00:00:00 --- log: started forth/02.01.06 04:16:34 --- join: XeF4 (~xef4@dsl-V-60.kotikaista.weppi.fi) joined #forth 06:33:09 --- join: Speuler (~l@c38038.upc-c.chello.nl) joined #forth 06:33:38 hoi 06:33:53 terve 06:50:53 --- quit: XeF4 ("...") 06:55:38 --- part: Speuler left #forth 11:26:25 --- join: I440r_ (mark4@pool-63.52.217.25.ipls.grid.net) joined #forth 11:55:42 --- quit: I440r_ (Read error: 104 (Connection reset by peer)) 12:02:41 --- join: I440r_ (mark4@pool-63.52.218.3.ipls.grid.net) joined #forth 15:23:29 --- quit: I440r_ (Read error: 60 (Operation timed out)) 19:03:37 * aaronl is away: I'm busy 19:04:09 --- join: edrx (~ochs@Panda.Math.McGill.CA) joined #forth 19:06:39 --- join: TheBlueWizard (TheBlueWiz@ip-216-25-205-190.vienna.va.fcc.net) joined #forth 19:06:44 hiya all 19:06:52 hi tbw 19:07:04 hiya edrx 19:07:48 --- mode: ChanServ set mode: +o TheBlueWizard 19:08:09 --- mode: TheBlueWizard set mode: +o edrx 19:08:47 I moved to Montrea; 19:08:50 montreal 19:09:58 * TheBlueWizard is a man...the /topic says c is for men...so c is for him!!! yaah! wait! I think this is a female only channel! better run!!! else I get my family jewel removed! ack! 19:10:22 ah...so how do you like Montreal so far? 19:12:23 there's a weird thing about that in my home page 19:12:23 http://angg.twu.net/test.html 19:12:23 very very nice 19:12:23 people are fantastic 19:12:23 where do you live? I'm sure you told me that 10 times, but I forgot... 19:13:51 hehe...I live in WashDC area 19:15:26 * TheBlueWizard peeks at the test.html...and sees a soporific message about drugs and castration 19:16:29 I'm starting to believe that the idea for that message was sent to me by The Being 19:17:33 whuh? 19:17:54 * TheBlueWizard notes that "whuh" is a fusion of "what" and "huh" 19:19:49 http://www.churchofeuthanasia.org/e-sermons/ogyrintv.html - search for "slogan" 19:22:32 ah... 19:23:08 * TheBlueWizard tosses Church of Euthanasia into a blender....*bwrrrrrr!* 19:24:35 ? 19:24:57 I consider that as trash 19:25:10 you're in a bad mood today 19:26:05 oh? I am not in a bad mood.... 19:27:20 * aaronl is back (gone 00:23:44) 19:27:32 hiya aaronl 19:27:42 hi aaronl 19:33:44 --- join: futhin (thin@24.66.212.245) joined #forth 19:33:48 hi all! 19:33:53 hi 19:34:28 how's it going edrx? 19:34:40 is the forth with you? :) 19:34:53 hello thebluewizard, you there? 19:34:55 hiya futhin 19:35:03 cool, how are you doing? 19:35:32 doing fine...though I am not looking forward to work tomorrow :P 19:35:37 futhin: ok - I moved to Montreal... because of that my Forth is taking a nap, but should wake up in a week or two 19:36:10 hmm, i'm moving downstairs, so yeah, my forth is taking a nap too :P 19:36:19 thebluewizard: what is your job? 19:36:27 futhin: :) 19:36:30 or you just want to avoid work? 19:37:07 edrx: come to calgary and we'll form a pair programming team and code a fully interactive collaborative universe in forth which will spontaneously become self-aware 19:37:27 futhin: how far is calgary from montreal? 19:37:39 my advisor is there now, btw 19:37:48 I work in IT....dealing with M$ products, and do some paper pushing as well 19:37:59 hm.. 1700 kilometers? 19:38:18 financial advisor? 19:38:55 futhin: I don't think it would be easy for me to go, my budget is student-sized 19:39:09 hmm, same here :) 19:39:13 er 19:39:19 actually, more than 1700 kilometers 19:39:22 i was thinking winnipeg 19:39:53 futhin: no, I'm here studying maths, actually doing research... 19:40:22 academic advisor 19:40:25 research on math? better be quantum mechanics math or it's a waste of time! :P 19:40:49 hehe! 19:40:59 well, opinions diverge... I think that anything related to physics is a waste of time... 19:41:05 ooh 19:41:11 what math do you do specifically? 19:41:33 * TheBlueWizard laughs 19:41:45 heh 19:42:26 that reminds me of Richard Feynman's disdain for pure math guys..."Waste of time!" 19:43:08 i like both 19:44:28 you like both math and physics? cool! me too! 19:44:45 a language that can be used to represent the "skeletons" of certain proofs - something that is much more succint than a fully formalized proof, but that sort of captures the intuition of what goes in the proof better than the full version... and also there is (i.e., will be, it's far from ready) an algorithm that fills up the missing details automatically 19:45:08 ,ljkmhgbf 19:46:52 futhin: edrx is researching in Category Theory area...that is one area that I happen to know a bit about...pretty powerful stuff 19:47:30 though it can get damn abstracting real fast hehe 19:48:09 TheBlueWizard: I'm trying to make it more accessible too :) 19:49:33 I understand that :) 19:49:41 TheBlueWizard: the first "working" version of the language for skeletons of proofs is exactly for category theory... :) 19:49:52 yeah, we've talked about it... 19:51:02 and I'm using Forth to implement it on a computer :) 19:51:20 hmm...interesting.... 19:53:47 not that I've gone very far on the implementation, but it is evident that Forth is the right tool to write the parser, build the data structures, and maybe generate code in other languages 19:55:27 hmm...a choice of language can be darn important...sometimes a certain language makes the implementation a snap...others would give you the equivalent of wearing an agonizer like in a certain Star Trek episode about alternative universe :) 19:55:34 --- join: k (~k@a161161.upc-a.chello.nl) joined #forth 19:55:40 hiya k 19:56:04 hi TBW 19:56:15 hi k 19:56:25 --- nick: k -> topaz8 20:00:52 --- part: topaz8 left #forth 20:01:20 gotta go...bye all 20:01:37 me too 20:01:41 night everybody 20:03:22 bye edrx 20:03:32 --- part: TheBlueWizard left #forth 20:24:37 --- quit: edrx ("[x]chat") 20:29:53 --- quit: futhin () 21:29:04 --- join: I440r_ (mark4@pool-63.52.217.64.ipls.grid.net) joined #forth 21:48:41 * aaronl is away: I'm busy 21:57:19 * aaronl is back (gone 00:08:38) 22:44:26 --- join: Speuler (~l@c38038.upc-c.chello.nl) joined #forth 22:44:40 g'day 22:57:06 hi ppl 22:57:13 im working at ski world :P 22:57:19 in the rental department :P 22:57:20 bleh 22:57:24 not coding ::P 22:57:39 reprogramming the rental computers ? 22:57:42 ah 22:58:15 no - handing out ski boots to teh skiers :P 22:58:57 i'm just trying to optimize julian day calculation, zellers congruence 22:59:21 got it to about 200 bytes (32bit threaded) 22:59:34 still quite a lot 22:59:34 :) 22:59:58 well, was 300 when i started 23:00:17 and looked more readable then 23:01:12 yes 23:01:14 exactly 23:01:21 : jd3 23:01:21 >r 9 + 12 /mod 1- >r 26 * 434 + 10 / + 23:01:21 2r> + 100 /mod >r dup 4 / r@ 4 / r> 2* - + + + 23:01:21 7 mod ; 23:01:21 optimized code is NOT readable 23:01:27 thats why god invented comments :P 23:01:54 comments are for weenies :) 23:01:58 just like docs 23:02:22 "t'was hard to code, why should it be easy to read" 23:02:27 heheh 23:03:12 comments don't help with that kind of words 23:03:38 just comment what it does then - not how :P~ 23:03:38 there's no good reason to do things this way :) 23:03:41 that works for me heh 23:03:51 i beg to differ 23:04:07 its like "why do you cllimb mountains.... "because they are there" " 23:04:14 we do it the hard way because its more fun that way 23:04:39 it does things that way because like this it works, 23:05:02 bla bla bla \ without this the code won't function 23:05:03 bla bla bla \ without this the code won't function 23:05:12 hehe 23:05:28 oh - is my itterative version of fc in teh title correct ? 23:05:47 could be. not tested ? 23:05:51 nope 23:06:06 and 5000000000000 fc wont crash me :P 23:06:11 mine i mean 23:06:16 can we make a contest for smallest zeller's congruence ? 23:06:20 it would the recursive one :)_ 23:06:29 i dunno what that is :) 23:06:42 if someone has a smaller one, everybody is helped 23:06:57 it gives you the day of week from date 23:07:05 oh 23:07:33 often, only without oct15,1582 ... feb2600 23:07:35 i could never code that heh 23:07:38 within... 23:08:03 there was a jump of weekdays in 1582 23:08:12 there was ? 23:08:15 makes alg more difficult to catch that 23:08:16 y ? 23:08:29 cause of change of date calculation 23:08:35 oh 23:08:39 to julian system ? 23:08:58 a correction to the lost days because of julian inaccuracies 23:09:25 aha 23:09:53 or , extra days, can't remember 23:10:29 i test them for all dates between oct15,1582 and feb2600 23:10:55 have three algs compare outputs 23:11:05 in a loop, for each day 23:11:41 only of no dicrepancy, do i adopt a modification 23:11:45 if 23:13:55 hrm 23:14:07 i would never be able to figure a way of doing it :) 23:14:42 some tweaking of traditional approaches can yet you quite an end 23:16:28 the 200 bytes wall is resisting 23:17:07 34 forth words 23:17:11 no loop 23:17:24 7 additions 23:17:38 9 literal (counted as words) 23:17:50 3 /mod and mod 23:18:10 3 >r and r> 23:18:24 i would use a !> in place of teh r> and r> 23:18:40 just do !> a constant and then name teh constant in place of the r> 23:18:45 to a value ? 23:18:50 NO 23:18:51 values suck 23:18:54 to a constant :P 23:18:57 i break rules 23:18:59 same problem 23:19:02 constants ARENT!!! hehe 23:19:03 non-reentrant 23:19:07 not realy 23:19:14 because in place of all literals i would use constants too 23:19:22 a, literals 23:19:22 that would drcrease the object size too 23:19:50 if i define constant 434 for single use, object size grows 23:20:11 26 is also not a good reuse candidate 23:20:17 foo 23:20:58 w/o reuse, constants are worse than literals 23:21:06 ya 23:21:53 forgot that :P 23:22:47 how about a shortest shell sort competition :) 23:23:00 ok 23:23:09 i would do it in pure asm 23:23:19 trivial things are always shorter in asm 23:23:26 and my shell in asm is about 20 ytes i thin 23:23:27 i forget 23:23:51 i got a hi-level shell sort in the lib. 23:24:15 helps me to sort right away when compiling for another cpu 23:24:29 no need to implement on target first 23:25:00 argh i forgot i exploded the loop on the shell sort hehe 23:25:04 sorted about 25000 items/sec on 386 23:25:05 its larger than it needs to be :P 23:25:25 radix/counting would do faster 23:25:32 ive yet to see a faster sort than radix counting 23:26:30 hash sort on ultra long word vector arithmetic machines 23:27:03 well if we had infinite memory then a PURE counting sort would be the absolute fastest heh 23:27:18 but for 32 bit words we would need 256 megs of memory just for the counts heheh 23:27:28 or more- i forget exactly 23:27:51 AND, you'd need more sophisticated means to locate the next element in the homing area 23:28:30 well, links may do 23:29:33 for compactness, one could consider byte code too 23:30:14 zeller3 would be 'bout 40 bytes 23:30:23 difficult to do in asm ... 23:31:19 zeller3 is this one: 23:31:25 : jd3 23:31:25 >r 9 + 12 /mod 1- >r 26 * 434 + 10 / + 23:31:25 2r> + 100 /mod >r dup 4 / r@ 4 / r> 2* - + + + 23:31:25 7 mod ; 23:31:53 that would be fairly difficult to do in asm but not impossible 23:32:03 the /mod and / are realy teh same function 23:32:12 the / is a wrapper for /mod 23:32:25 code it in asm wouldnt realy be that difficult 23:32:31 but i doubt the result would be smaller 23:32:42 because YOU arent counting the size of the code for /mod in teh forth version 23:32:51 or teh size of the code for >r 23:32:53 etc 23:33:07 i say "system need >r anyway" 23:33:15 correct 23:33:17 BUT 23:33:33 if we were to be honest we would count teh size of ALL required code to execute that funciton 23:33:44 including the os 23:34:37 heh 23:34:39 no 23:34:50 nor teh bios :P 23:37:12 i'd count out the reusable words 23:37:18 reused 23:37:34 as those don't exist exclusively for the function to implement 23:39:00 :) 23:39:34 yes, another byte saved 23:39:46 (at least, potentially ...) 23:39:58 heh 23:40:00 can get literal 434 to half ... 217 23:46:27 ? 23:46:36 217 is a byte 23:46:42 434 are two bytes 23:47:30 oh, i have written a NAND in brainfuck 23:48:10 if you can build a NAND, you can build a computer too 23:48:43 what about a forth interpreter in brainfuck :) 23:49:09 now THAT would be something for a change 23:50:17 hehehe 23:50:24 or in bash script 23:50:29 i wanted to do a forth in bash 23:50:35 then write a bash interpreter in forth 23:50:41 and have them interpret each other heh 23:51:35 did you have a look at tkgate ? 23:51:43 logic simulator 23:52:46 has an emulated serial port too, so you can stick your keyboard on the circuit 23:52:57 comes with a sample cpu. 23:53:08 and a file. to load into emulated ram 23:53:17 with program for emulated cpu 23:53:23 whats tkgate ? 23:53:35 gate logic emulator 23:53:47 digital breadboard 23:53:53 aha 23:54:05 can use FETs instead of gates too 23:54:45 i started to build a cpu :) 23:54:54 want to run forth with it :) 23:57:36 heh 23:57:59 here's my brainfuck and: >>[-] >[-] <<< [>>+>+<<<-] >>>[-<<<+>>>]<<[>+>+<<-] >>[-<<+>>]<[-[->+<]] 23:58:15 and the inverter: - 23:58:15 i cant read brainfuck heh 23:59:05 i wrote a small brainfuck interpreter this morning 23:59:14 'bout an hour work 23:59:33 : "<" swap 1- dup 0< abort" pointer underflow" swap ; 23:59:39 : ">" swap 1+ dup memsize >= abort" pointer overflow" swap ; 23:59:53 : "-" over memory -1 add.b ; 23:59:54 : "+" over memory 1 add.b ; 23:59:59 --- log: ended forth/02.01.06