←2005-11-18 2005-11-19 2005-11-20→ ↑2005 ↑all
00:08:03 -!- puzzlet has joined.
00:27:38 <calamari> does Haifu in 2005 seem right ?
00:43:27 <kipple> no, but I could be wrong...
00:47:02 <kipple> it was added to the language list in wikipedia in sept. 2003
00:47:15 <kipple> http://en.wikipedia.org/w/index.php?title=List_of_esoteric_programming_languages&oldid=1406250
00:47:32 <calamari> ok
00:47:54 <kipple> I suspect it is from 2000-2001 as most of DMM's languages
00:48:30 -!- cmeme has quit ("Client terminated by server").
00:49:05 -!- cmeme has joined.
00:49:07 <calamari> I sent an email to David Morgan-Mar as I noticed several langs by him I didn't have dates for
00:49:22 <kipple> yeah, that's probably the best way to do it :)
00:49:27 <calamari> so hopefully I'll be able to get better dates for his langs
00:49:37 <calamari> :)
03:21:55 -!- sekhmet has quit ("omg it's lvm2!").
03:45:46 <Sgep> Night all
03:45:54 -!- Sgep has quit.
04:18:43 -!- CXI has quit ("If you're reading this, it's probably xchat's fault.").
04:19:16 -!- CXI has joined.
04:52:29 -!- kipple has quit (Read error: 110 (Connection timed out)).
05:09:36 -!- calamari has quit ("Leaving").
06:32:59 -!- sekhmet has joined.
07:09:44 -!- kipple has joined.
07:20:17 -!- cmeme has quit (Network is unreachable).
07:22:07 -!- cmeme has joined.
07:22:11 -!- Arrogant has joined.
07:33:53 -!- sekhmet has quit ("leaving").
07:47:19 -!- sekhmet has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:16:15 -!- calamari has joined.
08:33:35 <calamari> uncovered a lang from the sange.fi mailing list: http://esoteric.voxelperfect.net/wiki/Dumbf%2Ack
08:41:55 <kipple> nice work detective :)
08:52:59 <calamari> seems that I had just forgotten about it... because I later post my approval of it hehe
08:54:53 <kipple> well there are so many bf-clones. can't expect to remember them all
08:59:11 <calamari> new lang http://web.archive.org/web/20011107102634/http://www.guldheden.com/~sandin/when/When.txt
08:59:31 <calamari> err I mean new in 2001 ;)
09:07:51 -!- Keymaker has joined.
09:14:31 <calamari> re Keymaker
09:14:53 <Keymaker> hello
09:15:11 -!- Arrogant has quit (Read error: 113 (No route to host)).
09:19:31 -!- Arrogant has joined.
09:31:36 -!- Keymaker has left (?).
09:46:02 -!- Arrogant has quit ("I AM QUIT HAVE A NICE DAY").
09:49:47 -!- jix has joined.
10:20:04 -!- J|x has joined.
10:22:53 -!- jix has quit (Nick collision from services.).
10:22:55 -!- J|x has changed nick to jix.
11:18:58 <calamari> wow.. 4am and I'm still not done with 2001.. oh well :)
11:19:09 <calamari> bbl
11:19:13 -!- calamari has quit ("Leaving").
11:46:18 <jix> i'm working on bfgentext again
11:47:52 <jix> i need to do something until my gp2x arrives (it's on the way from korea to UK atm... UK to germany is the next stage. Somewhere in germany to me is the last stage)
12:56:22 <jix> my sort is 2 bytes longer than the shortest i found :(
13:03:36 <jix> my version: >>,[[-[->>+<<]+>>]<+<[-<<]>>,]+[>[-<.>]<[->>+<<]>>+] (using byte counting)
13:03:55 <jix> Daniel B Cristofani's version: >>,[>>,]<<[[-<+<]>[>[>>]<[.[-]<[[>>+<<-]<]>>]>]<<]
14:00:52 -!- lindi- has joined.
17:04:45 -!- shmeebegek has joined.
17:05:05 <shmeebegek> hello
17:06:19 -!- sekhmet has quit ("memtest!").
17:15:24 -!- Sgep has joined.
17:40:11 -!- shmeebegek has quit (" Try HydraIRC -> http://www.hydrairc.com <-").
17:42:08 -!- Sgeper has joined.
17:47:53 -!- nooga has joined.
17:47:55 <nooga> hi
17:48:17 <nooga> who knows something about graphs?
17:56:23 <jix> hmm
18:25:10 -!- nooga has quit (Read error: 110 (Connection timed out)).
18:34:58 <jix> bfgentext is about getting usable
18:38:15 <jix> bbl
19:18:45 -!- calamari has joined.
19:31:13 <jix> moin calamari
19:44:02 <calamari> hi jix
19:52:32 <jix> my first lhs is done 90%
19:56:19 <calamari> cool
19:56:37 <jix> lisp is really a cool language
19:56:56 <jix> somewhat esoteric but fast and has really cool features
19:57:02 <fizzie> Scheme is nice too.
19:57:22 <jix> yeah but i need speed and there are better lisp implementations for osx
20:04:01 <calamari> is it possible to create a link to a category in an article, without becoming that category?
20:04:30 <calamari> I could copy the entire url but that seems bad
20:04:55 -!- Keymaker has joined.
20:05:19 <calamari> hi Keymaker
20:05:26 <Keymaker> hello
20:06:09 <jix> Keymaker: i'm modifying my bfgentext to support wrapping and unwrapping code (but only 8-bit because non 8bit is stupid for text output)
20:06:27 <Keymaker> ok
20:06:34 <Keymaker> btw, what is lhs
20:06:36 <Keymaker> `?
20:06:40 <jix> left hand side...
20:06:50 <Keymaker> i didn't get it :)
20:06:53 <jix> not important unless you know how the program generates the bf code
20:07:04 <Keymaker> ok
20:07:15 <jix> it does it in 2 passes.. one is the lhs the other one is the rhs
20:07:43 <Keymaker> bfgentext?
20:07:57 <jix> my bf output text generator
20:08:07 <jix> like textgen.java but in lisp and a bit more flexible
20:08:15 <Keymaker> yes, but you're talking about bfgentext, right?
20:08:23 <jix> yes
20:08:26 <Keymaker> ok :)
20:08:29 <Keymaker> i was a bit confused
20:11:33 <jix> calamari: does textgen.java use wrapping or nonwrapping bf?
20:11:54 <calamari> non-wrapping
20:12:12 <jix> ah
20:12:40 <jix> i first wrote wrapping code only but it was only a little change to allow wrapping in the written part too
20:13:24 <calamari> to do wrapping I'd need a way to compute the result of a wrap very quickly
20:13:35 <jix> &0xFF
20:13:53 <calamari> I think at one time I'd figured it out, but then I've forgotten
20:15:25 <calamari> jix: for example ---[>----<+]>
20:15:43 <calamari> -3 -4 1
20:16:13 <jix> that (3*-4)&0xFF
20:16:42 <jix> right
20:16:45 <jix> bbl
20:18:28 <calamari> ---[>+<-------]> = 219
20:19:51 <calamari> ahh this is better.. no 1: ----[>---<--]>++ = 136
20:24:48 <calamari> I believe the equation is 254 mod x = 252
20:30:47 <calamari> I know of no O(n) solution to that.. there are algorithms to find it but they take several steps or there may not be a solution
20:30:57 <calamari> err I meant O(1)
20:31:15 <calamari> anyhow.. :)
22:10:10 -!- graue has joined.
22:10:21 -!- graue has quit (Client Quit).
22:19:53 -!- Keymaker has left (?).
22:22:55 <jix> calamari: if you want tables for set_to_x[sub_y>add_z<] you could take a look at my bf2a.rb look up tables
22:24:45 <jix> calamari: is your textgen.java still under 'active' development?
22:30:53 -!- Taliesin41 has joined.
22:31:05 -!- Taliesin41 has changed nick to KevinN.
22:31:05 <calamari> nope
22:31:26 <calamari> but it is gpl, so you may change it however you like
22:33:59 <jix> hehe no my bfgentext isn't for from complete... i was just interested whether you are still trying to improve it or not
22:40:38 <calamari> probably not.. but who knows.. someday it may become very interesting to me again.. can't predict these things :)
22:51:04 <KevinN> calamari, you're a pro, aren't you?
23:11:13 <jix> KevinN: a pro?
23:11:35 <KevinN> a professional...
23:12:19 <jix> a professional what?
23:13:10 <KevinN> a pro in matters of language-design and stuff...
23:19:27 <KevinN> I think that every turing-complete language can be converted into a turing-tarpit consisting of only 2 operators...
23:37:22 -!- kipple has quit (Read error: 110 (Connection timed out)).
23:37:56 <calamari> KevinN: I think I'm far from being a pro.. but thank you for the compliment :)
23:38:12 <KevinN> you're welcome...
23:38:19 <KevinN> oehm... well...
23:39:02 <KevinN> to come to the interesting part: I think the way I've found does work for every language there is so far...
23:39:31 <calamari> what have you come up with?
23:41:45 <KevinN> I call it IM (or: "In-Memory")...
23:41:56 <KevinN> you have to think of it as a kind of brainfuck...
23:42:10 <KevinN> however, in a special sense of way...
23:42:31 <KevinN> you take language A that allows N different characters as input...
23:43:16 <KevinN> your source is written on a tape... its pointer can only be moved to the right and the value of the current cell can only be incremented by one...
23:43:53 <KevinN> the minimum value of a cell can be zero (no character) and the maximum value can be N
23:44:11 <KevinN> now you write the whole source onto the tape...
23:44:50 <KevinN> when you're tape-writing-source (in the so-called IM-language) comes to its end, the source that was written ONTO the tape is executed...
23:45:45 <KevinN> to make it simple: you write for example C - code onto the tape and that C - code is executed after writing the code onto the tape...
23:47:29 <calamari> if the original source is executed, is'not more than 2 operators?
23:48:54 <KevinN> well... I think operator is the wrong word... operand should be more correct...
23:49:12 <calamari> for example I could make a language that you just said B for the "code", but then fed a bf program to standard input.. now all B programs are alike and one characters but the language is really defined by the interpreter of B
23:49:39 <KevinN> it consists of the > - operand (move right) and the + - operand (increment)
23:49:55 <calamari> ok
23:50:16 <calamari> loops?
23:50:30 <KevinN> everything is done in-memory... ;)
23:50:45 <KevinN> so you need some kind of second interpreter or so...
23:51:11 <calamari> right, but having a second interpreter (IMO) means that whatever the 2nd interpreter does added to the operators
23:51:13 <KevinN> one that understands the IM-source (>+++++>+++>+++++ etc.) and one that understands that source that's written on the tape...
23:51:55 <KevinN> naaah... don't think so...
23:51:57 <calamari> Why not describe the source in terms of 0 and 1, it is the same :)
23:52:20 <KevinN> you see... that's the basis...
23:52:22 <calamari> then you have binary, which is how it is represented
23:52:33 <calamari> (even normally)
23:52:39 <KevinN> yes... in some way or another...
23:52:57 <jix> now the real genetic algorithm...
23:55:35 <calamari> KevinN: I think a lang can be devised with 3 operators that perform the essentials (iteration, infinite memory, etc), but with two I think it is impossible without enforcing some outside design on the two.. for example with iota and jot, they are decoded according to lambda calculus, so that a single 0 or 1 does not perform a single task, it does several depending on context.
23:56:36 <calamari> with bitchange or bf, each instruction performs a well defined task that does not change function depending on positioning, only on state
23:57:22 * GregorR is shocked as calamari bursts into flames!
23:57:42 <calamari> GregorR: I did?! :)
23:57:45 <calamari> lol
23:57:50 * GregorR enjoys abusing /me :P
23:58:02 * jix doesn't
23:58:07 * calamari tries to parse that sentence
23:58:24 <GregorR> Anybody want to chat on DirectNet 8-D
23:58:41 <GregorR> 8-D being a punctuation mark with roughly the same meaning as '?'
←2005-11-18 2005-11-19 2005-11-20→ ↑2005 ↑all