00:00:27 <Lyka> who is 78.245.243.132?
00:00:36 <oerjan> esolangs.zem.fi is a temporary domain name fizzie set up
00:02:24 <oerjan> how does he manage to use an IP which doesn't exist...
00:04:46 <Lyka> who's Koen in IRC?
00:04:48 <oerjan> @tell fizzie whatever you did to make esolangs.zem.fi work has made esolangs.org partly broken
00:05:01 <oerjan> Lyka: Koen_ usually, but he's not always here
00:05:25 -!- hilquias has quit (Remote host closed the connection).
00:05:34 <oerjan> confusingly, Koen without _ is someone else iirc
00:06:32 -!- ZombieAlive has joined.
00:07:21 <oerjan> @tell fizzie e.g. https://esolangs.org/wiki/Special:RecentChanges has stylesheet links mentioning //esolangs.zem.fi/
00:07:52 <HackEgo> [wiki] [[User:Lesidhetree]] N http://esolangs.zem.fi/w/index.php?oldid=43049 * Lesidhetree * (+79) Little crappy intro
00:17:21 <Lyka> Simplest "Cat" Program I can think of, terminating on Ctrl-@ (aka NUL): !A000SIB1[01_SOB1SIB1]01_Q000@
00:20:55 <Lyka> i think equivalent to void main(){char a=0,b; b=getc(); while(a != b){ putc(b) ; getc(b); } }
00:21:11 <Lyka> assuming i used getc() and putc() right
00:21:45 <Lyka> void main(){char a=0,b; b=getc(); while(a != b){ putc(b) ; b=getc(); } }
00:22:48 <oren> is a bowl of ice cubes a good heat sink?
00:24:30 <HackEgo> [wiki] [[Talk:Fourfuck]] http://esolangs.zem.fi/w/index.php?diff=43050&oldid=43048 * Lesidhetree * (+176) added a little disclaimer
00:24:54 <oren> We're trying to charge an old laptop without overheating it
00:25:15 <oren> Lyka: int main()
00:25:39 <Lyka> int main(){char a=0,b; b=getc(); while(a != b){ putc(b) ; b=getc(); } return 0;}
00:27:05 -!- oerjan has quit (Quit: Z).
00:29:16 <oren> So I have a metal bowl filled with ice cubes on top of the charger brick
00:31:46 <Lyka> Can you guess what this does? ![00=]00=Q000@
00:32:14 <oren> loops infinitly?
00:32:24 <Lyka> was it that obvious?
00:32:55 <Lyka> Can you guess what this does? ![00=SIB1SOB1]00=Q000@
00:32:59 <oren> Well [00= means while(c1 == c1)
00:33:33 <oren> that is a cat program
00:33:51 <Lyka> that runs forever
00:34:05 <oren> (From my working memory of your language's spec)
00:34:38 <Lyka> https://dl.dropboxusercontent.com/u/98841263/fourfuck%20language/Fourfuck%20v0.0-pre_alpha-0003b.pdf
00:35:10 <Lyka> i have that in front of me
00:36:50 <Lyka> it an now run commands stored in memory
00:38:32 <Lyka> though i have not yet debugged it
00:44:39 <Lyka> !A000A101A20DSIB2[02_SIB3SOB3+010]02_Q000@Hello, World!
00:44:49 <Lyka> is that cheating?
00:50:59 <Lyka> !A000T0D0A000T3D0Q000@Hello, World!
00:56:56 <Lyka> fine... !A048A165A26cA36cA46fA52cA620A757A86fA972AA6cAB64AC21AD00AE01AF0D[DF<SORD+DED]DF<Q000@
01:07:43 <Lyka> I guess this is what you wanted: !A000A101A200A348C230+010A365C230+010A36cC230+010A36cC230+010A36fC230+010A32cC230+010A320C230+010A357C230+010A36fC230+010A372C230+010A36cC230+010A364C230+010A321C230+010T3D2Q000@
01:11:03 <Lyka> well, i forgot to made the T--- command i need, after the subsequent number shift, the T3D2 would become T4D2, and I would use T502 instead of it.
01:13:02 <Lyka> but, yeah, that's the most efficient way to code "Hello, World!" into the code itself
01:15:41 <Lyka> Ignore the spaces: !A000A101A200 A348C230+010 A365C230+010 A36cC230+010 A36cC230+010 A36fC230+010 A32cC230+010 A320C230+010 A357C230+010 A36fC230+010 A372C230+010 A36cC230+010 A364C230+010 A321C230+010 T3D2Q000@
01:24:04 <oren> It should be possible to do direct translation to C
01:29:31 -!- lleu has quit (Quit: That's what she said).
01:31:10 <FireFly> b_jonas: I stumbled upon this, which reminded me of your pondering of infix operators for min/max in C a while ago: http://blogs.msdn.com/b/oldnewthing/archive/2015/05/25/10616865.aspx#10616991
01:33:54 <oren> http://hastebin.com/raw/yoboboqami
01:36:52 <oren> Firefly: hehe 2s complement screwery
01:39:38 -!- hilquias has joined.
01:40:16 <oren> explanation: -x is ~x+1 so -~x is ~~x+1 = x+1
01:41:07 <Wright> Oh, it expired. Never mind
01:42:01 <oren> -x-1 = ~x so ~-x = --x-1 = x-1
01:42:29 <oren> also it doesn't look like a tadpole in my font
01:42:41 <oren> beacuse ~ is high
01:44:56 -!- Sprocklem has quit (Ping timeout: 272 seconds).
01:47:03 <oren> it looks like a tadpole in monofur though
01:49:53 <FireFly> oh, that's what a tadpole is. oh, that's where poliwag's name comes from apparently
01:50:49 <int-e> Wright: it's mostly recovered (as far as I can see, whois lists the right DNS servers now, but the information needs to propagate to the .org dns servers)
01:51:55 <int-e> actually, no, it has to propagate from those to the rest of the net
01:52:28 <oren> it works here in canada
01:55:35 <oren> in the meantime you can use the esolangs.zem.fi
01:56:33 -!- Sprocklem has joined.
01:57:56 -!- ZombieAlive has quit (Remote host closed the connection).
02:03:14 <int-e> oh fun, 199.249.112.1 is routed to different hosts for different sources... makes sense, but it makes the result of dig any esolangs.org @199.249.112.1 non-reproducible.
02:04:00 <int-e> (it's wrong for me, still returning {ns53,ns54}.domaincontrol.com. as DNS servers)
02:08:04 -!- rdococ has joined.
02:08:41 <Lyka> https://dl.dropboxusercontent.com/u/98841263/fourfuck%20language/Fourfuck%20v0.0-pre_alpha-0003c1.pdf
02:09:04 <Lyka> (for thoise bored sous who keep track of my stuff)
02:10:09 <Lyka> Hello World: !A000A101A200 A348C230+010 A365C230+010 A36cC230+010 A36cC230+010 A36fC230+010 A32cC230+010 A320C230+010 A357C230+010 A36fC230+010 A372C230+010 A36cC230+010 A364C230+010 A321C230+010 T602Q000@
02:10:25 <Lyka> seems simple...
02:10:40 <int-e> a small sample... http://sprunge.us/bIRg (austria, austria, germany, netherlands, canada)
02:11:56 <int-e> So it looks like the first two are actually reaching the same server, but the rest are different.
02:14:56 -!- rdococ has quit (Ping timeout: 246 seconds).
02:26:24 -!- int-e has set topic: John Nash's beautiful mind has reached its final equilibrium | The Collatz files | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/ | http://esolangs.org/ is or will be back; try http://esolangs.zem.fi/.
02:31:37 <zzo38> I like to use URIs as identifiers in some stuff (as properties in extensible data files, to identify audio plugins, and other stuff), but some people like to use only HTTP URIs and really you can use other schemes too such as urn:uuid: and a lot more; it doesn't really matter
02:32:24 -!- hilquias` has joined.
02:34:09 -!- hilquias has quit (Ping timeout: 265 seconds).
02:52:23 <Lyka> trying to make a program that will output the numpers 0 to 999. problem is, the math operations are all for unsigned chars...
02:55:05 <Lyka> i used x00 - x63 in a single byte for 0 to 99
02:55:52 <Lyka> x0000 to x03E7 ain't that easy with 8-bit memory
02:58:23 <Lyka> 0 to 99 ! A000A101A264A330A40AA564A600A700 [03< {04> /047%746+366SOB6 |000 }000 %046+366SOB6 P0A0 +010]03< Q000 @
02:58:29 <pikhq> zzo38: URIs are fairly simple to parse and quite well supported. (though for identifying properties a URN might be more appropriate)
02:58:38 <pikhq> zzo38: So yes, that overall makes sense.
03:00:01 <zzo38> If you have a MAC address but no domain name or permanent IP address or anything like that, you can still generate a UUID.
03:01:00 * Lyka realizes that, not only does fourfuck no longer have anything to do with brainfuck, it may soon have 7 or 8 character command/argument blocks, as the arduino is a 16-bit environment
03:01:23 <pikhq> zzo38: An RNG will also suffice.
03:01:48 <zzo38> Yes, that also works, but not as good in my opinion.
03:02:05 <Lyka> so much for the name of my anguage meaning anything...
03:02:44 <pikhq> Frankly an RNG isn't at all worrying -- 122 actually random bits are basically *not* going to collide.
03:13:42 <zzo38> If you make up the Magic: the Gathering card like Notion Thief but reversed, should power/toughness/cost be changed a bit?
03:16:28 <HackEgo> [wiki] [[Talk:Fourfuck]] http://esolangs.zem.fi/w/index.php?diff=43051&oldid=43050 * Lesidhetree * (+547)
03:18:16 <Lyka> Okay, as the language seems to increasingly become less and less like Brainfuck, with the 4-character command/argument blocks becoming a pain due to anticipating needing 16-bit integers for the exact project it was designed for, I am gonna have to switch active development to a six or eight character version for intended use. I plan on coming back to developing Fourfuck as a simplified variant when I have finished the project this was designed for
03:18:31 -!- supay has quit (Ping timeout: 272 seconds).
03:19:40 <Lyka> i might need to store UTF-16 in the intended project...
03:22:00 <Lyka> ike my note said on the talk page, i'll simplify whatever i make back into 4 chars
03:23:02 -!- supay has joined.
03:24:43 <zzo38> You should not use Unicode
03:25:14 <zzo38> (And, even if you do, UTF-16 isn't really the best way to use Unicode, anyways.)
03:26:05 <pikhq> In general you should use Unicode if at all applicable, but UTF-16 is uniquely terrible and should only be used for compat purposes.
03:26:09 -!- variable has quit (Ping timeout: 246 seconds).
03:26:23 <Sgeo> pikhq, not as bad as UCS-2 which afaik is what Javascript does
03:26:30 <pikhq> Fuck UCS-2 so hard.
03:26:48 <zzo38> In general you should use ASCII if possible.
03:26:52 <Sgeo> Rust has a "WTF-8" encoding for UCS-2 compatibility
03:27:05 <Sgeo> https://simonsapin.github.io/wtf-8/
03:27:10 <Sgeo> "WTF-8 (Wobbly Transformation Format − 8-bit) is a superset of UTF-8 that encodes surrogate code points if they are not in a pair. It represents, in a way compatible with UTF-8, text from systems such as JavaScript and Windows that use UTF-16 internally but don’t enforce the well-formedness invariant that surrogates must be paired."
03:27:23 <pikhq> In general you should use UTF-8, which for most code is "use ASCII and don't do things that break UTF-8".
03:27:48 <zzo38> For programs that use Unicode, yes UTF-8 is much better than other encodings since then it works even if you are using only ASCII.
03:28:23 <zzo38> Sgeo: I thought Wikipedia says CESU-8 is encoding individual surrogate codes in UTF-8?
03:28:56 <zzo38> Still, VGMCK supports both that as well as proper UTF-8 (this is because the actual output file contains text in UTF-16 format)
03:29:31 <pikhq> zzo38: WTF-8 is different -- it's UTF-8 except it defines a representation of invalid UTF-16.
03:30:18 <pikhq> It's not UCS-2 compat, it's "dumb shit that acts like UTF-16 units are Unicode codepoints" compat. :)
03:31:06 <zzo38> I recommend in any program that uses input normally in ASCII but writes output into a file that contains UTF-16 text, that it would support both CESU-8 and proper UTF-8.
03:31:46 <zzo38> (In other programs that need Unicode I do not recommend adding support for CESU-8 since it isn't important.)
03:32:03 <pikhq> Supporting CESU-8 as well as UTF-8 without it being an explicit option (i.e. --enable-cesu8 or some such) is a bad idea though.
03:32:31 <pikhq> (non-obvious behavior with two non-identical strings mapping to the same string can be real dangerous)
03:35:45 <zzo38> It depends much on the program, I think.
03:38:02 <zzo38> It is not necessarily a bad idea. In the cases where it is, there is probably no point supporting CESU-8 anyways if you can just use an additional filter program to convert it at first.
03:39:17 -!- password2 has joined.
03:41:08 <zzo38> In the cases where I used it, there is no such danger.
03:47:41 -!- KnightArm0 has joined.
03:53:02 <oren> utf-8 is a good transport and file format, but for internal purposes Unicode-aware porgrams should use WTF-32
03:53:22 <pikhq> Why? What does UTF-32 buy you?
03:53:43 <oren> the ability to count characters, move them around, etc, easily
03:53:57 <pikhq> "Count characters" For what purpose?
03:53:59 <zzo38> I think it depends on the program!
03:54:18 <pikhq> Also, not really: a Unicode character is composed of one or more codepoints.
03:54:23 <zzo38> Many programs that have some support for Unicode don't need to count characters or anything else like that
03:55:22 <zzo38> Many of my programs do not support Unicode at all and I am not going to add any support. A few do, but only do what they need to do with Unicode, and not more than that.
03:56:29 <zzo38> For example, my RDF Turtle parser library has some support for Unicode; specifically, it allows Unicode in identifiers and will decode \u escapes into UTF-8.
03:58:14 <zzo38> (Allowing Unicode in identifiers is part of the specification of RDF Turtle syntax; I do not really like it much or recommend using that feature when it can be avoided, but it is there for completeness.)
03:59:02 <oren> Well, UTF-32 essentially reduces the complexity of reading a character to fread()ing 4 bytes
03:59:25 <pikhq> But why are you reading 'a character'?
03:59:35 <zzo38> oren: Yes, if that is what you need to do; like I said it depend on the software. I find that is rarely necessary anyways.
03:59:38 <oren> and allows to seek N characters through a string by adding n*4
03:59:51 <pikhq> Also, not really because a character is not 4 bytes.
03:59:52 <zzo38> Reading bytes tends to work better, and is compatible with ASCII.
04:00:02 <pikhq> A Unicode character is one or more codepoints.
04:01:00 <oren> pikhq: for eaxample?
04:01:05 <zzo38> Unicode has a lot of stupid stuff, really.
04:01:57 <pikhq> Consider LATIN CAPITAL LETTER A + COMBINING GRAVE ACCENT.
04:01:59 <zzo38> For my own stuff I generally prefer PC character set.
04:02:02 <pikhq> That is a single character.
04:02:09 <pikhq> It is two codepoints.
04:02:10 <oren> Are you claiming that formally, the combining characters are not characters?
04:02:58 <pikhq> Essentially. Or, at least, if you treat them as "characters" you're dealing with a very different concept of "character" than the end user is likely to.
04:03:26 <oren> For practical purposes, I have no toruble selecting a combining character and pasting it somewhere...
04:04:15 <pikhq> That's not "two characters" to most users that's "an A with a `" and if you treat it as an A followed by something that modifies it into an "A with a `" you're gonna have some fucking weird behavior.
04:04:27 <Sgeo> Does "right arrow" move by one character or one codepoint?
04:04:56 <zzo38> That's stupid stuff in Unicode.
04:05:17 <pikhq> zzo38: Sure, it would be vastly easier if they didn't put in any of the combining stuff.
04:05:31 <pikhq> This is a legacy of them trying to fit everything in 16 bits.
04:05:40 <zzo38> Even then, there is much stupid stuff remaining.
04:05:49 <oren> `unidecode y ̄
04:05:50 <HackEgo> [U+0079 LATIN SMALL LETTER Y] [U+0020 SPACE] [U+0304 COMBINING MACRON]
04:06:22 <pikhq> If you didn't have the space there the macron would be above the y.
04:07:02 <oren> Right, and I find it annoying that I can't put my cursor between them.
04:07:17 <zzo38> oren: Yes I agree too
04:07:24 <zzo38> That should be a property of the font metrics and not of the character set anyways.
04:07:28 <pikhq> You're gonna find similar stuff with decomposed hangul. :)
04:08:05 <zzo38> Complex scripts, combining characters, text direction, ligatures, etc all of that should be defined in the font metrics only.
04:08:20 <pikhq> (hint, nobody things of the individual jamo as "characters", but they can totally be represented as individual codepoints!)
04:08:22 <zzo38> And then they can use whatever character encodings you want.
04:08:58 <oren> right, it will be a problem if Korean user types a word, gets the last jamo wrong by typo and can't adjust it by simply pressing backspace
04:09:28 <zzo38> Yes, it is right it is the problem
04:12:03 -!- KnightArm0 has quit (Ping timeout: 264 seconds).
04:14:48 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
04:17:34 * Lyka makes a daily system backup
04:18:50 <oren> Irssi appears to do what I proposed: treat combining characters like other characters. if I type ñ and then backspace, it deletes the tilde but not the n
04:20:04 <oren> however, I can't put by cursor inside ñ. So it does half of what I think it should do
04:20:36 <pikhq> Now find a Spanish speaker that doesn't know about Unicode and see if that behavior is what they think it should be.
04:22:08 <oren> Right, but that is the fault of unicode by making the wrong behaviour easy to implement
04:22:28 <zzo38> Use the precomposed character then if you are Spanish like that
04:22:51 <pikhq> zzo38: Unfortunately for you, OS X prefers decomposed characters. :)
04:23:15 <zzo38> There are at least 100 problems with Unicode, I think..............
04:23:30 <pikhq> And it's still the least awful solution.
04:24:21 <zzo38> No it isn't. What TeX does is better
04:24:37 <pikhq> Uh, doesn't TeX not even support multilingual code that well?
04:24:49 <pikhq> Erm, multilingual text
04:25:21 <zzo38> Actuall TeX does have stuff for multilingual text, you can use different hyphenation pattern for each language in one file, and you can make the font for different language text too
04:26:35 <zzo38> DVI can even use 32-bit character numbers, and so can METAFONT
04:26:53 <zzo38> So it is more than Unicode and it mean you can even do made-up languages
04:29:29 <Lyka> so i am gonna try and figure out how a six-character or 8-character version of fourfuck would work
04:30:30 <Lyka> as, like i probably said, the arduino is 16-bit, why am i making an 8-bit language?
04:32:37 <ais523> someone writes a JIT for BF and posts it to reddit, and Urban Müller (or someone impersonating him) turns up to say hi
04:33:07 <zzo38> ais523: I don't know which one; ask Muller
04:33:09 <ais523> could easily be an impersonator, actually
04:33:38 <zzo38> Do you like my "deanonymizing operator" and "cons operator" extensions I added into RDF Turtle syntax?
04:34:46 <zzo38> For example now (1 2 3 4) and (1,(2,(3,(4,())))) is same thing.
04:36:38 -!- Lyka has changed nick to Lyka|Away.
04:36:57 <zzo38> Or you can make a "incomplete list" by (1 2 3 4,)
04:37:28 <zzo38> And a list can "loop" such as (!_:x 1 2 3 4, _:x)
04:38:16 <zzo38> (You don't have to loop from the beginning; you can loop from in the middle too)
04:59:50 <zzo38> With pure tokenizing highlighters it looks difficult to properly highlight IRC because I do not see how you can highlight the command name (or number) in this way. Well, maybe if you are using regular expressions that support looking behind you at what has already been parsed, then maybe you can, but I am not sure.
05:03:10 -!- Wright has quit (Ping timeout: 272 seconds).
05:06:57 -!- MDude has changed nick to MDream.
05:24:03 -!- Sprocklem has quit (Quit: [).
05:25:12 -!- rdococ has joined.
05:44:22 <rdococ> can you do loop like stuff in a mathematical language with assignment but not loops? I know I'm being really vague, but - say I have multiplication - can I, without loops, do exponentation by a variable value?
05:44:42 <zzo38> I think you are too vague
05:46:22 <rdococ> okay-- can I - without a built-in exponentation function - without using any loops - and only using numbers and enumerable types - do exponentation by a variable value?
05:54:07 <ais523> do you have logarithms?
05:54:18 <ais523> or, well, you still need /one/ exponentiation for that to work
05:54:30 <ais523> but in general, no because of computational complexity problems
05:54:50 <ais523> you can't produce a number of size O(2^n) in time O(1) without some operation that grows numbers exponentially
05:55:00 <ais523> and the lack of loops means you can't write programs slower than O(1)
05:55:22 <ais523> *grows numbers exponentially or faster
05:55:36 -!- password2 has quit (Ping timeout: 252 seconds).
06:05:03 <rdococ> what I decided to do is add unconditional goto
06:05:38 <rdococ> (since everything is a number you can do logic with 0's and 1's)
06:05:44 <rdococ> (in the language I'm making of course)
06:06:06 <izabera> http://iq0.com/notes/deep.nesting.html do you guys know this?
06:10:24 <rdococ> or c++ or whatever or c with pants
06:12:43 <oren> I've always done it that way with multiple returns
06:13:18 <zzo38> Maybe I will read it later
06:16:29 <oren> oh wow this is longer, it goes into many ways to simplify code
06:24:07 -!- Lyka|Away has changed nick to Lyka.
06:24:20 <Lyka> https://dl.dropboxusercontent.com/u/98841263/fourfuck%20language/Fourfuck%20Octopus-0000.pdf
06:24:42 <Lyka> (the octopus is just a random word i picked)
06:25:23 <Lyka> it's a different approach than the one i had been planning
06:25:59 <Lyka> i had to modify some commands to allow for optional 16-bit arithmetic
06:28:44 <Lyka> also, i looked at the backlog. nobody seems to say goodnight here.
06:31:16 <oren> I usually fall asleep with no warning
06:33:09 <zzo38> I do not find "goodnight" necessary on computer much
06:42:30 <zzo38> In Windows you can still count how many lines in a file: find /v /c ""
06:43:06 <Lyka> i'm gonna get something to munch on. maybe it will help me pass out
06:46:32 <rdococ> forget it I'm making a BF derivative -.-
06:47:21 <zzo38> Is it common to use a goto command in parsers more than in other programs?
06:48:22 <ais523> or to generalize, heavy use of goto is common in state machines
06:48:28 <ais523> and parsers are one of the main applications of state machines
06:50:50 <quintopia> i don't know why careful use of goto in other settings has not come back into vogue since "'GOTO considered harmful' considered harmful" came out. It seemed a pretty good argument to me.
06:51:45 -!- hilquias` has quit (Ping timeout: 265 seconds).
06:52:55 <Lyka> GOTO is not harmful?
06:53:14 <zzo38> GOTO is not necessarily harmful.
06:58:40 <Lyka> in c/c++, is this possible: int main(){f1();return 0;} void f1(){f2();} void f2(){f1();}
06:59:41 -!- variable has joined.
06:59:46 <Lyka> i mean, that sort of thing
07:01:01 <zzo38> If you declare the functions at first then you can do like that, although such thing isn't going to work.
07:01:11 <zzo38> It will just run out of stack space
07:01:31 <Lyka> i wouldn't do that program, of course
07:01:31 <zzo38> Either that or get into an infinte loop without using any stack space
07:01:35 <zzo38> This depends on the compiler
07:03:06 <zzo38> I think the only time I ever used setjmp in a C code is this: while(setjmp(exception_buffer)); (I do mean exactly that, including the semicolon)
07:03:41 <Lyka> the R--- set of commands in my language reads 4 chars and runs them as if it were a command
07:03:45 <ais523> zzo38: I normally put the semicolon on the next line when doing that
07:04:16 <zzo38> I wasn't sure whether or not it is allowed to reuse the same buffer, and got conflicting answers, so I did it like that rather than setjmp by itself.
07:04:19 <ais523> also a) I think it's equivalent to setjmp without the loop (/unless/ you're copying exception_buffer elsewhere and then jumping from the new location); b) if you want to be compatible with old versions of C you should compare to 0 explicitly
07:04:23 <Lyka> so it's menu --- run_commands --- menu ---
07:05:09 <Lyka> wel, it compiles for the arduino, but i have not yet tested it
07:05:09 -!- variable has changed nick to trout.
07:06:11 <HackEgo> [wiki] [[Talk:Folder]] http://esolangs.zem.fi/w/index.php?diff=43052&oldid=42543 * Rdococ * (+154) /* Name dispute */
07:06:31 <zzo38> ais523: O, you put in next line, OK, but in this case I am not copying the buffer at all; see above why I did that
07:07:00 <zzo38> (I have other times too used a while loop with no body, but this is the only time I used setjmp at all)
07:08:06 <oren> Ok so I just learned something. In addition to chmod there is a command chattr which, rather than being a chat program, is used to set extra attributes to ext2 files.
07:08:40 <zzo38> What extra attributes is it?
07:08:46 <oren> So if you don't know about chattr, and there is a file which has been chattr +i on it
07:08:52 <oren> you can't delete it
07:09:15 <oren> i for immutable
07:11:12 <oren> a file with i in its lsattr output can't be modified in any way, even by root
07:11:36 <oren> and only root can take off the i with chattr -i file
07:12:45 <zzo38> OK, now I know that
07:19:07 <Lyka> sorry i talk a lot about the language i am making due to having nothing better to do.
07:19:50 <Lyka> i'm thinking out loud, but am willing to hear comments
07:21:10 <zzo38> That is fine with me
07:22:06 <Lyka> seems like i keep kiling chat whenever i talk about octopus ( octopus is the clean alias of fourfuck's crrent development branch )
07:22:19 <zzo38> There is no point posting all of your thinking about it on here unless you are willing to hear comments, which you are, so it is OK
07:23:57 <zzo38> People that discuss other thing can do so it doesn't seem a problem to read different set of message mixed up in this way (it occurs even in books), but if someone does have a problem it is possible to program the computer to temporarily suppress such messages if they are interfering with the others.
07:26:17 <Lyka> if someone needs the channel's soapbox and i am aware, of course i will yield to the while they are on it
07:26:47 -!- ais523 has quit.
07:30:40 <zzo38> I thought you wanted to talk a lot about the language you are making due to you have nothing better to do; so, if you have something to write, do so.
07:30:59 <zzo38> Even if nobody comments right now, it would be logged
07:31:42 <Lyka> i can only say so much
07:32:27 <Lyka> i am making the language due to having nothing better to do
07:32:47 <Lyka> i talk about th language to get feedback
07:33:01 <Lyka> sorry if there was confusion
07:33:25 -!- Phantom_Hoover has joined.
07:35:48 -!- e_svedang has joined.
07:37:04 -!- trout has quit (Ping timeout: 264 seconds).
07:37:32 <zzo38> Well, post here; if I have any comment I may write some, or possibly someone else might. However it might come late but that's OK because we have logs
07:37:48 <zzo38> rdococ: What kind of epic idea is that?
07:38:22 <rdococ> an idea for an esolang
07:39:01 <rdococ> I wont spoil the details but it's going to be called And.
07:41:03 <Lyka> https://dl.dropboxusercontent.com/u/98841263/octopus%20language/Octopus-0000.pdf
07:41:20 <Lyka> https://dl.dropboxusercontent.com/u/98841263/octopus%20language/octopus_0000a.ino.txt
07:41:40 <Lyka> i should go to sleep soon
07:42:27 <Lyka> since chat's logged and stuff, you don't have to wait for me to wake up to comment
07:44:37 -!- Lyka has changed nick to Lyka|Away.
07:44:54 <zzo38> What is all that stuff at the top for?
07:59:45 <rdococ> it is at least as good as a push down automaton
08:04:53 <fizzie> @tell oerjan I'm not surprised. Anyway, the real domain seems to be back, so I undid the MediaWiki changes.
08:09:56 -!- Lyka|Away has changed nick to Lyka.
08:12:55 <Lyka> zzo38: what stuff?
08:12:59 <HackEgo> [wiki] [[And]] N http://esolangs.org/w/index.php?oldid=43053 * Rdococ * (+1708) /* And */finished typing it for now
08:13:46 <HackEgo> [wiki] [[And]] http://esolangs.org/w/index.php?diff=43054&oldid=43053 * Rdococ * (+6) /* Cat Program */
08:14:45 <HackEgo> [wiki] [[And]] http://esolangs.org/w/index.php?diff=43055&oldid=43054 * Rdococ * (+2) ...
08:15:06 <zzo38> All of the const char cx00[17] PROGMEM = " "; and so on
08:15:39 <Lyka> it's an arduino thing
08:15:49 <Lyka> cxarray is in flash memory
08:18:53 <Lyka> cxarray is part of the program, but never copied to ram
08:20:40 <Lyka> program as in the interpreter setch
08:23:09 <Lyka> zzo38: does this make sense to you now?
08:26:28 <Lyka> that cxarray[256] is a array of 256 17-byte char strings?
08:28:56 <Lyka> well, pointers to 17-byte char strings
08:28:57 <mroman_> are morphisms that are both epimorphisms and monomorphisms isomorphisms?
08:31:04 -!- Patashu has joined.
08:32:45 <zzo38> Lyka: I know that, but among other things they are all const and all spaces too
08:33:05 <mroman_> are there any other endomorphisms other than the identity function?
08:33:34 <mroman_> seeing as they are f: X -> X
08:35:05 <Lyka> zzo38: cause i haven't filled them in yet?
08:36:04 <Lyka> did i not say that cxarray is read-only?
08:36:11 <zzo38> Lyka: O, that's why, OK
08:36:43 <zzo38> I know it is read-only; you told me that and I can see it from the program too, but I don't know much about Arduino stuff or about your program, therefore it seem strange to me at first.
08:36:52 <mroman_> or is f(x) = -x and endomorphism as well?
08:37:05 <zzo38> mroman_: Yes there are other endomorphisms
08:39:09 <zzo38> Although it depends what category.
08:39:13 <zzo38> In some categories there aren't any
08:39:14 <shachaf> mroman_: http://ncatlab.org/nlab/show/bimorphism
08:41:47 <zzo38> mroman_: In that example though, if you have f: Integer->Integer then yes it can be
08:44:20 <zzo38> shachaf: I just changed that mage a few seconds ago
08:44:33 <zzo38> Because it said "Contemts" by mistake instead of "Contents"
08:44:46 <oren> is there a standard for naming of mp3's
08:45:30 <oren> should the filenames be Artist - Album - Song.mp3 or the reverse?
08:45:45 <zzo38> oren: I think that depends what you want to sort by
08:46:07 <zzo38> (Also MP3 isn't very good; FLAC is much better)
08:46:33 <oren> well yeah... mp3's, flacs, wav's etc
08:47:10 <oren> the extension isn't the point, that already has a standard :)
08:49:52 <zzo38> Although arranging the stuff in that filename in my opinion mostly depend how you want to sort by; I prefer filenames without spaces though. However you might also like to organize in a SQLite database then you can easy sort by whatever you want to
08:51:48 <oren> A database is a good idea. Or if there is a separator in each filename like Artist-Song then we can sort using sort -t- -k2 or the like
08:52:34 <zzo38> Yes, that is another way
09:01:45 <zzo38> And can help even if you do have a database, perhaps
09:02:35 <zzo38> Although - might not be best one depending on if the artist's name has a hyphen, too (unless you change it for purposes of the filename, which is also possible; the correct name can be placed into the database instead).
09:29:16 -!- rdococ has quit (Ping timeout: 246 seconds).
09:52:35 -!- e_svedang has quit (Ping timeout: 272 seconds).
09:54:37 <mroman_> oren: Artist/Album/Song.mp3 ;)
10:05:13 <b_jonas> nah, just don't put literal names of artists or albums or songs in filenames, beacuse those strings can be crazy
10:05:50 <b_jonas> you can put sanitized versions of them in the filename if you wish, as long as you keep them short, not have strange characters, and make sure the filename is unique.
10:06:10 <b_jonas> Keep the actual names somewhere else, such as in the file metadata or in separate files, whether text or database or whatever.
10:06:45 <b_jonas> Making the filenames sort in the way you want to play them is a good idea though.
10:06:58 <b_jonas> Helps playing whole album sorted on mobile phone.
10:13:03 <FireFly> Sure, they should be in the metadata as well
10:13:36 <FireFly> as long as the filenames are descriptive, I'm happy
10:18:40 -!- oerjan has joined.
10:19:24 <lambdabot> fizzie said 2h 14m 30s ago: I'm not surprised. Anyway, the real domain seems to be back, so I undid the MediaWiki changes.
10:21:12 <fungot> mroman_: mr president, on a permanent basis, developing employability, seeking alternatives to closures, then the commission would like to reiterate my thanks for the comments made today by the rapporteur, per gahrton, and now we are of the opinion of the wto that it is clear that the european parliament has new opportunities, for example, we have reached a consensus, mostly with countries with which we have acquired a living c
10:21:22 <fungot> Selected style: irc (IRC logs of freenode/#esoteric, freenode/#scheme and ircnet/#douglasadams)
10:21:25 <fungot> mroman_: i'm researching scheme for a while until this thing clears up. maybe as a list
10:25:12 -!- boily has joined.
10:26:15 <oerjan> fungot: are you using natural clarification or some chemical?
10:26:15 <fungot> oerjan: by a simple macro; what advantage does being able to solve this
10:26:33 <oren> Alternatives to closures?
10:26:58 <oerjan> "Many substances have historically been used as fining agents, including dried blood powder,[5] but today there are two general types of fining agents — organic compounds and solid/mineral materials."
10:27:54 <oerjan> "Because potassium ferrocyanide may form hydrogen cyanide its use is highly regulated and, in many wine producing countries, illegal."
10:28:33 <Taneb> oerjan, is Potassium Ferrocyanide an alternative to closures?
10:29:25 <oren> crap. I have no less than four copies of the same song. mtimes: 2006 2011 2013 2014
10:29:35 <oerjan> Taneb: i dunno; my TC proof for potassium ferrocyanide is not yet very far
10:31:42 <oren> I guess I really like E-Type - Life.mp3, aka etype life.mp3 aka life by etype.mp3 aka Etype: life (CD ver.).mp3
10:35:41 <oren> Hey E-Type is from sweden! I didn't know that!
10:36:08 <oerjan> a surprising number of people are swedes
10:37:41 <Taneb> The winner of this year's Eurovision Song Contest, for example
10:38:05 * oerjan didn't watch, as usual
10:38:07 <Taneb> And at least two people in this very IRC channel!
10:38:29 <Taneb> FireFly, my statement is still true!
10:38:44 <FireFly> That statement of yours is /also/ true
10:38:51 <oren> FireFly has established a better bound on it though
10:39:05 <FireFly> You are very truthful today, Taneb
10:39:22 <Taneb> oren, did you know that Graham's number is at least 11!
10:39:24 <boily> hellørjan. you're all having a very fungottian conversation this morning.
10:39:24 <fungot> boily: http://java.sun.com/ j2se/ 1.4.2/ docs/ latest/ html/ r5rs-z-h-7.html%_idx_138 put
10:39:31 <Taneb> It might even be more than 12!
10:39:44 <boily> fungot: no, no java yet for me, at least for the next two hours.
10:39:44 <fungot> boily: you should focus on the feature differnce, not the
10:39:50 <oerjan> boily: fungot seems to have gone into the wine business
10:39:51 <fungot> oerjan: i might as well as the type-checking and deconstruction. the pre-scheme compiler. http://www.bloodandcoffee.net/ campbell/ txt/ fnord and this:
10:40:10 <oerjan> boily: although e still manages to use scheme for it
10:40:47 <fungot> FireFly: fnord big lexical factor was miranda a couple of hours
10:40:56 <boily> Château de Scheme 2015. Type checked and well rounded.
10:41:27 <oerjan> Taneb: are you referring to graham's humongous number or to what he was actually approximating
10:41:39 <oren> I have three copies of Masterboy - Show Me Colours.mp3
10:41:41 <FireFly> Speaking of Eurovision, I think I heard an E-type song in our ESC qualifier competition once
10:42:03 <oerjan> because with those things, the lower bounds are ridiculously lower than the upper ones, it seems
10:42:25 <Taneb> At the time, the lower bound was 6 and the upper bound was Graham's number
10:42:44 <Taneb> Now the lower bound is 13 and the upper bound is the Hales-Jewett number
10:43:15 <oerjan> FireFly: don't worry, he's just using the blood and coffee as a fining agent hth
10:43:19 <oren> Is there a list of songs that have won somewhere
10:43:49 <oren> oh wiki has it
10:43:54 <oerjan> his wine will be a hit with the sleepless vampire segment
10:45:41 <oerjan> fungot: wait, are you male or female
10:45:41 <fungot> oerjan: if you look at the source. when you try to say something
10:45:50 <fungot> https://github.com/fis/fungot/blob/master/fungot.b98
10:47:00 <Taneb> oerjan, I do not believe fungot has a gender
10:47:00 <fungot> Taneb: i just threw away a lot, not use the site rules as a way to transmit information!
10:48:16 <boily> fungot is a fungot.
10:48:16 <fungot> boily: " metamodel"?? haha.
10:48:41 <oerjan> fungot: i found your source unenlightening on the matter tdnh
10:48:41 <fungot> oerjan: that's already there.))
10:48:59 <boily> fungot: flblblblblbl!
10:48:59 <fungot> boily: " write search terms in box. click search.") well, i
10:50:23 <oren> if fungot is hebrew it could be plural male
10:50:23 <fungot> oren: so you want to))
10:52:54 <oren> because -ot is how you make plural on male nouns apparently
10:53:21 <boily> shellochaf. expert advise on fungot's gender?
10:53:22 <fungot> boily: while foo fnord. _) a _, _ cannot be evaluated by the outer procedure. you then remove x1 from the set of polynomials is dense in the space
10:53:35 <boily> (or genders, because apparently the 'got is plural.)
10:53:50 <boily> Taneb: as a gender pluralist, any idea on fungot?
10:53:50 <fungot> boily: make the keys those small rubbery things
10:54:20 <Taneb> fungot, what would you say your gender or genders are?
10:54:20 <fungot> Taneb: ( scariest thing i could see some use for it. :p
10:54:50 <Taneb> boily, fungot is unnerved by the concept of gender, and is terrified that they can see some use for it
10:54:51 <fungot> Taneb: i think the avoidance is just for nomic-sh. i'm not set at all on that page are funny.
10:56:15 <Taneb> boily, I think that says it all
10:59:43 <boily> Tanebs Are Almost About ABCs?
10:59:56 <Taneb> These acronyms are a bit confusing
11:00:26 <Taneb> I am fairly sure there is only one Taneb
11:00:40 <Taneb> I mean there used to be a racehorse called Taneb back in the 20's or something
11:01:36 <Taneb> Phantom_Hoover, shared account with my brother
11:01:49 <Taneb> At the time he had the nickname "Neb" because it was Ben backwards
11:02:11 <Taneb> I tacked the first three letters (reversed) of my name onto that, "Tan"
11:03:27 <boily> I Don't See Why It Could Also Abort.
11:03:39 <Taneb> I do sometimes wish I could articulate acronyms
11:03:57 <oerjan> Taneb: not such a good start
11:08:04 <oerjan> "I don't see what is confusing about acronyms" hth
11:08:04 <Taneb> I have an exam in 6 hours about computability and complexity
11:08:19 <Taneb> Finally putting my years here to good use
11:08:27 <oerjan> Taneb: look at the bright side, it's pretty sure to terminate
11:09:07 <Taneb> It's a 90 minute exam, I should hope so
11:09:17 <Taneb> Jeez, I've been here for like 5 years
11:09:35 <Phantom_Hoover> Taneb, ah yes, i had the same experience when i took automata and formal systems last year
11:09:46 <oerjan> ...so why do i keep thinking of you as a newbie...
11:09:55 <Taneb> oerjan, because I'm like half your age
11:10:44 <b_jonas> oerjan: because he sometimes asks beginner questions here?
11:10:51 <Taneb> And also I kind of feel like a newbie wherever I go
11:10:55 <Taneb> It's an experience I enjoy
11:11:40 <Taneb> I sometimes get lost for fun
11:11:50 <Taneb> It's oddly liberating
11:11:59 <Taneb> ...I may be an odd person
11:12:47 <boily> wait wait wait. oerjan is twice Taneb?
11:12:57 <Taneb> boily, roughly, I think
11:13:01 <Taneb> oerjan, how old are you?
11:13:12 <oerjan> 44, for about another month
11:13:26 <Taneb> oerjan is 2*Taneb + 4
11:13:49 <oerjan> so in two years you can stop being a newbie
11:14:47 <Taneb> By then I may have almost graduated!
11:15:07 <mroman_> why again did I need inlining for my static typed stackbased programming language
11:15:12 <b_jonas> oh, the esolangs.org wiki is up again at the original address, great
11:15:31 <oerjan> mroman_: performance hth
11:15:44 <mroman_> it had something to do with it not type checking unless inlined
11:16:05 <oerjan> b_jonas: technically i never noticed it stop working, although its CSS got screwy because of fizzie's temporary workarounds
11:16:17 <mroman_> http://codepad.org/DtRAbW0q
11:16:39 <mroman_> probably because you could not assign a unique type to the function "mother"
11:16:49 <mroman_> because it would accept both hans susi and hans fritz
11:17:16 <oerjan> mroman_: oh right, statically typed stack is tricky when you have eval-like commands, i remember the old CAT discussions
11:17:29 <oerjan> i don't think he ever completely solved it
11:18:01 <oerjan> b_jonas: i did notice the discussion thank you very much
11:18:12 <oren> Ok, I have like ten songs under both DJ S3RL - song name and S3RL - song name
11:18:41 <oerjan> there's should be some ISSN
11:19:05 <mroman_> oerjan: I solved it by inlining
11:19:23 <mroman_> @= does not really define a new function
11:19:44 <mroman_> like #define mother parent female
11:19:53 <mroman_> because you can't assign mother a type
11:20:00 <mroman_> but you can use "parent female"
11:20:17 <b_jonas> I still don't really understand the scoping rules of metafont. It's confusing. Is there a good description of it somewhere?
11:20:26 -!- boily has quit (Quit: COTYLEDON CHICKEN).
11:20:39 <mroman_> http://codepad.org/bxNCwGbU <- that's the full code btw
11:20:40 <b_jonas> Like, when is a tag token used for its name, and when is it used for its current assigned meaning?
11:21:03 <b_jonas> Does the latter matter only when it's the first component of a variable name, or also when it's an index?
11:21:11 <oerjan> mroman_: you'll probably get problem with recursion, then...
11:21:30 <mroman_> some functions can't be recursive
11:22:18 <oerjan> in fact if you don't have recursion, hindley-milner types are exactly equivalent to simple LC types + pervasive inlining
11:22:33 <mroman_> Looks like I've implemented a "poly" flag
11:22:50 <b_jonas> `` mf <<<'\ show 2+2' # do we have mf available in the sandbox?
11:22:51 <HackEgo> bash: mf: command not found
11:22:55 <b_jonas> `` mpost <<<'\ show 2+2' # do we have mf available in the sandbox?
11:22:56 <HackEgo> bash: mpost: command not found
11:23:21 <oerjan> mroman_: do you have polymorphic types, if not that'd seem related (although not enough to solve typing in the presence of closures and eval)
11:24:22 <oren> In 2005 what *was* www.ginogina.ca?
11:24:57 <mroman_> oerjan: http://codepad.org/GM3Ik2p4
11:25:11 <mroman_> if I read my code correctly $- delays type checking until later
11:25:19 <oren> Supposedly I got a few songs from there that appear not to exist anymore. Hell the artists aren't showing up at all
11:25:30 <mroman_> but you have to specify every combination of acceptable types
11:25:45 <oerjan> mroman_: your language looks like a weird cross of stack and prolog/mercury
11:26:18 <mroman_> well you can do some prolog like stuff in it
11:26:25 <mroman_> if it type checks it means "true" :)
11:26:36 <oerjan> oh so that's all compile time?
11:27:02 <mroman_> http://codepad.org/oFsuRpEu
11:27:42 <oerjan> hm ISSN already means something
11:28:20 <mroman_> mind you that "true" and "false" are both types here
11:28:36 <mroman_> upper case letter are type placeholders
11:28:47 <oerjan> b_jonas: i think zzo38 knows metafont hth
11:29:08 <mroman_> (dup :- A -> A A; swap :- A B -> B A and so forth)
11:30:20 <b_jonas> mroman_: how do you define functions that invoke other functions?
11:30:53 <mroman_> := defines the function body
11:31:00 <oren> I can;t believe how hard it is to get info on something that existed only 10 years ago
11:31:16 <mroman_> or :- A true -> true; or := swap pop
11:31:26 <b_jonas> oren: yeah... some things are underdocumented on the internet
11:31:41 <mroman_> is like or :: a -> True -> True in Haskell
11:31:53 <mroman_> except Haskell doesn't really allow this
11:31:59 <mroman_> and in Haskell True isn't a Type
11:32:45 <oerjan> oren: http://web.archive.org/web/20051219014320/http://www.ginogina.ca/content.php?about
11:33:17 <mroman_> and I apparentely only implemented type checking so far
11:34:25 <oren> So I guess I must have had an account there and downloaded mp3's from links that were posted?
11:35:09 <oren> Or maybe my friends did and gave me the mp3's? who knows?
11:35:26 <mroman_> with :- and := the type checker will try to check the type of the function against the type you try to give it
11:35:51 <mroman_> with $- the type checker won't do that but only check the types in calls to that function
11:36:29 <mroman_> foo $- -> false; foo := true; would type check as long as you never call foo
11:36:56 <oerjan> mroman_: True can be a promoted type with the DataKinds extension hth
11:37:18 <oerjan> it doesn't have values though
11:37:56 <oren> looking at that web page makes me want to cry, the internet used to work fine without all this JQuery CSS3 HTML5 bullshit
11:38:16 <mroman_> mainly useful for things like mother := parent female where the function on the left can't be assigned "a single type"
11:38:34 <mroman_> which either has to be inlined through a macro or by defining it with $-
11:38:49 <mroman_> which tells the compiler to not type check mother := parent female
11:38:53 <mroman_> but type check calls to it
11:39:00 <mroman_> er.. *tells the type checker
11:41:16 <mroman_> oerjan: cat has a successor called kitten now
11:41:39 <mroman_> probably not by the same author though
11:44:26 <oerjan> btw i misspoke, hindler-milner without recursion is equivalent to duplicating let definitions, not inlining them
11:45:16 <oerjan> basically, you let each use site have its own type for a let-defined variable, but it still needs to have a type
11:47:06 <mroman_> e.g. if mother is used, you infer from the context what type it should have
11:47:51 <mroman_> what I could've done instead is that the type checker automatically creates overloaded versions of mother with all combinations of accepted types
11:49:39 <mroman_> I'm not sure if that works with recursion though
11:50:38 <oerjan> mroman_: kitten seems to be by evincar who was a regular here for a while, cat was by christopher diggins and my memory is vague on whether he came here or whether i just saw him elsewhere
11:51:08 <mroman_> It surely works if you can't overload the return type :)
11:52:13 <oerjan> and the cat language site seems to have vanished
11:55:40 <mroman_> I was just lazy to implement it
11:58:12 <oerjan> <pikhq> Frankly an RNG isn't at all worrying -- 122 actually random bits are basically *not* going to collide. <-- * imagines a far future in which civilization is destroyed by an unexpected hash collision
11:59:32 <mroman_> if(hash(time()) == 0xEAFFF44789ABCD17DBA) { /* start war on 6571-10-10 */ }
12:00:11 <mroman_> who had known that 6000-8-8 would also produce the same hash .
12:00:57 <mroman_> I guess it's semi-expected
12:02:09 <oerjan> i was imagining more like a future where we're all living as uploaded minds in computronium
12:02:47 <mroman_> and every individual is stored under the key hash(individual.dna)?
12:02:50 <oerjan> and the world computer uses hashes for security
12:03:05 <mroman_> somebody is up for being erased and replaced by somebody else
12:03:25 <mroman_> that way you know that your population won't grow infinitely
12:03:52 <oerjan> well the hashes were more than big enough when the system was designed, you see
12:03:55 <mroman_> someday a newborn cyberbaby will overwrite somebody
12:04:25 <mroman_> nobodys gonna use that not-enough condoms!
12:04:57 <mroman_> I don't know what you're plans about that are in the computronium.
12:05:11 <oerjan> in fact they still _seemed_ to big enough. nobody actually expected the hash collision.
12:05:57 <mroman_> but hash collisions are an inherent property of hashing
12:05:57 <oerjan> and it wasn't in something as mundane as a single person's hash, it was in a security proof for the fundamental OS
12:06:32 <mroman_> Doesn't git sorta have this problem?
12:06:47 <mroman_> I haven't read about what bad things will happen if a hash collides
12:07:14 <mroman_> I possibly have but already forgotten it.
12:07:14 <oerjan> yeah there are lots of places where it would be bad, surely
12:07:44 <mroman_> "I'll be already dead by then so who cares" - L. Torvalds
12:08:57 <oerjan> that can't be an actual quote, too polite
12:10:42 <oerjan> you are both welcome to prove me wrong with actual links
12:12:44 <mroman_> I'm not going to google "fuck l. torvalds"
12:13:00 <oren> https://www.youtube.com/watch?v=_36yNWw_07g
12:13:33 <oren> thats what came up, torvalds saying "Nvidia, fuck you"
12:14:59 <oren> https://lkml.org/lkml/2012/12/23/75 "Mauro, SHUT THE FUCK UP!" -- L. Torvalds
12:15:45 <HackEgo> [wiki] [[Microscript]] http://esolangs.org/w/index.php?diff=43056&oldid=42965 * SuperJedi224 * (+21) The imports disappeared.
12:15:55 <oerjan> oren: that was coherent, doesn't count hth
12:23:09 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
12:23:13 <oren> your compiler is pure
12:23:14 <oren> and utter *shit*.
12:23:27 <oren> -- L. Torvalds, to the GCC team
12:24:28 <oerjan> at least it's pure, has to count for something
12:25:24 -!- Patashu has quit (Ping timeout: 240 seconds).
12:25:52 <HackEgo> [wiki] [[Microscript]] http://esolangs.org/w/index.php?diff=43057&oldid=43056 * SuperJedi224 * (+0) Fixed an implementation error.
12:31:15 <oren> a search for linus torvalds shit led to
12:31:27 <oren> Mormonism as a religion is a fairly close second to the Scientologists in the race to "Batshit Crazy" -- L. Torvalds
12:33:59 <oren> . Hurd will be out in a year (or two, or next month, who knows), -- L. Torvalds 1991
12:34:28 <oren> I think it has a beta now, right?
12:37:15 <oren> In short: just say NO TO DRUGS, and maybe you won’t end up like the Hurd people. -- L. Torvalds 2001
12:42:00 <oren> If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce. -- L. Torvalds
12:43:24 <HackEgo> [wiki] [[Microscript]] http://esolangs.org/w/index.php?diff=43058&oldid=43057 * SuperJedi224 * (+89) /* Example programs */
13:01:11 -!- FreeFull has quit (Read error: Connection reset by peer).
13:02:24 -!- FreeFull has joined.
13:02:29 -!- FreeFull has quit (Changing host).
13:02:29 -!- FreeFull has joined.
13:08:37 <mroman_> arguably intel architecture is probably worse than it could be
13:11:09 <APic> And Cyrix was the Hell of a Mess.
13:11:46 -!- oerjan has quit (Quit: bring your elves).
13:14:46 -!- GeekDude has joined.
13:25:31 <mroman_> what's the opposite of altruism?
13:26:56 <fizzie> Antonyms of noun altruism
13:26:57 <fizzie> altruism, selflessness
13:26:58 <fizzie> Antonym of egoism (Sense 2)
13:27:01 <fizzie> =>egoism, egocentrism, self-interest, self-concern, self-centeredness
13:28:42 <fizzie> 1. malevolence, malignity -- (wishing evil to others)
13:28:42 <fizzie> 2. malevolence, malevolency, malice -- (the quality of threatening evil)
13:29:10 <mroman_> I'm looking for a word for "doing good without external motivation"
13:30:09 -!- `^_^v has joined.
13:30:40 <mroman_> that is, if such a pure thing exists
13:30:44 <mroman_> and isn't some form of compensation
13:31:17 <fizzie> "The opposite" is not really well-defined, since it's unclear whether the opposite should invert the "good" part, the "external motivation" part, or both.
13:49:09 <mroman_> doing good without egoistic benefits, doing evil without egoistic benefits.
13:50:30 <mroman_> kinda like where mother theresa is helping children just for the sake of helping them (altruistic)
13:50:46 <mroman_> father theresa is just robbing people for the sake of robbing them (....?)
13:53:24 <mroman_> or punching them in the face
13:53:29 <mroman_> just doing general evil of some sort
13:53:50 <mroman_> or polluting the environment
13:54:20 <Taneb> Pointlessly malicious?
13:54:27 <Taneb> I do not know if such a word exists...
13:54:49 <mroman_> If there's such a thing as pointlessly kind
13:54:56 <mroman_> then there must be a pointlessly malicous
13:55:04 <fizzie> Perhaps not, but "maltruistic" would sound good.
13:55:22 <fizzie> Even if the construction would make no sense.
13:55:57 <mroman_> People seem to think that you can do good without involving your ego
13:56:09 <mroman_> but refuse to believe that you can do evil with the same motivation
13:56:50 -!- Wright has joined.
13:57:53 <mroman_> but "maltruistic" would make a good word
13:59:11 <mroman_> although there are already some google results for it
13:59:31 <mroman_> and it seems to be defined as "fake altruism"
14:03:35 -!- Wright has quit (Excess Flood).
14:03:42 <mroman_> I'm also a terrible philosopher.
14:03:51 -!- Wright has joined.
14:04:34 <mroman_> fungot: Are you interested in Philosophy?
14:04:34 <fungot> mroman_: i've been thinking about fnord, other than gambit doesn't have a whois referral pointing at a dead-end position in life.
14:07:39 -!- TieSoul has joined.
14:12:36 -!- Wright has quit (Excess Flood).
14:13:08 -!- Wright has joined.
14:26:03 <mroman_> fungot: Do you know Peter Popoff?
14:26:03 <fungot> mroman_: any particular code you were reading was not the connotation i had a netgear first, but for ash it happens in other areas
14:31:53 -!- Sprocklem has joined.
14:38:10 -!- Weloxux has joined.
14:45:42 -!- hjulle has joined.
15:17:53 <fizzie> I liked the "whois referral pointing at a dead-end position in life" part.
15:18:00 <fizzie> Although it's perhaps a bit cruel.
15:18:13 <fizzie> fungot: You'll need to have some niceness programmed in you.
15:18:13 <fungot> fizzie: cannot open input file: invalid argument"
15:20:22 -!- KnightArm0 has joined.
15:21:46 <FireFly> fungot: I think people would appreciate if you refuted the argument instead of just dismissing it as invalid
15:21:46 <fungot> FireFly: looking at it... it has a typo. i'll fix that
15:22:24 <FireFly> oh no. fungot's growing self-sentient, fixing typos in its own source
15:33:47 <HackEgo> [wiki] [[And]] http://esolangs.org/w/index.php?diff=43059&oldid=43055 * Rdococ * (+886) /* Examples */
15:49:10 -!- Weloxux has quit (Quit: Leaving).
16:07:41 <Lyka> hopefully fungot is not synet.
16:07:42 <fungot> Lyka: 1-n words of mostly human language. i was thinking about that the next pixels according to the error somehow
16:09:41 -!- Sprocklem has quit (Read error: Connection reset by peer).
16:11:33 -!- KnightArm0 has quit (Remote host closed the connection).
16:13:11 -!- Sprocklem has joined.
16:19:36 -!- bb010g has quit (Quit: Connection closed for inactivity).
16:33:56 -!- rdococ has joined.
16:47:50 <zzo38> I remember I had a book once that described a chess variant known as "Emperor Wars". (I don't know if it may have been the only copy (it was made of plain paper, probably printed by computer or typewriter, did not mention any author's name or copyright notices, had hand-written corrections in it), and I don't know where it is now.)
16:59:22 <int-e> fungot: tell us more about those pixels
16:59:22 <fungot> int-e: unleash your creative side shine, while painting and marking your keyboard the way you put that in your average defun?
16:59:32 <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
17:02:26 <tswett> Hey, people who have [y] in their native languages!
17:02:36 <tswett> To me, an English speaker, [y] sounds like a cross between [u] and [i].
17:03:30 <tswett> Would you say the same?
17:32:53 <J_Arcane> tswett: speaking from Finnish, I wouldn't quite say that.
17:35:11 <nortti> y and i are articulated in the same place, and the height and roundedness of y and u is the same
17:47:47 -!- rdococ has quit (Quit: Page closed).
18:32:11 <Lyka> https://dl.dropboxusercontent.com/u/98841263/octopus%20language/octopus_0000b%20commands.pdf
18:33:01 <Lyka> https://dl.dropboxusercontent.com/u/98841263/octopus%20language/octopus_0000b.zip (the arduino source code. a *.ino is ascii text)
18:35:32 -!- lleu has joined.
18:35:32 -!- lleu has quit (Changing host).
18:35:32 -!- lleu has joined.
18:59:55 -!- hjulle has quit (Ping timeout: 256 seconds).
19:08:53 -!- bb010g has joined.
19:09:42 <Lyka> crap...have to double the amount of cache variables
19:11:15 -!- lleu has quit (Ping timeout: 256 seconds).
19:12:10 -!- oerjan has joined.
19:26:14 <fizzie> tswett: At least I'd put [y] "between" [u] and [i] if I had to sort them on a "scale". And also what nortti said.
19:28:25 <fizzie> oerjan: The problem you mentioned was kind of a known one, in the sense that I set up the backup address knowing people using https:// would have certificate problems, and sort of vaguely expecting it might cause something like that even when the regular domain name returned. Anyway, things should be back to normal now.
19:29:40 <fizzie> (I could've generated an esolangs.zem.fi certificate and have it serve that to anyone using the backup address and supporting SNI, but that felt like quite a hassle for up-for-less-than-a-day workaround.)
19:30:07 <oerjan> as a norwegian, i'd put [y] between [ʉ] and [i]
19:31:08 <oerjan> and [ʉ] between [y] and [u]
19:33:42 <oerjan> and i suspect swedes do the same
19:44:32 <mroman_> fungot: so... how's that Philosophy going?
19:44:32 <fungot> mroman_: you can get a correct result, namely ( 3 ( l 3 i 1)
19:51:48 <fizzie> Apparently it's solved.
19:54:52 <oren> So I think the standard should be Artist - [Album -] Song [feat. Vocalist] [(DJ Blah mix)] [(from Game)].mp3
19:55:20 <oren> where [] indicates that it might not be there
19:58:38 <oren> I'm not sure if that will eliminate all conflicts, but it should come close
20:07:12 <fizzie> I doubt that'd work for classical music very well.
20:07:39 <fizzie> Although I guess you can stuff multiple things in the 'Artist' and 'Song' fields.
20:08:50 <fizzie> It also doesn't really disambiguate between different remasterings of the same song. Unless the "DJ Blah mix" field counts for that, too.
20:09:29 <oren> I guess you could put in (Radio Mix) or whatever
20:10:27 -!- hilquias has joined.
20:11:14 <oren> right now I'm just manually changing things to roughly that format... I'm not sure how to account for songs that I literally have zero information... the mp3 has no info and is named the url of a long deleted youtube video
20:11:35 <fizzie> Play it back and run SoundHound on it.
20:11:45 -!- Sprocklem has quit (Quit: /me makes emself sparse for a while).
20:12:33 <fizzie> I'd say "run Google Music Search on it" since I kind of (very tangentially) work on that thing, but I can't honestly recommend it all that much -- it didn't even recognize this year's Eurovision songs! (SoundHound did.)
20:19:27 <oren> Oh geez... now that I'm playing it. it's a nightcored version of whatever the hell it is
20:21:15 <oren> so I guess I can put it on my 3ds, open soundcloud on the samsung, and play it at various speeds until it recognizes it?
20:21:58 <oren> (The Nintendo 3DS has the ability to adjest the speed of a song with the stylus)
20:26:47 -!- Sprocklem has joined.
20:30:28 <oren> Hm... and searching for lyrics "jungle jungle jungle jungle jungle jungle jungle jungle jungle" doesn't work either
20:37:01 <oren> It's a song from the Bollywood horror film "Agyaat"
20:38:10 <HackEgo> [wiki] [[And]] http://esolangs.org/w/index.php?diff=43060&oldid=43059 * Rdococ * (+18) edited examples
20:38:51 <HackEgo> [wiki] [[And]] http://esolangs.org/w/index.php?diff=43061&oldid=43060 * Rdococ * (-3) /* 99 bottles of beer */
20:39:00 -!- Koen_ has joined.
20:51:29 -!- hjulle has joined.
20:57:07 -!- ZombieAlive has joined.
20:57:33 -!- Phantom_Hoover has joined.
21:03:57 -!- Patashu has joined.
21:20:17 -!- Sprocklem has quit (Ping timeout: 256 seconds).
21:23:01 <Phantom_Hoover> man why did i avoid reading about homotopy type theory for so long
21:23:53 <oerjan> oh no they got Phantom_Hoover
21:25:43 -!- `^_^v has quit (Ping timeout: 276 seconds).
21:25:56 <shachaf> oerjan: that was pretty predictable hth
21:26:07 <shachaf> my oerjan simulator is complete
21:26:08 <oerjan> shachaf: i thought so too
21:26:58 -!- Patashu has quit (Ping timeout: 265 seconds).
21:27:24 <shachaf> oerjan: why do you hate hott
21:30:06 <Phantom_Hoover> oerjan, do you not then know why there are so many fucking polynomials
21:31:06 <Koen_> when I was in high school I thought polynomials all looked either like x^2 or x^3
21:31:22 <Taneb> I think I have done quite well on my computability and complexity exam
21:31:34 <Taneb> Phantom_Hoover, you can have polynomials over any semiring
21:31:40 <Taneb> There are a lot of semirings
21:31:44 <Taneb> Hence a lot of polynomials
21:31:50 <Koen_> then one day I realized x(x-1)(x-2) had three roots and you could not have three roots while looking like x2 or x3
21:32:07 <Phantom_Hoover> Taneb, yes but that doesn't explain why they're all tied up in knots
21:32:18 <Koen_> I've never believed in santa claus so I guess that realization was kind of the big thing
21:34:45 <oerjan> the three roots of santa claus
21:41:32 <oren> Koen: Most of them do in physics
21:42:27 <Koen_> believe in santa claus?
21:42:35 <Koen_> yeah physicists have all kind of weird beliefs
21:42:54 <HackEgo> [wiki] [[List of ideas]] M http://esolangs.org/w/index.php?diff=43062&oldid=43035 * Rdococ * (+144) /* Game */lol
21:45:26 -!- ZombieAlive has quit (Remote host closed the connection).
21:47:02 <oerjan> Phantom_Hoover: it's the followers - their empty eyes collapsed into a point
21:51:32 <Phantom_Hoover> otoh you get articles from them talking about how hard it's been to define the n-sphere or prove that the torus is the product of two circles
21:51:52 <Phantom_Hoover> and it does not inspire confidence in its practicality
21:52:43 <Koen_> the very straightforward parametric equation of the torus is literally a product of two circles
21:52:49 <Koen_> how is that hard to prove?
21:53:42 <oerjan> presumably homotopists don't accept that definition
21:54:39 <Koen_> is it too constructivistish? :(
21:54:57 <Phantom_Hoover> you define the torus in hott in terms of its homotopies
21:55:45 <Koen_> so I guess what they meant is "it's been hard to prove that our definition of the torus defines in fact a torus"
21:55:59 <Phantom_Hoover> so it's a point a, two loops p and q from a to a, and a homotopy h from q.p to p.q
21:58:57 -!- variable has joined.
22:00:35 <Taneb> Hmm, I am a little annoyed
22:00:54 <Lyka> was that you who helped out with the fourfuck page?
22:01:39 <Taneb> Phantom_Hoover, because people are complaining that the exam I just did was way too hard
22:01:52 <Taneb> It was a little more difficult than the past papers, sure, but not that much more
22:02:35 <b_jonas> Phantom_Hoover: no. computational complexity exam or some such thing I believe.
22:02:58 <Taneb> Phantom_Hoover, computability and complexity, so, yes, #esoteric
22:03:24 <Taneb> Questions involved "Is n^n in O(n!)? (3 marks)"
22:03:45 -!- hilquias has quit (Remote host closed the connection).
22:03:56 <Taneb> Phantom_Hoover, the answer is "no
22:04:34 <Taneb> I am not sure what Simpson's theorem is
22:06:16 <b_jonas> Taneb: people are always complaining about every exam
22:07:16 <zzo38> In my opinion LADSPA is a bit too simple and LV2 is a bit too complicated but I have the new idea of something similar actually
22:09:01 <Koen_> that's a bit overkill
22:09:33 <Koen_> n! is n * things smaller than n and n^n is n * things not smaller than n
22:10:35 <Taneb> Although the last question was a bit nasty, I thought
22:11:16 <Taneb> "Language L_1 is in P, and Language L_2 is neither \emptyset or \Sigma*. Prove that L_1 can be reduced to L_2 in polynomial time"
22:11:16 <Lyka> i have a hello world, a 0 to 99, a 0 to 999, a fibonacci below 65536, and a non-terminating cat
22:13:45 <Lyka> i'm strting to doubt that the language is a brainfuck-derivitive
22:15:16 <Koen_> Lyka: hmmm are you talking about fourfuck? as far as i'm aware the author hasn't disclosed much more than "it's loosely based on brainfuck" and "commands are four characters long" so I don't know how you would have all those programs
22:15:19 <Phantom_Hoover> Koen_, just showing it's smaller isn't enough, you need to show it overwhelms constant multiplication
22:15:40 <Koen_> Phantom_Hoover: one of the smaller factors is a 1
22:15:42 <Koen_> that's much smaller
22:15:56 <shachaf> Is that like http://spl.smugmug.com/Humor/Lambdacats/i-dVj9xxz/1/O/recurcat.gif ?
22:16:00 <Lyka> Koen_: fourfuck is my language
22:16:30 <shachaf> Or like http://cameronhunter.github.io/flight-edge/ ?
22:16:59 <Koen_> Lyka: well, the things I added are mere suggestions; if this is your language, feel free to edit the page as much as you like
22:17:10 <Koen_> including by removing or adding categories
22:18:07 <Koen_> Taneb: correct me if I understand your problem wrong
22:19:01 -!- Frooxius has quit (Quit: *bubbles away*).
22:19:21 <Koen_> Taneb: but "L_1 can be reduced to L_2 in polynomial time" is the same as "if I have a machine M2 that recognizes L2, then there exists a polynomial-time machine M such that M \circ M2 recognizes L1"
22:19:36 <Koen_> or something like that
22:19:45 <Taneb> I may have misremembered the question
22:19:52 <Koen_> and we already know L1 is polynomial-time
22:19:59 <Koen_> so you can just discard your M2 machine
22:20:40 <Lyka> which should i do: leave the page alone and focus on getting the language to work for things other than fibonacci output and hello world? or delete the page and remove the link for now?
22:21:13 <Lyka> i don't work well having to document every step
22:21:14 <Koen_> well in my humble opinion, designing a language is much more interesting that editing a wiki page
22:21:19 <Taneb> Koen_, hang on, let me check I am not misremembering the question, because that is backwards from what I put
22:22:00 -!- Frooxius has joined.
22:22:30 -!- variable has changed nick to trout.
22:22:44 <Lyka> without objection, i'm gonna delete the page from the wii and remove the link in language list
22:23:51 <Lyka> th language is changing too much
22:25:55 <Taneb> Koen_, http://i.imgur.com/QC3ecg3.jpg
22:26:27 <Koen_> okay, well I hold by what I said
22:26:48 <Koen_> "L1 is P" means you can solve L1 in P-time
22:26:52 <Taneb> I sort of interpreted it the other way round to you
22:27:24 <Taneb> Because L2 is non-empty, there is at least one string in it, say "a"
22:27:35 <Taneb> Because L2 is not full, there is at least one string in it, say "b"
22:27:46 <Koen_> not in it* but okay
22:28:08 <Taneb> Now, for a string w, if we define w' := if w is in P then a else b
22:28:15 <Taneb> (which we can do in polynomial time)
22:28:23 <Taneb> Then w is in L1 if and only if w' is in L2
22:28:55 <Taneb> That is what I said
22:29:19 <Taneb> (in my exam. that is
22:29:24 <Koen_> can you check "if x is in P" in polynomial time?
22:30:07 <Koen_> well okay that sounds great
22:31:09 <Taneb> It seems to use every detail given in the question, which is a good sign
22:31:18 <HackEgo> [wiki] [[Fourfuck]] http://esolangs.org/w/index.php?diff=43063&oldid=43047 * Lesidhetree * (+0) Removing my name...
22:31:51 <HackEgo> [wiki] [[User:Lesidhetree]] http://esolangs.org/w/index.php?diff=43064&oldid=43049 * Lesidhetree * (-21) Removing my name...
22:32:08 <Taneb> ...64% of this exam was "What does this Turing machine do?"
22:32:55 <Taneb> MDream, in more detail than that, unfortunately
22:33:04 -!- MDream has changed nick to MDude.
22:33:19 <Koen_> what value??? that's the question
22:33:41 <Taneb> Some of them instead recognized a language!
22:33:55 <Koen_> Taneb: I guess I didn't really know the formal definition of language reduction, only the idea of problem reduction
22:34:08 <Koen_> Taneb: they computed a boolean value didn't they
22:34:49 <Koen_> well "accept / reject / don't halt" isn't strictly speaking boolean maybe
22:35:08 <Taneb> Koen_, the one that recognized a language was total
22:35:47 <Koen_> they they did do compute a value!!
22:36:09 <Taneb> Anyway, I need to revise for my Groups, Rings, and Fields exam on Thursday...
22:36:29 <Koen_> surprise interrogation!
22:36:40 <Koen_> what's a syllow subgroup
22:37:08 <Taneb> It's a subgroup of order p^n where p is prime and n is the multiplicity of p in the order of the group!
22:37:10 <Koen_> or sylow or howevermany L there are
22:38:06 <Taneb> A subring I of a ring R such that for a in I and b in R, ab is in I
22:39:01 <Taneb> A ring whose units are all elements other than zero
22:39:26 <Koen_> does "unit" mean "inversible element"?
22:39:38 <Koen_> I thought the unit was the neutral element for multiplication
22:39:43 <Taneb> a is a unit if there exists a b such that ab = 1
22:39:53 <Taneb> Koen_, I'm just using the definitions I'm given, I am afraid
22:39:59 <Koen_> yup that's good enough
22:40:11 <Taneb> (a and b are in the ring, of course)
22:40:30 * oerjan carefully points out the ring should probably be commutative
22:40:48 <oren> Lord of the rings
22:40:49 <Koen_> I'm guessing some courses assume rings are or are not commutative so that would be the afraidful definitions
22:41:16 <oerjan> my guess was the same except with "fields"
22:41:47 <Koen_> Taneb: you should probably know the chinese theorem as well
22:41:58 <Koen_> it's not gonna be in the exam but it's a fun story for parties
22:42:21 <oerjan> are you missing a remainder or is this something i haven't heard of
22:42:31 <Taneb> oerjan, I think you *might* get commutativity for free?
22:42:48 <Taneb> You need commutativity
22:42:49 <oerjan> Taneb: no, there are "skew fields" or "division rings"
22:42:59 <Koen_> err the french name is "théorème chinois", that might not be the english name though
22:43:19 <Taneb> oerjan, you don't if it's finite, apparently
22:43:42 <Koen_> oerjan: it says if you've got an unknown number N but you know two remainders of N by say a and b (and a and b are distinct and big enough) then you can calculate N
22:44:00 <Koen_> and the legend says it was used by a chinese general to count his soldiers
22:44:01 <oerjan> Koen_: "chinese remainder theorem"
22:44:10 <Koen_> then I was indeed missing a remainder
22:44:33 <Taneb> I am going to bed now
22:44:51 * oerjan used that on the one question he managed to solve when in the IMO
22:45:02 <Taneb> Thanks for the pop quiz, Koen_, it did help
22:45:35 <Koen_> they're is probably more to know about silow subgroups than their definition, though
22:46:27 -!- boily has joined.
22:47:06 <oerjan> also the chinese remainder theorem is an important (although not the hard) part of the theorem that you can do integer division in logarithmic space
22:47:10 <lambdabot> CYUL 262200Z 23015KT 30SM FEW050TCU BKN240 28/17 A2997 RMK TCU1CI6 SLP150 DENSITY ALT 1500FT
22:47:27 <lambdabot> ENVA 262150Z 27010KT 9999 FEW035 BKN049 09/05 Q1010 RMK WIND 670FT 28012KT
22:49:44 <Koen_> oerjan: erm we learned how to do integer division in primary school didn't we?
22:49:53 <oerjan> Koen_: not in logarithmic space
22:50:12 <Koen_> well we used digits
22:51:01 <oerjan> yes, a quadratic amount of them, i am guessing
22:52:01 <oerjan> (although with mutable memory you can relatively easy reduce it to linear)
22:53:52 <Koen_> hey, I was wondering whether studying mixed time/space complexity was a thing?
22:54:33 -!- lleu has joined.
22:54:42 <Koen_> there are a lot of problems that can be solved in either very small space or very small time but the usual method to reduce space is to keep forgetting what you've already done and redoing everything a quintillion times
22:54:53 <oren> Hmm, mutable paper would be nice, like if erasers ersaed, as opposed to just spreading the graphite acros the page
22:55:28 <Koen_> aquarel erasers do
22:56:14 <oerjan> Koen_: hm i am thinking of path reachability, it can be done in polynomial time or in log^2 space
22:56:28 <oerjan> but can you get both at the same time?
22:56:49 <Koen_> yes I think that one was the problem that made me wonder
22:56:49 <oerjan> (it's the canonical NLOGSPACE-complete problem)
22:58:57 <Koen_> oren: http://en.wikipedia.org/wiki/Kneaded_eraser
22:59:49 <Koen_> it's more suited in painting when you want to erase the sketch without leaving pencil marks
23:03:34 -!- hjulle has quit (Ping timeout: 256 seconds).
23:04:47 -!- Lyka has changed nick to Lyka|Away.
23:10:12 * Sgeo saw CESU-8 today :(
23:12:53 -!- Koen_ has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?").
23:19:30 -!- hilquias has joined.
23:19:35 <oren> Sgeo: my condolences
23:20:09 -!- trout has quit (Quit: 1 found in /dev/zero).
23:20:56 <oren> Cones with very large cone angles make good paper fans
23:22:59 -!- variable has joined.
23:23:19 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
23:31:11 -!- gde33 has quit (Ping timeout: 265 seconds).
23:33:17 <zzo38> Is this document clear to you? http://zzo38computer.org/textfile/miscellaneous/rdf/c_rdf
23:41:38 -!- gde33 has joined.
23:43:29 <boily> Sgeo: Sgello. what's a CESU-8?
23:45:40 <zzo38> CESU-8 is encoding the UTF-16 codes of Unicode texts into UTF-8.
23:51:15 <oren> In other words, codes outside BMP end up as 6 bytes instead of 4.
23:55:39 <zzo38> Yes, that is how it is work
23:56:18 <pikhq> It's the UTF-8 encoding algorithm on UTF-16 units. Mostly a consequence of bad UTF-16/UTF-8 implementations and backwards compat with 'em.
23:58:08 <lambdabot> KATL 262352Z 00000KT 10SM -RA FEW005 BKN120 BKN150 OVC200 20/18 A3021 RMK AO2 SLP221 P0002 60108 T02000178 10289 20194 55007 $
23:58:44 <zzo38> A program I have written called "utftovlq" has no built-in CESU-8 support; however, you can convert between proper UTF-8 and CESU-8 by using a pipe of the program twice. If you tell it to convert UTF-8 to UTF-16 and then back again, you get proper UTF-8 out. If you tell it to convert UTF-8 to UTF-16 and then RAW-16 to UTF-8, the output will be in CESU-8 format.
23:59:45 <quintopia> (the rain is incessant this part of summer. thunderstorms daily.)