00:00:00 --- log: started forth/02.08.31 00:00:10 with "registers" 00:00:18 How do they GET to the stack? 00:00:37 how do they get 2 to register ax on the x86 architecture? 00:00:38 The X18 doc only tells me about 27 forth primitives. 00:00:52 5 bits each, you can fit 3 of then in a 16-bit word. 00:01:00 thin: 0xb8 0x02 0x00 00:01:27 that's x86 asm converted to hex ? 00:01:43 have you checked http://d-1-200.dhcp-149-159.indiana.edu/misc/p21faq.html 00:01:46 That's x86 machine code. 00:01:51 I will check. 00:02:09 How are the instructions encoded? 00:02:13 "How are the instructions encoded? 00:02:15 "How are the instructions encoded?" 00:02:20 that's a header in that faq 00:02:25 that's what you want isn't it 00:02:51 try searching ultratechnology for "instruction encode" 00:02:59 or some variation 00:04:48 I'll just use my own thing then... 00:05:47 robert: yeah, there doesn't really seem to be a whole lot of info on _how_ to implement machineforth's architecture on chips.. they don't give that much detail on the internals.. 00:06:36 robert: hm, didn't mean to ride you or be rude or whatever.. i'm half hyped and half tired heh 00:07:46 thin: eh, change ur nick back, plz! 00:07:51 You're not rude, only confused and talking too much. 00:08:04 onetom: nahh, this nick is truth 00:08:26 thin: paysan also has 4th chip documentation, test board and so on 00:08:55 thin: no, its just the crippled version of our favourite nick of yours ;p 00:09:56 but it's more concise ;) 00:09:59 it's thinner ;) 00:10:18 but also less fu ;) 00:10:19 thin represents thin apis, thin/tight code 00:10:24 put it back, plz 00:10:26 thin philosophy 00:10:30 nahhh ;P 00:10:39 thin 0wnz0rz! 00:10:45 every1 like that 1! am i right, others! hey! 00:11:22 heh 00:11:26 futhin was a nice & unique, complete name while 00:11:37 thin is just a usual, everyday, well known word.. 00:12:34 i wonder about futhin with umlauts on the u 00:12:40 the two dots on the u.. ? 00:12:44 heh 00:12:54 nevermind 00:14:02 Heh. 00:14:05 Fythin... 00:14:11 Sounds odd ;) 00:14:18 Mr. füthin is here! 00:14:29 I have umlauts in my name :D 00:14:33 Östling >:) 00:14:56 is there umlauts in "uberman" ? 00:15:05 the u in "uberman" is pronounced oo 00:15:16 * onetom needs more sleep. bbl 00:15:21 futhin is supposed to be pronounced foothin 00:15:38 thin: do u have a sound card? 00:15:41 übermann is pronoynced like yyberman. 00:16:00 thin: German 'u' is pronounced like 'o'. 00:16:04 thin: i can record u the pronounciation 00:16:14 thin: And ü is pronounced like y 00:16:41 (Not "why", but like 'iii') 00:17:01 Damn americans, can't pronounce anything :P 00:17:15 :) 00:17:33 thin: btw where does this uberman come from 00:17:41 german 00:17:44 it means "superman" 00:17:47 its either übermensch oder superman 00:17:52 but not like superman the comic hero 00:18:02 but like the man who uses his potential fully 00:18:05 focuses 100% 00:18:14 rises above himself.. 00:18:27 like a forth coder?? ;P 00:18:52 thin: i know phylosophy. we learnt that in highschool.. 00:19:38 thin: so what about the snd card? 00:19:44 I'd say philosophy isn't something you _learn_, almost everything is philosophy. 00:20:06 Robert: mm, but u can learn it too 00:20:29 nope, my speakers are screwed 00:20:43 certainly there is phylosophy behind everythig, but still.. 00:20:54 robert: depends on which definition you use :P 00:21:01 there r patterns u can learn 00:21:06 onetom: everything.. thinnnng 00:21:15 2 b able 2 look behind things 00:21:15 :P 00:21:30 FUthin: :pp 00:22:08 i know u as futhin for half a year 00:22:20 so pl dont rechristian urself 00:22:29 heh 00:22:45 go to bed :P 00:22:51 admit it 2 urself, it was a bad try 00:22:56 2 chg ur nick 00:23:07 ;) 00:23:33 dream w fml & 4thchips, futhin! ;) 00:23:50 --- quit: proteusguy (Read error: 110 (Connection timed out)) 00:24:27 :) 00:25:10 i like thin, i never should've added on the fu ;P 00:27:37 but u were futhin ever here on #forth werent u? 00:27:53 dont lie, i could chk the logs! :) 00:29:55 pardon? your question doesn't make grammatical sense.. 00:30:34 thats true :) 00:30:56 im a bit tired 4 speaking in english grammatically 00:31:30 when u 1st joined 2 #forth ur nick was futhin, wasnt it? 00:32:30 and u havent changed that ever except a few days b4 00:32:35 ago 00:32:45 b4/ago -- which 1 2 chose? 00:33:16 both work.. ago is a little better 00:33:26 thx 00:33:29 more idiomatic i suppose 00:33:37 but forget idiomatics 00:33:40 use b4! 00:33:41 :P 00:33:42 :D ok 00:33:51 yeah i was futhin 00:34:12 and now i'm thin 00:34:27 and like 6 new people have come to this channel in the last 7 days 00:34:29 :D 00:34:36 coincidence? nahhh ;) 00:39:51 http://sec.dunasoft.com:9673/forth/FrontPage -- updated according 2 ur invention ;) 00:40:12 it is! sure. :) 00:40:21 u know that joke: 00:40:26 what is it?: 00:40:55 u throw it up its white 00:41:01 and if it falls back 00:41:49 hey, i'm gonna change my nick to futhin, but only because i need to op myself to change the topic 00:41:50 the lamp is at the neighbour is lit up 00:41:59 --- nick: thin -> futhin 00:42:00 lollll 00:42:01 :pppppppppppp 00:42:13 --- mode: ChanServ set +o futhin 00:42:51 --- nick: futhin -> thin 00:43:57 --- topic: set to 'stack shuffling is drastically reduced with proper factoring, therefore locals are slower. GRR! | http://www.msmisp.com/futuretest/Forth's_Dilemma.htm | x86 Linux Forth coded in asm - http://isforth.clss.net | home of forth - http://www.ultratechnology.com' by thin 00:44:13 --- quit: Robert (Read error: 110 (Connection timed out)) 00:45:13 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #forth 00:49:42 --- quit: I440r (No route to host) 01:01:09 --- quit: thin ("ta ta!") 01:13:47 http://www.rfc1149.net/pictures/pigeon.png :D 01:20:42 --- quit: Soap` (Read error: 104 (Connection reset by peer)) 01:38:56 --- join: thin (~thin@h24-64-175-61.cg.shawcable.net) joined #forth 01:39:22 --- quit: CrowKilr ("User pushed the X - because it's Xtra, baby") 02:36:20 --- join: ASau (ASau@158.250.48.197) joined #forth 02:36:34 Good afternoon. 02:36:41 I'm back again. 02:36:48 Any news? 03:30:48 Hi :) 03:53:15 Hi, Robert! 04:28:59 --- quit: Fractal (Read error: 60 (Operation timed out)) 04:52:12 --- join: Fractal (aht@h24-77-171-228.ok.shawcable.net) joined #forth 05:27:02 thin! 05:27:19 thin: Any news on contest? 05:28:27 * ASau hear voice in the head: "Kosil Yas' konyushinu..." 06:12:48 What contest? 06:21:52 http://icfpcontest.cse.ogi.edu/ 07:29:09 --- quit: Miciah (Read error: 104 (Connection reset by peer)) 07:31:59 --- quit: Robert (Remote closed the connection) 07:32:15 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #forth 07:46:26 --- join: Miciah (~Didle@NBN-TNT2-pool1-87.coastalnet.com) joined #forth 08:36:36 asau: no news 08:36:55 asau: i imagine mslicker is coding or something.. 08:37:16 Maybe he's sleeping... 08:37:58 I slept today if I'd to meet my friend. 08:38:05 I slept today if I'd not to meet my friend. 08:38:35 As a result I'm reading now. 08:38:50 I'm not able to code. 08:38:59 No. 08:39:06 I can code. 08:39:22 asau: why not code top-down.. just code how the bot deals with moving around and stuff, in pseudocode 08:39:26 But with _NO_ pleasure AT ALL. 08:39:41 example code: 08:39:41 items? 08:39:41 botsadjacent? wateradjacent? scoreitemsonground scoreinvitems > if drop then 08:39:54 where drop is drop all items 08:40:05 and then the next turn, pick up items of highest score 08:40:09 thin: I understand this way is also useful 08:40:12 maximizing efficient 08:40:15 maximizing efficiencty 08:40:16 and very good. 08:40:17 er 08:40:19 maximizing efficiency 08:40:36 But I do not want to code. 08:40:49 why not? 08:40:54 not in the mood? 08:41:00 Yes. 08:41:04 problem is not interesting? 08:41:07 Not in the mood. 08:41:20 the problem isn't very interesting i don't think :( 08:41:20 No! 08:41:26 I'm just tired. 08:41:46 what do you know about graphs and such for route planning ?? 08:42:10 Oh, I studied this 08:42:16 about 3 yr ago 08:42:28 or 4-5 yrs? 08:42:39 I can try to recall smth. 08:43:20 As I remember 08:43:52 it's unable to find step-by-step answer for route. 08:44:12 The minimal length - OK, but not steps. 08:45:17 The general method for doing this 08:45:35 is to invent your own system of penalties 08:45:43 and minimize them 08:46:22 What penalties to assign I don't know. 08:46:31 Unable to think so deep now. 08:46:59 I'm listening music. 08:47:44 Belarussian vocal&instrumental ensemble (VIE, in Rus.- VIA) "Pesnyary". 08:47:51 www.pesnyary.com 08:48:47 One of the bests their disks, "The Enchanted One". 08:50:26 "Zaczarovanaja" 08:51:54 thin: If you have concrete question, ask. 08:52:43 I'd recommend "take-wide" class search. 08:53:13 what's that? 08:54:53 :))) 08:54:53 The real solitude: 08:54:53 You're not delivered any letters. 08:54:53 Even spammers do not send any to you. 08:54:53 :) 08:54:54 The real depression: 08:54:59 You don't know where to go. 08:55:45 What do you mean under "that"? 08:56:48 Possibilities: 08:56:58 that = solitude 08:57:08 that = depression 08:57:19 that = search-in-wide 08:57:44 ? 09:03:51 no 09:03:55 that = thing 09:04:02 that = pointer to thing.. 09:04:16 ASau: how wide is ur outgoin net connection? 09:04:17 what's that = what is that = what is that thing? 09:04:23 hello thin 09:04:28 hi onetom 09:05:05 onetom: I've not understood. 09:05:30 thin: What thing? 09:05:38 asau: what is "take-wide" class search? 09:05:44 the immediate thing 09:06:04 that is, what you just said "I'd recommend "take-wide" class search." 09:06:14 I mean you're not diving, but look around. 09:06:18 and "what's that?" asks about the immediate stuff.. 09:06:25 the topic, etc 09:06:40 First, look around. Then go further. 09:08:39 ASau: what type of net connection do u have? outgoing kbit/s? 09:14:58 I have not measure. 09:15:20 asau: regular modem? cable modem? dsl? 09:20:19 onetom: you're asking him the question out of the blue.. it's kinda suspicious :P 09:43:24 --- join: tcn (tcn@tc4-login20.megatrondata.com) joined #forth 09:49:21 tcn :) 09:49:36 hey 09:50:14 Good evening! 09:50:48 How does your prog. run? 09:51:48 tcn's or mine= 09:51:50 tcn's or mine? 09:56:06 Both. 09:57:34 heh. i'm just getting started. 09:57:51 you're player too, robert? 09:59:28 are you in the contest? 10:02:22 Nope. 10:02:27 What is this big contest, really? 10:02:38 I'm coding on my own stuff... 10:02:42 oh hehe 10:03:23 well i've gotta get together the basic networking code.. never done that before 10:03:35 not in C, always in perl or something lame like that :) 10:03:48 * Robert never used Perl. 10:06:12 --- join: mslicker (~maslicke@64.27.199.31) joined #forth 10:06:52 Hi mslicker. 10:07:02 Hi Robert 10:07:08 Good evening! 10:07:16 How does your prog. run? 10:07:28 Not good, 10:07:46 I can't boot my linux system. 10:08:02 Hmm. 10:08:33 the disk was already flakey, but now decides not to boot. 10:08:49 In afternoon I tried to make linux to run in 90 MB avail. on my machine. 10:09:03 No place to install. 10:09:13 -> NUL 10:09:22 O! 10:09:28 NUL only! 10:09:40 NUL! 10:09:55 is the ONLY place to install it. 10:12:41 I think colorForth would have have been very good for this, 10:13:00 --- join: CrowKiller (Forther@Ottawa-HSE-ppp3654508.sympatico.ca) joined #forth 10:13:18 The program was taking on a declaritive sytle. 10:14:00 Which tells me, these super abstract functional languages, do not have much over forth. 10:14:17 for the contest? 10:14:35 yes, icfp 10:16:23 They have much over Forth in the way of memory management. 10:16:45 which wouldn't be useful for this task 10:16:57 FP-ers do not need to monitor memory usage. 10:17:17 I don't think it garbage collection is needed here. 10:17:26 just keep alocating. 10:17:27 tcn: If we take the problem closer 10:17:38 asau: is that way the test server is over 1 meg and doesn't work? 10:17:42 we'll see much of AI task here. 10:18:05 you have 64 megs 10:18:13 that should be plenty 10:18:14 80 10:18:14 --- join: hi_imok_how_r_u (~hko@81.25.34.16) joined #forth 10:19:02 oh yeah, for this task, just allocate the map, etc. just once, and then some temporary stuff each turn 10:19:03 Good evening, OK-er! ;) 10:19:15 I bet the tasks are quite hard... 10:19:28 >:D 10:19:33 I.e. nothing for me. 10:19:50 hard to win, but a competive entry should not be bad. 10:20:07 just have to make sure you follow the protocol. 10:20:11 --- nick: hi_imok_how_r_u -> divgrad 10:20:19 tcn: If the density of population of robots in world be high 10:20:27 most entries will be eliminated, because of bugs. 10:20:30 task require AI. 10:20:32 hehe 10:20:43 there is no AI 10:21:07 however, there is natural selection.. 10:21:37 hi, ASau 10:21:41 My English is not so good for I explain why AI. 10:22:16 --- part: divgrad left #forth 10:22:33 Good evening, divgrad! 10:23:17 Where have this Laplasian disappear?! 10:23:39 a rule based aproach is what you need. 10:24:12 not quite sure how to do that in forth, but I've heard it is possible. 10:24:14 mslicker: A rule based approach is a rudiment of intellect. 10:25:06 I.e. you can't _think_, so you replace _real_thought_ by set of rules. 10:25:10 computers just follow rules.. humans have billions of years of evolution.. 10:25:33 35000 years of Homo Sapiens. 10:25:58 H.S. is only 35000 years old. 10:26:08 at most 40 000 10:26:21 an AI robot can't learn enough to compete with rule-based robots in the number of games in the tournament 10:27:17 You're given 1 sec/move on 1,5 GHz CPU for adaptation of you rules. 10:27:47 yes you could switch stratagies, mid play. 10:27:49 If you've made a kind of evolutionary automaton. 10:28:10 switch to a different set of rules. 10:28:27 see how well that works. 10:28:40 maybe evolve the strategy. 10:28:51 basically depends how many robots there are, how close, and if you have a package 10:29:01 This adaptation is a must for AI prog./sys. 10:29:51 lisp would have an advantage for this. 10:30:19 M.S.: Of course. 10:30:36 and almost every functional lang. 10:30:51 no, lisp is special 10:31:04 functions and data are the same. 10:31:45 you can evolve the data/function 10:31:52 FPL is not and FPL, if it cannot construct functions on-the-fly. 10:32:01 It is by definitions. 10:32:35 FPL means functions be equal in rights to data. 10:32:50 no, not really 10:33:12 some languages are purely rule based, like prolog 10:33:19 Show me an FPL that contradicts. 10:33:29 Prolog is not an FPL. 10:33:50 ProLog = Programming Logic(ally/s) 10:33:54 other languages have high order functions, but dat is seperate. 10:34:27 No doubt LISP is the lowest level FPL. 10:34:48 lambda calculus, does not imply data of any kind. 10:35:27 maybe forth is, I think 10:35:29 An "operation" of "apply" imply functions and data. 10:35:39 then lisp 10:35:50 Functions are applied not to functions but to data. 10:36:37 Lambda calculus even uses formal parameters. 10:36:46 lambda calculas has no distinction. 10:37:05 numbers are encoded as lambda expressions. 10:37:35 It's shown that lambda calc. may be reduced to no formal parameters 10:38:14 There is a concatenative/combinatoric way to do things. 10:38:42 of course functional languages, encode numbers in an efficient way. 10:38:57 not as functions. 10:39:11 J.Backus, inventor of FP, used purely functional lang., based on composition. 10:39:32 Not on lambda's. 10:39:42 did it also have application? 10:39:49 Yes. 10:39:52 IIRC. 10:40:17 The forth model is a compositional one. 10:40:28 See Joy. 10:40:45 A Forth-like (but) functional lang. 10:40:47 yes, Joy is the one that points this out. 10:41:10 functional language is loosely defined. 10:41:28 I know, that Forth gives a way to program in a functional style. 10:41:33 few, are pure. 10:41:42 One thing it lacks. 10:41:58 Automatic heap memory manag. 10:42:52 that could be added, 10:43:18 If a function decides to build an n-dim array 10:43:28 Forth will have nothing to do. 10:43:47 Forth will have only one way -- to stop. 10:44:18 But if we add this feature to Forth, will it remail being Forth. 10:44:42 "That's the question." (C) Shakespare. 10:45:12 well, when Chuck solves a problem he builds a language to suit. 10:45:36 Well. 10:46:06 He solves not a contest's kind of problems. 10:47:46 These "..FP" meaning "Functional Programming" are not a joke or smth like. 10:48:48 They mean such problems (at our level of knowledge) requre functional programming techniques. 10:50:34 heap memory management can only be at edit time, unless you want to loose precious compile and run time simplicity and speed 10:51:14 forth is truly one big program, it emulates perfect hardware for the job at hand using the primitives the host give to it 10:51:23 --- join: mslicker0 (~maslicke@64.27.199.31) joined #forth 10:51:50 so please pretty please dont ever talk using forth for heap management, thats your job ;p 10:51:56 connection trouble. 10:53:29 forth at our stage need a good "dictionary scanner": the ebst way possible in forth to search through a dictionary ame of source only 10:53:57 damn epic4! 10:54:07 --- quit: mslicker (Read error: 60 (Operation timed out)) 10:54:11 so that compilation can be made only after the program is constructed from source fragments 10:54:35 --- nick: mslicker0 -> mslicker 10:54:38 referenced from the programmer's dictionary, or as i like to think, his spellbook ;p 10:54:40 ASau: the priv conv appeared in the #squeak window, so i missed everything else happening here :) 10:55:22 ASau: I think forth is just a capable as these languages, 10:55:48 just the aproach is different. 10:56:29 Forth doesn't start with so many libaries, 10:56:46 the aproach is to build a language to suit the problem. 10:56:54 no fixed assumptions. 10:57:15 maybe libraries would be useful in a contest like this, 10:57:32 where the emphasis is on getting to the solution quickly. 10:57:47 CrowKiller: u r right regarding the heap mngmnt, BUT 10:58:54 CrowKiller: when some1 says "4th needs ..." that means we need a proven implementation of ... built on top of the words of the ancient 4th model 10:59:51 eg, i liked the double linked listd kind of heap mngmnt of pascal 11:00:10 s/listd/list/ 11:00:11 CrowKiller: A straight q: DO YOU KNOW ANYTHING EXCEPT FORTH? 11:00:49 but there is a similar solution for gforth by wil baden or a similar guy 11:00:53 CK: Have you programmed a legendary Eliza-like robot? 11:01:08 an intersting thing, is that colorForth has the possiblity of code evolution. 11:01:48 mslicker: ? could u explain it more? 11:01:48 the preparsed block format, and instant compilation, make this possible 11:02:15 mslicker: cf is the evolution direction 11:02:26 onetom: He means he is able to change source and dyn. link it in. 11:02:27 mslicker: or u r talking about genetic programming? 11:03:02 ASau: sounds good, but i havent read about this on CF pages yet.. 11:03:24 CK: An unhumble question. 11:03:50 so could some of u support me w urls on this topic? 11:04:20 How do you think, does a scientist know how much resources he'll need next month? 11:05:17 ?? 11:05:18 That is the first difference of intelligent thing/creature of unint. 11:05:37 what? 11:05:54 what resources? memory, speed? 11:05:55 An automaton won't be able to behave right in future. 11:06:31 Because future is not present: it can, may and will change everything. 11:07:08 whom does the HE referes to in the sentence w the word "scientist"? 11:07:20 Well. 11:07:39 HE refers to the scientist himself? 11:07:41 I "ate a dog" in such disputes. 11:07:56 It's rather familiar. 11:08:14 r u tryin 2 compare a robot w a scientist or what? 11:08:36 An automaton is assigned a capabilities of intellect. 11:08:40 i feel i wont get the point :) 11:09:08 assign requires a TO clause... 11:09:12 I say: my English is not enough to explain this well. 11:10:01 but its not too far from enough ;) 11:10:07