00:00:00 --- log: started forth/02.12.21 00:10:08 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 00:16:36 --- quit: Serg_Penguin () 00:21:06 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 00:34:03 --- quit: Serg_Penguin () 01:04:23 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 01:07:23 --- quit: Loke3 (Read error: 104 (Connection reset by peer)) 01:15:22 --- join: joonas (jpihlaja@kruuna.helsinki.fi) joined #forth 01:15:54 hey joonas 01:16:29 hei 01:16:39 what's going on? 01:17:29 ?? 01:17:41 i'm testing trash cdr's 01:17:41 What's the topic of discussion, if any? 01:17:55 forth programming language 01:18:12 :) 01:18:22 I had that coming. :) 01:18:45 chan is quite dle so glad see u 01:18:55 _I_dle 01:19:35 everyone here knows my kbd is wreck :( and can't change coz it's at work :((( 01:20:27 so what's u'r experience in Forth ? 01:20:55 How do you mean? I'm not a professional Forth programmer, just a hobbyist. 01:21:25 And yours? 01:21:49 logic gamez, tiny utils 01:22:16 < 1 year 01:22:58 ah... now I understand your question. I mainly use forth to write small utilities. Been using it for about two years now, on and off. 01:23:55 --- join: Loke3 (martinbl@as2-1-1.kt.g.bonet.se) joined #forth 01:24:06 Do you have much experience in asm coding for x86? I'm looking for some opinions as to the most portable assembler for Linux/Windows environments. 01:24:49 my ASM exp -> 0 01:24:53 I've not done any asm programming for years, and am looking to start again. 01:25:50 for what hell do u need portable asm, if asm proggies aren't portable by definition ? 01:26:15 Serg_Penguin: some notes on color4th: 01:26:20 12-21 08:53:49 < onetom> hmm.. iwas thinkin about a simple flux editor 01:26:20 12-21 08:54:11 < onetom> plain text editor (during showering :) 01:26:20 12-21 08:54:38 < onetom> and realized that u can change the name of any definition 01:26:23 12-21 08:55:07 < onetom> & u dont have 2 search&replace all its occurances 01:26:42 imeant: further references 2 that word 01:26:57 but gotta go now 01:27:02 you think of strange things while showering :) 01:27:11 comment would b welcome 01:27:23 good luck, didn't grasp what u said at 4 what purpose :((((( 01:27:24 lament: yeah, iknow :DD its not 2 common 01:27:43 Serg_Penguin: read intro 2 aha on ultratech 01:27:59 serge_penguin: different kind of portability... as I'm only planning to target x86 at the moment, I'd like to use an assembler that works painlessly with both Linux and Windows. 01:28:19 Serg_Penguin: then u gonna understand what is a "half-compiled" / "compressed" src is 01:28:23 So far, I'm most impressed by nasm. 01:28:28 http now blocked, will read later if remember 01:28:57 blocked http? how disgusting... use a proxy :) 01:29:20 -> joonas i was about to advice nasm too, but my opinion dunno mean much coz i dunno write in asm 01:31:00 -> onetom i won't use 'compressed' src, only maybe as 'object' or half-exe format, NOT meant 4r editing but 4 fast dynamic linking 01:34:00 --- quit: Serg_Penguin ("i have 2 reboot...") 01:36:45 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 01:36:49 re 01:37:15 serg_penguin: What does "re" stand for? 01:37:34 "re" = i'm back 01:38:04 Like "Re-Enter"? 01:38:08 y 01:38:29 thanks 01:38:30 so what did u wrote last, in Forth or anything other ? 01:39:40 Lately I've been writing some code to analyse the accuracy of some methods of solving ordinary differential equations. 01:40:10 Using OCaml for that. 01:40:25 8-{P} bletch, i dunno like maths 01:40:58 me - many text utils and 2 incarnation of Sokoban, the game 01:41:52 unicode2ascii (RUS), email harvester, dissociated press(text mangler) 01:42:26 2nd sokoban - all in one-line words 01:42:27 I did Soko too, with the nethack 3.3.1 levels. I could never solve them in real play so wanted to practice some without bug bears and wolves fighting me. :-) 01:42:50 Still can't solve one of the last levels. :-/ 01:43:04 so write solver :) 01:43:18 good idea.. 01:43:42 or better write _maker_ 4 soko puzz 01:44:10 give it size and difficulty and get fresh maze 01:44:28 i'll think on it 4 my Soko-]I[ 01:46:45 How does your unicode2ascii represent non-ascii characters? 01:48:52 it works just like codepage converter, but w/ many tables, switched by 1 char 01:50:23 do u grok ? 01:50:52 no, not really... I meant how does it represent, say a chinese glyph as ascii? 01:51:05 it's only RU 01:51:22 bit RU codepages have RU letters >12 01:51:28 >128 01:51:32 8bit 01:51:44 ah, so you're converting to Latin-14 (or whatever Latin version had RU chars in it.) 01:52:02 cp866 :) 01:52:38 q&d hack to read damn Word docs on non-Win'able boxes 01:53:21 if char has no representation, it's just swallowed 01:54:13 ok 01:55:30 or to clear macro virii :( 01:56:48 so what about soko _maker_, do u think it's good idea ? 01:57:34 It might be useful for a small soko game, so the levels don't have to be included. 01:57:50 not only small but endless... 01:58:10 :) But the game file itself would be small. 01:58:13 it's no interest solving 'em next time 01:59:20 mine is 26k w/ forth 01:59:29 hit ` and u are at console 01:59:38 How much of that is devoted to the levels? 01:59:54 and 6k - no forth, UPXed 02:00:12 What's a UPXed thing? 02:00:22 exe file compressor 02:00:28 ok 02:00:58 50 400 * 3 / 02:01:14 50 levels 02:01:17 20x20 02:01:27 6 cells in 16word 02:02:07 oops, + 100 bytes = coordinates 02:03:20 * joonas goes to find out what VxD is 02:03:49 6766 bytes - i ran game, hit ` and calculated in Forth console 02:04:07 VxD kinda like win98 kernel modules 02:04:29 fu..get it, windoze suxx 02:04:42 * joonas returns, dissapointed. 02:07:25 back 02:13:31 --- quit: Serg_Penguin ("has to go down and up..") 02:13:47 * joonas wonders how a Sokoban solver could be made efficient 02:16:47 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 02:17:24 re 02:18:53 The basic algorithm might be something like this: 1. If we've won, then stop. 2. find all boulders that can be pushed in any direction. 2. for each possible boulder and direction, push it and recurse to 1. 02:19:35 It should detect and avoid cyclic states in step 2. 02:22:07 I imagine that this simple algorithm would be really slow for even moderately small levels, because it basically searches all possible moves. The search tree should be pruned somehow. Any ideas? 02:23:05 play thinkflly easy levels and notice how _yu_ do solve 02:23:58 2) avoid moves what do stuck bombs 02:24:16 and make puzz unsolvable 02:25:20 hm... I wonder if a strategy that works backwards from victory towards the initial configuration wouldn't work just as well. 02:26:10 no 02:26:17 make 'position evaluation' function - sum of distances from bolders to places 02:26:40 current position may be unsolvable even if no bombs stuck 02:28:16 and first try moves what increase evaluation 02:28:49 But how do you prove that a position is unsolvable? 02:29:29 12-21 10:32:49 < Serg_Penguin> -> onetom i won't use 'compressed' src, only maybe as 'object' or half-exe format, NOT meant 4r editing but 4 fast dynamic linking 02:29:30 _without_ trying all possible moves in the search tree below that position. 02:29:49 dunno know, but some are obvios, like bomb cornered or 4 bombs in square 02:29:56 Serg_Penguin: "not meant 4 editing"? what do u mean by that? 02:30:22 chess are not _fully_ analyzed but comp beaten Kasparov 02:30:33 * joonas sits back to think soko-like thoughts 02:30:40 _full_ analysis is _not_ needed 02:32:08 onetom: how the hell will u edit 'compressed' src by u'r editor of choice? 02:32:20 how will u grep it ? 02:32:31 u dont have 2 02:32:55 * Serg_Penguin likes using common tools 02:33:06 grep is just a dumb tool searching src 02:33:21 it finds not only what u were seraching 02:33:32 coz it doesnt know about the structure of the text 02:34:08 u dont have 2 choose an editor but write 1 02:34:20 so thats gonna b the best 4 u 02:35:10 2 write a color4th editor - suit the 1 of flux 2 ur needs - is not 2 difficult 02:36:11 anyway, this is so easy 2 convert a "compressed" color4th src in2 html: 02:36:12 http://www.oakland.edu/~maslicke/colorforth/cf2html.c 02:36:16 * Serg_Penguin doesnt go as far as Chuck - writing all tools myself 02:36:33 do u think is really hard 2 write an editor 4 it? 02:37:00 do u think its... 02:37:10 i wanna write editor 4r normal text to fit my 1-liner style 02:37:51 of course the editor is a lil bit less compelx in that case 02:38:06 but u loose a lot of other benefits 02:38:33 and what do i lose ? 02:38:39 a lot :) 02:39:06 example ? 02:39:32 what do u loose if u write a vectorgraphic image directly in postscript instead of -say- coreldraw? 02:40:27 ???? dunno grasp, dunno do much vector gfx 02:40:30 u loose the benefits of that the system knows a lot about the src. 02:40:52 it "understands" the src already @ edit time 02:41:14 so it can help u autocomplete words 02:41:38 --- quit: Loke3 (Read error: 113 (No route to host)) 02:41:43 it has the ability of changing word names in 1 place 02:42:11 so s/.../xx/g functionality is avoided 02:42:31 the compiler gonna b more faster 02:42:39 and what if i make new name longer ? 02:42:40 and simpler of course 02:43:07 dunno 02:43:29 probably the words after it in that block shift away 02:45:24 shift and break ref's to them, or need to scan everything to fix ref's - no much simpler 02:45:31 Serg_Penguin: I'm not at school saturdays. 02:45:42 Serg_Penguin: Especially not on christmas vacation :) 02:45:57 sleeping till noon :))) ? 02:46:02 Yeah ;) 02:46:13 i'm sleepy too 02:46:25 I've been sleeping way too little the last few days. 02:46:49 And that just ends up with that I don't know who or where I am in the afternoon. 02:47:13 me too - love and hacking is BAD00Mish mix 02:49:12 onetom: if i would write so big proj to wrry abut compile speed, i'll compile incrementally, saving Forth system w/ partly compiled proj 02:49:14 scan everything? 02:49:25 just that block 02:49:30 its 1k 02:49:49 doesnt requires 2 much processing speed... 02:49:51 so are ref's segmented, like block:offset ? 02:50:05 any comp can easily do the calculations while u type 02:50:41 hmm... it seems a dumb question 2 me 02:50:52 ??? 02:51:04 what does "segmented" mean? its just .... 02:51:20 how can i explain, let me c... 02:51:36 block:ofs = block*1024+ofs 02:51:48 ok 02:51:51 = blk << 10 + ofs 02:52:02 so what? why is it segmented? 02:52:42 to have some slop between words, would i want insert in middle 02:53:30 its only segmented if u wanna consider it structured like blk:ofs 02:54:13 anyway, i won't use 'packed' src - i'm happy w/ traditional 02:54:20 sure 02:54:33 a lot of ppl r happy w c 02:54:34 C 02:54:49 not me :) 02:55:03 c==fu+++bar 02:55:11 2 gochas in 1 line 02:55:22 utter obfuscation 02:55:32 extraneous typing 02:55:34 hehe 02:55:42 CRT estate waste 4 indents 02:56:52 C is moby lossage 4 me 02:56:57 i just want 2 show u a paralellizm between the ppl stucked 2 C & u "stucked" 2 traditional src 02:57:13 i dont even use compressed src yet 02:57:25 but it seems a pretty advanced idea 02:57:44 & its not even complicated! 02:57:52 & its still not even complicated! 02:58:20 & could give u very nice benefits 02:58:59 --- quit: Serg_Penguin ("i gotta reboot....") 03:02:24 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 03:02:33 re 03:02:44 ack 03:02:52 hey serg 03:02:55 hi 03:03:04 so 03:03:06 hi 03:03:26 iwas about 2 tellin ya that i already had compiletime issues 03:04:09 gforth was terribly slow on a 386 & it took ~8secs 2 recompile my program on linux 03:04:21 how big project ? 03:04:32 BUT not this is the only point of fast compiles 03:04:40 gforth - bletch ! 03:04:41 it was THAT big :) 03:05:07 it had a lot of tables 03:05:19 and what about comments in packed code ? 03:05:20 data tables described in 4th 03:05:40 comments r stored "uncompressed" 03:05:48 so pre-compile tables and save 'em w/ Forth 03:05:56 & they r immediately skipped @ compiletime 03:06:05 no need 2 parse them that time 03:06:38 Serg_Penguin: hehe. i had 2 recompile all the time because of the data tables 03:06:57 ??? 03:07:02 coz i represented the data not in simple data structures 03:07:07 they changed ? 03:07:09 but directly in code 03:07:16 of course 03:07:40 they were changing when i was developing 03:08:14 but w ultrafast compile time u can choose much simpler programming solutions 03:08:28 so u can even compile interrupt routines on the fly 03:08:58 & this can simplify code a lot 03:09:57 but u should read the intro 2 aha article.. 03:10:09 its worth 2 read it. a lot.. 03:10:15 but 'compressed' is opaque format.. 03:10:23 so gimme exact link pse... 03:10:27 sec 03:11:31 ..and if u send u'r src to other guy who hasn't utils to view it... 03:11:38 http://ultratechnology.com/ahatalk.htm 03:11:40 thats it 03:11:51 its a lil bit hard 2 understand 03:11:59 ..and how u'll suck in src from outer world ? add traditional compiler ? 03:12:15 so i also read the discussion between jeff & sean 03:12:43 & that has enlighted these fantastic ideas iwas talking about above 03:13:22 u can easily implement a traditional compiler 2 03:13:43 in fact flux also has EVALUATE 03:13:54 i mean not traditional to code 03:13:55 what understands ans forth in plain txt 03:14:04 i mean text<>compressed 03:14:12 ? 03:14:45 these things r already pretty well thought 03:14:57 traditional<>compressed(colorforth'ish) 03:15:28 certainly further research is still necessary but its worth 2 read & start thinking about these technologies... 03:15:37 * joonas remembers he's supposed to be out christmas shopping 03:15:47 There's a ham out there with my name on it. 03:15:58 bye, and a merry christmas to all. 03:15:59 --- quit: joonas ("ircII2.8.2-EPIC3.004+Kasi --- Bloatware at its finest.") 03:16:00 umean how can u represent color4th in plain txt? 03:17:27 btw, only the last 1/5 of ahatalk.htm talks about this "compressed src db" 03:17:47 the other stuff just prepares it 03:19:42 downloaded 3 big articles 03:19:52 will palm & read 03:19:53 which 1s? 03:21:46 damn ! 3d didnt saved... 03:22:10 ahatalk.htm 03:22:11 enthflux.htm 03:23:33 --- quit: Serg_Penguin ("down and up..") 03:24:06 im very curious how impressed u gonna b :) 03:27:03 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 03:27:07 re 03:27:17 im very curious how impressed u gonna b :) 03:27:28 so which is the 3rd article? 03:27:49 oops, i remembered, t was link to some ads 03:28:01 my brain jams now... 03:34:38 i slept 2 damn little 03:41:19 me 2 03:41:24 * onetom -> bed 03:41:32 cu soon 04:19:00 --- quit: Serg_Penguin ("wll go down and up..") 04:23:25 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 04:23:31 re 04:28:54 re 04:29:42 Hey onetom :) 04:31:12 whats up? i asked u some questions last day u didnt answer afair 04:31:26 but i can remember. they werent important 04:33:17 Me? 04:33:38 07:03 < onetom> hey robert! thx 4 that great great great grand jpeg decoder code & 4 the page itself! 04:33:41 07:04 < onetom> it has some info on running chucks 4th in bochs 04:34:00 I only see that... 04:35:01 ah, thats it :) 04:35:14 and my note about the editor 04:36:23 08:58 < onetom> hmm.. iwas thinkin about a simple flux editor 04:36:23 08:59 < onetom> plain text editor (during showering :) 04:36:23 08:59 < onetom> and realized that u can change the name of any definition 04:36:23 09:00 < onetom> & u dont have 2 search&replace all its occurances 04:36:45 yup 04:36:46 I see... 04:37:17 dont u think it has a severe impact on code quality? 04:37:27 What has? 04:37:45 naming things the right way counts a lot 04:38:12 Yeah... 04:38:15 : x .... ; 04:38:16 : z .... ; 04:38:17 it u cant corret naming mistakes easily u wont, just because 04:38:18 : v4 .... ; 04:38:20 etc. ;) 04:38:52 the algorithm is independent of the names 04:38:53 yeah, names must be short because if I have to type a lot of useless characters I just won't code it at all =) 04:39:15 na, stop. 04:39:24 Or you do like on IRC, x :P 04:39:35 XeF4: yes I see what you mean 04:39:36 i missed sg probably, but i dont know what r u talkin about 04:39:44 onetom : but u'r view of algorithm _is_ 04:40:21 Serg_Penguin: sure. & thats why the renames r so important 04:41:31 i give names alike asm mnemonics 04:42:12 like !r+ 6*+ lvld ... 04:42:31 lvld? 04:42:36 load something something descriptor? 04:42:45 level load, in game 04:43:02 more decode than load 04:43:08 brb 04:56:06 back 04:56:57 'tis odd to be back in civilization 04:57:17 Were you visiting Sweden? 04:57:30 z 04:57:36 Korppoo (little Finnish with a shipyard and not much else) 04:57:43 Finnish island, even 04:59:52 :) 05:00:06 Korpo in Swedish 05:00:46 * Robert thinks of a bird. 05:06:20 could be, the place is mostly Swedish-speaking 05:06:35 supposedly there is a village there, but I never visited 05:06:53 and the shipyard is full of old men who just work there 05:10:34 --- join: Speuler (~Speuler@mnch-d9ba413f.pool.mediaWays.net) joined #forth 05:18:07 hi 05:18:14 'morning 05:18:54 * Speuler feels like being in GMT-6 05:20:30 --- quit: Serg_Penguin () 05:23:07 --- quit: Speuler (Remote closed the connection) 05:34:51 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth 05:42:03 z 05:45:19 zz 05:52:39 it wazzzzzzzz just like manual keepalive to avoid NAT dropping TCP link 4 nactivity :) 05:56:36 thought so 05:57:13 an E is the shortest.... in morse ;) 05:58:24 in RU it should be O 06:03:32 really? how come? morse is not international? 06:07:15 it IS i8l, but O is most frequent ltr in RU, not E 06:07:30 so morse is sub-optimal 4 ru 06:10:16 ah, got it now :) 06:12:50 Samuel morse was total loser 4 not adding redundancy of speech to his code 06:13:22 what sort of redundancy? 06:14:05 extra info to correct errors by deep layers of mind 06:14:21 ah right, but that is already present in language 06:14:37 like we overlook typos, not even thinkfully guessing what was typed 06:14:45 like in that last sentence, 'that' or 'is' or even both could be absent and it would still make sense 06:14:58 how does that change when it is encoded as morse? 06:15:42 chars longer than needed (code length) and harder to mix (code distance) 06:17:07 i mean rednundancy not only of language, but of speech 06:17:08 eg, so if a bit or two is missing you can still guess the letter most of the time? 06:17:18 yeah ! 06:17:30 missing=mixed 06:18:27 hm yes 06:18:35 like: E --. 06:18:43 O -.- 06:18:49 prefixes to specify vowel/consonant, voiced/unvoiced, etc would do nicely 06:18:54 but those are different between languages 06:19:04 and NO -.. or ..- 06:19:31 I could probably come up with a coding that would correct errors with Finnish text, but that isn't enough for an international code =D 06:20:07 general rule - cover any one sign in any latin letter, and u still can read 06:20:27 and u still can tell letters apart 06:22:46 --- quit: Serg_Penguin () 06:23:24 --- quit: lament ("mental mantle") 09:03:42 --- quit: proteusguy (Read error: 54 (Connection reset by peer)) 09:29:27 --- join: Speuler (~Speuler@mnch-d9ba413f.pool.mediaWays.net) joined #forth 09:56:49 rebol was forth-inspired somebody said here a while ago ? 10:08:56 Hrm. 10:09:08 grep says gilbertbsd mentioned it last. 10:09:19 And I haven't seen him in days. 10:09:36 ah. i just came across a version for my pocket machine 10:09:43 so i thought i'll put it on 10:13:52 What _is_ rebol? 10:14:20 programming lang/envi said to facilitate network programming 10:15:28 OK :) 10:15:28 bbl 10:33:43 --- join: wossname (wossname@HSE-QuebecCity-ppp81820.qc.sympatico.ca) joined #forth 11:17:38 --- nick: wossname -> haxint3 11:25:24 --- quit: haxint3 ("WTF@!$!$!@~@~# p0rtscan(!$$)") 13:32:12 --- join: Klaw` (anonymouse@ip68-4-77-1.oc.oc.cox.net) joined #forth 14:11:49 --- join: TheBlueWizard (TheBlueWiz@ip-216-25-205-135.vienna.va.fcc.net) joined #forth 14:11:54 hiya all 14:15:41 Hi :) 14:15:47 hiya Robert 14:39:18 hi TBW 14:39:33 hiya Speuler 14:39:36 mr dobson ? 14:39:39 seen I440r? 14:39:47 not today 14:39:50 mr dobson who? 14:40:15 chris ? 14:40:59 just wondered whether icq TheBlueWizard and you are the same person. 14:43:15 oh...hmm...nope...I don't have ICQ (wait, there are only ICQ numbers, right?) 15:32:42 well...gotta go...bye all! 15:32:50 --- part: TheBlueWizard left #forth 15:42:47 --- join: Xuz (aemerson@bgp01132867bgs.ypeast01.mi.comcast.net) joined #forth 16:31:01 --- quit: Xuz ("[BX] We are BitchX of Borg. You will be assimilated. Using ircII is futile.") 19:42:46 --- quit: Speuler () 20:00:16 --- join: ASau (~asau@158.250.48.197) joined #forth 20:00:57 --- quit: ASau (Remote closed the connection) 20:01:22 --- join: ASau (~asau@158.250.48.197) joined #forth 20:01:35 Good morning! 20:02:32 Is anybody alive? 20:03:06 Well, I see noone. 20:03:08 Bye! 20:03:09 --- quit: ASau (Client Quit) 21:02:12 --- quit: TreyB () 21:18:24 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 22:22:29 --- join: lament (~lament@h24-78-145-92.vc.shawcable.net) joined #forth 22:49:33 --- join: proteusguy (~username@65.191.88.177) joined #forth 23:47:19 --- quit: proteusguy (Read error: 54 (Connection reset by peer)) 23:47:31 --- join: proteusguy (~username@65.191.88.177) joined #forth 23:59:59 --- log: ended forth/02.12.21