only against typical ones?
16:35:46 <b_jonas> sure, then it is certainly possible
16:35:52 <ais523> actually, if you look at Legacy Omnitell, that's pretty close to a universal combo setup as is
16:35:52 <b_jonas> in Vintage, definitelyu
16:35:57 <b_jonas> probably in Legacy too
16:36:08 <Vorpal> ais523, why markdown? :(
16:36:24 <ais523> Vorpal: because I'm still an esolanger :-)
16:36:33 <Vorpal> Hm
16:36:42 <b_jonas> can you combo into a state where you can even let the opponent take any number turns alternating with you, and he still can't do anything to disrupt you?
16:36:47 <ais523> I did acknowledge that it was an inappropriate syntax
16:36:53 <b_jonas> or, say, at least a few thousand turns
16:37:10 <ais523> b_jonas: yes, the normal approaches involve either controlling the top of the opponent's library, or Shared Fate
16:37:13 <Vorpal> ais523, I just want to see Literate StackFlow now.
16:37:23 <b_jonas> ais523: oh, or just repeated controlling the opponent's turn
16:37:26 <ais523> (the Shared Fate Modern deck is probably tier 3, but hilarious)
16:37:31 <ais523> b_jonas: then the opponent isn't really taking turns
16:37:39 <b_jonas> yeah, true
16:37:44 <b_jonas> hehe, "normal approaches"
16:37:49 <b_jonas> sounds funny in this context
16:39:23 <Vorpal> ais523, it would seem these kinds of things would break tournaments a lot, not people trying to make the game undecidable, but combos you can't break out of would. That is why fighting games have combo breakers for example...
16:39:53 <ais523> Vorpal: in practical tournament play, the opponent normally concedes when you do that
16:39:58 <ais523> combo breakers exist but they're counterable
16:40:16 <Vorpal> Also presumably you would have to build your deck for that
16:40:23 <ais523> (e.g. in Legacy play, Force of Will is commonplace not because it's an actual good card, but because it's one of the few cards that can escape an opponent's combo)
16:40:24 <b_jonas> Vorpal: the cryptographic thing is definitely more complicated than what you could do in a tournament
16:40:29 <b_jonas> an ordinary infinite combo isn't
16:40:41 <Vorpal> Hm
16:40:51 <ais523> every now and then a tournament player says "show me the combo" and forces the other player to play it out
16:41:06 <ais523> and sometimes they screw it up (especially if they got the deck from the Internet and haven't practiced it much)
16:41:11 <Vorpal> b_jonas, From an outsider perspective (who never played the game) this would seem a major flaw in it
16:41:25 <b_jonas> ais523: definitely, that happens even in moderately complicated decks
16:41:46 <b_jonas> even I have built decks where I have to practice a lot before playing, or else I make a ton of mistakes
16:41:47 <Vorpal> <ais523> every now and then a tournament player says "show me the combo" and forces the other player to play it out <-- is that a move? Forcing the other player to play a combo???
16:42:02 <b_jonas> whereas some decks are so simple you can play them without paying much attention
16:42:03 <ais523> Vorpal: no, you do it when the other player has started a combo, as the alternative to conceding
16:42:07 <Vorpal> Ah
16:42:11 <ais523> b_jonas: not really, even simple decks have lots of decisions
16:42:21 <ais523> Burn is often considered a very simple deck but you still have to choose whether to aim at players or monsters
16:42:26 <ais523> err, creatures
16:42:33 <b_jonas> ais523: sure, but they're either easy decision, or decisions that don't impact the player much
16:42:44 <b_jonas> I wasn't thinking of a burn deck, but more of a creature-based deck
16:43:04 <b_jonas> but sure, even that can sometimes have lots of decisions about which creature to play first or how many creatures to attack with
16:43:12 <ais523> or whether to hold them back about wraths
16:43:16 <ais523> *against wraths
16:43:21 <ais523> and how to play around counterspells
16:43:24 <b_jonas> yes
16:43:35 <b_jonas> the metagame matters a lot in how simple it is
16:43:42 <ais523> admittedly, midrange vs. midrange is something I find really boring to watch (especially bad as it's what Wizards has been pushing the most recently)
16:43:48 <ais523> there are still lots of decisions, just ones I don't find interesting
16:44:19 <ais523> Vorpal: there's some Literate StackFlow lower down the page
16:44:43 <b_jonas> well, Wizards always has to push the game in a direction that many decks are simple to play, or else the whole game gets too complex for tournaments
16:44:52 <b_jonas> and less enjoyable
16:44:52 <Vorpal> ais523, I saw that a bit later yes
16:49:20 <ais523> b_jonas: anyway, what we need to do wrt StackFlow
16:49:24 -!- `^_^v has joined.
16:49:34 <ais523> is a) fix the current bug in the construction
16:49:44 <ais523> b) see if we can find some way to observe the halting oracle outcome from within the game
16:49:53 <ais523> are there any effects that replace draws /other than/ Shaharazad?
16:50:07 <ais523> (and Enter the Dungeon, which doesn't exactly count)?
16:51:18 -!- atrapado has joined.
16:52:07 <ais523> huh, do you shuffle your deck at the start of a subgame?
16:52:27 <b_jonas> ais523: you man replace "the game is a draw"?
16:52:31 <ais523> b_jonas: yes
16:52:40 <ais523> we already have lich's mirror for replacing losing
16:52:43 <b_jonas> as opposed to replacing "draw a card"
16:52:51 <ais523> right, replacing "draw a card" is easy
16:53:02 <ais523> but in this case, I particularly want to replace an infinite loop
16:53:11 <b_jonas> I think there deliberately isn't a way to replace "the game is a draw", because that would cause a problem if there's an infinite loop or if all players lost at the same time
16:53:30 <ais523> well shahrazad replaces a draw in the subgame by halving each player's life total in the main game
16:53:48 <b_jonas> but Shahrazad is banned in all formats, isn't it?
16:53:59 <b_jonas> it's as if it was silver or gold bordered now
16:55:35 <oerjan> @messages-
16:55:35 <lambdabot> boily said 5h 53m 5s ago: fungot answers me. fungot is great. fungot fungot fungot ♪
16:56:48 <Vorpal> ais523, what is a sub game in MtG? What it sounds like? (Pausing the game and playing a separate game, then resuming the main game, modified in some way based on the sub game outcome?)
16:57:02 <ais523> Vorpal: yes, that
16:57:08 <ais523> Wizards quickly realised it was a bad idea
16:57:15 <ais523> and banned it in every ruleset they support
16:57:21 <ais523> (well, not /that/ quickly, but they have done)
16:57:22 <Vorpal> ais523, it would make the game awfully complicated
16:57:40 <ais523> it's fun to do once, but players were repeatedly starting recursive subgames in order to run out time in the round during tournaments
16:57:40 <Vorpal> Well, it already is
16:57:50 <ais523> as a method of avoiding losing or to maintain a lead
16:57:58 <Vorpal> ouch
16:58:26 <oerjan> @tell int-e <int-e> ithm effective <-- you must not have seen fungot's response hth
16:58:26 <fungot> oerjan: some experimentation with various envelope rates, then drops to an i/ o
16:58:26 <lambdabot> Consider it noted.
16:59:16 <Vorpal> ^style
16:59:16 <fungot> Available: agora alice c64* ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp youtube
16:59:26 <Vorpal> ^style c64
16:59:27 <fungot> Selected style: c64 (C64 programming material)
16:59:28 <Vorpal> Ah
16:59:40 <Vorpal> ^style discworld
16:59:40 <fungot> Selected style: discworld (a subset of Terry Pratchett's Discworld books)
16:59:42 <Vorpal> hi fungot
16:59:43 <fungot> Vorpal: ' don't mention it. do not meddle in the affairs of priests, with dios in front of his colleague. ' lots of people with nothing to do with it.
17:00:44 <b_jonas> I wonder if you could lock a player's turn down while still allowing him to draw any card from his deck and keep any seven in their hand. You have to destroy all their mana sources, donate an Agressive Mining, get ten copies of Sphere of Resistance and Suppression Field each so he can't do anything even if he discards seven spirit guides, but maybe there's still a way to break out of that.
17:01:09 <b_jonas> Luckily Suppression Field and Sphere of Resistance work for cards in any zone
17:01:31 <Vorpal> jeez, MtG is complicated.
17:02:00 <b_jonas> Vorpal: which part? (all of it, I know)
17:02:20 <ais523> b_jonas: do any of those cards block triggered abilities?
17:02:23 <Vorpal> b_jonas, well what I seen of it. All the terminology.
17:02:26 <Vorpal> For a start
17:02:33 <ais523> I'm thinking of discarding a card with an on-discard triggered ability to hand size
17:02:44 <b_jonas> ais523: oh, right, madness cards with a trigger
17:02:45 <b_jonas> true
17:02:54 <ais523> there are cards with madness 0, although your sphere might block that
17:03:07 <ais523> but you could discard a dredge card and then start dredging narcomoebas
17:03:32 -!- Frooxius has joined.
17:03:35 <b_jonas> Narcomoeba... I didn't know that one
17:04:05 <ais523> it's actually one of the main cards in both a legacy deck /and/ a vintage deck
17:04:11 <b_jonas> ais523: how about if you throw in a Yixlid Jailer?
17:04:15 <ais523> also extended, back when that existed
17:04:35 <ais523> strangely enough the combo was legal in standard for a while but nobody realised it was good at the time
17:05:38 <b_jonas> but yes, Madness could cause problems
17:06:02 <b_jonas> I'm not sure, but I think Sphere of Resistance does block madness
17:06:13 <b_jonas> but wasn't there some other trigger on discard somewhere?
17:07:01 <ais523> so the problem with getting shahrazad to work is you somehow have to set the combo up inside the subgame without any player decisions
17:07:09 <ais523> which will be awkward, because of things like mulligans
17:07:15 <b_jonas> what ability was it that replaced discarding anything with a move that causes the card to be not revealed?
17:07:32 <b_jonas> ais523: yes, that's what I said, you can't force much to happen inside the subgame
17:08:16 <b_jonas> you can set the subgame up if you're allowed to make decisions, but you can't force a computation in it by setting up earlier
17:08:22 <ais523> bleh, you do shuffle your libraries at the start of the subgame
17:08:30 <b_jonas> of course you do
17:08:41 <ais523> and I can't think of any way for the main game to influence the subgame, except by influencing card availablility
17:09:22 <b_jonas> right, you can influence who the two players are, their starting library, and maybe their sideboard? I don't know how subgames work now
17:09:28 <ais523> are there any Conspiracies that prevent mulligans?
17:09:36 <ais523> you can influence the sideboard via wishing cards out from it in the main game
17:09:45 <ais523> it's a separate zone from exile, though
17:09:51 <b_jonas> ais523: I mean the sideboard of the subgame
17:10:03 <ais523> not sure whether main-game cards count as subgame-sideboard, subgame-exile, or neither (although I'm pretty sure it's not subgame-exile0
17:10:08 <b_jonas> can't a wish in the subgame wish something from some zone of the main game?
17:10:26 -!- ineiros has quit (Ping timeout: 240 seconds).
17:10:40 <ais523> " All objects in the main game and all cards outside the main game are consi dered outside the subgame (except those specifically brought into the subgame)"
17:11:06 <ais523> looks like every zone in the main game is "outside the game" for the subgame
17:11:13 <ais523> meaning that you can subgame-wish cards out of maingame-exile
17:11:20 <ais523> but can't subgame-riftsweep them from there
17:11:24 <ais523> (or anywhere else)
17:12:17 <b_jonas> hehe, there are rules about subgames in main games that are Planechase, Vanguard (seriously, isn't that completely impossible?), Commander, and Archenemy, but none for Conspiracy... no wait, I'm just reading a too old ruleset
17:13:00 <b_jonas> I think wishes and similar in the subgame can bring cards from the main game to the subgame
17:13:11 <ais523> b_jonas: paper vanguard exists
17:13:25 <ais523> also, I'm reading the current version of the comp rules
17:13:30 <ais523> none for Conspiracy, you're right
17:14:14 <ais523> hmm, you could probably actually pull that off, too; isn't there a conspiracy that lets you add another booster pack to the draft, and is there any restriction that it's a conspiracy pack?
17:14:51 <b_jonas> hmm
17:15:16 <ais523> err, Conspiracy doesn't seem to be on Gatherer
17:15:17 <ais523> the set, taht is
17:15:35 <ais523> hmm, or not in the search
17:15:39 <ais523> I can find conspiracy cards via their names
17:16:41 <b_jonas> if you make two levels of subgames, then concede the middle level game, what happens with the cards of the innermost level game? do those cards still count as being in the middle level game, in which case they are moved to the exile of the main game by 715.5?
17:17:36 <ais523> ooh, yes, Lore Seeker
17:17:52 <ais523> allows you to use Alpha cards and Conspiracies in the same format (3×Conspiracy draft)
17:18:32 <ais523> this means, in particular, that the subgame/conspiracy interaction is possible in a sanctioned format
17:18:46 <b_jonas> ais523: I'm not sure, aren't banned cards still banned?
17:18:47 <ais523> because I don't see Shahrazad on the list of cards banned in Conspiracy draft
17:18:56 <b_jonas> hmm
17:19:22 <b_jonas> it's not banned in Conspiracy? isn't the Conspiracy ban list derived form some other ban list?
17:19:40 <ais523> Conspiracy's a draft format
17:19:43 <ais523> those tend not to get bans
17:19:50 <b_jonas> ais523: yes yes, but still
17:20:16 <b_jonas> ais523: wait, can this allow you to add an Un-set booster, and play silver bordered cards from it?
17:20:39 <b_jonas> mind you, literal Alpha boosters are very expensive these days
17:20:49 <ais523> b_jonas: I was thinking more "can you add a Beta booster and play /ante/ cards from it?"
17:21:05 <ais523> although I guess they'd just be removed from your deck
17:21:09 <b_jonas> ais523: what would that buy you in a limited game?
17:21:14 <b_jonas> even if it wasn't removed
17:21:20 <b_jonas> you'd just get a few more cards from that limited pool
17:22:12 <ais523> b_jonas: it'd buy you a 39-card deck
17:22:21 <b_jonas> ais523: there's already a Conspiracy card for that
17:22:25 <ais523> although there are easier ways, /especially/ in Conspiracy (where you can have a 35-card deck legitimately)
17:22:32 <b_jonas> exactly
17:23:28 <ais523> I can confirm that nowhere on the Wizards website does it give a banned list for 3×Conspiracy draft
17:23:32 <b_jonas> whereas if you could bring an Un-set booster in, that could potentially influence future games with the same opponent using Time Machine
17:24:14 <ais523> are Un-cards even Magic cards, legally?
17:25:16 <b_jonas> ais523: see rule 108.2, I think they are
17:25:22 <b_jonas> they're just banned in all formats
17:25:35 <b_jonas> gold bordered cards are too, but you can't get those from a booster
17:26:13 * ais523 checks the MTR
17:26:36 <b_jonas> mind you, I don't think Time Machine actually works in the rules
17:26:45 <Vorpal> WHat is "Un-set"?
17:27:00 <Vorpal> What*
17:27:02 <ais523> I was thinking that Un-cards might technically work the same way as token cards, ad cards, etc.
17:27:10 <ais523> Vorpal: a joke set that looks sort-of like a real Magic set
17:27:17 <ais523> but the cards have silver borders and abilities that don't work within the rules
17:27:28 <ais523> they were printed by the same company who makes Magic
17:27:44 -!- danofthedeep has joined.
17:27:51 <ais523> `welcome danofthedeep
17:27:52 <HackEgo> danofthedeep: 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.)
17:28:54 <b_jonas> hmm
17:28:59 <ais523> wait what?
17:29:14 <ais523> AFAICT the MTR allows checklist cards from Innistrad, but not from Magic Origins
17:29:15 <b_jonas> "101.1. Whenever a card’s text directly contradicts these rules, the card takes precedence."
17:29:57 <b_jonas> is there anything that says that the text of other objects (tokens, copies of spells, emblems, etc) or players also take precedence?
17:30:08 <ais523> actually MTR 3.5 implies that an Origins DFC has to be represented by an Innistrad checklist card in the deck
17:30:16 <ais523> players don't have text
17:30:26 <b_jonas> ais523: ok true, players can't
17:30:32 <b_jonas> but how about other objects?
17:30:45 <ais523> "Players may not use errors or o missions in Oracle to abuse the rules. The Head Judge is the final authority for card interpretations, and he or she may overrule Oracle if an error is discovered."
17:30:52 <ais523> oh, I guess this is an old MTR copy
17:30:58 <ais523> it talks about when Magic 2015 becomes legal
17:31:04 <Vorpal> ais523, ah
17:31:53 <b_jonas> I'd like to find a rule that has the intent that rules other than that of the Comprehensive Rules and that of objects already in the game con't modify the game state, so rules you just make up or text on the Time Machine of the previous game can't modify the game.
17:32:13 <b_jonas> I'm quite sure you can't just make up rules and say that they influence the game.
17:32:14 <ais523> b_jonas: there's an un-card that undoes errata
17:32:17 <ais523> which is sort-of the opposite of what you want
17:32:58 <b_jonas> ais523: no, I'd like to prove that Time Machine, Double *, and Ass Whuppin' doesn't actually work in the games targetted
17:33:20 <b_jonas> so they can't influence non-un games per Comprehensive Rules, despite that Maro's un-judgement says they can
17:33:32 <ais523> Ass Whuppin' only hits silver-bordered cards for a reason
17:33:43 <b_jonas> oh yes, probably not Ass Whuppin' then
17:33:50 <b_jonas> but how about Time Machine and Double *
17:34:51 <b_jonas> mind you, I think even if they work, an opponent can always break out of a multi-game cascading effect by conceding one game very early
17:35:06 <ais523> oh yes, conceding is a choice
17:35:23 -!- danofthedeep has quit (Quit: Textual IRC Client: www.textualapp.com).
17:35:48 <b_jonas> but still, it shouldn't influence a serious game even if the opponent forgets to play a normal game after the last un-game in which you Double Play-ed him
17:36:08 <b_jonas> s/normal/casual non-un/
17:36:48 <ais523> <Bill Gates> Like almost everyone who uses e-mail, I receive a ton of spam every day. Much of it offers to help me get out of debt or get rich quick. It would be funny if it weren't so irritating.
17:37:18 <Vorpal> ais523, heh
17:37:30 <b_jonas> what? doesn't he get spam to buy rolex or viagra?
17:37:52 <Vorpal> I don't get much spam. Maybe one a week or so
17:38:00 <b_jonas> most of the spam I get is actually academic spam, because that goes through spam filters more easily than other spam
17:38:14 <Vorpal> b_jonas, academic spam?
17:38:27 <int-e> any *eroxy spam?
17:38:27 -!- bb010g has joined.
17:38:49 <Vorpal> int-e, what is that?
17:38:57 <b_jonas> Vorpal: spam inviting you to publish in a scientific journal or apply to a scientific conference, sent to people with their name and email address taken from scholar articles, with "Professor" always appended to their name
17:39:02 <Vorpal> b_jonas, ah
17:39:05 <ais523> b_jonas: this was in the earlier days of spam, I think
17:39:06 <int-e> Vorpal: oh there are so many conferences and journals where "peer-review" is handled by handing over a sufficient amount of money
17:39:20 <Vorpal> int-e, oh dear
17:39:23 <ais523> incidentally, spam is now a much smaller fraction of Internet traffic than it once was
17:39:23 <int-e> Vorpal: I'm getting spam from Engineroxy and Computeroxy all the time.
17:39:34 <ais523> I think once it was over 90%, but it fell under 50% recently
17:39:35 <Vorpal> int-e, never heard of them
17:39:38 <int-e> The Internet is for Porn!
17:39:40 <b_jonas> ais523: well of course. with all the high resolution videos people download
17:39:46 <int-e> Vorpal: ok, just curious since the topic came up
17:39:47 <ais523> probably not because spam has slowed down, but because it's bandwidth-efficient and the bandwidth use of everything esle has gone up
17:39:48 <b_jonas> it's hard to send so many text emails
17:39:51 <b_jonas> "text"
17:40:22 <b_jonas> you know, text and html and embedded images and trojans and malware triggering a security vulnerability, but those are generally all small
17:40:27 <Vorpal> I usually get spam for rayban glasses in my gmail, about one every few weeks.
17:40:29 <Vorpal> That is about it
17:40:32 <Vorpal> kind of weird
17:40:36 <b_jonas> the actual malware payload is either small or is downloaded separately
17:40:54 <b_jonas> the spam emails definitely don't contain tens of gigabytes of videos
17:41:02 <Vorpal> inded
17:41:07 <Vorpal> indeed*
17:41:18 <b_jonas> that would fill up your inbox so they couldn't send you the next spam email
17:41:45 <Vorpal> b_jonas, also gmail has a 25 MB / mail limit iirc
17:42:11 <b_jonas> yeah, all mail servers have some size limits
17:42:49 <b_jonas> but a hundred spam mails throughout a week, 25 MB each, is also too big
17:43:17 <Vorpal> yes
17:44:06 <int-e> have you actually seen such big spam?
17:44:41 <Vorpal> b_jonas, I use a separate email for each site I register with. That way I can black list an email if it starts getting spam, also complain to the web site in question
17:45:04 <Vorpal> int-e, it would expend too much resources on the spammer's server as well
17:45:50 <Vorpal> There should be a spam opt out. Saying "I'm not this stupid, just stop sending me the bloody stuff, it is pointless"
17:45:54 <Vorpal> Wouldn't work of course
17:46:24 <int-e> stop reading email altogether... unfortunately, that's increasingly impractical
17:46:47 <Vorpal> int-e, well, as I said, I don't get much spam. Maybe one a week at most
17:46:53 <Vorpal> So it isn't much of an issue for me
17:48:00 <ais523> b_jonas: aha, in a Conspiracy×3 tournament, you're not allowed to use boosters that weren't given to you by the tournament organizers
17:48:14 <ais523> meaning that if a Beta booster gets in there somehow, it's the TO's fault
17:48:43 <ais523> however: "Other than basic land, only cards from the expansions of the boosters opened (and only cards opened in that player’s pool) may be used in a player’s deck."
17:48:56 <ais523> this seems to imply that cards from a Beta booster are legal, if you get one into the draft somehow
17:49:31 <ais523> oh, haha, they closed it specifically: "Because it was designed specifically for multiplayer play, the use of Conspiracy boosters in sanctioned, competitive Limited-format tournaments (Sealed Deck and Booster Draft) is not permitted."
17:50:30 <ais523> a cop-out, but I guess that works
17:52:13 -!- XorSwap has joined.
18:07:42 -!- oerjan has quit (Quit: leaving).
18:25:06 -!- ^v has joined.
18:46:23 -!- XorSwap has quit (Ping timeout: 244 seconds).
18:47:14 -!- hppavilion[1] has joined.
18:47:24 <hppavilion[1]> What could we do for Esoteric Libraries
18:47:35 <hppavilion[1]> I like the concept of Material Measurements
18:56:29 <HackEgo> [wiki] [[Fish]] http://esolangs.org/w/index.php?diff=44568&oldid=44143 * 76.164.8.130 * (+0) /* Movement and execution */ Moved the '#' mirror to be on the same line
18:57:51 <HackEgo> [wiki] [[Fish]] http://esolangs.org/w/index.php?diff=44569&oldid=44568 * 76.164.8.130 * (+0) /* Literals and operators */ Made it so code blocks dont wrap
19:00:56 <HackEgo> [wiki] [[Material Measurements]] N http://esolangs.org/w/index.php?oldid=44570 * Hppavilion1 * (+1391) Created page
19:02:47 -!- mauris_ has joined.
19:03:08 -!- XorSwap has joined.
19:03:26 <HackEgo> [wiki] [[Material Measurements]] M http://esolangs.org/w/index.php?diff=44571&oldid=44570 * Hppavilion1 * (+21) Removed word "metric" for tax purposes, extended words
19:05:45 -!- mauris has quit (Ping timeout: 265 seconds).
19:09:03 <hppavilion[1]> How about an OO Combinatory Logic language? xD
19:10:03 <HackEgo> [wiki] [[Material Measurements]] M http://esolangs.org/w/index.php?diff=44572&oldid=44571 * Hppavilion1 * (+3) Fixed a piece of code
19:13:52 <hppavilion[1]> Anyone have any MM ideas?
19:19:31 <MDream> Inches are usually divided down to the 1/16th.
19:20:09 -!- MDream has changed nick to MDude.
19:20:54 <MDude> I've been wanting to have some kind of measurement converting system, but I didn't have any measurements to use first so I didn't work on it.
19:23:28 <Vorpal> MDude, hm? for what purpose?
19:23:59 <Vorpal> If you just want to convert there are plenty of utilities that do that
19:24:07 <MDude> For just general things related to measurements?
19:24:18 <Vorpal> MDude, apt-get install units?
19:24:24 <MDude> I don't see your point.
19:24:26 <Vorpal> Command line
19:24:37 <Vorpal> for converting SI to other unit systems
19:24:50 <MDude> Why would I want a uitility when I'm talking about coding?
19:24:53 <Vorpal> Oh okay
19:24:58 <Vorpal> I misunderstood you then
19:47:29 <hppavilion[1]> I got Prealgebra working for LIME
19:48:03 <hppavilion[1]> !
19:49:59 <Phantom_Hoover> what the hell is prealgebra
19:51:19 -!- evalj has joined.
19:53:09 <b_jonas> ais523: um, Conspiracy always worked that. it was designed as a casual format, like Archenemy and Planechase and Commander, so it was never intended for serious tournaments. that's what they always claimed about that product from the start.
19:53:22 <ais523> b_jonas: boring :-(
19:54:49 <b_jonas> ais523: still, at least Archenemy and Conspiracy and Shahrazad are officially supported by the Comprehensive rules, and can get oracle text errata, unlike un-cards and gold bordered cards
19:55:01 <b_jonas> (the gold bordered cards that aren't also printed as black bordered, that is)
19:56:00 <b_jonas> ais523: also, I was thinking about this finite but large combos thing, and I'm starting to understand why you can get large ones, though I couldn't make a combo that's as large as the ones you mentioned
19:56:37 <b_jonas> I can get a two up arrowed one, and could probably get a three up arrowed one if I tried hard enough
19:57:10 <b_jonas> and I can at least imagine that you could get a five or six up arrowed one if you research and work on it carefully, though it would be very hard to avoid infinite combos
19:57:33 <b_jonas> I think Djinn Illuminatus helps a lot
19:57:55 <hppavilion[1]> Phantom_Hoover: Well, prealgebra is usually a class in schools to teach about algebra, but in THIS case I'm using it to refer to arithmetic with variables
19:58:34 <hppavilion[1]> It's an API I designed to make Set-builder notation work well in my Set Theory editor
19:58:48 <b_jonas> ais523: and yes, Doubling Season is a very good idea too
20:00:58 <b_jonas> wait wait
20:01:23 <b_jonas> isn't there any spell or ability that simply lets you choose a target sorcery or instant spell from the stack and copy it, without allowing you to change the targets?
20:02:20 -!- AnotherTest has quit (Ping timeout: 244 seconds).
20:07:30 <b_jonas> oh, Spitting Image looks very useful
20:07:46 <b_jonas> because with 40 lands in your deck, you could play a single Spitting Image 40 times
20:07:51 <b_jonas> and you can get each of them copied a lot
20:08:22 <b_jonas> s/play/cast/
20:09:35 <b_jonas> hmm, though I'm not sure you could get it copied enough times
20:12:56 <ais523> b_jonas: all spell copies I know also retarget the copies
20:13:29 <b_jonas> Radiate lets you start growing a bit faster than just a Twincast, but I don't think it matters much asymptotically
20:21:25 <hppavilion[1]> You know what'd be interesting?
20:21:38 <Phantom_Hoover> a scarf made of pasta?
20:21:39 <hppavilion[1]> A Linguistics-Centric Programming Language.
20:21:42 <hppavilion[1]> Phantom_Hoover: You win.
20:22:27 <hppavilion[1]> Like a free Mathematica for programming linguists
20:22:44 <hppavilion[1]> Do we know any linguists?
20:23:34 <Phantom_Hoover> augur?
20:23:42 <augur> hello!
20:24:01 <Phantom_Hoover> O.o i thought you'd gone into the Long Lurk
20:25:37 <augur> Phantom_Hoover: i lurk, but im always on IRC
20:25:44 <augur> and always at my computer :)
20:27:19 <b_jonas> most of the simple combos are either only exponential, or infinite, or don't work
20:28:52 <b_jonas> oh.... Panoptic Mirror sounds interesting
20:29:09 <augur> hppavilion[1]: what do you mean? can you elaborate?
20:29:36 <b_jonas> if you have say 60 or 100 turns, you can _cast_ an instant or sorcery 60 or 100 times
20:29:38 <hppavilion[1]> augur: Well first I need to find a Linguist willing to help.
20:29:41 <b_jonas> that sounds very useful
20:29:50 <hppavilion[1]> augur: THEN I let them talk about what they want in a language
20:29:58 <b_jonas> just let your deck run out after 100 turns
20:30:09 <hppavilion[1]> augur: THEN I spend a few years figuring out what that stuff means and what the stuff important to that means
20:30:17 <hppavilion[1]> augur: THEN I learn C++ better
20:30:23 <hppavilion[1]> augur: THEN I make the language
20:30:36 <FireFly> I don't think you'd necessarily have to do them in that order
20:30:37 <augur> hppavilion[1]: well, im a linguist, and im happy to talk
20:30:41 <hppavilion[1]> Excellent
20:30:49 <hppavilion[1]> Do you also do programming?
20:32:13 <hppavilion[1]> That'd be pretty useful.
20:33:08 <hppavilion[1]> augur: Are there any cool Linguistics words we could name this after temporarily?
20:33:17 <augur> i do programming, yes
20:33:25 <hppavilion[1]> OK
20:33:26 <augur> hppavilion[1]: first you have to tell me what "this" is :P
20:33:26 <hppavilion[1]> Excellent
20:33:32 <hppavilion[1]> xD
20:33:52 <hppavilion[1]> The idea is, in the abstract, an attempt to make a linguistics-oriented programming language
20:34:26 <hppavilion[1]> Sort of like the Wolfram language, but for Linguistics instead of Math
20:34:56 <hppavilion[1]> Mostly just to see if it's possible, partially for someone to actually maybe someday use, with a hint of boredom resolution for me
20:35:16 <augur> what does that mean tho
20:35:31 <hppavilion[1]> augur: Well I don't understand Linguistics enough to figure that out xD.
20:37:11 <hppavilion[1]> If I understand it properly, it'd have built-in interfaces for creating formal grammars, for making translators between languages, etc.
20:37:17 <hppavilion[1]> I have to switch classes now
20:37:20 <hppavilion[1]> Be back in a bit
20:37:29 <Phantom_Hoover> you do not sound contemptuous enough of wolfram language
20:37:32 <Phantom_Hoover> it's nothing special
20:38:44 <Phantom_Hoover> it's an idiosyncratic language with some nice symbolic manipulation features and a big standard library
20:40:35 <b_jonas> surprisingly many cards convert damage dealt to something to that many tokens
20:40:50 <b_jonas> as in, you get N tokens when N damage is dealt
20:41:07 <b_jonas> Druid's Call is probably the most versatile, unless you want to deal damage to a player
20:41:43 -!- hppavilion[1] has quit (Ping timeout: 256 seconds).
20:42:46 -!- gamemanj has quit (Ping timeout: 240 seconds).
20:43:20 -!- bb010g has quit (Quit: Connection closed for inactivity).
20:44:08 -!- XorSwap has quit (Quit: Leaving).
20:46:28 -!- hppavilion[1] has joined.
20:52:09 <Vorpal> <Phantom_Hoover> a scarf made of pasta? <-- cooked, not cooked but fresh, or dried?
20:52:52 <Phantom_Hoover> why would you even ask this
20:53:16 <Vorpal> Phantom_Hoover, because I'm trying to figure out how it would work, what sort of material properties it would have
20:53:58 <Phantom_Hoover> i didn't really think through the details of the pasta scarf
20:54:08 <Vorpal> Phantom_Hoover, ah, fair enough
20:54:12 <Phantom_Hoover> it wasn't an exercise in worldbuilding
20:54:20 <Vorpal> touché
20:54:44 <olsner> you don't need to construct the whole world, just the pasta scarf
20:54:53 <Vorpal> good point
20:55:00 <olsner> or were you intending for a world where pasta behaves specially?
20:55:03 <Phantom_Hoover> what kind of culture would wear the pasta scarf though
20:55:11 <Phantom_Hoover> what's their level of food technology
20:55:19 <Vorpal> Phantom_Hoover, Italian plumbers?
20:55:25 <olsner> presumably at or beyond the making of pasta
20:55:26 <Phantom_Hoover> is the pasta scarf everyday wear or a badge of office
20:55:34 <Phantom_Hoover> Vorpal, ah, of course
20:55:39 <Phantom_Hoover> pasta because he's italian
20:55:49 <Phantom_Hoover> scarf because... plumbers... always wear scarves?
20:56:01 <Vorpal> Phantom_Hoover, not sure
20:56:02 <hppavilion[1]> augur: I'm back
20:56:09 <Vorpal> But yes Italian certainly
20:56:15 <hppavilion[1]> But I'm in class, so I won't be on too much for a bit
20:56:43 <Vorpal> Phantom_Hoover, anyway, wrt everyday or badge of office, that would depend on how practical it was, which again goes back to the material properties
20:57:24 <Phantom_Hoover> evidently the Scarf Culture must be agrarian to obtain the grain needed for the scarf
20:57:42 <Vorpal> Honorary paraphernalia does not have to be as practical or durable as every day clothes
20:57:57 <Vorpal> Well yes
20:58:27 <Phantom_Hoover> or perhaps they are a nomadic tribe, weaving pasta scarves from the pillaged wheat of their tributaries
20:58:37 <Vorpal> Phantom_Hoover, But it could be a sort of carnival cloth, for the harvesting festival.
20:58:39 -!- lleu has joined.
20:58:39 -!- lleu has quit (Changing host).
20:58:39 -!- lleu has joined.
20:59:01 <Vorpal> Phantom_Hoover, thus it would not need to be practical as such
20:59:14 <olsner> `? brickbrain
20:59:14 <HackEgo> brickbrain? ¯\(°​_o)/¯
20:59:16 <olsner> `? brickbrained
20:59:17 <HackEgo> brickbrained? ¯\(°​_o)/¯
20:59:21 <olsner> `? brick
20:59:21 <HackEgo> Brick goes in brain. The statutory punishment for perpetrators of brainfuck derivatives.
20:59:36 <augur> hppavilion[1]: so depending on what you want, something like that might already exist
20:59:49 <hppavilion[1]> Possibly xD
20:59:53 <Phantom_Hoover> augur what do you think the word for a scarf made of pasta would be
20:59:54 <Vorpal> Phantom_Hoover, oh, instead of shrunken heads or other trophies? That would work yes
20:59:55 <hppavilion[1]> I should probably check
20:59:57 <augur> except that Wolfram Lang ties into all sorts of other stuff like the whole suite of crap that Mathematica has
21:00:04 <hppavilion[1]> True, true.
21:00:27 <augur> but you might want to look into GF / Grammatical Formalism
21:00:36 <hppavilion[1]> Can't find anything
21:00:48 <augur> Grammatical Formalism, by Aarne Ranta
21:00:56 <augur> theres a whole page, youll find it in google
21:01:46 <Vorpal> Phantom_Hoover, well, that was a fun exercise in creative thinking. Need to sleep now though
21:02:15 <Phantom_Hoover> yes vorpal
21:02:23 <Phantom_Hoover> no scarf
21:02:25 <Phantom_Hoover> only dreams now
21:02:32 <Vorpal> How poetic
21:02:41 <Vorpal> Though I could dream about scarfs perhaps
21:07:12 <hppavilion[1]> augur: I use Duck Duck Go, and it didn't come up. Amazon was a bust too.
21:07:17 -!- Patashu has joined.
21:07:50 <augur> Phantom_Hoover: i dont know, pasta scarft?? who knows
21:08:02 <augur> hppavilion[1]: https://www.google.com/search?q=grammatical+framework+ranta
21:08:15 <augur> http://www.grammaticalframework.org/
21:08:59 <hppavilion[1]> Huh
21:09:28 <hppavilion[1]> augur: I still feel like developing my own xD
21:09:34 <hppavilion[1]> Can I ask you for help when I get to that part?
21:10:37 -!- `^_^v has quit (Ping timeout: 256 seconds).
21:11:37 <augur> hppavilion[1]: i guess, but its gonna be a huge pain in the ass for you
21:11:52 <augur> im implementing something similar to Ranta's GF, for my wanna-be AI startup
21:12:54 <hppavilion[1]> I know xD
21:13:06 <hppavilion[1]> Not about the second part, but the first one uyes
21:18:48 <hppavilion[1]> It'd be a pretty useful thing for TBGs, I suppose...
21:29:53 -!- hppavilion[1] has quit (Ping timeout: 256 seconds).
21:43:07 -!- evalj has quit (Remote host closed the connection).
21:43:57 -!- Patashu has quit (Ping timeout: 268 seconds).
21:46:54 -!- atrapado has quit (Quit: Leaving).
22:07:07 -!- hppavilion[1] has joined.
22:08:11 <hppavilion[1]> What other applied-mathematical fields could we make Programming Languages about?
22:08:32 <hppavilion[1]> (Applied-mathematical measn things like Linguistics in this concept)
22:08:58 <hppavilion[1]> I suppose operator overloading is like Ring Theory if you manually preserve closure
22:15:42 -!- mauris_ has quit (Ping timeout: 265 seconds).
22:15:57 -!- lleu has quit (Quit: That's what she said).
22:20:27 <hppavilion[1]> How does one define a ring syntactically
22:20:28 <hppavilion[1]> ?
22:20:42 -!- J_Arcane has joined.
22:24:03 -!- ineiros has joined.
22:32:48 -!- hppavilion[1] has quit (Ping timeout: 264 seconds).
22:45:40 -!- hppavilion[1] has joined.
23:03:25 <tswett> Ahoy.
23:03:31 <tswett> Define a ring syntactically, you say?
23:03:48 <tswett> I'm not sure that my idea of what "syntactically" means is the same as your idea.
23:05:39 <ais523> well, variable names in many languages have the syntactic behaviour of a set
23:05:50 <ais523> because it doesn't matter which order you define them in and you can't define them twice
23:06:22 <ais523> (this becomes more obvious if you look at the grammar)
23:06:47 <ais523> (although from the grammar, it looks more like a multiset, because uniqueness of names is normally implemented via freshness conditions)
23:28:49 <HackEgo> [wiki] [[J--]] http://esolangs.org/w/index.php?diff=44573&oldid=44020 * LegionMammal978 * (+1) /* Hello World */
23:32:25 <FireFly> Oh, not a J derivative
23:42:19 -!- JuanDaugherty has joined.
23:44:01 -!- JuanDaugherty has left ("Exeunt").
23:44:09 -!- XorSwap has joined.
23:54:30 <ais523> idea: a language that is a derivative of itself
23:54:33 <ais523> err, not including Feather
23:56:34 <MDude> A language whose description is itself an enlessly complex recursive document?
23:56:38 -!- hppavilion[1] has quit (Ping timeout: 272 seconds).
23:57:09 <ais523> a language that is defined as being like itself, but with exceptions, perhaps
23:57:24 <ais523> in a way that allows the entire language to be deduced via a fixed point operator
23:59:51 <\oren\> im home
←2015-09-30 2015-10-01 2015-10-02→ ↑2015 ↑all