00:00:00 --- log: started forth/02.05.18 00:02:55 --- join: davidw_ (~davidw@adsl-209-109.38-151.net24.it) joined #forth 00:03:15 --- nick: davidw_ -> davidw 00:07:36 hi davidw 00:07:40 im about to test isforth in fbsd heh 00:07:48 im hoping ive fixed it@ 00:27:14 --- quit: I440r ("Reality Strikes Again") 00:46:35 --- join: Soap` (flop@210-55-148-151.dialup.xtra.co.nz) joined #forth 01:12:54 --- join: adu (~andrew@dsl-64-130-166-225.telocity.com) joined #forth 01:14:02 wow lots of ppl 01:54:21 --- quit: adu () 02:40:03 morning :) 02:40:56 --- nick: zZzZz_bert -> rob_ert 03:43:37 --- quit: oink (carter.openprojects.net irc.openprojects.net) 03:43:37 --- quit: davidw (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: air|cabin (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: Soap` (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: _MrReach_ (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: onetom (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: goshawk` (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: Etaoin (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: deltab (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: gforth (carter.openprojects.net irc.openprojects.net) 03:43:38 --- quit: jcassady (carter.openprojects.net irc.openprojects.net) 03:43:39 --- quit: rob_ert (carter.openprojects.net irc.openprojects.net) 03:43:39 --- quit: cdesousa (carter.openprojects.net irc.openprojects.net) 03:43:50 --- join: Soap` (flop@210-55-148-151.dialup.xtra.co.nz) joined #forth 03:43:50 --- join: davidw (~davidw@adsl-209-109.38-151.net24.it) joined #forth 03:43:50 --- join: oink (~ziga@owl.cuckoos.net) joined #forth 03:43:50 --- join: _MrReach_ (~mrreach@209.181.43.190) joined #forth 03:43:50 --- join: onetom (tom@adsl52029.vnet.hu) joined #forth 03:43:50 --- join: gforth (guest@adsl52029.vnet.hu) joined #forth 03:43:50 --- join: air|cabin (brand@12-254-199-50.client.attbi.com) joined #forth 03:43:50 --- join: rob_ert (~robert@h237n2fls31o965.telia.com) joined #forth 03:43:50 --- join: goshawk` (goshawk@panix1.panix.com) joined #forth 03:43:50 --- join: Etaoin (~david@ljk26.sat.net) joined #forth 03:43:50 --- join: deltab (~deltab@216.234.235.21) joined #forth 03:43:50 --- join: jcassady (~jcassady@adsl-64-171-52-226.dsl.snfc21.pacbell.net) joined #forth 03:43:50 --- join: cdesousa (syf@faeldryn.demon.nl) joined #forth 04:25:04 --- join: gforth0 (guest@adsl52043.vnet.hu) joined #forth 04:25:04 --- quit: Soap` (Read error: 104 (Connection reset by peer)) 04:31:34 --- join: Speuler (~l@a161161.upc-a.chello.nl) joined #forth 04:31:45 --- join: erlichr (~erlichr@24.42.170.144) joined #forth 04:37:09 --- join: onetom_ (tom@adsl52043.vnet.hu) joined #forth 04:43:37 --- quit: onetom (Read error: 110 (Connection timed out)) 04:43:37 --- nick: onetom_ -> onetom 04:43:47 --- quit: gforth (Read error: 110 (Connection timed out)) 04:43:47 --- nick: gforth0 -> gforth 04:43:58 Hrm. 04:44:23 Isforth prints a space after the number when I use dot :-/ 04:47:40 --- quit: erlichr ("Leaving") 04:53:14 Hrmm.. 04:53:20 How do I convert a number to a string? 04:53:26 Is there any standard way? 04:59:21 rob_ert: dont think so :( 04:59:32 Hrm :-/ 05:00:40 rob_ert: look 05:00:44 g4 see . 05:00:47 onetom: : . 05:00:48 onetom: s>d d. ; 05:00:51 g4 see d. 05:00:54 onetom: : d. 05:00:55 onetom: 0 d.r space ; 05:01:01 g4 see d.r 05:01:04 onetom: : d.r 05:01:05 onetom: >r tuck dabs <<# #s rot sign #> r> over - spaces type #>> ; 05:01:47 and there u go, "<# #s #> type" is the construct 4 u 05:01:58 the <<# is a gforth extension 05:02:03 over <# 05:02:14 Hmm 05:02:18 I'll check. 05:02:25 g4 123. <# # # # # # #> type 05:02:28 onetom: 00123 05:02:44 g4 123. <# # # char - hol # # # #> type 05:02:47 onetom: in file included from *the terminal*:-1 05:02:47 onetom: /tmp/fsock-sh-server.request.tmp:46: Undefined word 05:02:47 onetom: 123. <# # # char - hol # # # #> type 05:02:47 onetom: ^^^ 05:02:47 onetom: Backtrace: 05:02:49 g4 123. <# # # char - hold # # # #> type 05:02:52 onetom: 001-23 05:03:26 so mind that it converts the digits in a reverse order 05:04:09 :-) 05:05:20 its terribly not convenient i think :( 05:05:33 and <# r not nestable 05:05:35 and so on 05:07:04 output routines have also been kept minimalistic :/ 05:07:28 Hehe 05:07:33 Well, I'm leaving now 05:07:36 thanks for your help 05:13:22 onetom: you still can process the result string from <# ... #> to work on digits in a reverse fashion 05:26:08 --- join: mur (ammu@baana-62-165-189-112.phnet.fi) joined #forth 05:37:49 hi mur 05:38:15 from the country of the_linux 05:42:50 onetom: i have an implementation of <# # #s hold #> which don't use a buffer, but work exclusively on the stack. those are nestable. 05:43:36 onetom: #> copies to buffer, but #>type doesn't (outputs chars from stack) 06:00:35 --- join: Fare (fare@samaris.tunes.org) joined #forth 06:01:07 --- quit: davidw (Read error: 113 (No route to host)) 06:15:34 --- quit: onetom (Read error: 104 (Connection reset by peer)) 06:16:30 --- quit: gforth (Read error: 104 (Connection reset by peer)) 06:17:02 --- join: gforth (guest@adsl52043.vnet.hu) joined #forth 06:29:35 --- join: onetom (tom@adsl52043.vnet.hu) joined #forth 06:35:34 --- nick: mur -> murfood 06:45:04 Hmm... does anyone know if Mark will release the socket code for IsForth? 06:46:48 --- nick: murfood -> mur 07:00:58 --- quit: Speuler (Read error: 113 (No route to host)) 07:06:36 --- nick: mur -> murBBL 07:37:46 --- nick: murBBL -> mur 07:41:36 --- quit: _MrReach_ (Read error: 104 (Connection reset by peer)) 08:20:19 --- quit: Fare ("Connection reset by pear") 08:29:21 --- join: CrowKiller (Vapo_Rulez@cnq5-233.cablevision.qc.ca) joined #forth 08:29:49 hi 08:29:56 h 08:30:32 is Mr Moore supposed to be returning today, or was that rescheduled/cancelled? 08:30:50 wasny it 18th`? 08:31:40 he's supposed to come by today 08:31:55 mur: today is the 18th...at least my time 08:32:03 my = by my 08:32:30 crow: thanks! 08:32:31 but I know nothing more about it ;p 08:35:48 no problem =) 08:37:05 goshawk` m,aybe eu time 128th? 08:37:15 oh 08:52:03 --- join: gforth0 (guest@adsl52014.vnet.hu) joined #forth 09:01:29 --- join: Fare (fare@samaris.tunes.org) joined #forth 09:01:39 --- quit: Fare (Read error: 104 (Connection reset by peer)) 09:02:58 --- join: onetom_ (tom@adsl52014.vnet.hu) joined #forth 09:04:40 --- join: GilbertBSD (~gilbert@m153.max3.dacor.net) joined #forth 09:04:53 is Chip chuck coming over today? 09:05:43 hope so ... 09:07:20 http://www.digdug.cx/quotes/?5542 but it is just funny 09:09:58 --- quit: onetom (Read error: 110 (Connection timed out)) 09:09:59 --- nick: onetom_ -> onetom 09:10:35 :)) 09:10:39 --- quit: gforth (Read error: 110 (Connection timed out)) 09:10:39 --- nick: gforth0 -> gforth 09:10:58 damn, i have even given some password to nickserv, what a dumbass i am ;) 09:12:50 yeeah, how irresponsible u r :) 09:22:03 http://www.digdug.cx/quotes/?195 09:24:10 How do I create a string? : string create allot does> ; didn't work... 09:26:28 And also, how is S" etc. used? 09:42:54 hi Robert 09:43:03 Hi :) 09:43:20 how would you create a string in Asm? 09:43:44 --- join: tathi (~josh@ip68-9-58-81.ri.ri.cox.net) joined #forth 09:44:20 Uhm 09:44:27 In the data section, do this: 09:44:36 mystring db "Hello, world!",0 09:45:34 so why can't you do the same in forth? 09:46:05 s" Blah" doesn't work in IsForth 09:46:12 And I don't know how to do :-/ 09:46:24 does your forth do it? 09:46:30 g4 see s" 09:46:31 tathi: You don't happen to know how strings are handled in isForth? 09:46:33 GilbertBSD: noname : 09:46:34 GilbertBSD: 34 parse /line min >r s"-buffer i cmove s"-buffer r> ; 09:46:34 GilbertBSD: lastxt 09:46:34 GilbertBSD: noname : 09:46:39 GilbertBSD: My forth can do it, yes. 09:46:48 But my forth can't do much else, so I don't use it ;) 09:46:50 so do it in your forth then! 09:46:59 g4 see s"-buffer 09:47:03 GilbertBSD: Variable s"-buffer 09:47:47 g4 s" blah " 09:48:13 gforth: : x s" Blah" type ; x 09:48:16 rob_ert: Blah 09:48:19 :) 09:48:33 g4 .( Why don't you just do this ) 09:48:36 GilbertBSD: Why don't you just do this 09:49:00 unless you want to do things to the X 09:49:25 Uh 09:49:31 I don't want to _print_ a bloody string. 09:49:37 I want to declare/define/whatever. 09:49:43 so don't make it bleed then! 09:49:44 duh. 09:49:48 ." blah" is enough for me when it comes to printing. 09:51:35 rob_ert: so why can't you create s" in isforth? 09:52:09 Well... Mark thinks s" is evil 09:52:16 So I wanted to know how his solution worked. 09:52:32 and you don't think S" is evil so go with your non-evil 10:01:50 --- part: GilbertBSD left #forth 10:49:54 Hrmm 10:50:38 I don't want turnkeyed apps to contain every damn word... it's like compiling in the entire glibc in your C programs! 10:51:30 Hrmm... and I certainly don't want his ASCII art ;) 10:58:11 --- join: Speuler (~l@a161161.upc-a.chello.nl) joined #forth 10:58:31 Hey Speuler. 10:58:38 Do you know how to handle string in IsForth? 11:05:01 --- join: oner (oner@Ba71e.pppool.de) joined #forth 11:05:17 :) 11:05:55 .(hello) 11:06:23 Nono 11:06:25 .( Hello) 11:06:30 gforth: .( Hello) 11:06:33 rob_ert: Hello 11:06:35 :) 11:06:40 ok! 11:06:45 gforth: .(Hello) 11:07:02 Hmm.. he didn't like that :) 11:07:14 --- join: thefox (fox@adsl-209-182-168-45.value.net) joined #forth 11:07:18 Hey :-) 11:07:22 heyhey 11:07:32 Hmm.. was it today Chuck was coming? 11:07:55 note for oner: Chuck Moore is the creator of forth ;-) 11:08:00 rob_ert: in file included from *the terminal*:-1 11:08:00 rob_ert: /tmp/fsock-sh-server.request.tmp:46: Undefined word 11:08:00 rob_ert: .(Hello) 11:08:00 rob_ert: ^^^^^^^^ 11:08:00 rob_ert: Backtrace: 11:08:05 ok ;) 11:08:12 oner: See, that didn't work :) 11:08:30 hi the_rob 11:08:35 Hi Speuler. 11:09:15 missing space 11:09:48 I know. 11:09:54 ah 11:09:55 gforth: .( Hello) 11:09:55 rob_ert: Hello 11:10:01 .(hello) 11:10:13 Just wanted to show him it doesn't work like that :) 11:10:25 You don't happen to know if isforth has any support for strings, Speuler? 11:10:45 not that i know but shouldn't be too much effort to add it 11:11:17 Well, no... 11:11:39 I should bug I440r to add it now, so I don't have to reinvent the wheel :) 11:11:53 ok 11:13:02 what sort of string support do you need ? 11:13:27 Uhm 11:13:34 Just so I can play with 11:13:51 like with s" filename" open ? 11:13:55 Need to be able to do something like s" my.file" 11:13:58 Yeah. 11:14:12 does it have ," ? 11:14:49 * rob_ert checks. 11:15:18 Thanks Speuler :) 11:16:01 maybe create foofile ," filename.ext" would do 11:16:52 --- quit: oner (Read error: 104 (Connection reset by peer)) 11:17:10 or : open" bl word count open ; 11:17:13 sorry 11:17:20 or : open" ascii " word count open ; 11:17:29 open" filename.ext" 11:17:43 Nah, he uses raw syscalls. 11:18:07 and so ? 11:18:27 hidden behind open 11:18:40 Not hidden :-) 11:18:45 open" just parses a string and passes it to open 11:18:46 Hm 11:18:49 Uhm 11:18:52 n/m 11:18:58 I misunderstood you 11:19:22 btw, he uses the '"' syntax for charachter literals. 11:20:09 or decimal : open" 34 word count open ; 11:21:02 --- join: gforth0 (guest@adsl52005.vnet.hu) joined #forth 11:21:08 --- join: svara (~svara@pD950B1DD.dip0.t-ipconnect.de) joined #forth 11:21:26 Hi svara... 11:21:34 rob_ert! 11:21:38 Back from your garden? 11:21:45 heh 11:21:56 gforth0: 1 dup base ! . 11:22:36 Evil. 11:23:22 sif: 1 dup base ! . 11:23:24 Speuler: Word not found: base 11:23:56 Hehe 11:24:29 1 dup base ! . 11:24:29 *the terminal*:1: Pictured numeric ouput string overflow 11:24:40 that's gforth 11:24:48 Yes. 11:24:54 * rob_ert test isforth 11:25:08 It segfaults :) 11:25:16 --- join: I440r (~mark4@1Cust232.tnt2.bloomington.in.da.uu.net) joined #forth 11:25:37 rob_ert test isforth 11:25:37 It segfaults :) 11:25:42 hi i440 11:25:43 r 11:25:48 svara = new #forth person ? 11:25:51 missed segfault by 10 seconds 11:25:53 Nah 11:26:05 svara's just a german dope addict, coming here to get some company. 11:26:18 Anyway... I had a bunch of questions for IsForth... 11:26:22 Forgot most of them :) 11:26:31 about IsForth* 11:27:56 Oh, right... when will you release your socket code? 11:28:02 lol 11:28:11 segfault ? 11:28:17 oh 11:28:17 not sure 11:28:24 It seems pointless that me or someone else would have to rewrite it... 11:28:45 (It'll go faster since we don't have to reverse-engineer glibc to do it ;D) 11:28:49 rewrite what ? 11:28:56 sockets code. 11:28:57 isforth or sockets 11:29:00 lol 11:29:05 Not the entire IsForth 11:29:08 lol ive been doing that heh 11:29:26 right now im trying to add dns queries 11:29:35 you have a need for sockets ? 11:29:56 futhin has a need for them for his mud 11:30:02 Well 11:30:07 bongo who/what was segfault ? 11:30:10 I like playing around with stuff. 11:30:12 hehe 11:30:18 1 dup base ! . 11:30:26 *evil* 11:30:33 lol 11:30:49 answer has to be zero 11:30:57 any other answer will be wrong 11:31:05 the highest digit is always 1 less than the base 11:31:05 Well... hell of a lot of zeroes it seems :) 11:31:21 so in base 1 only zero exists heh 11:31:34 Yeah.. 11:31:40 ya this world is FULL of alot of zeros 11:31:51 bill gates 11:31:58 And all number expect zero will fuck things up :D 11:32:00 bill clinton 11:32:06 hrm alot of them seem to be called BILL!!! 11:32:12 1 base ! 0 . 11:32:13 0 11:32:14 :) 11:32:38 neway i dont know if the sockets code is going to be released to the public or not 11:32:47 Hrm.. 11:32:48 it MIGHT be commercial only code 11:32:50 Oh, well... 11:33:07 Then I'll have to go the long way ;) 11:33:20 --- quit: onetom (Read error: 110 (Connection timed out)) 11:33:22 --- quit: gforth (Read error: 110 (Connection timed out)) 11:33:23 --- nick: gforth0 -> gforth 11:33:36 i MIGHT release it as a "for non commercial use only" commercial users can pay me $2045902790 gazillion dollars first 11:33:42 :D 11:33:47 Hmm 11:33:50 Another thing... 11:34:18 What should I modify so stand-alone programs don't show your violet ASCII art? :) 11:34:36 its magenta!!!!!!!!!!!!!! 11:34:43 heh 11:34:49 statoff turns the status bar off 11:34:49 erm 11:34:56 hang on you can turn the hello off too 11:35:02 --- join: onetom (tom@adsl52005.vnet.hu) joined #forth 11:35:12 look in hello.f 11:35:19 And the timer? 11:35:20 * Speuler imagines the queues of the fortune-500s struggling for i440r's socket code 11:35:20 Okie 11:35:36 Will I have to recompile the isforth executable? 11:37:54 Hmm... so, what definition shows all the crap? 11:38:27 speuler !!! exactly!!! hehe 11:38:47 is magenta used because of its resemblance to purple ? 11:38:50 to juse the sockets ? 11:38:58 yes :) 11:39:02 ah 11:39:18 i actualy modified the palette for eterm/linux console to make the magenta less PINK!!! 11:39:21 What definition shows all the crap? 11:39:25 Hehe 11:39:29 magenta by default is a girlie pink colour 11:39:30 I'm color-blind anyway 11:39:53 so why'd you call it magenta if it's purple ? 11:40:02 robert you dont want the hellp message when you run isforth ? 11:40:14 because magenta IS purple lol 11:40:43 purple has more blue/less red 11:40:49 I440r: I don't want it in turnkeyed apps. 11:41:01 speaking of colors, 've seen a funny tshirt text: 11:41:09 aha - it wont be if you dont include it in them 11:41:12 (white letters on black tshirt) 11:41:22 I440r: Hum... how do I do then? 11:41:29 "i only wear black until they found something darker" 11:41:30 for a turnkeyd app you would make a copy of isforth.f and comment out all the things your app doesnt use 11:41:36 I440r: And - how do I shrink the size of the turnkeyd apps? :) 11:41:38 Oh 11:41:50 I see, 11:41:54 Thanks :) 11:42:11 then you would create a stripped down version of isforth 11:42:13 You should REALLY write some docs :) 11:42:19 robert heheh ya 11:42:21 i know :P 11:42:27 What's left to do really before you can start documenting? 11:42:29 i need some help doing that 11:42:33 * Speuler whistles innocently 11:42:39 Hmm.. if I'm bored enough...maybe ;) 11:42:45 i volunteer bongo!!!! :) 11:42:52 XD 11:42:55 heh 11:43:40 robert once isforth has an assembler/meta compiler ALOT of the stuff that is currently in the KERNEL will be removed to extensions 11:43:46 making them optionable 11:43:55 :)) 11:44:00 this will make turnkeyd apps ALOT smallker 11:44:02 smaller 11:44:03 Code faster! 11:44:08 but i need an assembler :( 11:44:17 it's two words, *a* *lot* :-) 11:44:30 deltab alot is a valid word 11:44:44 a lot is an area 11:44:52 more properly an alotment 11:44:54 deltab: Tell onetom how to spell, too :) 11:45:29 onetome has alot of abreviations i dont understand so i dont always get exactly what he is saying 11:45:41 lot 11:45:41 n 1: (often followed by `of') a large number or amount or extent: 11:45:41 "a batch of letters"; "a deal of trouble"; "a lot of 11:45:41 money"; "he made a mint on the stock market"; "it must 11:46:16 neway i gtg do some work... (deltab the word "alot" IS a valid word :P) 11:46:20 ill idle here 11:46:39 :) 11:46:43 Idle king. 11:46:44 I440r: show me evidence :-) 11:47:19 deltab: Is IRC logs evidence enough? 11:47:25 no 11:47:54 :-) 11:48:03 Ask onetom about english spelling. 11:48:24 He thinks the shortest and most unreadable spelling is the one to be used :) 11:58:31 --- join: Shrdlu (~david@ljk2-5.sat.net) joined #forth 11:59:21 --- quit: Etaoin (Killed (NickServ (Ghost: Shrdlu!~david@ljk2-5.sat.net))) 11:59:22 is futhin going to be handling this, today? 11:59:29 --- nick: Shrdlu -> Etaoin 12:00:13 --- join: oner (oner@Ba779.pppool.de) joined #forth 12:01:45 goshawk`: No idea :) 12:01:51 delightful :P 12:02:00 Ask futhin ;) 12:03:53 hah, yeah =) 12:03:55 w/c 12:04:31 There was still a bunch of questions to bug Chuck with, right? 12:04:44 well, that was the thing...I haven't talked to futhin since the interview 12:04:54 --- quit: tathi ("leaving") 12:05:02 and I didn't see him make much of a stink about this upcoming session 12:05:07 like he did witht he last one 12:05:31 Hmm' 12:05:44 Maybe because he doesn't want #forth to get flooded again? 12:05:55 Last time there was hell of an activity here. 12:07:35 well, it was announced on the newsgroup that there was a follow-up session 12:07:43 but nothing was made of it after that 12:07:57 actually, "announcement" is too much emphasis 12:08:09 it was kind of dropped in the posting casually 12:08:14 so, that might explain it :P 12:08:22 :) 12:08:51 Is there anything interesting going on on comp.lang.forth? 12:09:01 * rob_ert wonders if it's a good place to hang. 12:09:18 hmmm 12:09:47 I am not sure "hang" is the word I'd use 12:10:14 I think occasionally it does cover some interesting material, so it is worth it 12:10:20 Okay. 12:10:22 :) 12:10:22 I don't personally participate, though... 12:10:29 I don't have the energy to do that 12:13:29 Hehe 12:13:39 I'm too lazy for forums too, actually. 12:14:15 --- quit: oner ("bxmircxchatwtf") 12:14:37 * Speuler lets todays event for what it is and goes to cinema instead 12:14:44 baai 12:14:50 Bye ;) 12:14:58 take care =) 12:15:12 may the forth ... 12:15:19 jedi action 12:15:21 :)) 12:15:26 Star wars II? 12:15:31 That's what you're aiming for? 12:15:45 yep 12:16:25 :) 12:16:53 noticing a recent interview for Mr. Lucas regarding his new movie 12:17:06 I couldn't help but notice that the last movie seemed to have been a kick in the shins for him 12:17:46 he doesn't seem to have the "confidence" he had before the release of the last one 12:18:37 there was a lot of laughter in the theater at times that were clearly not intended to be funny. 12:18:57 ouch =) 12:18:59 :) 12:19:04 --- join: kc5tja (~kc5tja@stampede.org) joined #forth 12:19:09 Hi 12:19:31 Howdy 12:19:35 hi 12:23:46 --- nick: jcassady -> jim 12:24:18 re... just checking in; what's the time frame? 12:29:02 thefox: did ya just see it yesterday? 12:29:19 yes 12:29:47 how did it compare interest-wise to I? 12:30:48 (I;m considering waiting for III, and then see them both then) 12:31:48 I'm pretty sure there's some reviews on the interweb 12:32:19 I enjoyed it, more special effects, more spectacular scenes. I was not much impressed by the first generation digital production, it might look better on a digital projector but the resolution clearly seemed lower. Annie had grown up a bit, not much, and I hate seeing people walk away from spaceship crashes. 12:33:25 haha 12:34:45 im waiting for the matrix reloaded 12:36:08 --- quit: Speuler (Read error: 113 (No route to host)) 12:38:00 does someone here worked with x86 microcode before? 12:39:35 jeff fox! hi dood 12:39:36 Do you mean microcode or assembler? 12:39:51 im sorta afk rite now - hardly workin! 12:40:31 microcode like the code updtaed by the bios into the procesor core 12:40:53 I440r: :) 12:41:46 ive read that you can define instructions on a pentium class machine by changing the microcode 12:42:06 crowkiller you can - the linux kernel even allows you to do that :) 12:42:09 so im trying to get info 12:42:19 sounds very silly for joe blow to recode the microcode on a pentium tho :) 12:42:39 but do you have documentation on writing microcode 12:42:42 or reading it? 12:42:42 my father tels me you could even re-microcode a 6502 :) 12:42:47 not sure how you would do that tho 12:42:49 Heh 12:43:07 Optimize your C64 :D 12:43:33 really? I thought the 6502 was too simple to have a microcode 12:44:33 deltab corse it had microcode - it is one of the best microprocessor designs EVER 12:44:35 I440r: did you get anywhere with your forth-in-a-module? 12:44:50 jim not yet but its planned - the more i learn the more i like the idea 12:45:02 --- join: BadCog (spamtrap@1Cust252.tnt3.beaverton.or.da.uu.net) joined #forth 12:45:03 having isforth as a kernel loadable module wont be that difficult i dont think 12:45:25 what for ? 12:45:34 but theres alot of stuff isforth needs. im working on sockets code for it, a text windowing system 12:45:43 it needs an assembler, meta compiler and debugger etc 12:45:54 you could have a fairly strong ability to debug and create other kernel modules 12:45:55 too much for one person to do it all 12:46:18 it doesnt need a metacompiler... but you could have that 12:46:22 jim being able to single step through the kernel would be sexy 12:46:28 Hehe 12:47:05 jim i know it doesnt NEED one - but with an assembler and meta compiler alot of what is currently IN the kernel can be removed and become an option 12:47:11 maybe wouldnt need to check kernels by the stanford checker anymore 12:47:53 jim have you read the todo file in my distro ? 12:47:59 it gives my .plan heh 12:48:05 1: finish isforth 12:48:11 2: convert linus to forth.... 12:48:12 heh 12:48:44 well, if you could create a situation where the headers were separated into userspace, you could then make kernel modules very easy to make and save 12:49:17 jim heaers are already seperate... if the isforth kernel becomes a module it will have 2 modes of compilation 12:49:21 1: compile into user space 12:49:25 2: compile into kernel space 12:49:34 im thinking... 12:49:38 and in terms of CM's question of "What is the future of forth", your project could be part of the answer 12:49:58 :-) 12:49:58 thats what im hoping but like i said... too much for ONE person 12:50:23 i dont think isforth is THE future for forth but im hoping it becomes a big part of the future of linux 12:50:45 my .plan is only PARTLY in jest 12:51:03 well, I think he would enjoy hearing about the idea 12:51:03 Hehe 12:51:14 I440r will soon rule the world :} 12:51:38 or at least its internals :) 12:51:51 robert if i did the first thing i would do would be to create a national receprocity law (did i spell that rite?) 12:52:03 i.e. if you can carry in state X you can carry in all other states too. 12:52:16 yue spelt rite rong :) 12:52:17 for me the future is simply a linear interlaced array of words and dictionary entry point adresses, and the smallest interpreter possible for tokenized forth 12:52:27 jim i spel alot rong :P 12:53:28 wel, alot as almoste spelt rite; yue ned uh spase 12:53:32 crow my forth background makes me distrust anythin except direct or indirect threadings where indirect should only be used on a harvard architecture but TOKEN threading is looking nicer and nicer the more i build isforth 12:54:22 now, I have to find somewhere that teaches Yang style tai chi short form... 12:54:29 isforths unextended kernel is 15356 bytes currently which to me is HUGE for a forth 12:54:55 him do wing chun - if you can find a sifu who will teach it to you that is 12:55:18 most wont teach you the whole thing... mook jong or plub blossom steps etc 12:55:21 those are BIG secrets 12:55:34 I don't want to know secrets :) 12:56:00 the whole system can be learned in about a year - wing chun is VERY minimalist 12:56:01 I want to have something that improves my health and balance by at least one step :) 12:56:05 takes years to master of corse... 12:56:21 judo would to that :P 12:56:22 I440r: 15KB of code sounds perfectly normal for a 32-bit kernel. 12:56:51 kc5 yes its that 32 bits part thats making it big - which i STILL dont like - token threading would allow me to decreas that size 12:56:58 but would also decrease my address space 12:57:08 the one thing that stops me converting to token threading. 12:57:15 well, that would also make it slower 12:57:26 its already not fast 12:57:38 isforth memory is all r,w,x 12:57:46 I440r: Most Linux computers can deal with 15kB... even if you're ashamed, it's still useable :) 12:57:49 i hope to be able to implement one day a kernel in less than a k, with all apps in source format, so the kernel would really be only a virtualization layer, so the code could run on every 32 bit platform avaible 12:57:56 which means we are fetching data from pages we just executed. BIG cache hits for that :) 12:58:19 JJ was talking about a token threading whereby the tokens would be standardized, hence creating a form of binary compatibility 12:58:21 * kc5tja restarted FS/Forth System 1 Release 3 using a Machine Forth varient. 12:58:26 crow do it for the 8051 - i have an unfinished DIRECT threaded 8051 forth :) 12:58:38 i was writing an 8051 simulator so i could debug it too 12:58:42 also unfinished 12:59:22 dang, futhin isnt here, I wanted to change some of my questions 12:59:24 actually right now my BIGGEST problem is not having a paid job :P 12:59:42 and I can't be here (unfortunately) 12:59:43 whats the eta for chippies arrival ? 13:00:00 where is he now? at home? 13:00:03 I have 12:53PM here, so in another hour or so. 13:00:35 jim i dunno 13:01:14 --- part: I440r left #forth 13:01:14 --- join: I440r (~mark4@1Cust232.tnt2.bloomington.in.da.uu.net) joined #forth 13:01:15 --- mode: ChanServ set +o I440r 13:01:18 --- mode: I440r set +o thefox 13:01:20 --- mode: ChanServ set -o thefox 13:01:24 argh 13:01:29 he set strictop on again 13:01:30 * kc5tja is in Pacific time zone, so he'll be here in about an hour or so. 13:01:32 bad chanserv 13:01:37 bad BAD chanserv 13:01:56 I am currious, last time did Chuck say anything that offended any of the readers? I noticed a few people leaving in the middle. I also wonder if there way anything that Chuck said that people just didn't understand. 13:02:03 Oh, I didn't even see thefox join in. 13:02:12 --- mode: I440r set +o thefox 13:02:24 not 2 me 13:02:24 he's been here since 11 am PST :) 13:02:55 I think the only thing people might be perturbed about is his brevity. But, as far as I can see it, it's their loss. There's no rule on IRC that says you must be verbose. 13:02:56 hmmm... 13:03:32 ok, here's my take... 13:03:55 there was a lot of question and being verbose isnt an option most of the time in more than 20 person irc interviews 13:04:15 what if futhin doesnt show up 13:04:17 panic! 13:04:30 CrowKiller: Chuck is a very slow typer from what I've observed. Had he typed as fast as I do, he could have been as verbose as he desired. 13:05:03 if i typed slower i mite type better but it still wouldnt improve my spellin so... bleh :P 13:05:14 i can type 35 WRONG words per minute :) 13:05:25 * kc5tja can type over 90 correct words per minute. 13:05:46 build an IRC client in Forth with a built in spell checker :^) 13:05:46 I can type about 10 words per minute. 13:06:03 Since I have to erase every word 10 times to get it right. 13:06:04 badcog actually i plan on writing an irc SERVER and client 13:06:08 - his comment about portability and the implication that had on job security was at diametric odds to his comment that he was glad/ok with the fact forth got out... I was originally under the impression that he was originally opposed to forth becoming public at the outset of FIT 13:06:14 where irc doesnt use a spanning tree network 13:06:40 kc5tja: did you saw his keyboard layout for his colorforth?? no wonder he type slowly on a standard keyboard ;pppp hehe 13:06:46 (I'm pretty good when I'm warmed up and used to the kbd) 13:07:11 CrowKiller: That keyboard has the potential to be even faster than existing Dvorak layouts. The world's all-time fastest typers *ALL* use Dvorak layouts. 13:07:27 jim portability isnt all its cracked up to be... its one of isforths design goals to NOT be portable :) 13:07:46 CrowKiller: I'm actually rather intrigued by his keyboard layout. With only 27 keys on it, to get any kind of speed, you must end up chording, and chorded keyboards are the fastest there are. 13:08:02 tho - porting to freebsd isnt that big a deal - it would just mean a different syscall include and dfferent syscall numbering 13:08:40 Portability is critical in today's world of extremely diverse processor environments that all must work together. 13:08:41 i mite not be able to be here for this :( 13:08:59 my father and i are setting up a gun store - he got is lisence to manufacture arms :) 13:09:20 and we have a commercially zoned second building right next to our house 13:09:26 2 story cement block... 13:09:29 perfect for a gun store 13:09:55 Chuck hasn't been using chorded key input for many years. But when using the function keys in OKAD he moves a lightning speed. It is several times faster than any mouse driven interface. 13:10:30 - in an env such as this with so much unix-based projects and with unix's maxim of "A) there's a good language for anything you want to do; B) but if not, create a new one" as background, CM's comment that forth is the best for anything seemed "uninformed", shall we say... having said that, I do understand that he's saying it' 13:10:32 mouse interfaces are very clumsy - specialy touchpads 13:10:39 s the best -for him- 13:11:08 thefox: Actually, mouse input and keyboard input are complimentary towards each other. Nothing beats a mouse for when you have to say, "That right there," -- it's far faster to just point at it and click a button. But for most things you do with the computer, a keyboard ultimately is faster. 13:11:36 That's why I dislike MacOS and a large amount of X11 applications -- they are so heavily dependent on the mouse that it's frustrating. 13:11:37 mouses are good for drag and drop coding! 13:11:42 :) 13:11:44 I thought Chuck's comment about Forth being the best language would not be understood by most people. It is one of those things that needs a little explanation. His short version is I wouldn't do it that way. 13:11:49 Drag-n-drop and range selection are ideal applications for a mouse, too. 13:13:01 yes, having been involved with forth for many years, I do understand that; he has a habit of saying "it takes practice to use forth; I'm the one with the most practice" 13:13:48 forths learning curve is the opposite to c's. in c you lear alot very quickly - i.e. any idiot can learn c. forth has a steep INITIAL learning curve that seperates the wheat fromthe chaffe 13:13:50 Forth, easy to learn, difficult to master 13:13:56 thefox: At work, I use a very minimal stack environment to implement the configuration file system for a semiconductor verification tool I've written. It saved me days, if not weeks, in writing a stupid parser for it. But, everyone who uses it has a strong distaste for it. 13:14:07 badcog no. forth difficult to lear at first - very very easy to master 13:14:16 once you get over that initial leraning curve 13:14:18 Chuck will solve problems using Forth, just Forth. The environments that he constructs will have a minimal of introduced problems on top of the real problem that he wants to solve. Other people will import thousands of other problems by using other peoplle's code and extra layers of abstraction. 13:14:20 the opposite is tryue for c 13:14:57 my favorite question. how many C coders do you know with a 100% written from scratch C compiler ? 13:15:12 I440r: I agree with badcog. Basic Forth is rather trivial to teach. In my documentation for my verification tool, I detail the essential qualities of Forth in about three pages of text. People here are writing their own configuration scripts now. They don't realize it, but they're programming in a highly specialized dialect of Forth. 13:15:47 that's why I tell anyone who might have interest in forth, "You need approx one month of solid 8-hour-per-day coding in forth to get good at everything... when you come out of your hole, you'll pretty much understand everything about it" 13:15:58 kc5 the individual primatives are easy to learn. its the RPN and how to "think" in forth that are the initial difficulties 13:16:12 I440r: I seem to have no difficulty in teaching it. 13:16:20 mostly, I feel the real uses of forth are embedded apps, and training people to program in asm 13:16:32 i.e. nameless variables that are on the stack - learning to use THOSE instead of named variables etc 13:16:40 In fact, my recent experience pretty much follows suit with both Elizabeth Rather's experience and Jeff Fox's experience in teaching Forth to new-comers. 13:16:45 there are many ways in Forth to build your castle, the difficulty lies in how to build the best castle 13:16:46 kc5 then i put to you that maybe you are a good teacher 13:17:00 I440r: There's your first problem. DO NOT THINK IN TERMS OF VARIABLES. That *WILL* confuse people. 13:17:24 kc5 i dont - but a beginner will 13:17:33 BadCog: the good thing is that castles made of sand fall in the sea, eventually... you blow away your castle many times and make new ones 13:17:36 In fact, the best way to teach Forth is to never acknowledge the existance of any other language, until it's absolutely required. 13:17:43 i think in terms of parameters to words 13:18:01 I440r: Not in my experience. If they do, "You simply don't need them. Forth takes care of all that stuff for you." They're usually happy with that answer. 13:18:03 (my last had a Jimi Hendrix lyric) 13:18:43 Another way to teach Forth is by way of Data Flow concepts -- graph the word out in a tree on a white-board, and show how data flows from the inputs to the outputs. 13:18:59 --- join: tathi (~josh@ip68-9-58-81.ri.ri.cox.net) joined #forth 13:19:03 If they can visualize plumbing, which many and most do, then they can usually grasp the use of the stack. 13:19:10 my father wrote a coltrol aapplication for some networked control cards (CAN) where there wasnt a single variable in his code 13:19:17 The two points that Chuck made last time that I thought were most important were when he said that he didn't do structures 'that' way, and that there were really not lower and upper layers to his code. That was the thing that I found most different about his style of doing Forth. 13:19:17 not even a send/recieve buffer 13:19:22 everything was done on the stack 13:19:54 Cool :) 13:20:03 thefox - "i dont do structures that way " which way? 13:20:16 thefox: I can't program without structures. The complexity of the problem(s) I'm trying to solve mandates that I use structures. 13:20:45 perhaps everyone should have a copy of the log of CM's conversation from last time at their disposal :) 13:20:57 its good to seek out alternatives to structures and to variables, it will lead to more thoughfullness in programming 13:21:09 jim its on the net in various locations including jeff fox's site :) 13:21:24 we should consider topicing that 13:21:33 --- join: mslicker (~mark@64.27.199.31) joined #forth 13:21:44 BadCog: For relatively trivial software, structures aren't needed. For anything that involves manipulation of real-world data, data which must be structured, you just plain have to use structures. 13:21:51 From what I have seen it is one of the biggest problems that people have with Forth. Chuck's approach to programming is to first think through the data structures long before coding anything. That is about the most important thing, data structures and scaling. 13:21:58 You'll never find an FEA application written in pure ML, for example. 13:22:13 thefox sounds like JSP :) 13:22:25 JSP? 13:22:32 java server pages? 13:22:32 jacksons structured programming 13:22:37 ahh 13:22:41 its an old method of coding 13:22:59 if you follow the rules exactly given the same problem as me and i also follow the rules properly 13:23:01 Forth provided a dictionary and database mechanisms in the kernel. Forth was short for a Fourth Generation Language where you define your own database structures and words to access then at the 'lowest level' 13:23:05 we will end up with IDENTICAL code 13:23:08 not similar... identical 13:23:09 based on the Jackson Five and Under crowd? :) 13:23:28 jo named after its inventor who i belive made alot of money with it heh 13:23:28 Once you have done this you can write code where structure access is free most of the time. 13:24:01 thefox: Absolutely. If you have only a single instance of a given structure, then the code is correspondingly trivial. But if you have multiple instances, then you need field accessors. 13:24:19 but wouldnt a compiler for some "other" language do that anyway? 13:24:23 For example, let's say I intend on writing a small DSP application using Forth. 13:24:38 Converting a bitstream into a BPSK modulated audio sample. 13:25:03 The opposite idea is to mimic C style structures in Forth and use the sort of horrible late binding code in the Forth Scientific Library to make data structures that are many orders of magnitude less efficient than they should be. Chuck doesn't do C Style data structures in his programs. 13:25:03 This application is very simple and trivial. 13:25:04 futhin isnt here - he isnt on icq - and i prlly cant be here either - got work to do 13:25:33 forth is to me much like the missing link between lambda calculus (thought structure factored to the metal) and more "structured" programming language, spoken languages and thought patterns 13:25:34 thefox: Can you give an example of what you mean by "late binding"? 13:25:38 thefox i think thats what i just added to isforth - its about 15 lines of code 13:25:44 not seen the stuff from fsl 13:25:52 * kc5tja can't live without C style data structures. 13:25:58 But I don't do late-binding either. 13:26:10 (unless it's called for by the problem being solved) 13:26:45 how do you do C style structures in forth? 13:26:56 mslicker: I'll post code. It's four lines of Forth code. 13:27:10 mslicker: : STRUCTURE ( -- offset ) 0 ; 13:27:20 by late binding structures I mean that addresses are always calculated at runtime, even when they 13:27:35 'afternoon folks... (btw, before I go, I want to point out that I let John Cassady know about this event and irc channel... part of his history is he did the 8080 forth assembly listing for the FigFORTH forth implementation team) 13:27:36 mslicker: : FIELD ( offset size -- offset' ) CREATE OVER , + DOES> @ + ; 13:27:39 could be resolved at compile time and compiled as literals immediately. 13:27:46 thefox yea - thats sometimes required because you dont know the base address till run time 13:27:48 mslicker: : END-STRUCT ( offset -- ) CONSTANT ; 13:27:55 OK, three lines of code, but that's pretty much it. 13:28:01 kc5tja: I prefer to use struct-offset.... 13:28:11 So I don't have to calculate offsets manually. 13:28:35 thefox: That's true if and only if you know a priori the address of the structure you're accessing. in the real world, that happens less than 2% of the time. 13:28:43 isforth compiles structure elements as consts - in isforth a const is compiled as a literal and they will eventually be discarded at turnkey 13:28:47 kc5tja: I see.. 13:28:58 1440r, but you are not talking about the way the way that Chuck does it. You are talking about where there are many more problems introduced by having to use a foreign file system rather than your own internal Forth database the way Chuck does. 13:29:02 mslicker: For example: 13:29:06 mslicker: STRUCTURE 13:29:15 mslicker: 1 CELLS FIELD String.Address 13:29:23 mslicker: 1 CELLS FIELD String.Length 13:29:30 mslicker: END-STRUCT SIZEOF-String 13:29:38 mslicker: CREATE foo SIZEOF-String ALLOT 13:30:11 thefox - what about data thats created dynamically at run time - like say a chess games search tree. 13:30:15 mslicker: : $@ ( str -- caddr u ) DUP String.Address @ SWAP String.Length @ ; 13:30:29 kc5tja: I think I agree with Chuck ;) 13:30:34 each node of the tree has an offset within the memory buffer and each element of the node has an offset from there 13:30:45 mslicker: : $! ( caddr u str -- ) 2DUP String.Length ! NIP String.Address ! ; 13:30:50 We had a three year war over structures at iTV. FIG President and team of ANS Forth experts copying 'C' code vs FIG VP, team of MachineForth programmers, and Chuck. I have heard it all over and over. 13:31:15 to me its a useless debate 13:31:24 thefox i agree - if it can be computed at compile time it should be but there are times when it cant 13:31:29 mslicker: Well, I'm sorry you feel that way. Structures exist for a reason. They solve a very real and useful problem. 13:31:35 Let people choose :) 13:31:36 It was a good way to bloat up sections of code by 100 and slow them down by 10,000 times. 13:31:36 or is there a way to transform any problem so that it can ? 13:31:53 thefox: I vehemently disagree with that, 100%. 13:32:03 Structure offset calculation takes, at most, 1 clock cycle. 13:32:25 If you can resolve the exact storage address at compile-time, then do so. But you'll find nearly always, that you just can't do that. 13:32:58 Chuck has the ability to write a completely new OS for any programming problem he chooses, and thus, he has ultimate control over everything. He knows everything a priori. 13:33:03 I do not have that luxury. 13:33:03 if the data is created at compile time you can. if it is created at run tim you cant 13:33:07 Nearly always you just can't do that if you don't use Chuck's style, that is true. 13:33:46 I don't consider it a luxury, I consider it Forth. 13:33:55 thefox: No, it's a luxury. 13:34:06 My use of Forth at work is every bit as valid as Chuck's. 13:34:28 But I can't just say, "Oh, I'll rewrite this MIPS processor's BIOS just so I can pre-compute all possible combinations of all possible structure field offsets." 13:34:29 No way. 13:34:52 That'd put me *so* far behind schedule, alone, let alone its other impracticalities, that I'd get fired right away. 13:35:04 thefox - i messaged you in private a while back... 13:35:45 everyone codes the way THEY think. which is one of the reasons i think forth should be totally without a standard 13:36:22 I think a good standard would be very helpful for Forth. ;-) 13:36:38 I440r: The thing is, use of C-style structures is not a bad thing. Not even close. And it doesn't slow down runtimes 100,000x, nor does it complicate the interface any more than anything else would. 13:36:54 kc5 which is exactly why i added them to isforth 13:36:55 Like any tool, it must be applied with skill. 13:37:09 Do not blame the tool for the user's ineptitude. 13:37:24 what would standardize? it seems forth in it's natural state is constantly changing. 13:37:24 Because that dirties those of us who DO use it correctly and well. 13:37:29 my biggest issue with traditional forth is its lack of seperation for compile time words and run time words. when you turnkey you have a hell of alot of FLUFF left in the application 13:38:08 I440r: Are you using "unified" vocabularies, or "split" cmForth-style vocabularies? 13:38:18 My opion is that C style ... is not a bad thing, for C. 13:38:24 * kc5tja is going with cmForth style vocabularies. They're just plain easier for me to handle. 13:38:26 --- join: geqo (~geqo@as7-1-5.kp.g.bonet.se) joined #forth 13:38:28 isforth is eventually going to put each vocabulary in its own section 13:38:30 Hej. 13:38:45 and the compile time vocabularies are going to be transient - stripped at turnkey 13:38:46 thefox: I am surprised you and Mr. Moore didn't propose your own "standard", perhaps as a response to ANS 13:38:54 this is a long way off tho 13:38:55 =) 13:38:55 hej rob_ert 13:39:03 i cant do that with nasm building the kernel 13:39:13 any more than 2 sections and nasm goes fubar 13:39:43 thefox: I am inclined to disagree. C style structures, which really isn't "C-Style at all"; it's just plain common sense and is how the CPU itself computes effective addresses, is applicable to any language you care to apply it to. 13:39:47 it was bad enough just getting nasm to link on various vocabularies heh 13:40:06 I was a strong proponent of ANS Forth for years and used to argue with Chuck with all the arguements that people always make. My experience on the FIG board, and at iTV changed my opinions. 13:40:15 thefox: If you have a large quantity of instances of a data structure, and you cannot compute, at compile-time, which ones will be accessed, you have no choice but to do the calculations at run-time. 13:40:51 --- part: geqo left #forth 13:40:52 thefox i code on instinct mostly, if it FEELS right it IS right - my gut reaction to ans forht was identical to my reaction to c 13:40:52 As FIG VP I saw a lot of stuff behind the screen that made me see ANS very differently. 13:41:00 and i HATE c more and more every day 13:41:37 i tried to learn ANSI forth, and I wasnt able to do it lol 13:41:43 thefox ans is an attempt at politically correcting forth 13:41:45 imho 13:41:52 i simply cant and dont want at all to code in this monstrosity 13:42:01 kc5tja: you ignored what I said about how much of the time it can be free, but only if you do it the thoughtful way. 13:42:15 I440r: _why_ is C that bad? 13:42:24 ans forth is a lot more complicated than colorForth 13:42:29 I440r: Won't you admit that it can be used for a lot of things? 13:42:36 i4: I think it was more of answer to exploit Forth's potential commercial value 13:42:42 thefox: No, I didn't. It's just that the type of programming I do, and the overwhelming majority of others I work with and have worked with in the past, you have no a priori knowledge of when or where a structure is going to be used. 13:42:45 robert c iteslf isnt. its all those idiots who think they can code it - and c encourages bad coders and very bad coding style 13:43:14 there isnt a c coder i know who knows how to factor code properly OR WHY! - and almost NONE of them know what commenting is 13:43:29 Hehe 13:43:31 I440r: If you write C code clearly enough, you don't need commenting. 13:43:36 * rob_ert feels guilty. 13:43:37 I440r: I fall into that category. 13:43:42 cut and paste coding 13:43:47 So do I. 13:43:55 Hehe 13:43:55 kc5 i could not possibly disagree with you more on that one. ALL code needs to be commented 13:43:59 * kc5tja follows the principles of extreme programming vehemently. 13:44:16 the comments are NOT for you. they are for the poor shmuck that needs to understand that which is naturall to you 13:44:17 I440r: Why just repeat the code in comments, when the code is patently clear in its intent? 13:44:17 I don't factor alot in C programs... not as much as I do in assembly language. 13:44:37 I440r: You're demonstrating your ignorance of the concept. 13:44:40 kc5 i think we had this arguement before and failed to resolve it :) 13:45:02 kc5 there is NO SUCH THING as 100% self commenting, self doccumenting code 13:45:03 I admire I440r's habit of commenting, but I'm too lazy to do it myself :) 13:45:07 In my software, there isn't a single C function that exceeds 16 lines. 13:45:18 any coder who thinks there is is a fool 13:45:18 ok i have an idea for you people, why dont we try to solve problems instead of discussing about some imaginary ones? precahing by example is far more convincing and clearly show what work what doesnt 13:45:23 I440r: And I posit you are wholly incorrect. 13:45:35 kc5 lets agree to disagree again :) 13:45:39 i4: I think you can have your cake and eat it too...it just requires a new approach 13:45:42 Instead of arguing with me, pick up a good book on Extreme Programming, or visit some of the extreme programming websites, and do your own research. 13:45:46 CrowKiller: shut up. this is the internet. we argue. 13:45:47 Hehe... I actually got an error that my C function was too big to compile :)) 13:45:49 and us here apparently do not have that approach 13:46:10 Etaoin: :) 13:46:12 well go argue, but I know this is useless and I wont participate lol 13:46:17 kc5 much of what people now call extreme programming is the way ive been doing it from day 1 13:46:18 If it's one thing I abhorently hate more than anything else, it's blind ignorance. 13:46:29 which is not meant to be criticism of anyone, simply a statement 13:46:33 I440r: And if you still need comments, then you're not doing it good enough. Keep trying. 13:46:42 I don't think compromises are always inherent 13:46:43 kc5 :) 13:47:14 I think it just generally requirs someone who cares enough about something and is not tied too much to conventional bias 13:47:34 For example, Chuck's IDE software written in Color Forth was pretty clear to me, once I got used to his brevity in naming his words. (Knowing that ColorForth has horrible limitations on word names helped explain why he's so brief too.) 13:47:53 I figured it out without ever having to read his shadow screen. 13:48:01 if you have thunked about your code enough to explain it to someone else you have thunked about it enough to write it 13:48:04 period. 13:48:25 taking the time to explain yuor code HELPS you write it 13:48:28 kind of annoying how it resets the machine if I press backspace at the wrong point though 13:48:30 I440r: And just as Chuck can eliminate run-time structure field resolution, so I too can write software that is 100% self-documenting. 13:48:47 You're willing to believe everything Chuck does, but somehow I'm inferior for having a different skill? What hypocracy. 13:48:48 kc5tja: I haven't felt limited by colorForth naming. 13:49:28 I440r: And if you look at Chuck's shadow screens, he does not explain what his code does. 13:49:28 hangn on testiing syntax for mode change command 13:49:36 --- mode: I440r set +m 13:49:39 --- mode: I440r set -m 13:49:58 mslicker: If you're used to 2-5 character long names, fine. I regularly use names exceeding 16 to 32 characters. 13:50:11 (even in my C code) 13:50:12 kc5tja: yikes! 13:50:24 kc5 neither does isforth. something i very much hope to change 13:50:39 one of my goals is to have every single word in the isforth dictionary explained IN FULL 13:50:45 with REAL examples of their use 13:50:56 --- join: BadCog2 (spamtrap@1Cust161.tnt3.beaverton.or.da.uu.net) joined #forth 13:50:56 mslicker: :) 13:51:25 * kc5tja also has complete unit test suites for all his written C code, and I'm developing similar techniques for use in his Forth code. 13:51:34 Applying XP techniques to Forth is a natural. 13:51:54 Especially since Forth encourages the short code sequences that XP promotes. 13:52:19 --- quit: BadCog (Read error: 113 (No route to host)) 13:52:31 --- mode: thefox set +m 13:52:43 --- mode: thefox set -m 13:52:45 Heh 13:52:53 Having fun, thefox? 13:52:53 :) 13:52:54 just testing 13:52:55 kc5tja: names can actually be longer, but for searching purposes the prefix must be unique. 13:53:11 im teaching thefox all about POWER and CONTROL in irc channels 13:53:19 :) 13:53:19 lol 13:53:19 we will cover kick and ban on the next lesson :) 13:53:26 "How to be a dictator." 13:53:28 chippie is here 13:53:33 --- join: chipChuck (chipChuck@user-38lc091.dialup.mindspring.com) joined #forth 13:53:35 --- mode: I440r set +o chipChuck 13:53:36 Hi there. 13:53:39 hey hey 13:53:43 :) 13:53:44 welcome Mr. Moore 13:53:45 =) 13:53:47 mslicker: Yes, I know. However, I like it when all characters are accounted for in word resolution. For example, in my latest Forth project, I have two words: ForEachOperand: and ForEachOpcode: -- will these be considered unique in ColorForth? Doubtful. 13:53:48 * mur ! :) 13:53:51 Hi 13:54:03 ok futhin is a no-show 13:54:06 and i might have to leave :( 13:54:38 mind if I message you, i4? 13:54:40 Great... :( And futhin is the one with the list of questions too. I suppose this is as good a time as any for Chuck to ask us questions now. :) 13:55:00 ColorForth only checks 1st word (6 chars?). Could check more. 13:55:03 --- join: gforth0 (guest@adsl52042.vnet.hu) joined #forth 13:55:26 chipChuck: Oh, you should have seen the arguments flying here before you arrived... ;D 13:55:27 How many of you write Forth code? 13:55:29 i believe futhin had a list of unasked question 13:55:37 guilty! 13:55:40 * kc5tja does 13:55:44 I do 13:55:44 * rob_ert is trying ;-) 13:55:46 me 13:55:52 I don't 13:55:57 i think thefox does too :) 13:56:07 jim does but i believe he is afk 13:56:19 I am trying to write my own "punctuated Forth" using Machine Forth concepts under a DOS environment though; I've never written a target compiled Forth environment yet. This will be my first attempt. 13:56:33 I don't believe I am good enough at Forth to say that "Forth" is what I write =) 13:56:42 goshawk`: Hehe ;) 13:56:43 im writing a linux x86 only direct threaded forth... 13:56:47 not minimalist tho 13:56:47 to call it Forth is possibly insulting 13:57:04 The only reason I'm using Machine Forth-ish techniques is because I'm extremely unhappy with the (lack of) quality of the most recent NASM release. 13:57:05 goshawk`: Forth is a wide concept, hehe 13:57:15 but minimalist is one of its design goals - make everything an option 13:57:16 --- join: TheBlueWizard (TheBlueWiz@ip-216-25-202-3.vienna.va.fcc.net) joined #forth 13:57:16 --- mode: ChanServ set +o TheBlueWizard 13:57:21 Seems a majority of those talking 13:57:40 TheBlueWizard: is also a forth coder... 13:57:41 Yes. 13:57:58 to date I only wrote a 3 instruction "toy" forth, but its working 13:57:59 Any money-making applications? 13:58:17 no 13:58:20 Nope. 13:58:21 not yet but im writing sockets code for isforht that i hope can make me some 13:58:25 unfoutunately no, I keep my eyes open. 13:58:42 chipChuck: Not yet. I have written a nano-Forth for use as a configuration file system for a chip verification tool I've written. Helps keep me employed. Does that count? :) 13:58:44 I'm not even sure I'm old enough to start a company. 13:58:45 never had made any money making apps on my own 13:58:46 isforth even (mental note - must learn to type) 13:59:28 I'd love to sell music using the Forth software (and eventually hardware) that I use to perform with 13:59:59 --- join: onetom_ (tom@adsl52042.vnet.hu) joined #forth 14:00:02 goshark': have any Mp3s we can listen to? 14:00:04 Hi onetom_ 14:00:06 I'm happy just to use Forth. But it'd be nice to be profitable. 14:00:10 bad: I am afraid not 14:00:28 * kc5tja does have plans for using Forth in the amateur radio community in the construction of high throughput digital equipment for use on 2m, with significantly reduced costs (current 19.2kbps equipment for 2m band is on the order of $800!!!) 14:00:36 --- join: futhin (futhin@dial-208.ocis.net) joined #forth 14:00:39 * onetom_ is just arrived back from wine festival ~:) 14:00:51 Hi futhin! 14:00:54 * I440r slaps futhin arround with the 83 standard 14:00:55 --- mode: ChanServ set +o futhin 14:00:56 hiya futhin 14:00:57 yer late 14:00:58 Maestro! 14:01:05 heh :) 14:01:24 * onetom_ nods toward futhin 14:01:25 how long have we been here? :) 14:01:26 we seem to be doing ok without +m here 14:01:39 he's been here for a few minutes 14:01:41 yeah, it seems that way 14:01:42 Currently, Chuck is asking us questions. 14:01:47 ah 14:01:58 what's the question? 14:02:01 Any interest in wireless? 14:02:07 How many of us have written money-making software with Forth? 14:02:12 I rather like that forum, as long as it interests him 14:02:25 i have - working for joy mining machenery. my father has... 14:02:26 chipChuck: Define "wireless." That necessarily falls into the amateur radio category if it operates on ham radio frequencies. 14:02:38 UWB 14:02:40 Only wireless device I have is a radio... 14:02:52 * kc5tja sorely dislikes the concept of UWB. 14:03:07 The idea of using spark-gap techniques today is most unpalettable to me. 14:03:09 rob_ert: i also have tin-whistle. its also wireless ;) 14:03:13 --- quit: onetom (Connection timed out) 14:03:13 --- nick: onetom_ -> onetom 14:03:45 Especially since it'll hit the HF spectrum the hardest, which is where I spend most of my radio-listening time. :( 14:03:49 siemens is going to be using celphones in their power meters... 14:03:51 --- quit: gforth (Connection timed out) 14:03:51 --- nick: gforth0 -> gforth 14:04:08 Low power doesn't interfer 14:04:18 onetom :) 14:04:34 chipChuck: It does if enough people use it. The study of spread spectrum technology has proven that. 14:04:34 Seems like a good application for 25x 14:04:46 x25 ? 14:04:57 25x chips 14:04:59 My chip with 25 computers 14:05:00 oh 14:05:06 By "enough people," I mean with sufficient density to cause a relatively large field strength in the local area. 14:05:07 rite 14:05:29 chipChuck: I find the 25x to be a natural for simulating neural networks... 14:05:38 is 25x actually working yet? 14:05:50 No. Still vaporware 14:05:56 Hmm... what do you think those 25 chips will be used for, chipChuck? 14:06:05 What do neural networks simulate? 14:06:10 Any application the average programmer will come close to? 14:06:32 I'd like an all-band radio 14:06:36 chipChuck: Neurons (nerve cells, especially sensory and pattern recognition) 14:06:40 I would use a 25x for genetic programming research 14:06:53 I have an idea, make goggles which have the 25x built in, which are a portable virtual reality video game 14:07:00 Hehe 14:07:16 would be a waste 14:07:19 har har, sudden silence :) 14:07:20 BadCog2: that sounds cool! 14:07:21 Yeah. Life can be virtual reality 14:07:29 unless you could sell a gazilliuon of those and make chuck rich :) 14:08:00 The smallest production run is 20,000 14:08:23 And the cost per chip in tha tproduction run is probably $10K, so it's well outside of "normal" funding sources. 14:08:24 make it wireless and make it double as a cell phone, so when the phone rings, the game automatically pauses 14:08:28 chipchuck is 25x designed but just not produced ? 14:08:28 What kind of companies have bough forth chips earlier? 14:08:46 NASA 14:08:51 Mr. Moore, how much would a production run of 20,000 cost? 14:08:52 Patriot Scientific 14:08:58 Cute...any larger amount? 14:09:05 $100,000. $5 per chip 14:09:16 chip: thanks 14:09:17 Oooo...I was dead on the money... 14:09:40 chipchuck if you had 100k would 25x be a reality right now ? 14:09:58 chipchuck: i have an interest in osdev, and that is why i came across forth and liked it for what it was.. it seems to be an ideal language for me & my philosophy. I am interested in a desktop os that both programmres and average users would be comfortable with 14:10:07 I440r: Thinking about donating? ;-) 14:10:14 if i had 100k - yes 14:10:16 I'm guessing $200K for 1 prototype runs 14:10:18 I440r: :) 14:10:26 Sorry, 2 prototype runs 14:10:43 chipchuck: what do you think about a desktop os with forth as the language? any problems? 14:10:58 Well, it's almost a certainty that you can use the 25x to make your circuit simulation software 25x faster. :) 14:11:10 I'd be great, except for running existing applications 14:11:17 Hehe 14:11:26 yeah, i'm not concerned about existing applications 14:11:29 Well, some software needs to be rewritten from scratch anyway ;) 14:11:30 * TheBlueWizard has some money, but is saving toward buying a house...but if there is a "guaranteed" investment, then.... :) 14:11:31 that stuff can be recoded better 14:11:34 I'd like to genetic program the c18 core 14:11:37 MS-Word isn't really minimalistic. 14:11:43 but for the N64, they used Realtime recompilation technique to make the emulation possible 14:11:46 chipchuck you could write a virtual (name the processor here) for 25x :) 14:11:51 i think in forth it would be doable 14:11:51 Actually, I have thought about writing a "macro-scopic" circuit simulation in Forth -- one that emulates circuits at the PCB level, with a bent on analog circuitry for RF use. 14:12:24 tbw nothing is guaranteed- dont gamble unless you can afford to lose :) 14:12:32 Forth's good for simulations 14:12:40 i4: but wouldn't that contradict having 25x in the first place? 14:12:56 chipchuck: do you think that i can keep the forth language for the os "low-level" and keep it easy to program for, for the end programmers and users.. it seems to me that i "need" to start incorporating high-level features.. 14:12:57 --- mode: I440r set +c 14:12:59 I've got simulator for 25x on PC 14:13:16 I440r: No color spam? :( 14:13:25 I440r: I know...I can't afford to lose, so.... 14:13:29 --- join: bpt (bpt@user-2ivf1pm.dialup.mindspring.com) joined #forth 14:13:39 What I find most intriguing about your simulation strategy is that you actually model the actual electrons themselves, and not something more abstract. that is, you actually account for individual charges throughout the circuit. I find that to be quite fascinating. I never thought of doing that; I was thinking of using something more like nodal analysis (the traditional "school-book" way of doing circuit analysis). 14:13:49 chipchuck: is that released for colorForth, or something else? 14:13:51 a forth micro could have forth as the os, but then have apps written in C, or Basic or whatever the programmers like 14:13:55 if i win the lottery ill donate - hmmm i have to buy tickets to win... right ? :) 14:13:58 programmers are customers too 14:14:02 I440r: ;) 14:14:03 futhin: you could use Forth as sort of a shell 14:14:13 C calling Forth OS, how would that work? 14:14:18 chipchuck i want isforth to become a linux shell option :) 14:14:33 Hehe... that's a nice goal, I440r. 14:14:40 BadCog2: Any number of mechanisms will work for that, including assembly language thunk layers (the same basic process of how a Forth program calls a C program, only in reverse) 14:14:47 kc5tja: the higher level simulations have artifacts, of unknown consequence 14:14:57 chipChuck: You said forth is good for simulating, what are the weaknesses of forth, as you see it? 14:15:10 I could release 25x simulator, but no demand so far 14:15:37 chipChuck: dont care about the lack of demand 14:15:42 forht traditionally doesnt seperate out creating words (compile time items) from run time 14:15:43 chipChuck: So, realistically speaking, OKAD is a finite element analysis type simulator then, where the smallest element is the "cell". Quite fascinating. I wonder how I can apply that to a more 3-D world, where coils and variable capacitors and the like will be... 14:15:56 Greg Bailey has a Forth os for PC 14:16:13 chip: I'd be interested, if that counts for anything 14:16:15 yes there are some forth oses out there.. 14:16:30 rob_ert: If Forth had a weakness, I'd correct it 14:16:30 futhin: You don't happen to know where I can find it? 14:16:33 chipChuck: any piece of ur works can inspire other forth programmers 14:16:50 brb 14:16:53 would that OS be a decendent of Polyforth? 14:17:01 chipchuck its possible that a 25x simulator might generate interest for the real thing - but dont hold your breath :( 14:17:29 back 14:17:29 Remote descendent of Polyforth. Search for Athena 14:17:54 does a turing award bring any money for the person awarded ? 14:17:58 chipchuck: what are your opinions about "high-level" language features and forth os? it kind of seems to me that it is necessary for me to incorporate a whole bunch of "high-level" features.. is it a bad thing? can i avoid it in a general purpose os ? 14:18:07 chipChuck: most of of wouldnt even dare 2 contact w u, coz u r so mythical 14:18:08 The only flaw in Forth, is the everyone doesn't love it 14:18:10 theres a thread on CLF asking if you have ever been nominated for one 14:18:35 chipChuck: so who could we dare 2 ask u things like 25x simulator :) 14:18:42 forth has a high learning curve.. even though it is very simple in many ways, it is a complex paradigm 14:18:43 chipChuck: The only flaw with those who don't love it, is they're afraid to learn something new. 14:18:44 chipchuck that was one of the questions i was going to ask - what can be done to make forth more universally acceptable - thats one of isforths goals 14:19:21 Forth needs good, universal applications 14:19:34 such as? 14:19:37 and "tractor apps" to boot 14:19:49 Don't know about Turing 14:19:59 make throw away Forth paperback computers to make Forth universally available. 14:20:07 Web browser 14:20:12 chipChuck: can this *need* b satisfied without a forth standard? 14:20:13 planned in isforth 14:20:14 chip: I am not sure if you think it needs it, but do you plan on documenting colorForth more extensively on your website? 14:20:18 I440r: Say... are you trying to get Chuck interested about IsForth, or? ;-) 14:20:25 If Forth was obviously better, peoply would opt in 14:20:29 web browser, ftp client, irc client, irc server - all of the above 14:20:41 Recyleable Portables for like $30 each 14:20:55 Yes, documentation on my list 14:20:56 thefox: I see your F21 being critical in that aspect. They are small, cheap, and if you want good graphics, you can gang the chips up to provide 4096 colors, as well getting triple the compute power for free. 14:21:28 Hey Fox! 14:21:32 chipchuck: i am sure some people here, myself included, would be interested in helping you document forth, your philosophy, colorforth site, etc.. 14:22:05 yeah the huffman tables on the site are erronous, tehres some errors in it, 3 inversions 14:22:08 futhin: yes 14:22:11 Needs doing. Thing is, in 30 years many people have written much copy. And here we are 14:22:11 Perhaps we can get Jef Raskin to work on such a project too -- his Canon Cat project proved that Forth can be used in *precisely* the manner that thefox mentioned. 14:22:22 Jef Raskin also really likes Forth, to boot. :) 14:22:37 *YOUR philosophy* <-- this one is the most importants things 2 document, ithink.. 14:22:39 the copy isn't very accessible.. i am disappointed with the resources on the internet 14:23:14 Doesn't need more words. Needs better words 14:23:20 yes, agreed 14:23:46 when i say "accessible" i also mean, in my mind, that the words out there are not very accessible to people's minds.. 14:23:56 My philosophy is ruthless simplicity 14:24:14 chipchuck: what are your opinions about "high-level" language features and forth os? it kind of seems to me that it is necessary for me to incorporate a whole bunch of "high-level" features.. is it a bad thing? can i avoid it in a general purpose os ? 14:24:15 sure, but u should teach it by examples 14:24:43 thats what he is doing: colorforth is an example 14:24:44 Give me an example of high-level feature 14:24:45 youthink c is? with all those gazillion libraries with a gazillion #defines and inter dependancies between this .h and that .h - c is a monster to learn 14:24:51 i sudied it and its really simple 14:24:54 most c is closed source 14:25:01 i say the linux kernel is closed source 14:25:04 TOO complex 14:25:20 chipChuck: coz, u use ur philosophy all the time. u shouldnt try 2 explain what is it, just show some results of it 14:25:22 I440r: C is easy to learn; hard to use. Contrary to popular belief, C and the library are distinctly separate. 14:25:29 Yeah. Open source is a con if you can't read the whole thing 14:25:39 My earlier Dolphin attempts were written in C, and not a lick of ANSI Standard libraries were to be found. 14:25:41 kc5 i refer you back to our earlier arguement where you argues just the opposite :) 14:26:08 I440r: I did not argue the opposite: I argued that Forth was easier to teach than C. That's not the same argument. 14:26:26 chipchuck c tries to cover far too many bases - i want alot of what c offers but without the bloat in complexity 14:26:33 chipchuck: garbage collection, features for the lazy coder.. etc.. 14:27:01 like a text user interface, sockets, xlib interface but without a gazillion .h files that #include each other ad infinitum 14:27:06 Companies make money by making things complex 14:27:07 I440r: 14:27:17 "To drop a man in the middle of the Atlantic Ocean and tell him 14:27:17 he is at liberty to walk ashore, would not be more bitter irony 14:27:17 than to place a man where all the land is appropriated as the 14:27:17 property of other people and to tell him that he is a free man, 14:27:17 at liberty to work for himself and to enjoy his own earnings." 14:27:18 --Henry George, Social Problems. 14:27:34 chipChuck: Unfortunately, there is some requirement for that. If they didn't, their competitors would probably drive them out of business. 14:27:36 chipchuck: i agree with onetom.. i think there should be more examples out there.. hopefully usable ones 14:27:40 I440r: (re: Linux kernel too complex) 14:28:03 chipchuck: it would be lovely to see some of your code, like the 1-2k gui 14:28:03 OKAD is my best example, but it's still proprietary 14:28:11 yeah 14:28:24 and the gui is probably owned by itv.. 14:28:40 No. It's all owned by me 14:28:54 i was gona say - rewrite the gui :) 14:28:55 chip: I vaguely recall something about you releasing OKAD to the world 14:28:57 the colorforth source is avaiable on the net, i find it a concrtete example 14:29:04 chip: or am I being fanciful? 14:29:43 Not yet, but it's a possibility 14:29:47 i think that one good example of your code, something that is relatively useful as a model (like the gui) would be very helpful in showing everybody what forth & simplicity _really_ means 14:29:49 oh, ok =) 14:30:01 chip: OKAD and the 25x simulator are both for colorForth? 14:30:03 I like his IDE block, personally. 14:30:07 chipChuck: you need to be a better marketing man - SELL OKAD - :) 14:30:09 i do too 14:30:14 furthin: OK was published, the smallest GUI, and you may be confusing the 1-2k GUI that I wrote for iTV and others with Chuck's GUI in OKAD II. 14:30:14 That's the smallest IDE disk driver I've ever seen. ;D 14:30:19 but it's short, and it's not really that usable to me 14:30:32 i'm interested in a bigger application 14:30:36 futhin: It's plenty good enough for implementing blocks. 14:30:40 like a gui, as an example 14:31:16 futhin: a web browser would be a good example 14:31:20 the ide driver isn't much of an application to me.. just a few lines of code.. doesn't show me ways of solving problems, how do deal with lots of different things, etc.. 14:31:32 To me, a gui means point and grunt. Very hard to program in 14:31:42 But here's the crux? What is an "application" in Forth? 14:31:49 drag and drop coding --- the way of the future! (lol) 14:31:51 gui must not be windows, it can be much else 14:31:56 agh 14:31:58 theres no application, only words 14:32:01 kc: heh, true :) 14:32:01 dont have to be windowing :) 14:32:08 or maybe archived dictionaries ;p 14:32:09 all the same :) 14:32:12 * kc5tja agrees with mur 14:32:22 no such thing as a forth program. just --- forth. you extend it with what you want. 14:32:42 I did a home theater controller for my laptop. Released with some version of colorFroth 14:32:47 mur im doing a text windowing system in isforth - a curses replacement :) 14:33:04 i have an idea of how to combine a gui and command line, so that you can get away with never touching the mouse 14:33:06 I440r: Great :-) 14:33:08 actaully current gui consept is very hard to use, but we have learned it.. many users 14:33:24 i have a better gui concept ;P 14:33:31 the more simple and less features there are visible the easier it is to use 14:33:37 chipChuck: I'd like to implement a BPSK modem in Forth for my laptop, to see if I can get 19.2kbps throughput on the 2m ham radio band. Currently, the highest one can get is 9600bps using very specialized, very expensive radios. I'm looking to change all that. 14:33:40 i have very many good gui consepts 14:34:03 kc5: sounds good 14:34:20 mur: "concept" 14:34:23 I still like the idea of voice control 14:34:27 agh, rob_ert! :) 14:34:34 chip: do you plan on improving colorForth's compatibility with more hardware? 14:34:39 I'd like the CPU to do as much of the work as possible (just the opposite of "common sense," since it's just a prototype), but that's largely only because I am not made of money. :) 14:34:42 futhin: can you explain your GUI concept? 14:34:45 (or time for that matter) 14:34:47 mur: You can't spell like in finnish :) 14:34:50 chipchuck i dont think i would like that - i dislike telephones etc heh - if it doesnt have a keyboard.... :) 14:35:01 There was a good suggestion for booting colorForth on the mailing list 14:35:15 Voice recognition is one such application for neural networks, which brings us back to the 25x... :) 14:35:28 rob_ert yes but finnish hasnt' got all those tempus rules (any form of verb can be used) and graphemes (one letter is one sound) 14:35:32 chipChuck: have you heard of the ShortTalk system? 14:35:36 I'll improve compatibility if I get the right suggestions/ideas 14:35:38 i think 25x could be used by mars electronics with their bill acceptors too 14:35:39 chipChuck: voice-controlled editing 14:35:56 they use nural nets exclusivly now for recognition - still do validation the hard way tho 14:36:12 bpt: similar to plan 9 gui 14:36:20 ShortTalk no 14:36:34 i dont like plan 9 - i like windowmaker - very nextish 14:37:04 Voice could replace colorForth keypad. No good for text, tho 14:37:13 chipChuck: http://www.research.att.com/~klarlund/ShortTalk/ 14:37:22 plan 9 has a command line prompt doesn't it? (unless i'm messing up which gui i'm thinking of) 14:37:25 * kc5tja dislikes Plan 9 too. Although I'm all for function before form, there still has to be *some* kind of form! Plan 9 has none at all. That's why I like AmigaOS Classic's GUI -- very simple, very clean, very functional, all in the right proportions. 14:37:41 futhin: can you describe it? (I haven't used Plan 9) 14:37:48 Plan 9 is like the next generation of Unix. 8(1/2) is the GUI for it. 14:37:51 i only read about some ideas about plan 9, haven't actually seen it 14:37:58 workbench is absolutely the best gui ive ever used! 14:38:19 kc5 is it replacing x? - i think x is in its dying throws 14:38:24 I440r: A shame my Amiga mouse is fucked up then ;-) 14:38:26 futhin: okay, but can you describe /your/ GUI system? 14:39:08 chipchuck are you currently making any money with forth yourself ? 14:39:23 No money at the moment 14:39:26 I440r: No; AmigaOS Classic is dead. No longer being supported, except to prolong life as a basis for AmigaDE. 14:39:31 I'm on sabatical 14:39:54 me too. for the past 9(ish) months. not a single contract 14:40:03 one fone interview - but no jobs 14:40:16 economy really bad now 14:40:35 which is why my father and i are trying to set up a gun smithing operation here 14:40:54 sounds great, good luck! 14:40:56 We're selling house, reorganizing life to gain time and money 14:41:01 anyoe want to by some hand made 1911 grips ??? :) 14:41:15 Heh... are there any rich forth coders out there? :-) 14:41:19 I440r: Don't like guns :) 14:41:23 ive been selling stocks at a loss just for food - i hope your not THAT poor 14:41:32 There were rich coders. 14:41:45 I440r: :( 14:41:50 Where have all the coders gone... 14:41:51 linus t isnt poor 14:41:55 The closest to a rich Forth coder is Elizabeth Rather, I suspect. 14:42:00 I meant forth coders, I440r. 14:42:03 and his codeing style and choice of language just plane sucks 14:42:12 linus works for some company 14:42:41 Linus currently works for Transmeta. 14:42:44 linus works for Transmeta 14:42:58 tbw but linux IS making him rich 14:43:00 Some of Forth Inc's customers made millions 14:43:35 Oh, here's a question I've been meaning to ask. How long did it take you to write and debug OKAD or OKAD-II? 14:43:37 chipChuck: are you thinking of fedex and others? 14:43:55 brb - coffee... 14:44:03 OKAD till several years. OKAD II several months 14:44:33 * futhin is thinking of asking some of the questions left over from last interview 14:44:34 I440r: linux is indirectly making Linus rich, yeah....though not as wild as say, B.G. :) 14:44:36 OK, so I have about two years to look at in developing the circuit simulator then. 14:44:45 Yeah. FedEx, King Rihad, Cybek 14:44:51 Chuck, you mentioned that you used F-PC as a PC Forth. Didn't you get a free copy of PolyForth when you left Forth Inc.? 14:44:51 perhaps rewriting a thing is faster than writing it in the first place, you already have a path defined. Another reason to rewrite that code! 14:44:59 futhin: I personally would be more interested in hearing Mr. Moore ask us questions, if this is what he would like to do 14:45:30 kc5: OKAD was sourceless. OKAD II is colorForth 14:45:33 BadCog2: Yes, absolutely correct. My next Dolphin is a 0.5 release, and it's looking better and better all the time. 14:45:50 --- quit: gforth (Connection timed out) 14:45:52 --- quit: onetom (Connection timed out) 14:46:14 Q: what do you consider the most difficult aspect/challenge when it comes to forth programming? 14:46:16 fox: not really. I had a copy, but no authorization 14:46:28 lol 14:46:43 * kc5tja ponders what resolution to make the voxels in his circuit simulator... Unlike chip simulators, this simulator must be 3D... :/ 14:46:45 tbw directly, indirectly - who cares - is making him rich :) 14:47:09 The challenge of coding is to understand the problem 14:47:15 Independent of language 14:47:17 I440r: hehe 14:47:27 The only way to do this is to code, then recode 14:47:36 Only Forth makes that feasible 14:47:51 I440r: but you're basically happy using Linux as versus the other operating systems out there, so Linus earns his due :) 14:48:25 chipchuck: how about specific to forth? anything hard about forth itself? 14:48:31 In the case of Forth programming, that means factoring 14:48:49 Stack management 14:48:52 tbw he DID earn it yes - but so has chippie here - he should be rolling in it 14:49:09 chipChuck: Well, I think it would be a good idea, prototype if you will, to just create a 2-D model, where only PC board traces are emulated, and not worry about the third dimension until I'm pretty happy with the PC board emulation. :) 14:49:09 Newbys put too much on the stack, then too little, finally just right 14:49:22 I440r: Linux is more popular than forth :) 14:50:10 heh, i'm at the stage of trying my hardest to put too little on the stack 14:50:13 i4: Linus' goals dropped a lot closer to the tree, though 14:50:14 chipChuck: how much should one put on the stack (on average)? 14:50:15 is that a bad thing? 14:50:24 futhin: Hehe :) 14:50:26 kc5: actually 2.5D, since boards have few layers 14:50:31 he originally only set out to build a more sophisticated version of Minix :P 14:50:40 does too little mean that i'm wasting my time? 14:51:09 that isn't anything even resembling an evolutionary step, imo 14:51:14 chipChuck: From the PC board perspective, that's true. However, I'll also be needing to simulate things like capacitors and coils. In an RF environment, exact placement of such components can make or break a circuit. 14:51:15 usually the greatest plans are too great to work as planned.. 14:51:21 Pass 2 parameters on stack. More can be nested for later use 14:51:27 mur: :) 14:52:31 Same with variables. Use with discretion, but don't avoid 14:52:54 I440r: Have you asked Chuck about his opinions about the original forth VARIABLE against the solution you used in IsForth? 14:53:07 not sure he is aware of my solution 14:53:12 i define var and const 14:53:15 Not aware 14:53:18 variable and constant both exist too 14:53:31 a var is a better name for a "value" it is more descriptive of its function 14:53:40 its definition is identical to a constant 14:53:50 a const is similar but all consts are state smart 14:54:00 i.e. they compile literals 14:54:01 in arm any const can be a variable 14:54:05 in RAM 14:54:18 what i see is diffulct in forth is all those stack management commands like TUCK etc. tehy're hard to remember as they have nothing in common. hard especially for nonenglish speaking, though i suspect that english speaking have hard to remember which certain words they were.. also i see that what might cause it's use to be less than it coudl be is that it uses text marked variables less than some other languages.. 14:54:29 my plan is to have the conts defined in the compiler vocab which will be discared on turnkey 14:54:35 mur: They're few enough to learn. 14:54:43 yep, but still 14:54:49 2 const af_inet 14:55:00 mur: But I guess you could try to make a more minimalistic and simplier variant, if you wanted. 14:55:02 : blah .... af_inte ... ; 14:55:05 compiles a literal 14:55:10 mur: Not sure how readable that would be. 14:55:13 Only necessary words are DUP DROP OVER PUSH POP 14:55:20 when you turnkey the definition for af_inet itself is deadwood 14:55:31 mur: A friend of mine once wrote a "stack manipulator compiler" for PygmyForth once. You used it like this: STACK-OP MyDUP (a-aa) 14:55:34 i dislike push and pop - they arent descriptive of whats happening 14:55:37 i prefer >r and r> 14:55:50 also prefer !> not TO 14:55:54 mur: if it calls for a lot of stack dancing to get things done, it can be hard to follow, yeah....of course the usual "advice" is to rewrite that....and that can get tiresome sometimes 14:56:15 mur: It would create a CODE word MyDUP, then compiled the code that implemented the a-aa stack effect. So, he could implement TUCK like this (ab-bab) :) 14:56:37 * kc5tja flat out *never* uses VALUEs or TO. 14:56:50 If it is a variable, I declare it as such, and always use @ and !. 14:56:53 >r and r> smack of jargon 14:57:00 they save you a @ 14:57:27 @ and ! distinguish Forth from other languages 14:57:33 they are more descriptive, ! store > to. store something To somewhere 14:57:40 I440r: How do I store something in a VAR? 14:57:42 the word TO is ambiguous 14:57:46 from 1 TO 10 14:57:50 !> 14:57:52 why does >r and r> smack of jargon? 14:58:03 they are cryptic to non forthers 14:58:05 I440r: Like: 1 myvar !> 14:58:14 i think ! and @ could've been named better.. like SAVE and GET 14:58:21 10 !> myvar 14:58:23 futhin: ! and ? :) 14:58:35 !> is parsing :) 14:58:37 ! has historical precedent in the Lisp community. 14:58:39 and state smart heh 14:58:58 I440r: Okay... is that slower/larger than '1 myvariable !' ? 14:58:59 people complain it's a write-only language, possibly because it looks a little like line noise :) 14:59:00 How do you spell them? >r was originally I am not sure about @ though. However, as someone else on the Forth newsgroup mentioned, the more often a word is used, the shorter it should become. And @ and ! are used quite often. :) 14:59:42 kc5tja: Yeah... :) 14:59:44 robert no speed difference nor size difference on the store. but fetches save you the @ in the code. the fetch still occurs, maybe a little faster 14:59:47 the xecutd form need to be small, but the compile time code can be as large as you want 14:59:49 but definatly smaller 15:00:00 I440r: Okay, nice :-) 15:00:08 CrowKiller: if you can discard all that compile time code when you turnkey yes! 15:00:17 one if my goals btw 15:00:25 chipChuck: I like >R and However, >R and R> does have a certain symmetry too. 15:00:46 transient vocabularies. exist at compile time, discarded at turnkey 15:01:06 chip: what was the issue about whether 'drop' should be included in 'if'? 15:01:09 I find the chipchuck philosophy of edit compile xecute pretty good: edit is the decision we take as programmers, compile is the automatic, preprocessor like decisions and execution is execution 15:01:11 kc5tja: it's not harder for me to type GET instead of @ and is more readable 15:01:21 ive got a areplacement for if else then too :P 15:01:36 ( --- f ) ?: true-word false-word blah blah 15:01:44 Hardware doesn't like to include DROP in IF 15:01:45 thats why we have good c compilers, because we have lazy c programmers ;p 15:02:12 I440r: You really like forcing people to factorize! 15:02:12 ?: MAKES you factor out the if and else parts :) 15:02:19 so does my case: definition :) 15:02:28 Yeah 15:02:35 But that's more understandable 15:02:39 robert i think its unforthlike to FORCE good practices on people - it wasnt my intention, it just happned that way heh 15:02:42 Would changing @ and ! make Forth more acceptable to people? I doubt it 15:02:53 Haha 15:03:01 --- quit: bpt ("bbl") 15:03:02 So do I ;) 15:03:16 Still feels a bit strange to factor out small IF THEN constructs, I440r. 15:03:25 It's a simple optimize to make X @ into one machine instruction 15:03:42 I440r: um...I'd say it'd be wise not to factor into atom-sized pieces....then I'd have to devise good names for all those pieces...yuck 15:03:45 my 3 instruiction forth is like the one from fran sergeant(pigmyforth) !,@ and # != store @=fetch #=jump or execute 15:03:50 yes - but if/else/then still exist in isforth 15:04:02 ! and @ are essential in their caliss forms 15:04:05 classic 15:04:07 tbw you could make them headerless :) 15:04:08 Large IF ... THEN constructs, especially nested, are unreadable 15:04:24 if/and and but loop nesting is very bad 15:04:32 I440r: with care, yes...but the code would get more unreadable IMHO 15:04:34 if (x=5) 15:04:37 What I'm especially interested in is IF-less coding practices. 15:04:38 .... 5000 lines of code 15:04:42 else 15:04:47 * kc5tja really likes the concept of coding without any embedded branches. 15:04:47 5000 lines of code 15:04:56 totally lost context on the if/else 15:05:24 Hehe 15:05:37 Simple code doesn't make decisions at execute time 15:05:39 Hmm... is there any clever solution to fix the naming problem? 15:05:58 If your imagination isn't that good, how should all those factored-out code pieces be named? 15:06:05 * TheBlueWizard wonders whether chipChuck is bored...figures he's seen those issues many times.... 15:06:14 * kc5tja wrote a program to convert lowercase ASCII letters to uppercase once, without ever using an IF statement anywhere. I was rather pleased with the results. 15:06:22 Yes, but not recently 15:06:27 chipie not everyone has the ability to simplify complex problems 15:06:33 kc5tja: :) 15:06:47 I have a question 15:07:05 futhin has a FEW he would like to ask too :) 15:07:12 Surely the less clever you are, the simpler your code? 15:07:19 To what extent do you feel that it is worthwhile to share Forth code examples on the internet. 15:07:23 chipchuck: how do you feel about the interview right now? 15:07:24 ( ch -- CH ) DUP 'a U>= 'z U<= OR $20 AND XOR 15:07:39 eh 15:07:51 i believe strongly in the saying "it takes a genius to be simple" 15:07:58 Doing great 15:08:05 the more clever you are, the simpler your code ;) 15:08:08 kc5tja: Bit arithmetics is nice if you know it, which I don't ;) 15:08:10 The problem with sharing examples 15:08:26 is that non-trivial code is interesting to only a few 15:08:35 rob_ert: It just takes experience. 15:09:23 For most of my applications, I write using IF and related constructs. But where speed becomes a critical issue, I'll work hard to optimize out any and all unnecessary branches, since that thrashes the CPU's branch prediction logic. 15:10:04 Here is the one spot where I will introduce comments into my code: I'll detail the original, "slow" but obvious, implementation, and explain the reason for replacing it with the more non-obvious, yet faster, implementation. 15:10:07 chipchuck: what about creating a little essay describing forth and the forth philosophy, and going into depth on a good forth code example.. 15:10:08 I'd like examples of coding classic algorithms 15:10:19 futhin are you aware of the origin of the word genius ? 15:10:24 :) 15:10:35 of course, einstein ;) 15:10:49 futhin its latin i believe for guardian angel 15:11:06 Nope. 15:11:14 It's latin for, "to beget" -- that is, to give birth to. 15:11:16 If I was a better teacher, I'd do that 15:11:26 But I'd rather write code than teach it 15:11:26 The root word for genius is "gene", as in genetics. 15:11:39 large examples, whole applications rarely get read because it is alot of work, it takes alot of interest and dicipline and intent 15:11:48 Chuck has said that he has bee providing examples and explanations for years and will continue to do so, few people pay much attention. 15:11:51 chipchuck thats my problem with isforth - no dox cus in too busy coding it heh 15:11:56 I like teaching, but I'm a useless teacher :-/ 15:12:19 thefox and what about you providing some examples??? :) 15:12:25 But small snipits describing good techniques might help other people good forth style, how to build better castles 15:12:29 It's foolish to think that Forth teaches itself 15:12:38 i think simple source code examples teach much more than descripting what it does. teaches in way that it'll be learnt 15:12:43 But it almost does, being interactive 15:13:01 chipChuck: The user must be receptive to it though. 15:13:04 They're quite often not. 15:13:13 People get gigabytes of tutorials from my web site each month. I will continue to add new ones and try to make them better. 15:13:16 mur: Also teaches you a lot of smart tricks 15:13:21 Ah, for an open mind 15:13:23 like examples on how to avoid control structures 15:13:33 thefox: May I ask where your site is? 15:13:37 chipchuck: your code can tell a story.. here's an idea: how about coding a little example forth application, and while you are coding, the text is being "recorded" all the stuff typed, and all the changes, are saved.. and then it can be played back, and people can see a little, how forth is coded & used 15:13:39 examples on how to minimize stack traffic, and such 15:13:47 colorForth compiler has few conditionals/loops 15:14:03 thefox how about putting those examples in physical form and getting o'reiley to publish them 15:14:06 I remember seeing Chuck's use of 64STARS, where 64STARS consists of 4 calls to 16STARS, etc. That changed my coding style substantially. 15:14:20 http://www.ultratechnology.com I have lots of html and video presentations and tutorials by Chuck and I. 15:14:43 The only problem with thefox's video website is that I can't download the videos fast enough... ;D 15:14:49 thefox: Nice, thank you. 15:14:55 I440r: publishing requires expense...similar catch-22 like with x25 fabbing :P 15:15:01 thefox: your site is the best I learned very much there 15:15:02 futhin: my coding style is a mess. Only the final result is publishable 15:15:05 I have a few of the disk with the movies on them 15:15:16 tbw o'reiley might publish - look at all the linux crap they publish :) 15:15:34 futhin: did you looked at colorforth sources? 15:15:47 well, Linux stuff numbers in millions now.... 15:15:48 crowkiller: the parts in forth 15:15:54 I440r: Heh, you really don't like Linux, do you? :) 15:16:03 a lot of the code is in asm :( 15:16:04 robert i love linux. its the ONLY os i use 15:16:06 * chipChuck getting a coke 15:16:11 chipChuck: Is it a mess, or is it just dense? There is actually a difference. From what I read of your IDE sources, it's very dense, but it makes a lot of sense. 15:16:21 futhin: I learned asm to be able to read it 15:16:22 coke will kill a rhino :P 15:16:46 kc5 nobody ever gets it right first time. most coders however accept their first attempt 15:16:46 --- join: ChanServ (ChanServ@services.) joined #forth 15:16:46 --- mode: carter.openprojects.net set +o ChanServ 15:16:46 --- mode: ChanServ set +l 83 15:16:46 who cares about asm? :P 15:16:51 I suppose there are those who think dense code is a mess though. I personally dislike writing dense code, but that's just my style. 15:16:52 * rob_ert  15:16:56 * I440r slaps futhin arround with the isforth kernel 15:16:56 so try to understand that first and the come again asking for Keystroke captures LOL 15:17:01 probably better than cigarettes though 15:17:06 I wrote an article for the Free Software Magazine this month on zero operand architecture and Forth. They sell over 5000 printed copies and I have no idea how many people in the Free Software community read it online. 15:17:26 fox: do you have a link? 15:17:27 BadCog2: Yea, tabacco shouldn't exist :-/ 15:17:30 thefox: Never heard of this magazine. Where do I find more information on it? 15:18:09 * chipChuck is back 15:18:14 wb :) 15:18:20 kc5tja: I think Slashdot mentioned that magazine 15:18:36 * kc5tja doesn't read Slashdot. 15:18:51 i read 2/3's of the /. interview with chuck :) 15:18:54 And I very actively try not to when it can be avoided. 15:18:54 * TheBlueWizard does :) 15:19:08 my fave part is the "ans doesnt describe forth but a language of the same name" heh 15:19:10 There have been four issues so far I think. There are various sites where you can find it. Search Google for the Free Software Magazine. I will post my own copy after they post it FSM. 15:19:34 chipchuck: would you be interested in coding a simple forth application that shows some of your coding style and tricks, problem solving approach, etc.. 15:19:57 there doesn't seem to be a whole lot of forth code on the internet.. well-written accessible forth code.. 15:20:04 yes - particulary how you avoid using structures.... 15:20:08 I do that all the time. Do you have a suggestion? 15:20:22 I440r: You should add a section with Chuck quotes only to your "quotes" file in IsForth ;-) 15:20:47 futhin: theres the colorforth source on merlintec, sercah "color forth source" on google 15:21:05 i dont read enough of the online stuff to be able to quote chuck, too busy coding - but fortune runs for every shell i open :) 15:21:06 uh, what kind of suggestion? 15:21:42 What simple application I should code? One problem is that I often simplify the application away 15:22:09 well what about putting the gui up online? 15:22:14 lol 15:22:19 simple application :) 15:22:29 One reason to read Forth code is the same as why I read Linux code. To find out how to use a different interface card 15:22:35 An example right in your face: kc5tja: minus to uppercase, just substract 32 from the ASCII char, no? 15:22:41 kc5: www.rons.net.cn/english/FSM/online, if you hadn't found that yet... 15:23:08 CrowKiller: Only if the charachter is in the right interval. 15:23:08 ugh i hate wading thru linux code - i would rather reverse engineer libc to learn how to do the syscalls than look at its sources 15:23:18 tathi: Thanks. I think I just found tha tsite, actually. 15:23:19 much more descriptive of what i need to do 15:23:26 I440r: Use the google instead :) 15:23:26 rob: the code of the conversion is just that 15:23:37 futhin: My coding style isn't like Chuck's, but I have a block editor I'd be happy to post. I just haven't had the time to do it just yet. 15:23:48 robert no - ill use ida-pro thats why i baughtn it 15:23:50 it's written for GForth though. 15:23:53 I'll post PPP as soon as I code it 15:24:01 --- join: gforth (guest@adsl52009.vnet.hu) joined #forth 15:24:04 C source code is ALWAYS obfuscated 15:24:14 I440r: What is that? IDA-pro... 15:24:23 Obfuscated code is job security 15:24:26 chipchuck: theres some good info in a msg i sent to the colorfoth mailist about embedded ppp 15:24:26 the interactive disassembler pro 15:24:35 Heh 15:24:36 chipChuck: I'm sorry. The effects of writing a PPP stack will pass as soon as the medication starts to take effect. ;P (As a person who's reviewed the X.25 protocol standard, I can feel for anyone who has difficulty implementing PPP) 15:24:38 chipchuck lol 15:24:39 I saw it 15:25:07 programming in forth seems to be more about coding for oneself.. if forth were geared more to coding for users, then perhaps more people would code in it.. code for user interfaces like gui, etc, are interesting for that reason.. 15:25:21 from my readme #define foo (blah << 6 && ((fudge >> 2) ^ (wtf << 3))) 15:25:32 ...where blah, fudge and wtf are equally indecernable... 15:25:33 if you were to reinvent TCP/IP ... how would it be different? 15:25:44 i think he answered that one 15:25:44 I440r: That's too readable. The variables are all longer than one character. :) 15:25:51 That's true. I code for myself. My PPP will interface to Earthlink only. But modifyable for others 15:26:10 oh i should be using variablel names like _ you mean? :) 15:26:17 I440r: Hey, I've seen it done. 15:26:23 so have i 15:26:51 I'd make TCP/IP headers shorter. Fewer options. But the concepts are valid 15:27:04 chip: i'm interested in seeing you put the gui code online, unless you are against that for whatever reason 15:27:09 tcp is nicely designed i think 15:27:20 * kc5tja sorely dislikes TCP, though I really like IP. 15:27:20 tcp/ip even 15:27:28 ip was a mistake 15:27:36 it should have been 64 bits from the start 15:27:36 No, IP is very cleverly designed. 15:27:41 Oh, I see. 15:27:51 IPv6 answers that issue, and it's actually even *simpler* than IPv4. 15:28:01 but in those times 64 bits would be overkill 15:28:03 and it will never be universally used 15:28:04 For his ham radio project, he intends on using IPv6 from the get-go. 15:29:03 it will be like that old joke in england. all cars will be driving on the opposite side of the road to be more in line with the rest of the world, the change will be gradual so people can get used to it 15:29:19 ipv6 is a non starter 15:29:26 futhin: That gui is so old, I'm not even sure what it was. Fox has a modern example. But I've no colorForth code yet. It'll be in browser 15:29:34 I440r: In the commercial industry, yes. Ham radio is not confined to commercial whims. 15:29:53 kc5 linux already supports ipv6 - not sure my isp does tho :) 15:30:01 kc5: most people do move with commercial whims in amateur radio, though 15:30:14 I440r: Who gives a hoot what your ISP or my ISP supports? I don't care. :) 15:30:25 :) 15:30:30 kc5: a lot has changed in the last 10 years or so even 15:30:32 goshawk`: Those are the same people who won't be using packet radio either. 15:30:58 Alan Kay once said, "The best way to predict the future is to invent it." 15:31:17 I'd like to know how to read my scanner on USB. 15:31:21 This is perfectly in line with Forth, perfectly in line with ham radio, and is perfectly in line with my direct observations of the sorry state of affairs in the amateur digital community. 15:31:46 If I can figure it out, that'd be a good example 15:32:00 Question to everybody: About how many instructions per second does a general-purpose personal computer need to perform in your opinion? Assuming everyone won't use bloated software, of course. 15:32:13 chip: would a USB printer be any easier? 15:32:24 usb isnt the problem - its figuring out the data that the scanner sends OVER usb thats the problem 15:32:27 is this for games or what? 15:32:27 chipChuck: Hmm...an interesting question. I'd like to recommend MindShare's "USB System Architecture" as a decent reference. It's explained a lot to me. 15:32:34 BadCog2: Nah... 15:32:34 my printer is a usb WIN-printer 15:32:37 linux doesnt support it 15:32:42 BadCog2: You use Windows for that ;) 15:32:44 Probably not. I have to learn about USB. Anyway, don't have such a printer 15:32:44 www.beyondlogic.org 15:32:46 I440r: It'll be SCSI's command set for scanners, quite likely. 15:33:07 chip: I'd be willing to provide mine (a cheapie, but still)...if that would be at all helpful to you 15:33:08 kc5 or a superset of that 15:33:34 but, if it isn't easier, nm then =) 15:33:37 i got my printer for free when i joined this isp 15:33:42 and a flatbed scanner too :) 15:33:59 * TheBlueWizard thinks many USB based devices have proprietary protocol trafficking thru the wire, so it'd be tough to reverse engineer the mess 15:34:17 isps are giving away free printers and scanners to people who join up? ;) 15:34:20 I440r :O 15:34:23 TheBlueWizard: Not true. The USB chipset handles the traffic control, not the devices. 15:34:30 gos: thanks, but I want to master my devices 15:34:49 the days of the free lunch are over, atleast until the next boom cycle occures 15:34:53 I440r: freebies? wow...n33t...but prolly winprinter and win specific scanner 15:35:26 kc5tja: I'm talking about the command stuff, not the traffic controlling 15:35:28 bah, i wait the tiem when all devices follow ONE solid standard.. :) 15:35:32 printer is win-printer - linux supports the PRINTER, just not the usb version of it 15:35:49 and i can alwasy run windows under vmware to use them :) 15:35:56 chipchuck: are you interested in being some kind of a regular to this channel? perhaps coming by once a month to keep track of the forth community pulse :) 15:35:59 I440r: heh, you half-lucked out on that :P 15:36:12 mur: It'll never happen, except in a Communist nation. 15:36:15 thefox: same question for you, are you interested in being a semi-regular? :) 15:36:18 i dont think #forth is in tune with the forth community pulse :) 15:36:21 i dont think this chan represent the forth community 15:36:32 CrowKiller: <-- what he said :) 15:36:32 A part of it. 15:36:33 i440r: i know, i exaggerated i suppose :) 15:36:43 only a little --> " heh 15:36:43 kc5tja i think it might. and will.. mostly, once it's developed. the standard.. 15:36:44 hey, futhin is just throwing a shameless plug for the channel 15:36:48 nothing wrong with that =) 15:36:49 I440r: That's because the Forth community doesn't have a pulse. 15:37:03 it does - but its a very quiet one 15:37:05 futhin: I'm really unsettled now. Invite me sometimes. Maybe when I resettled I can predict schedule 15:37:06 we should code more, argue less ;p 15:37:12 Maybe that would give people inspiration though. 15:37:13 and tends towards abominations like ANS 15:37:31 chipchuck: ok, that's fine, thanks :) 15:37:54 chipchuck i added you to the channel access list. chanserv will op you when you join the channel if you have authenticated 15:37:57 jeff fox too 15:38:02 #forth doesn't require any invitation....I just dive into #forth and chat away hehe 15:38:05 I can be a regular at the moment. 15:38:07 unanounced visits would be welcome :) 15:38:15 heh :) 15:38:22 I440r: Hehe... they'd be scared away if they came here a normal day ;-) 15:38:36 irc is one HUGE waste of time tho - if im on irc i cant code, i have to disconnect 15:38:40 I440r: Ok. Tho I don't know what that means 15:38:56 chipchuck it means you get channel operator status when you join 15:39:03 an @ in front of your name 15:39:12 So you can kick people you don't like. 15:39:16 it means you can remove abusing users 15:39:16 it makes you semi in charge of the channel (sort of :) 15:39:28 I440r: I'm sure he feels honored. 15:39:31 and change topic (which is on the top of the window) 15:39:39 its traditionally a way of recognizing a persons worth in the channel 15:39:44 Hehe 15:39:48 If you use a windowing IRC client, that is... 15:39:56 and make others operators and add voice 15:40:05 I think his name makes people trust hun, I440r. 15:40:07 I think he gets it... ;) 15:40:08 him* 15:40:14 chipchuck: you are using mirc? 15:40:19 Yes 15:40:35 thuogh, if someone unknown comes and asks to type /mode #forth (something) then shuodl you be alerted 15:40:45 What makes you prefer Windows over UNIX clones? 15:40:45 like: 15:40:46 Though it motivates me to write my own 15:40:51 Hehe 15:41:01 Isn't there any forth IRC client? 15:41:01 /mode #forth +o mur 15:41:06 woudl make me operator too :) 15:41:06 unix/linux is very unforgiving of the beginners 15:41:11 --- mode: I440r set +o mur 15:41:12 /mode #forth +q mur 15:41:13 rob_ert: Nope. The closest thing we have is I440r's bot code. 15:41:16 --- mode: I440r set -o mur 15:41:17 :) 15:41:19 heeh 15:41:23 kc5tja: Oh... 15:41:25 yes.. an irc client would be pretty easy and interesting to code 15:41:25 Heh 15:41:38 futhin my forth irc bot is basically an irc client 15:41:40 That really shows how little forth is used. 15:41:42 Windows was my shortest path. It still is 15:41:46 if i add a user interface to it thats exactly what it will be 15:41:48 * kc5tja has always found the IRC protocol to be unbelievably over-complicated. I really couldn't follow the RFCs for it. 15:42:08 I440r: Then do so! 15:42:13 I'd offer to try an irc client if tcp/ip support ever becomes available in colorForth 15:42:20 kc5 the protocol doesnt even follow itself in all cases 15:42:29 Haha. 15:42:31 some message bodies start with a : others done 15:42:41 kc: if you can telnet in and send message with PRIVMSG to the channel, it seems simple enough to me 15:42:43 different numerics have a different number of parameters on them... 15:42:45 IRC doesn't only ruin your life, it's impossible to code for, too! 15:43:00 rob: heh 15:43:03 the docs don't fully describe the protocol actually in use 15:43:21 * rob_ert goes to clean his glasses. 15:43:21 i need a better way to parse raw irc messages than im using right now tho 15:43:40 ok I have a question: do you people looked into gaming platform hobby dev??? they cost a lot less than a pc or a pda; imagine a Forth working on all the gaming platform: GBA, PS2, XBOX and Gamecube, in the order of fesability. 15:43:46 and the bot is on hold till i get the sockets code a little better 15:44:14 * kc5tja can't telnet into port 6667; it times out after trying to find my identd. 15:44:16 why is gamecube last? :-( 15:44:22 crow i tried to get bally gaming to let me use forth once - they wernt interested 15:44:31 proprietary drive and no docu on the hardware 15:44:36 one forth for all consoles simultaneously? 15:44:37 the gba 15:44:44 has a serial bootloading mode 15:44:49 when theres no cart 15:44:55 its truly great for homemade dev 15:45:01 I want a 42" plasma monitor, not an NTSC 15:45:08 a forth game console would be good 15:45:09 Crow: great idea. I have often thought that the best games would require programming for new reflexes for your droids or spaceships or whatever and of course in Forth. 15:45:24 chipChuck: Plasma is too power hungry for me. I prefer a scanning laser display. :) 15:45:24 imagien changing games 15:45:31 using forth to modify existing games 15:45:32 --- join: GilbertBSD (~gilbert@m183.max3.dacor.net) joined #forth 15:45:35 maybe a terminal emulator with SOUP capability would bring news and mail to colorForth 15:45:38 I would like to see a MISC baised Forth game console :^) 15:45:39 lol the possibilities are endless lol 15:45:40 hey hey GilbertBSD 15:45:47 hi MUrr 15:46:00 has chip chuck been here? 15:46:02 * kc5tja has a great idea for a MISC F21-based game console, which I shared with Jeff a couple of days ago. 15:46:06 Hi GilbertBSD. 15:46:07 he IS here 15:46:27 oh. 15:46:33 I thought of recoding early Arcade games in Forth. Don't know what it would take for modern games 15:46:44 a real graphics engine 15:46:46 Hmm 15:46:47 written in forth 15:46:49 there are some classic games that i would love to preserve :) 15:46:57 a gba multiboot serial uploader could be coded in colorforth, using minimal hardware 15:46:58 written in forth of course ;) 15:46:58 i want to recode the original settlers in forth 15:46:59 I want old Nitendo games in forth :-) 15:47:00 chip: the dissection of the actual game itself I think would be the hard part 15:47:00 for linux 15:47:07 You have three F21s dedicated to video generation. Using the RGBI outputs of each F21, you have red channel, green channel, and blue channel. One of the F21s will be responsible for synchronization of course. Then, using two more F21s, you can get stereo audio, also using the 4-bit output from the video port. 15:47:10 they are very well married to the architectures they were written for, often 15:47:16 A graphics engine that it would be easy to interface Forth into, fully modern, all of the effects 15:47:16 i440r: cool! that would be interesting :) 15:47:20 I really love those games, not too complex but still a lot of fun! 15:47:24 good games SELL 15:47:29 and make alot of money 15:47:33 Hmm... 15:47:34 Nah. 15:47:36 but the gaming industry is fickle 15:47:39 Modern games sell ;-) 15:47:42 Bad: the iTV Internet products could load machineforth apps over tcp/ip and run them. You could also use it as a server and FTP or Telnet into it over the internet. All Forth. 15:47:50 SMB3 still is funnier than most modern games. 15:47:50 a gba is a lot more expensive than a pda, and it got a colorscreen ;p 15:48:01 i440r: i tried settlers, but the graphics is so poor, and the user interface is a little awkward.. 15:48:05 Do any graphics boards support 16x9 aspect? 15:48:15 --- quit: GilbertBSD (Client Quit) 15:48:22 chipChuck: None to my knowledge. 15:48:23 some games support it in software 15:48:23 futhin the game supports a svga mode - 640x480x256 15:48:31 I think Macintoshes are the closest you'll find in that area. 15:48:32 and once you know the UI its PERFECT!!!! 15:48:40 heh 15:48:45 i STILl play the settlers and to me games have a shelf life of 2 weeks 15:48:45 --- join: GilbertBSD (~gilbert@m183.max3.dacor.net) joined #forth 15:48:50 the settlers was released in 94 15:48:55 I440r: Do you have any plans on adding some set of words to IsForth to do graphics? 15:49:04 i didn't read the manual, so i haven't really figured out settlers yet 15:49:17 robert i know ALOT of people who are experts at graphics engines. the maths is beyond me but the code isnt 15:49:24 kc5: yes F21 could do those things. 25x has some very similar capabilites. It would be pretty easy to generate HDTV data streams with 25x. 15:49:26 and i have alot of friends who can help 15:49:30 the best games are old, civ1 was released also in beginnign of 1990s 15:49:44 i will be doing a complete gfx engine eventually - i hope 15:49:54 i also still play civ 1 15:50:02 Would the bus interface of the 25x support the data rates for HDTV? HDTV is pretty ... high definition. :) And with high colors, that's a lot of bandwidth. 15:50:09 there is a NES emulator on the GBA laoding from the serial port 15:50:12 but i consider the original setters to be the best game ever coded. 15:50:13 I count Duke3D as a modern game, it's only 6 years old :-) 15:50:13 --- join: davidw (~davidw@adsl-209-109.38-151.net24.it) joined #forth 15:50:15 no need for carts 15:50:20 Hi davidw. 15:50:48 kc5tja have you seen HDTV european open equalent, DVB and MHP? 15:50:50 25x can handle huge data rates. The problem is memory to store frames 15:50:58 thefox: Does the F21 exist in a 64-pin DIP package at all? 15:51:13 68 pin clcc. 15:51:18 :( 15:51:22 chipchuck: maube a digital delay line, like an asynchronous output buffer or something? 15:51:22 mur: No. 15:51:28 chipchuck would the 25x require multi port ram or would just one of the computers be in charge of ram ? 15:51:44 --- quit: GilbertBSD (Client Quit) 15:51:54 kc5tja, mhp is actually java implementation, www.dvb.org and www.mhp.org 15:51:55 One computer runs ram. Or graphics buffer 15:51:56 chipChuck: That's why I'd rather have the three separate F21s. 15:52:14 mur: ??? HDTV is a broadcasting format, not a language-based technology. 15:52:29 No prob 15:52:36 so each of the 25 computers handles a different job? 15:52:39 kc5tja DVB is broadcasting format, mhp is opentv equ system 15:52:41 how did you decide on 25 ? 15:52:43 Right 15:52:50 mur: We don't have anything even akin to OpenTV. 15:52:57 25 fit on the smallest prototype chip 15:53:03 :) 15:53:03 mur: We don't have any market interest. The web is vastly superior to it. 15:53:14 har har 15:53:43 25 x computers are great in the field of cybernetic research, its the rhing theyre wainting for, a computer they can put under your skin 15:53:59 mur: You might laugh, but it's true. Our cable companies are trying hard to bring video-on-demand via TCP/IP, not through normal cable. That pretty much means HDTV is dead on arrival, if they manage to get it to work. 15:54:05 * chipChuck is experiencing sun dazzle. Will soon sign off 15:54:05 I describe 25x as a Forth cluster chip, 25 Forth cores. Where is Forth going? || parallel 15:54:25 kc5tja k 15:54:33 * kc5tja wants a 5x5 array of 25x chips... ;D 15:54:37 chipchuck ok 15:54:45 Well, nice to talk to you Chuck. Hope to see you again in a not too distand future. 15:54:52 distant evem 15:54:55 even even 15:54:59 bye :^) 15:55:01 * rob_ert hates his fingers. 15:55:11 Thanks for visiting. It's been a lot of fun, as always. 15:55:17 thefox paralell computing means more power but also more complex coding problems... 15:55:22 just to say to you Mr. Moore, in india someone like you could easly be a guru, you're a master for every person manipulating bits in our world lol 15:55:24 re 15:55:25 I want a few million 25x for myself. But I hate to be too greedy. 15:55:34 thefox: 15:55:40 lol 15:55:42 well guys 15:55:44 :) 15:55:46 i gotta get off the computer 15:55:49 CrowKiller: Heh.. 15:55:53 futhin: OK, bye :) 15:55:55 thefox if you get them im in for a job helping you code them :) 15:55:57 --- quit: chipChuck ("so long") 15:56:04 goodbye everyone 15:56:07 --- quit: futhin ("bbl") 15:56:09 nite futhin 15:56:18 jeez...didn't even get to say goodbye to the man 15:56:28 goshawk`: awww :( 15:56:29 i think i'll go to sleep 2 am anyway 15:56:37 mur: :) 15:56:37 hmm...no one ask when chipChuck will be back....guess it'd be a random event now :) 15:56:41 Night to you too. 15:56:48 take care =) 15:56:50 we must get something tasty 15:56:54 trying to :) 15:56:54 I440r I spent a decade studing parallel computing concepts and Forth tamed it to a few words. You can get everything for OCCAM, Linda, PVM, and more in a few lines of Forth. 15:56:55 not just ramblings 15:56:57 futhin will have to send a "pleeeeease" mail. 15:57:01 nite mur :) 15:57:02 thefox: Are the F21s available, and if so, for how much? Are they fairly easy to get going in a wire-wrapped environment? 15:57:08 unless we want him to go away for good ;p 15:57:20 thefox forth tents to tame alot of dragons :) 15:57:20 crow: agreed 15:57:52 thefox tends to tame a lot of dragons =) 15:58:06 * mur wants one dragon :) 15:58:14 I440r and goshawk` you have some? :) 15:58:31 Will anyone put up today's logs with Chuck and Jeff somewhere? 15:58:38 i have a hand drawn dragon curve fractal here - on paper 15:58:41 clog shoudl be logging it. 15:58:41 maybe someday I will grow up to be a dragon...but for the moment, no =) 15:58:44 took me 6 weeks to draw it :) 15:58:50 I440r :) 15:58:55 kc5tja: I mean, nicley cut logs. 15:59:00 its a maze for a game :) 15:59:15 When was he here the last time? 15:59:18 goshawk`: you want to become a dragon "for real"? 15:59:27 futhin was against advertising this event, so it wasnt as big as last time 15:59:33 was probably easier for chuck too 15:59:36 :) 15:59:40 When was he here the last time? 15:59:51 you missed it ? 15:59:53 I should check my logs. 15:59:55 No. 16:00:02 But I don't remember the date. 16:00:05 tbw: no...it was a poke at the fact that I'd like to really get somewhere with my life in the near future =) 16:00:13 robert: the 4th, I think 16:00:15 I440r: Of course I won't miss such an event! 16:00:22 tathi: OK, thank you. 16:00:24 goshawk`: ah...hehe 16:00:24 * kc5tja loves dragons. 16:00:34 * mur doenst eat dragons :) 16:00:48 kc5: F21 are still available. I have in fact given away more than I sold. All I have been asking for a dozen years was that people show a serious interest and prospect of actually doing something. I was told to sell them for a 10% profit but I doubted that anyone was willing to pay $5000 per chip. Those that I did not give away were sold for $999 or $800. 16:00:48 mm....dragons. 16:00:48 Neither do I. 16:00:51 But I collect them. 16:00:53 and a dragon seems appropriate...an untamed and unconstrained "wild thing" 16:01:00 i think chuck moores presence here makes my sitting in here for almost 2 years all by my lonesome a worthwhile time investment :) 16:01:06 not such a bad place to be :P 16:01:14 kc5tja: I didn't know you're a dragon fan....cool 16:01:30 thefox too of corse :) 16:01:40 I am a dragon whisperer. 16:01:44 I440r: Hehe :) 16:01:59 thefox: Oh, wow... :) Those are being sold for a lot more than I'd expected. But it makes sense, I suppose. 16:02:00 thefox i suggested after chipchucks last visit that we might also invite E.R. for an opposing viewpoint 16:02:00 I should follow your example and disconnect from IRC sometimes. 16:02:02 thefox: lol...now that's a novel twist 16:02:04 Damn additcive. 16:02:14 do you think she would come - would it be a worthwhile experiment ? 16:02:30 Dragon whisperer? 16:02:43 kc5tja: "Horse Whisperer" 16:03:00 TheBlueWizard: Still doesn't mean anything to me. 16:03:03 kc5tja: play on the idea of taming a dragon 16:03:18 ahh 16:03:22 I expected people to have software ready since I published complete simulators and development environments years earlier. Most people who wrote a little code got one for free. 16:03:26 Elizabeth seems a lot more connected with reality than Mr. Moore, if I may make a bold statement 16:03:35 of course one has to think two layer deep of humor to get it 16:03:54 she is - chuck moore is more connected withthe code 16:04:05 if it isnt perfect it isnnt finished. 16:04:20 david: the reality you speak of is kind of unfortunate, if I dare also make a bold statement 16:04:29 time to market gets a little delayed when you work that way 16:04:42 but the end product is alot better of corse 16:04:44 I440r: Yeah, tell me about it. 16:04:48 goshawk`: what - multi user systems? multiple processes? 16:05:00 chuck does R&D - ER sells what he invents 16:05:30 well, I was referring more to the politics of "business" 16:05:41 more than anything else 16:05:41 business is business 16:05:41 thefox - if they had them for free it might encourage them to use the real thing... 16:05:44 * mur is what finnish bears say remember ? ;) 16:05:44 so? 16:06:00 give away the development environment. sell the hardware 16:06:16 I440r: The development environment is freely available, IIRC... 16:06:22 well, I just firing out a quip...I think Mr. Moore is very connected to reality 16:06:28 * mur and good night , that is then :) 16:06:37 and greatly prefer his any day over that of Ms. [Mrs.?] Rather's 16:06:41 night mur! 16:06:45 just waiting one thing 16:06:45 bye mur 16:06:46 goshawk`: I agree 16:06:47 hehe 16:06:50 then i'll go 16:06:51 Natti mur 16:06:52 hyvaa yo 16:06:57 always the same thing :) 16:07:01 hm maybe we could design our own chips in fpgas or cplds, theres a lot of paper about that 16:07:01 i never leave when i say first time 16:07:03 can't type diacritics 16:07:06 then over and over again :) 16:07:24 david: Chuck likes to do the best job he can. He does more research than development but would like to sell a chip product, or actually he would like a customer to sell lots of the custom chips that he designs for them. ER is more into the reality of marketing and dealing with marketing scams-er-schemes that everyone else in the marketplace has to deal with. Chuck just does Forth as best he can. 16:07:37 TheBlueWizard that doesnt matter, in finnish if umlauts are not used they are written like a and o 16:08:04 mur: ah...thanks :) tho' it risks ambiguity 16:08:16 thefox: he seems to do an excellent job with some things. He strikes me as brilliant in some ways. In other ways, he doesn't seem very connected to today's computing reality... 16:08:22 thefox - do you think inviting E.R here would be worthwhile ? 16:08:25 thefox: The reason I was asking is because, if I get this BPSK environment running and a market for custom hardware opens up, I was considering the use of the F21 for its implementation. 16:08:33 and I realized I made a mistake...it should be hyvaa yota 16:08:47 i4: I am not sure it would be all that productive or informative 16:08:52 * TheBlueWizard must watch case agreement 16:09:00 thefox: But at $800 a chip, I don't foresee that happening, unfortunately. :( 16:09:11 tell me and i¨'ll go sleep 16:09:12 i'll tell it slowly then 16:09:21 15:24:11 Computers are so fast, anything should happen instantly 16:09:24 stuff like that 16:09:34 david: He's not, he is about ten years ahead of it. 16:09:36 oh, and my favorite: 16:09:37 TheBlueWizard correct :) 16:09:41 14:49:18 Portability is not important 16:09:42 14:49:24 Portability is not possible 16:09:42 14:49:42 Real applications are closely coupled to hardware 16:09:42 davidw: what's wrong with that? 16:09:53 davidw portability is a MYTH 16:09:55 i think the only thing really need to amke colorforth better is metacompilation: colorforth explained in colorforth, after the circle is completed you can start experimenting with colorforth a lot more 16:09:57 look at the linux kernel 16:09:58 I440r: bullshit 16:10:02 it isnt THE linux kernel 16:10:07 its the linux KERNELS 16:10:17 2985698236792 different versions of the kernel all interleaved into the same sources 16:10:23 Linux is a bad example, because it is necessarily tied to the hardware. 16:10:24 sorr. but thats NOT portable 16:10:31 portability is always possible if a certain level of abstraction is possible 16:10:39 A better example is something like a web browser, which need not be tied to the hardware. 16:10:39 so? on top of it, I can run orders of magnitude more code, portably. 16:10:42 and chuck dont like abstraction at all lol 16:10:52 neither do i 16:10:53 at least dont like this level of abstraction 16:11:00 kc5tja: and thus is not a real application, according to chuck's quote 16:11:22 maybe I'm taking his quote the wrong way or something 16:11:27 thefox: what do you think? 16:11:36 I440r: Hmm... if I were to port some IsForth code with 32-bit integers to a 16-bit environment, what should I think about? 16:11:51 kc5tja: web browser is perhaps a bad example if it is intended to run on multiple platforms like Win, Mac, Linux...they all require different API and stuff 16:12:08 TheBlueWizard: it's difficult, but it can be done 16:12:14 davidw his point was in response to my question. which i already knew the answer to. my question was related to the cost of starting over with your code for different architectures 16:12:15 The first 25x chips would cost about $2000 each. Anyone really seriously interested I would think would be interested in getting in the ground floor and being one of the experts when they start making millions of them. But everyone has been waiting for a decade for someone else to make the first million. I had five sources for the money to do a 20,000 unit production run and sell chips for $6 but all five sources fell through and for several yea 16:12:26 the overhead of reqriting an app is almost ZERO with forth 16:12:31 development time being minimal 16:12:31 TheBlueWizard: Not true. Sockets code is all the same on all three platforms. Only the user interface code is non-portable, and that's only because we don't have an "ANSI Standard GUI" yet. Everything else, however, is. 16:12:35 so who cares if its portable 16:13:11 thefox: How much does each chip cost, if you don't count one-time investments? 16:13:17 I440r: I think you are full of shit:-) No language lets you rewrite an app in no time 16:13:18 (cost to produce, that is= 16:13:33 I440r: read the "Mythical Man Month" or whatever it's called 16:13:35 kc5tja: um...I heard socket codes differ a bit on each platform...never seen the code tho'.... 16:13:35 argh, i can never go to sleep, he tells it so slowly :D 16:13:44 mur: :) 16:13:46 I440r: The problem is, and I can use my block editor as proof of this, is that after you "port" your application to the new Forth environment, you now have two separate code bases to maintain. It's not the same code base anymore. 16:13:50 davidw development time in forth is minimal was the point 16:14:07 thefox: market them to bioinformatics people, that's a current fad;-) 16:14:18 TheBlueWizard: For something relatively low-level like diagnostics, sure. But not for a web browser. Something as trivial as a TCP/IP connection to port 80 is easy as cake and extremely portable. 16:14:27 kc5 so lets have 500 different code bases all interleaved into the same source files just so we can have ONE code base ? 16:14:28 crow: Chuck doesn't like unnessecary abstration. The top of the Forth is suppose to be abstracted, what distinguishes Chuck's Forth IMO is that there isn't much difference between the bottom and the top, they look pretty much the same and there isn't much distance between them. 16:14:56 I440r: If I change one fuction definition in the Linux kernel, it affects all those kernels equally. The change is said to be portable. 16:15:05 robert: which chip? when? to whom? how many? 16:15:08 I440r: Sure, you have some machine-specific issues, but it doesn't dominate the source. 16:15:27 However, in my block editor, if I change my source for Pygmy, I have to make a similar, but "different" change to make it work in GForth. 16:15:30 kc5tja: for simple web browser, yeah...but modern web browser is expected to handle different protocols like https, ftp, whatever, so I suspect my assertion stands 16:15:48 TheBlueWizard that bottom and top look the same gave some some great ideas, ispiration, though i dont know waht it is yet :) 16:15:49 thefox: In very large quantities, and I meant your new varpourware chip :) 16:15:53 the linux kernel isn't even that great an example, because they do have development 'issues' 16:16:02 TheBlueWizard: It fails, and fails miserably. FTP, HTTP, etc. is implemented in the application, and NOT in sockets. Sockets is responsible for establishing the connection, and then stops there. 16:16:03 look at netbsd if you want super portable 16:16:05 I just wondered how much a chip like that costs to produce. 16:16:16 david: I mention that Enumera planned to do things like that and thus 25x evolved. The plan is a bit different now. 16:16:25 anyway enough forth fuss for today im gonna play to State of Emergency on PS2!!!!!!!!! 16:16:25 for isforth to work in linux AND freebsd the only difference would be the syscal interface 16:16:30 mur: what are you talking about? 16:16:33 its even betetr than GTA 16:16:40 see ya everyone 16:16:46 and the fact that syscalls in linux have different numbers than the same syscall in freebsd 16:16:49 bye CrowKiller 16:16:50 TheBlueWizard the bottom and top look really the same and such.. 16:16:51 btw, I just saw the ColorForth IDE driver... cute. 16:16:55 I440r: Thus, isforth is "portable" between Linux and BSD. 16:16:55 and linux has some syscalls freebsd doesnt and vice versa 16:17:00 seesja CrowKiller 16:17:01 --- quit: CrowKiller ("User pushed the X - because it's Xtra, baby") 16:17:01 kc5 not 16:17:01 kc5tja: ok...I buy your claim then :) 16:17:10 I440r: You just proved it yourself!! 16:17:18 * davidw doesn't care about isforth, because he uses PPC hardware 16:17:24 for there to be a freebsd version of isforth i would need a syscall nuymber include file 16:17:32 every single syscall would need to be defined from the start 16:17:34 I440r: The only difference between isforth is the syscall interface. That's it. Do you have to rewrite the whole of isforth just to port it to BSD? I don't think so. 16:17:35 i dont WANT that 16:17:43 the only syscalls isforth defines are the ones it uses 16:17:52 I440r: So what? 16:17:53 and the syscall numbers are HARD CODED 16:17:57 I440r: That's a technicality issue. 16:18:08 I440r: at compile time, though, I imagine 16:18:09 the extensions define new syscalls 16:18:11 as they need them 16:18:16 so that you can just recompile on different systems 16:18:27 the extensions would also need to be rewritten for freebsd 16:18:34 due to the syscall numbers being different 16:18:39 davidw: if someone write a translation of IsForth for PPC, then yeah you'd get it....and that is a good example of nonportability of IsForth (as long as it's written in assembly code) 16:18:39 i.e. its NOT portable 16:18:49 i will need a linux isforth and a freebsd isforth 16:18:55 robert: which chip is that? I guess I own some rights, not clearly specified yet, in 25x and Chuck did jokingly call it vaporware because it has only been tested in CAD and not fabed. But I programmed such simulated chips for years and I don't consider it vaporware if you can put down some cash and make them. 16:19:09 I440r: But the sources between the two are more than 75% equivalent, right? 16:19:18 quite alot more 16:19:22 Then it's portable! 16:19:30 its mostly portable 16:19:30 so the conclusion: there are different kinds of portability....we've learned something! :) 16:19:38 thefox: Yeah, you said that... but anyway, the 25x one was what I meant. 16:19:39 Just because "some" modifications are needed doesn't make it so utterly non-portable that it can't be ported. 16:19:43 but im not going to be inserting a gazilluoh #ifdefined linux 16:19:45 do this 16:19:48 thefox you shoudl write more i get good ideas all the tiem :) 16:19:53 #eles 16:19:55 do that 16:19:56 #then 16:19:59 FUCK that 16:20:26 tie that source code up into one HUGE gordian knot 16:20:31 I440r: That's not the only way to achieve portability, you know. 16:20:47 You have an awful lot to learn in the field of software engineering. :) 16:20:53 kc5tja: here here 16:21:06 a close mind is a wasted mind 16:21:08 kc5 agreed. ive only been doing it 20 years. i have NO formal qualifications 16:21:09 closed 16:21:11 I440r: Is for/next used in IsForth? 16:21:16 no 16:21:25 i disagree with chuck on the NEXT thing 16:21:33 NEXT has special meaning in forth 16:21:37 its part of the threading 16:21:39 I've been doing it for only 20 years, and also have no formal qualifications, But I do have experience. 16:21:48 for/next will be implemented when we have a better name for next 16:21:51 Okay... how is it usually defined, as immediate words or using the return stack? 16:21:58 It costs about $14,000 to have 25 25x chips fabed. Chuck estimated however $100,000 grand for prototyping 16:22:12 thefox twice! 16:22:19 he estimated two prototype runs 16:22:22 thefox: Is most of that cost in packaging and pinning the chips? 16:22:22 * davidw has being doing it for a lot less time, but enough to see that people who get fixated with one particular idea often get pretty sidetracked 16:22:32 Hmm... didn't you say something about $6/chip is you made 20,000? 16:23:05 210952Z $100,000. $5 per chip 16:23:08 So he is saying the first 25x would cost $4000, then $5 each for 20,000 more, $1 each for a half million more etc. 16:23:23 Nice. 16:23:31 rob_ert: That's how quantity discounts work. :) 16:23:35 Hehe 16:23:45 Well, just make someone buy 500,000 of them! 16:23:47 ;) 16:23:51 I was talking about F21 in 1998 at that point. Chuck was talking .18u 2002 costs. 16:24:05 This is Kinson Wong from Asia Magic (HK) Ltd company. 16:24:05 We got your company information from internet. 16:24:12 haha, i'm company nowdays :) 16:24:19 well many i have been called but not company :) 16:24:59 each run of F21 in .8u was about $10k another $600 for packaging. Today I can get wafers for the cost 16:25:25 thefox maybe you should get E.R. interestd in the forth chips :) 16:25:55 of prototype chips in 1998. $2 per chip with a minimum of 24 wafers instead of 25 chips. Packaging in Taiwan would add more. 16:26:06 * mur wonders why .8u sounds and looks intriquing 16:26:10 thefox: what about cells phones. That seems like a high volume market to target. I don't know what is involved techincally though. 16:26:56 .8u was available in 1994. It is almost free now in the third world. It can still deliver mips for pennies. 16:27:38 * kc5tja would love to design his own chips, but alas, it's way, way too expensive for a hobbiest. 16:27:48 I440r: long story. 16:27:55 kc5tja maybe not in the future 16:28:30 mslicker: iTV had investors Sanyo and L.G. Electronics and they wanted chips for cell phones etc. 16:28:35 i hope that chuck can be sucessful with his chips someday 16:29:31 what happened to that? 16:29:46 * goshawk` is lost for words 16:30:09 --- join: onetom (tom@adsl52009.vnet.hu) joined #forth 16:30:29 onnetommonom 16:30:40 thefox i sense that chuck and E.R are at odds with one another - her being more interested in market issues, him being more interested in R&D 16:30:44 hahhh! the net came back :) or ;((( ? 16:30:58 Because of the tile scaling in OKAD F21 could be manufactured in any size from 1.2 to .07u. This is also true for 25x or whateverX. State of the art today is about 5000 mips per microscopic piece of silicon. 16:31:03 i get the feeling he things she has bastardized his creations for her own profit? 16:31:08 ive missed chuck ;( ;(( 16:31:27 onetom: :( 16:31:33 onetom ya 16:31:39 rob_ert: If Forth had a weakness, I'd correct it 16:31:46 Favourite quote for today ;) 16:31:52 forth does have a weakness 16:31:56 its not accepted 16:32:05 He said that, too, yeah. 16:32:11 What to do about it, I440r? 16:32:15 thefox: if F21 was fabbed again today, how much would it cost and how much faster would it be? 16:32:16 thefox: while i was forced 2 b offline i was reading http://www.ultratechnology.com/forth.htm. there r some typos in it. 16:32:17 Revolution? ;) 16:32:19 The Asian economy collapsed in 98 and all their engineers got laid off or had to give up all retirement accounts and go back to work at home for half the money. Those were some of the iTV investors. 16:32:26 also - i think the non seperation of compile time creating words and those words they create is a problem 16:32:28 thefox: could u correct them? 16:32:31 I440r: maybe you should think about that, possibly considering the idea that the rest of the world is not completely full of morons 16:32:57 davidw when i see evidence of that..... :) 16:33:01 thefox: or u dont care? im thinking on translating them 2 hungarian. what do think? 16:33:10 The economy thing again 16:33:12 The prices are better and state of the art is about 20x what we were prototying in 98. 16:33:13 I440r: you say the same things as the scheme crowd 16:33:20 blah blah blah, scheme is the greatest blah blah blah 16:33:23 almost smells kinda fishy 16:33:34 davidw but i say it partially in jest :P 16:33:47 thefox: at this point, would you push for another shot at F21, or would prefer seeing 25x coming to life? 16:34:06 (and I don't just mean for Mr. Moore's sake) 16:34:10 onetom: what is the question? 16:34:20 I440r: forth has a lot of neat things going for it, but I think keeping an open mind and trying not to tilt at windmills is the best way 16:34:32 By the way, where can I find information about those forth chips? 16:34:36 I440r: pick something you can do hands down better in forth, and put all your weight behind it 16:34:51 davidw you name it. forth does it better 16:34:55 thefox mentioned he'd written software for development on them, too. 16:35:00 that is my honest opinion 16:35:22 gos: both. 16:35:22 i have worked in c, forth, assembler and a few other languages 16:35:22 thefox: onetom wanted to know if he could translate some of your articles at UT to Hungarian 16:35:34 none of them meet forths felxability 16:35:39 assembler is closest 16:35:39 thefox: 1: shall i point the typos or u dont care? 2: do u allow me translating ur articles 2 hungarian? 16:35:42 I440r: but it's not true, from a user point of view. Linux can do more than some vaporware forth os. mozilla beats a non-existant forth browser. does forth have an editor that competes with emacs? 16:35:42 thefox: and also append some corrections to you for the original articles 16:35:45 and is even better for some things 16:35:58 davidw forthos will always be vaproware 16:36:09 thefox: I meant if you had money only to do one at this time? 16:36:10 thefox: 3: do u think is it worth 2 translate it 2 other langs? 16:36:17 I440r: I guess that makes C better for OS's? 16:36:19 emacs is another bloated pile of (expletive deleted) 16:36:27 davidw no 16:36:32 onetom: certainly feel free, but I can't be of too much help. I know people doing translations sometimes have difficulting getting the meaning of the idioms. But feel free. I'll post cross links. 16:36:34 davidw: I implemented a VI-based block editor a while ago. 16:36:38 well, I prefer OS's that exist to fantasy OS's 16:36:47 there arent enough people coding forth to make a real forth os viable 16:37:00 I440r: I think one bright person can make a difference 16:37:03 davidw: In Forth. It's not as powerful as VI, because I don't need all of VI's abilities, but extending the editor is easy. You can even extend the editor after it's been loaded/compiled. 16:37:09 the computer world is full of them 16:37:13 davidw im hoping so 16:37:14 chuck moore is a good example 16:37:15 There are some German and Japanese translations of some of my pages that I know of. 16:37:30 kc5tja: sounds like a start 16:37:43 * davidw can't stand vi's key bindings, but I guess that's besides the point 16:37:53 davidw: I wrote it because I was pretty ticked off with Pygmy's block editor. :) 16:37:59 chuck moore has made a beautyful language but has had very small impact on the real world - he never got forth to be as universally accepted as C is 16:38:10 and forht is orders of magnitude better than c 16:38:11 It can accomodate Emacs-like key bindings too; in insert mode, it'll handle CTRL-keys the way it handles VI-like command keys. 16:38:13 in ALL cases 16:38:15 1440r IMHO forthos is a required component of Forth. 16:38:21 In fact, that's how it handles the ESC key to get back into command-mode. 16:38:29 I440r: so, if you base a business on forth, you will do orders of magnitude better. Prove it. 16:38:39 I440r: Historically the accepted languages had large corporations backing them. 16:38:43 there are clients that don't care what language you use 16:38:50 thefox a full blown operating system in forth that supports all the hardware linux does etc.... THAT i would love to see 16:38:57 thefox: do u have some spare capacity for answering some questions related 2 the meaning of ur documents (translation related issues r not a problem 4 me. i did lotsa translation) 16:39:02 when that is developed ill be doing rm -rf / 16:39:06 and installing it 16:39:12 DAY ONE of its release 16:39:41 mslicker because they could market them 16:39:42 I440r I didn't see your happy face, but I assume that you were joking. Talk to Jim Schneider if you were not. 16:40:06 onetom: sure 16:40:08 I440r: you can market anything 16:40:12 anyway, I am going bike riding tommorow, and I need to sleep 16:40:19 mslicker lets go back to 1980 16:40:23 Night davidw. 16:40:25 you by microsoft c version 1.0 16:40:34 ill buy microsoft forth version 1.0 16:40:57 thefox: advised email address 4 further possible contact? 16:40:59 if you upgrade at every upgrade frokm M$ you will have over 50 thousand dollars invested in that c compiler 16:41:04 ill have my original investment 16:41:16 because ill have fixed all the bugs ive found and added anything new i wanted 16:41:21 and removed any crap i didnt 16:41:33 onetom: fox@ultratechnology.com 16:41:39 Which necessarily means that there is no commercial incentive to use Forth. 16:41:41 * onetom sends deep sleep onto davidw 16:41:50 microsoft isnt interested in marketing somethign they cant make after-market profits on 16:41:54 The government and the corporations back what they back, and it isn't Forth 16:41:55 I440r: what microsoft does is entirely different thing. 16:41:57 thefox: thx 16:42:02 It's much too malleable. 16:42:11 mslicker it was an example but lets go further 16:42:16 I440r: I am sorry that you didn't make it to my last presentation to FIG. It was on the latest GUI. 16:42:16 when was microsoft founded ? 16:42:33 :) 16:42:45 I440r: 1975, if memory serves me correctly. 16:42:47 ive never been to FIG - im not even a member 16:42:53 and when was forth inc founded 16:42:57 sometime in the 80's 16:43:03 who has a larger market share ? 16:43:09 i4: 70's, actually 16:43:15 73 I think 16:43:15 forth inc is a very small company 16:43:24 I440r: I agree the goal of propietary software is to control you. 16:43:38 Microsoft wasn't using C back then, though. They were perveyors of BASIC interpretters. That alone, however, gave them a huge market appeal, because of all the up-coming 8-bit home computers of the day. 16:43:59 i would put E.R. up there with the TOP marketing people. and bill gates is the best marketing man ever 16:44:15 microsoft doesnt produce ANYTHING i would call marketable 16:44:19 not a FSCKING thing 16:44:20 Actually, no; he's the best legal tactician I know. 16:44:21 1440r: You can't be a member of FIG at the moment. It is virtually dead and they have not accepted membership dues for over two years now. I do attend SVFIG meetings, chapters are not legally part of FIG. I should have said to an SVFIG meeting. 16:44:31 I don't think MS got where it was without a lot of people around them being quite stupid all at the same time 16:44:46 so, I don't give them *that* much credit 16:45:14 thefox ok. - is this somethign that happens IRL or online ? 16:45:22 What about the PC vs 68k baised computers 16:45:25 Microsoft got so big for one reason only: illegal tactics. Requiring computer manufacturers to bundle MS-DOS, to pay extra royalties for those who don't use MS-DOS, and other illegal crap like that. 16:45:30 why did people choose the PC instead of the 68K? 16:45:47 Something to do with support of the government and the megacops I think 16:45:47 BadCog2: PC was originally going to be 68K based, actually, but because Motorola didn't have a chip delivered on time, Intel won out. 16:46:02 motorola produced microprocessors that are far superior to intels. intel produces the best microcontrolers ther are tho 16:46:10 except they are all subject to glitches of corse... 16:46:43 i dont like motorola controllers - i love the 8051 tho :) 16:46:48 I440r I prefer Chuck's microcontroller designs to Intels. 16:46:50 IBM started out building counting machines for the US census 16:46:51 I440r: Not quite. Motorola produces the best *everything* I know; but they can't deliver on time to save their lives. Even today, they have major shipment problems. Hifn, the company I work for, used to use Motorola as a fab. We don't anymore. There's a reason for that. 16:46:54 i didnt even like the f68hc11 :P 16:47:08 And Intel's microcontrollers SSUUCCKK!! 16:47:17 --- join: AlephNull (~blandest@h24-65-137-230.ed.shawcable.net) joined #forth 16:47:19 I mean, 12 clock cycles per 1 machine instruction cycle?!?! What the hell is that?! 16:47:30 kc5 enough 16:47:41 Greetings. 16:47:44 its not how good you are, its your position in the empire 16:47:48 The 68xx series of microcontrollers are based on their 6809 CPUs. 16:47:59 if you have position in the empire, you can afford to be sloppy 16:47:59 why go more when its not realy needed. a simple 8051 with 4k of rom and a 12 mhz clock can do anything 16:48:11 Hello AlephNull. 16:48:14 if it cant then they need better coders :P 16:48:14 I440r: 12MHz clock for 1MIPS performance. Real smart. 16:48:26 I'd rather go with a 6502-based system, and get 6MIPS performance for the same clock speed. 16:48:44 6502 is the best microprocessor EVER designed 16:48:48 i stated that earlier :) 16:48:54 It's a derivative of the 6809 though. 16:49:06 6811 and 6809 both are based on the 6800 architecture. 16:49:12 it even has a nice pipelined instruction queue - somethign the 8086/8088 never had 16:49:29 I'd rather go F21 or 25x than 8051, 68hc11, 6809, 6502, 8080-x86. Been there, done that. 16:49:32 in fact i believe it wastnt until the 386 that intel started pipelining the instruction queue ? 16:49:35 Actually, it doesn't -- that's one of the reasons why it was so much faster. 16:49:48 thefox you win :) 16:49:53 thefox: But not for $2000 a chip... :) 16:50:03 lol 16:50:13 brb i gotta make a pot of coffee 16:50:22 I440r: They had the instruction queue starting with the 8086, but it wasn't until the 80486 that pipelining came into effect. 16:50:34 if the F21 or 25x had any kind of serious production, it wouldn't cost that much 16:50:45 But it doesn't right now. 16:50:58 I have to look at things pragmatically, not in terms of what "might" have happened. 16:51:07 I was really hoping for the F21's success. 16:51:15 There is a *lot* you can do with it. 16:51:20 Well I remind people that the first Pentium runs always require many millions per chip. Even iapx432 had billions of dollars in development costs and I don't think they sold more than a dozen or so. 16:51:43 --- join: qless (~cerberus@mani.kobayashimaru.org) joined #forth 16:51:58 Hmm..not sure it was that low in qty, but I do know it was much, much less than 5000. :) 16:52:11 Hi qless :) 16:52:22 heya rob_ert :) 16:53:01 Still, the Itanium looks like it's another 432 effort -- really innovative chip, has bright future, but alas, nobody's using it. 16:53:03 --- join: onetom_ (tom@adsl52005.vnet.hu) joined #forth 16:54:30 I gotta go, thank you to the organizers of this event. 16:54:43 kc5 nobody wants to buy new shit thats goingot be obsolete on 2 years is why 16:54:44 take care, mslicker 16:54:45 =) 16:54:47 --- quit: mslicker ("[x]chat") 16:54:50 (irc servers reacheability interferes w my crappy net connection resulting a hardly usable env :( 16:55:03 --- join: gforth0 (guest@adsl52005.vnet.hu) joined #forth 16:55:08 who cares about 23487562893769432 GHZ when the code that runs on that beast is going to NEED it just to get by 16:55:21 I am not sure why so many people who have to write their own software stagnate with their PC's 16:55:25 :^) 16:55:32 But, they pay you to just learn their crap 16:55:34 just to "stay" compatible 16:55:38 even if it takes years 16:55:41 (i beg ur pardon 4 the lots of join/part msg) 16:55:42 I440r: The Itanium isn't that bad off. It's performance is comparable to a Pentium III at the very least, IV at best. 16:55:54 clock for clock, that is. 16:55:58 even for software that will never see the light of day, somehow "compatibility" seems to be figured into all of this 16:57:01 See, I'd liek to make a CPU that, like the Transmeta, is designed to emulate something like x86, but it is implemented as a stack architecture, and not as VLIW. 16:57:14 I love VLIW, but I just don't see it going anywhere significant. 16:57:41 Nuon seems to have crapped out 16:57:43 The Steamer16 home-brew CPU, for example, gets 80386-level performance out of a chip that has *less* gates in it than the Intel 4004 4-bit CPU. 16:57:58 they ran out of funds b4 they could hit critical mass 16:58:09 Nuon is what, the successor to Transmeta? 16:58:31 Its an imbeded solution for DVD players and Settop boxes 16:58:32 --- quit: davidw (No route to host) 16:58:39 It also doubles as a game machine 16:58:46 company VM Labs 16:58:56 There is even a fan group with a website 16:59:01 mostly old Atari Fans 16:59:10 OK, but I'm not sure exactly what differentiates Nuon from other technologies. 16:59:13 What makes it unique?> 16:59:17 You can actually go buy a DVD player with this chip in it 16:59:28 and download a development kit to play around with 16:59:49 VLIW 16:59:54 cheap 17:00:07 Player Samsung 501 17:00:12 Cost $200 17:00:17 Store Best Buy 17:00:19 This meeting today competed with the SVFIG May Meeting. Someone mentioned that when Chuck said the Min told him that the 18th would be a good day. 17:01:03 But this group was also much larger than the SVFIG meeting if it was like the last few. 17:01:04 Doh... :( Did he miss the meeting? 17:01:20 I have not DL the devkit and bought the player for lack of time/energy/whatnot 17:01:48 only 4 games released for it 17:02:16 BadCog2: Ahh. I am still in full support of stack architectures though. :) They're just as easy to implement (easier, actually -- less transistors), and get comparable performances if you have a well chosen instruction set. 17:02:37 I would have prefered MISC for the same purpose 17:02:47 What did people think of this chat session with Chuck vs the first one? Better? Worse? Just different? 17:02:58 i was not on the 1st one 17:03:41 Slightly worse. 17:03:44 But not much. 17:03:45 --- quit: gforth (Read error: 110 (Connection timed out)) 17:03:45 --- nick: gforth0 -> gforth 17:03:47 --- quit: onetom (Read error: 110 (Connection timed out)) 17:03:48 --- nick: onetom_ -> onetom 17:03:51 Largely, I was expecting more questions from Chuck. 17:03:58 The first one was moderated questions to Chuck, answers, not so many people making comments. I have a transcript at my site. 17:04:29 I wonder if he was slightly disappointed to hear that not many people are coding in Forth for profit. 17:04:45 thefox: my problem was the PR for this particular session 17:04:50 it didn't even come close to the first one 17:05:00 I think most people totally forgot that this one was even scheduled 17:05:07 He knows that a lot of people do, he just wanted to know who was who to better answer questions. 17:05:24 so, most people here, in a sense, were sort of not prepared for it because they were not even sure if it come off 17:05:31 * kc5tja skipped an aikido session to be here... :) 17:05:42 :) 17:05:49 * mur skipped sleeping to be here :) 17:05:53 Hehe 17:05:57 And apparently, still are. 17:05:58 :) 17:06:01 mur: Heh. 17:06:04 * I440r skipped working on gun-store to be here 17:06:05 lol 17:06:09 :) 17:06:25 * rob_ert lives here anyway, so he didn't skip anything :-/ 17:06:31 I also wanted to hear more questions from Mr. Moore 17:06:42 questions like what? 17:06:46 it started out that way, but that portion kind of peetered out 17:06:55 well, whatever he wanted to ask 17:07:14 perhaps that is a tall order 17:07:27 Well, I think that kind of necessitated us doing stuff for profit (e.g., so as to be more interesting to him). 17:07:27 but i prolly have to go do some work in a min... my father woke up heh 17:07:27 but, I sort of got the impression from the last session that he had things he wanted to know 17:07:32 and I wasn't sure if he got to that or not 17:07:43 thefox - do these fig meetings happen irl or online (if you answerd i missed it) 17:08:06 --- part: qless left #forth 17:08:09 I think people offered enough ideas that Chuck didn't have to fish for topics. 17:08:23 thefox: alright, then nm what I am saying then =) 17:08:35 if irl and i have the travelling money to make one ill come to one 17:08:38 I think I need to blow 17:08:50 Good night, folks. May the forth be with you tomorrow, too. 17:08:56 BadCog2: be regular in here - we need more forthers in here :) 17:09:05 svfig irl usually on the 4th saturday at Cogswell College in Mountain View they have a svfig page at forth.org 17:09:07 eyes getting tierd, need to go for a walk and breath some fresh air 17:09:17 argh 17:09:19 i shoudl go sleep 17:09:20 thefox: what about the FIG UK meetings? tell us about those 17:09:22 gn now 17:09:25 1440r I might, do you have meetings every Sat? 17:09:26 Natti mur 17:09:27 (if you don't mind) 17:09:28 --- quit: rob_ert ("Nothing is real.") 17:09:47 mur: good night! 17:10:13 BadCog2: I think that depends on what Futhin manages to schedule. 17:10:14 I am leaving, bye 17:10:19 hmm 17:10:26 I can't promise anything at all 17:10:29 just maybe 17:10:38 Long Live Forth 17:10:40 --- quit: BadCog2 () 17:10:41 #FIGUK is a chat room like this one. They meet the first Sat of each month. 17:10:57 what irc servers do they use? 17:11:02 gn 17:11:05 --- quit: mur ("MURR! end of file reached. continuing filling logs some other time.") 17:12:01 aha maybe #figuk and #forth should merge 17:13:27 Yes. For the first meeting I was in #FIGUK and invited them all to come over to #forth but we all had to use a different server to get here. These are the only two chat rooms I have visited. 17:14:00 brb 17:14:14 thefox: do you deem the crowds on both channels to be compatible? 17:16:03 this channel tends to BS at times 17:16:07 Yes. 17:16:15 thefox: I only ask 17:16:39 because I was curious about whether such a suggestion should be seriously proposed 17:17:01 back 17:18:25 they talk about figuk articles and projects and their own projects. People ask and answer questions. Not much arguing or BS. It was nice to know that we could have an unmoderated chat the way we did here. 17:18:33 i encourage light hearted banter in here - strict channel moderation is anal 17:18:37 Right now, my first goal is to get FS/Forth System 1 at least to the point where I can use it in place of PygmyForth for all of my coding needs. 17:19:04 when we talk forth we talk forth - when we dont,... we dont :) 17:19:06 thefox: neat 17:19:38 thefox i think the original meeting could have been unmoderated maybe 17:19:51 but i didnt want to overload chuck with love and admiration :) 17:20:04 I like the original meeting's format. Moderated was a good thing. 17:20:12 no matter how much its deserved :) 17:20:13 There were a lot of questions to be asked. 17:20:25 I tried the kill button on a pull down menu in my usenet reader but all it did was filter out someone's messages! ;-) 17:20:55 thefox: too bad "cancel" or "supersede" doesn't work =) 17:21:02 thefox is computer illiterate :) heh 17:21:39 thefox: is globali.us.ircnet.net one of the same IRCnet servers that #FIGUK use? 17:21:56 there are several IRCnet's around, so I am hoping to clarify 17:21:58 irc.stealth.net 17:22:06 i4: ok...the same 17:22:26 gos: I could look but it might disconnect me from this server if I to where I login/out of a server. email me. 17:23:13 I think i4 might have answered my question, but thanks 17:23:24 --- join: Fare (fare@samaris.tunes.org) joined #forth 17:23:47 im stealth.net is ircnet, ircnet #coders is where all the REAL coders hang :) 17:23:49 not undernet or dal :P 17:24:09 hi fare 17:24:14 --- mode: I440r set +o Fare 17:24:33 --- mode: I440r set +v clog 17:24:35 heh 17:24:39 --- mode: I440r set -o I440r 17:24:39 gos: I think that would work as a server. It looks like one I was using. email me to confirm. 17:25:02 irc chat in colorforth would be neat. ;-) 17:25:55 Yes. The ability to say : tf thefox ; would be quite handy for talking to a specific person. :) 17:26:21 a good place for FML 17:26:29 thefox: do you know the status of UDP/IP/Ethernet 17:26:39 for colorforth 17:27:05 Greetings, master fox. 17:27:07 thefox: done. thanks 17:27:19 thefox: I really think so, too 17:27:29 Chuck is starting with PPP to his ISP so that he can connect under colorForth, I am not sure where he will go with it next. 17:27:51 Thank you. 17:28:08 thefox: so is there any cross-fertilization between your Aikido activity and your programming activity? 17:28:25 * I440r should get back into wing chun 17:28:45 wing chun? 17:28:54 wing chun is the forth of the martial arts works 17:28:58 very minimalist 17:29:05 you learn ever basic move in the first form 17:29:11 can learn the entire system within a year 17:29:17 its a kung fu style 17:29:26 interesting 17:29:29 but no wavy arms all over the place 17:29:44 its where bruce lee started but he never learned it all 17:29:48 Certainly. One of the books that I intend to write will be called 'My Masters" Chuck will in there as one of the characters. Most of the principles are the same. Mindfullness is mindfullnes, efficiency is efficiency, truth is truth. 17:30:01 offered a million to yip man to teach him the mook jong. yip man refused 17:30:16 thefox: I'd be delighted to buy and read this book 17:30:29 does Chuck Moore practice any martial art? 17:30:34 Forth =) 17:30:35 forth :) 17:30:41 goshawk`: :) 17:30:47 i440r =) 17:31:03 I will say that as someone with decades of experience practicing and teaching a number of martial arts, and decades of practice of Forth with its creator for the last decade that I am often reminded of a quote by Daniel Dennet. 17:31:09 * kc5tja still prefers aikido. 17:31:25 Forth is definitely an art, but I wouldn't call it martial. 17:31:32 There is nothing that I hate worse than a bad argument for something that I hold dear. DD. 17:31:54 that one belongs in /usr/games/fortune 17:32:10 akido is an art i would be interested in learning 17:32:20 Fare: The underlying tenets are the same though. 17:32:24 Reminds me of a Bastiat quote: The worst thing that can happen to a good cause is, not to be skillfully attacked, but to be ineptly defended. 17:32:26 But martial arts discussions on the internet are much like discussions of programming languages, everyone seems to consider themselves a master. 17:32:52 thefox: are you familiar with Stanislav Grof? 17:32:54 lol ive been doing wing chun for 3 months (havent done any in about 9) 17:33:00 yea---- im a master :) 17:33:10 I certainly do not consider myself a master in either programming or martial arts. 17:33:30 Man, I've been practicing aikido for 2.5 years now, and I'm only now getting one principle of iriminage... :) 17:33:48 In the martial arts it is somewhat dangerous to spout off to people who have decades more experience than one because there always is a bottom line. But on the internet everyone has a right to an opion and a right to express it. 17:34:26 thefox bravado here isnt likely to do anything other than generate ridicule - IRL you can get a bloody nose :) 17:34:30 I just prefer aikido because I like it. It makes me feel good when I practice it. I love flying 10ft through the air when being thrown, landing, and getting back up for more. 17:34:43 I440r: Worse -- death. 17:34:58 But still if you taught Wing Chun and Aikido and Karate and other arts for decades you you really get to laugh a lot at what some people say about their arts. ;-) 17:34:58 kc5 thats why i pack a 45 :) 17:35:28 thefox: I'm sure. I have always wanted to take Tai Chi as well, even now. Both of those arts fascinate me greatly. 17:35:52 thefox: what do you think of the pace at which MISC evolves? 17:35:59 I'm not talking about the spiritual/stretch/yoga-esque Tai Chi -- that's just one small part of the whole of Tai Chi. 17:35:59 thefox i did judo when i was 12. for about a year. i did tak kwan do for about the same lenght of time. i have about 3 months of wing chun under my belt - i would never call mysel an expert at it 17:36:21 MISC sounded very exciting 10 years ago. But nowadays, it seems to have made little progress. 17:36:27 but of the three martial arts ive done ive learned MORe in the 3 months of wing chun than with the year or so of the other 2 17:36:37 better teacher or better martial art. one or other... 17:37:06 I440r: There is no "one" martial art. I would highly suspect better teacher. Stick with him/her. It sounds like you'll learn a lot. 17:37:35 * kc5tja would also like to learn iaido and kendo at some point too -- I love swordsmanship. 17:37:43 kc5 he is 2 hours away and ive no gas money. i would be there for 12 hours a day for the next 10 years if i was rich :) 17:37:52 Ah 17:38:27 he isnt just a good teacher, he is a good man 17:38:27 I have to say that as teachers go, until I met a handful of people on the net, I've had a lot of real poor examples to call "teachers" at just about anything 17:38:33 I feel like I am playing catchup :P 17:39:57 I440r: what about marksmanship? Ain't that another martial art you learnt? 17:39:58 kc5tja: iriminage was traditionally known as the twenty-year technique. 17:40:47 fare marksmanship is another martial art 17:40:56 i would consider playing the guitar a martial art too 17:41:01 or the violin/piano 17:41:14 maybe they are inverse martial arts 17:41:16 why martial? Do you use guitar/violin/piano in a fight? 17:41:21 the worse you are the more damage they do :) 17:41:41 goshawk`: My sensei won't admit, but he's one of the best aikido instructors in the country. He's been a personal favorite for Toyoda sensei for many, many years, up to his death. He, Katz sensei, and James Nakayama sensei are Toyoda sensei's Western Division instructors group. 17:41:46 they do call the guitar an axe you know :) 17:41:57 I440r: I know what you mean. I can dodge a guitar strike, but sometimes the sound they sent out!!!!!! 17:42:12 thefox: Yeah, tell me about it!! :) I fear what some of the other techniques will be like. :D 17:42:19 thefox i can butcher paganini's capriee #24 for ya :) 17:42:41 i play a mean rendition of joe starianis midnight tho (from surfing with the alien) 17:43:25 goshawk`: here is my sensei in action at a seminar at Chushinkan dojo: http://members.cox.net/kenshinkan/Events/Chushin701/subSide.htm 17:43:30 thefox i always considerd the biggest part of learning a martial art was learning how to NOT NEED it :) 17:43:37 I440r: don't make me cry. 17:43:54 thefox lol :) 17:44:19 gotta get with that finger tapping technique! 17:44:29 bleh - not touched my guitar in over 2 years 17:44:39 kc5: that's neat 17:45:10 Well as Lao Tse said, when in harmony... 17:46:13 thefox i sense you becoming more at one with IRC - dont start playing MUD games - they will suck you in and you will never escape :) 17:46:36 irc is the great time waster. alot of work never gets done in here :) 17:46:52 goshawk`: Now you understand what I mean when I say, "When I go flying through the air..." ;D 17:47:11 flying isnt the problem - its the landings 17:47:23 Macbeth sensei is a very fluid, very dynamic, very circular teacher, though he also teaches linear techniques when they are needed. 17:47:48 I440r: Learning to land isn't the problem. 17:47:57 I can imagine. And on that note, in harmony I hope, I will say that I should get going now. Do you want me to schedule a time or just drop by the channel at random? 17:48:09 random is good 17:48:14 take care, Mr. Fox 17:48:15 tho the channel has idle times 17:48:21 thank you for joining us today 17:48:33 thefox: Thanks for stopping by. It's always a pleasure. 17:48:45 an honor to be exact 17:49:12 thanks to everyone, best wishes. 17:49:16 --- quit: thefox () 17:49:45 :) 17:49:55 nice guy 17:49:58 ok we can all start misbehaving again now :) 17:50:44 well, I need a bite to eat 17:50:49 several, actually 17:50:52 take care, everyone 17:50:56 l8er dood 17:50:57 thanks for the links, kc5 17:51:05 have a good one, i4 =) 17:51:29 * I440r boots vmware and loads the settlers 17:51:35 its the only way i can play it on this laptop 17:51:37 :P 17:52:42 Check this out. This is at our dojo's 5 year anniversery. The mat...um...shows some wear and tear it seems. :) http://members.cox.net/kenshinkan/Events/Kenshin996/index.htm 17:52:57 This is Toyoda sensei conducting the instruction, BTW, not my sensei. 17:57:29 Doggone it, they still don't have any of the videos up on the website yet... GRrr... 17:57:30 Oh well. 17:57:56 Anyway, I'm going to get some food. 17:58:11 --- nick: kc5tja -> kc-food 18:02:37 --- quit: Fare ("Connection reset by pear") 18:11:37 --- join: bpt (bpt@user-2ivf097.dialup.mindspring.com) joined #forth 18:11:43 re 18:15:30 --- quit: tathi ("leaving") 18:20:09 * TheBlueWizard returns...processed gigabytes of data...and made a tiny burp...thank you :) 18:25:33 --- nick: kc-food -> kc5tja 18:25:50 I'm about to process gigabytes of data...and about the belch rather loudly. ;D 18:26:15 hehe 18:26:55 * kc5tja finds it interesting that Chuck Moore is into UWB radio. 18:27:00 * kc5tja wonders why... 18:34:09 what is uwb ? 18:35:57 --- quit: onetom (Read error: 110 (Connection timed out)) 18:36:25 A highly controversial method of transmitting information by radio. 18:37:02 --- quit: gforth (Read error: 110 (Connection timed out)) 18:37:15 UWB works pretty much like spark gap radio back in the late 1800s. Each spark would create a *huge* bandwidth signal -- reaching up to the gigahertz of bandwidth. 18:37:37 UWB uses specially timed pulses to achieve the same effect. 18:37:53 ultra wide band ? 18:38:03 hmm...why is it controversial? 18:38:07 However, because of the high bandwidth of the signal, it could potentially interfere with HF and some VHF radio receivers. 18:38:19 ah 18:39:31 This is a *really* sore point for amateur radio operators, because our radios need at least 120 to 130dB of gain just to be able to *barely* hear a morse code signal from across the country, let alone across the world. 18:39:40 (at 100W of broadcast power, which is the average) 18:39:45 I440r: remember I told you about Jet Thomas, a fanatic Forther? Well, I think I found a new email of his (I haven't tried that one yet): j2thomas@swbell.net...tell him John Rogers refers you to him :) 18:40:24 yea i remember you talking about him :) 18:40:34 :) 18:40:36 wouldnt it be better if you emailed him and invited him in here ? 18:40:42 YOU know him afterall... 18:40:44 UWB will pollute our HF spectrum to the point of being useless for world-wide communications anymore. 18:40:56 hehe...I will test-email then 18:41:00 kc5tja: what about ultra low frequency... 18:41:09 didnt tesla work on ulf ? 18:41:21 Only for power transmission -- not information 18:41:31 ULF is useless to transmit data with. 18:41:54 isnt there ulf radios used by the military ? 18:42:20 i think tesla wanted to deliver electricity to the end user via ulf signals directly through the ground 18:42:23 yes -- they work at 0.1 bps too 18:42:24 FOR FREE! 18:42:47 i think chuck moore is much like tesla 18:43:08 tesla had more patents than any other single person 18:43:17 and we use alot of his inventions every day 18:43:26 for instance. tesla invented radio, NOT marconi 18:43:31 Tesla's power system would never have worked. 18:43:47 --- join: gforth (guest@adsl52005.vnet.hu) joined #forth 18:43:51 kc5 teslas power system is STILL IN USE today at niagra falls :) 18:44:00 niagra mohawk power company... 18:44:09 the through the ground one im not sure about 18:44:14 Besides being a huge threat to human life, it would have caused ground heating (because ground has electrical losses), and that would have been an ecological disaster. 18:44:40 No, his Tesla turbines may be still in use, but I can assure you, that they're still producing 3-phase AC through wires and not through ground radiation. 18:44:55 * kc5tja builds Tesla turbines out of paper, BTW -- sweet turbine design. 18:44:59 kc5 yes - thats what i was refering to heh 18:45:11 i want a tesla coil! 18:45:27 I was talking about his Wardencliff project. That would have killed a lot of people (accidentally, but it would have) 18:46:10 kc5 i have a good web page in myu bookmarks about tesla. it attributes the thing in siberia to tesla 18:46:12 tungustra ? 18:46:16 At any rate, the very fact that we have AC power instead of DC power in our wall sockets is all because of Tesla. 18:46:24 Tunguska is not from Tesla. 18:46:27 yes. dc would never have worked 18:46:38 tunguska. - thats how ua spell it 18:46:53 i dont believe it is but this site mentions it as a possibility :) 18:47:05 tesla also invented radar!! 18:47:08 Eye witnesses to the event pretty much disproves it. 18:47:36 yea i think i saw the same doccumentray on it:) 18:47:54 Besides, when Tunguska occured, Tesla was in America. :) 18:48:39 kc5 new york to be exact 18:49:23 the story is he was trying to communicate with the people at the north pole but a bird flew into his ray and instead of sending a signal to the north pole his ray hit siberia :P 18:49:26 and blew up :) heh 18:49:54 it makes a nice story 18:50:03 the web page does have other info on tesla etc 18:50:08 and is a very intersting read :) 19:01:24 email sent :) 19:03:51 --- quit: gforth (Read error: 110 (Connection timed out)) 19:07:18 cool :) 19:07:33 lets see if it gets some posative response :) 19:08:29 good night everyone :) 19:08:41 good night oink 19:11:11 bye oink 19:11:29 nite oink - you gotta learn forth dood :) 19:11:49 AlephNull: what does your nick mean ? 19:12:02 ive seen that joke forth AlephNull bottles of beer on the wall etc 19:12:07 ive no idea what it means tho heh 19:12:13 I don't know. 19:12:47 I usually use 'Blandest' 19:13:24 I took Aleph from a chemical naming by Alexander Shlugin, not the the mathematical term. 19:13:59 --- join: gforth (guest@adsl52005.vnet.hu) joined #forth 19:14:08 It's a program that prints the song "99 bottles of beer on the wall." The goal is to do it in as few lines of code as possible. 19:14:25 AlephNull: so I take it that you already know the math definition of aleph null then? 19:14:45 I do not. 19:15:06 oh? well! I do!! :))) 19:15:11 heh, care to share? :> 19:15:17 sure! 19:15:35 but first, need a bit of background 19:15:56 not me 19:16:02 I'm sure you know what a set is...like {4,6,2} 19:16:03 no idea what it is heh 19:16:34 Aleph Null is the set of all positive integers. 19:16:35 all RED cars are in a set :) 19:16:38 Correction: 19:16:45 the cardinality of the set of all positive integers. 19:16:58 (that is, the number of elements contained in the set) 19:17:06 infinity 19:17:12 It's the first-order infinity. 19:17:20 There is a whole set of Alephs. 19:17:31 Aleph 1 is an even *bigger* number than Aleph null. 19:17:34 the size of a set S (written |S|) is defined to be the number of elements in that set. So, |{3,8,7}| is 3 19:17:45 this is all JUNK math lol. the distance from zero to one is infinite :P 19:18:04 I440r: Well, to be exact, it's Aleph 1. Funny you should bring that up. :) 19:18:14 heh 19:18:36 now consider the size of the set of positive integers....Cantor defined it as aleph null 19:18:36 there are those who think maths should be theoretical ONLY. practical math has no place in the universe 19:18:57 i would disagree. the point of doing MATHS isnt to do maths. 19:19:02 * AlephNull thinks 19:19:16 (thanks) 19:20:13 break something down into the pure theoretical side and to me all you have done is broken it :P 19:20:22 actually it is more formally defined in terms of power set 19:20:43 its all those practicall mathamtics gurus that i like :) 19:20:58 tbw i still want you to teach me maths however :) 19:21:07 i struggle at 1+1 still :) 19:21:14 r=1/r+1 19:21:18 lol 19:21:27 that sorta shit throws me 19:21:48 r is equal to one more than its own reciprocal - ugh 19:21:49 * TheBlueWizard slaps I440r with a diffy eq tome 19:21:49 --- join: njd (junk@njd.paradise.net.nz) joined #forth 19:21:54 heh 19:22:11 I440r: The practical mathematics gurus still have a solid foundation in the theory though. 19:22:57 yes 19:23:07 the theory is an aid to the practical 19:23:10 I440r: one of the solution to that equation is a golden number (phi)...a lovely number 19:23:16 yes i know 19:23:22 1.61803499..... 19:23:31 Well, if you multiply both sides by (r+1), you get r^2 + r = 1, which is a normal, quadratic equation. It's relatively easy to solve for r by completing the square, factoring, and voila. You can determine for what value r satisfies the equation. :) 19:24:08 kc5 i do it by successive approximation 19:24:10 call r 2 19:24:12 info on aleph null: http://mathworld.wolfram.com/Aleph-0.html 19:24:18 2 = 1/2 +1 19:24:19 erm nope 19:24:34 ok. feed right hand side back into left hand. and start over 19:24:41 repeat till it settles on some value heh 19:24:46 memorize the quadratic formula!!! IT'S EASY :) 19:25:21 tbw errr 19:25:22 heh 19:26:08 by the way, aleph null = |set of positive integers| = |set of all integers| = |set of all rational numbers| = |set of all algebraic numbers| 19:26:36 irrational numbers dont exist. therefore alphe null = the set of all numbers :) 19:26:49 tbw have you seen the mathematical proof that all girls are evil ? 19:27:15 yeah 19:27:21 hehe 19:27:24 stoopid, yet cute 19:27:25 gotta love that :) 19:28:01 and irrational numbers do exist...ancient Greeks proved that, much to their consternation 19:28:11 girls cost time and money 19:28:17 and everyone know time IS money 19:28:23 so girls = money ^ 19:28:33 and everyone knows that money is the root of all evil so 19:28:42 girls = squareroot of money squared 19:28:48 erm something like that 19:30:42 it is interesting to note that |set of all real numbers| does not equal aleph null! Cantor discovered this, and this triggered vehement attacks from Kronecker (a very smart math guy, but idiotic in other areas), declaring that Cantor was trying to play with "infinity" that belongs solely to God, and had many universities bar Cantor. 19:31:21 are you a student of math TheBlueWizard? 19:31:29 he has a masters in it 19:31:35 Cantor eventually went mad and IIRC, he killed himself...yet his work forms a key foundation for modern mathematics 19:31:36 he know a bit about math :) 19:31:53 the cantor set - very interesting 19:31:55 I do know a good bit about math :) 19:32:13 how do you find forth for playing with mathematical concepts? 19:32:23 oh yes, that Cantor set is quite interesting...has a number of unusual properties 19:33:02 and it can be drawn :) 19:33:09 if i can see it i can understand it 19:33:40 AlephNull: um...in the continuum realms (e.g abstract diffy eqns, topology, whatever), not much....you HAVE to use this 3 pounds of gray matter to do these kinds of math ... I like that :) 19:34:26 I440r: um...the approximation to a Cantor set can be diagrammed...but the "final" one cannot be drawn precisely 19:34:50 heh 19:34:56 NOTHING can be drawn exactly 19:35:12 is there a final one ? 19:35:12 gotcha (well, as much I can given my background) :> what do you use forth for? 19:36:14 I440r: look, Cantor set is the *limit* of >>>infinite<<< deletions of the middle thirds...GOT THAT NOW???? 19:36:45 ah, why didn't you just say so :p 19:37:33 tbw yes i know - 19:37:35 draw a line 19:37:40 I440r is just being idiotic in saying "is there a final one", as if it can be casually diagrammed as such 19:37:45 repeat it but delete the middle third heh 19:38:12 and keep going... 19:38:21 AlephNull: I thought you're familiar with Cantor set.... 19:38:33 --- join: sbk (~sbk@dsl-65-184-108-105.telocity.com) joined #forth 19:38:38 hi sbk 19:38:43 you code forth ? 19:38:49 My math is rather limited. 19:39:02 learning more about it, more like... 19:39:13 sbk thats cool too :) 19:39:23 the more newbies forht has the more experts it will have later 19:39:38 I think I saw (was it you?) your url at the chuck moore chat log 19:39:48 yeah...Cantor set is briefly mentioned in undergrad college math courses...but does get a bit more attention in grad math, esp. in topology 19:40:05 sbk chuck moore was in here again a few hours ago :) 19:40:43 TheBlueWizard: I see. I've only taken one undergrad math class so far. 19:40:43 tbw koch curve thinge is based on the cantor set isnt it ? 19:40:48 Cool :-) 19:40:55 i also see similarities with the dragon curve 19:40:56 is the log of today's chat with C.M. posted online anywhere? 19:41:06 alphenull clog sees all 19:41:12 do a whois on clog 19:41:19 clog == channel log bot 19:41:27 nice scoop on Cantor set: http://mathworld.wolfram.com/CantorSet.html 19:41:46 Much thanks. 19:41:56 I bookmarked that link TBW. 19:42:55 I440r: Koch curve is somewhat related to Cantor set, yeah....though it exhibits completely different characteristics...like for example Koch curve is connected while Cantor set is completely disconnected 19:43:19 tbw ya. 19:43:42 but i dont think koch could have visualised his curve without first seeing the cantor set :) 19:44:20 I feel so stupid. 19:45:38 incidentially, while the Cantor set (as viewed using a series of diagrams) looks like a "line" full of dots standing in isolation (and thus its length is zero...not hard to prove), its set-theoretical size is NOT aleph zero...in fact it has the same set-theoretical size as that of a set of real numbers! 19:46:20 I think Koch may arrive at his curve from another way, without needing Cantor set 19:46:23 how do you like that. you compliment people and they just ignore you. 19:47:29 heh 19:47:34 I see Cantor set as something of a Zen math: so empty, yet containing everything hehe 19:48:15 a minimalist maths ??? :) 19:48:24 god invented the integers 19:48:30 well, gotta go to bed...bye all 19:48:30 --- quit: bpt ("zzz") 19:48:34 tbw :) 19:48:50 nite nite tbw 19:48:54 good night TheBlueWizard 19:49:02 nite I440r and AlephNull 19:49:11 has svara talked in here ? 19:49:16 --- part: TheBlueWizard left #forth 19:51:06 oops 19:55:02 --- join: gforth0 (guest@adsl52025.vnet.hu) joined #forth 19:59:49 --- part: I440r left #forth 19:59:49 --- join: I440r (~mark4@1Cust232.tnt2.bloomington.in.da.uu.net) joined #forth 19:59:50 --- mode: ChanServ set +o I440r 20:00:02 feel the power! 20:00:05 --- mode: I440r set +v gforth 20:00:08 --- mode: I440r set +v gforth0 20:00:11 --- mode: I440r set +v sif 20:00:24 --- mode: I440r set -o I440r 20:00:36 yay 20:00:48 I feel it not flowing through me 20:00:50 heh 20:01:09 you guys talked a lot today 20:01:17 :) 20:01:19 --- quit: gforth (Read error: 104 (Connection reset by peer)) 20:01:20 --- nick: gforth0 -> gforth 20:02:13 is gforth a registered nick ? 20:02:15 at 11:21:34, svara talked 20:02:24 you could get gforth to use the ghost kill command 20:10:31 --- quit: sbk ("Bye") 20:12:25 --- join: Herkamire (~jason@ACA056B9.ipt.aol.com) joined #forth 20:12:30 herk! 20:12:56 hi :) 20:16:54 I noticed that the 25 cores in chuck's 25x chip don't have "swap" in their instruction set. they do have "over" though. 20:16:58 does that seem odd? 20:17:47 over rot drop :) 20:18:12 I'm not suprised it doesn't have rot :) 20:18:20 heh 20:18:56 over >r nip r> :P 20:18:57 over >r >r drop r> r> was the best I could come up with 20:18:58 lol 20:19:22 if the chip caches top of stack a nip is faster than a drop 20:19:29 or as fast maybe 20:19:52 in isforth a nip is faster than a drop i think 20:20:09 add esp, immediate being faster than a pop i THINK 20:20:21 his chip doesn't do nip. all instructions are one cycle 20:21:57 I440r: huh? 20:22:26 its faster to add an immediate value to the stack pointer than it is to pop an item off the stack into a register 20:22:42 isforth caches top of stack so nip is add esp, byte 4 20:22:47 drop is pop ebx 20:22:55 so a nip is faster - i THINK 20:23:45 ok. I'm not farmilliar with x86 assembly. I have a different meaning for ESP ;) 20:23:57 hehe 20:24:03 ESP - the 32 bit stack pointer 20:24:08 Extended SP 20:24:26 like ax, bx, cx, dx being extended to eax, ebx, ecx, edx 20:24:49 I don't see the point in all the names. why not r0, r1, r2, r3 ...? 20:25:52 because x86 has its roots in z80 :P 20:26:15 z80?? isn't that the chip in TI calculators? 20:26:18 i like 68k a0, a1, a2,..... d0, d1, d2..... 20:26:28 heh 20:26:33 no idea 20:26:41 its the chip in the kaypro II tho :) 20:27:21 tathi used to program is TI86 calculator, and I seem to remember him saying z80 a bunch. 20:28:24 :) 20:28:28 --- join: Soap` (flop@210-54-74-28.dialup.xtra.co.nz) joined #forth 20:28:31 i want an old hp calculator 20:28:40 if you could only have one be fast (and the other ~6 times slower) would you choose SWAP or OVER? 20:29:00 hrm. swap 20:29:03 welcome 20:29:08 swap dup >r swap r> 20:29:18 hehe 20:30:11 hrm - should isforth cache top of return stack too ? 20:30:15 could use edi for that 20:30:27 >r dup r> swap :-p 20:31:26 nah. make the primatives juggle shit too much 20:32:03 leave top of return stack uncached. x86 is too register starved and forth already eats esi, ebp and ebx 20:32:16 I440r: do you have any speed tests for your forth? (benchmarks) 20:32:41 yes 20:32:52 thers a bench directory that has a few in there 20:33:07 dont you have isforth ???? 20:33:15 * I440r slaps Herkamire arround with the isforth kernel 20:34:02 would it be hard to change it so it cached the top of return stack? 20:36:00 not realy 20:36:15 need to change nest, exit >r dup>r r> r>drop 20:36:17 thats about it 20:36:34 cool. well, if you get boored, try it :) 20:37:12 heh 20:37:40 the only benchmark i consider of any real use tho is the one that shows the difference between vars and consts in isforth 20:37:44 or constants and consts 20:37:51 var being the same as a constant :) 20:37:59 consts compile literals 20:38:09 literals are SOOOOOO much faster 20:39:53 cool... so you have a defining word (const) that creates immediate words that compile a literal? 20:40:48 yes 20:40:59 state smart immediat words :) 20:41:17 oh :) even cooler 20:41:29 so you can use it at runtime 20:41:56 I'd rather be messing with a colorforth though... 20:43:07 yes 20:43:14 it just returns its value 20:43:17 like a variable 20:43:24 0 const 0 20:43:26 1 const 1 20:43:37 those are the first two consts isforth defines :) 20:43:41 3 4 5 are next :) 20:43:45 --- quit: njd (Read error: 104 (Connection reset by peer)) 20:45:01 the PLAN is to have all const definitions compiled into the compiler vocabulary which will be discarded on turnkey 20:45:14 basically meaning you can give names to literals 20:45:43 the consts no longer exist at run time in a turnkeyd appp 20:46:00 can have huge libs of #defines taht dont cost you SHIT unless you actually use one :) 20:46:31 what does it mean to be "turnkeyd"? 20:46:49 it usually just means strip all the headers and save out the target executable 20:46:53 does 0 const 0 actually do anything useful? 20:46:59 yes 20:47:19 like get rid of the dictionary and save state? 20:47:24 when you type a 0 forth has to search the dictionary for it. if it doesnt find it it has to pass it to NUMBER 20:47:28 then number has to interpret it 20:47:31 all at compile time 20:47:46 having 0 constant 0 etc defined usually speeds up compilation alot 20:48:03 right, but if you put 0 in the dictionary, you have to search past it for every word you look up 20:49:07 heh forth traditionally defines 20:49:10 0 constant 0 20:49:12 1 constant 1 20:49:15 2 constant 2 20:49:21 and maybe 3 constant 3 20:49:24 but people probably use 0 and 1 enough to make it worth it. 20:49:39 yes 20:49:53 isforth is greedy and defines 0 thru 5 :) 20:50:07 actually we dont need to search through them unless the search hits that specific thread 20:50:20 0 1 2 3 4 and 5 might even be in successive threads 20:51:10 yeah, dictionary optimization. I haven't gotten around to doing anything special with the dictionary. 20:52:23 im actually NOT threadign yet 20:52:28 i mean hashing 20:52:33 so turnkey means to throw away the dictionary? 20:52:34 so everything is in one thread :) 20:52:41 sort of :) 20:52:46 it means to LOCK the application 20:53:13 ??? what's that? no more compiling?? 20:54:52 yes 20:55:10 that would be usefull for programming embedded devices. 20:55:41 and trimming down standalone programs.... does isforth create standalone programs? 20:55:49 yes 20:55:52 turnkeyed even 20:56:02 --- join: gforth0 (guest@adsl52031.vnet.hu) joined #forth 20:56:24 nifty. they would be x86 linux programs I presume. 20:56:45 yes 20:57:33 I440r: what have you gained by programming isforth in asm rather than C? 20:57:40 not using c 20:57:42 nuff sed :) 20:58:02 i HATE c 20:58:09 c has NO place what so ever 20:58:16 other than the circular filing cabinet 20:59:14 ok, so in asm rather than some other well optimized high level language (asuming there is such a thing) 20:59:20 also ive gained READABLE code 20:59:35 heh isforths kernel is not heavilly optimized :P 21:02:30 I have mixed feelings about optimization. 21:02:37 me 2 21:03:29 --- quit: gforth (Read error: 110 (Connection timed out)) 21:03:31 --- nick: gforth0 -> gforth 21:03:42 on the one hand I have a lot of fun with extremely computationally intensive stuff (fractals) 21:04:10 i do optimizing for the fun of it occasionally 21:04:15 and on the other, it seems that if you ignore optimizing and keep stuff simple, many projects would end up much faster. 21:04:26 try to save 1 clock cycle off a 3 itteration loop that executes once a month :) 21:05:01 if it's always three iterations, why loop? :-) 21:05:08 hehe 21:05:15 i TOLD you 21:05:18 for the FUN of it heh 21:05:50 STARS16 etc. 21:05:53 as long as there is Some way to get a small bit of code to run Really fast I'll be happy I think. 21:07:44 However, try simulating E- and B-fields of a complicated RF circuit in 3-space over the entire course of execution of the circuit, without optimization. 21:07:47 There is a time and place for everything. 21:07:48 also I think that optimizing is something that's better done After. I have a tendancy to try to think up optimization tricks before I write something. I'm starting to think that that is not a good idea. 21:08:12 One of the reasons why CPUs are so fast today is because of computer games and their heavy graphics requirements. Notice how as soon as graphics cards GPUs took over all the number crunching, how the CPU clock wars pretty much became a background issue? 21:08:33 :) 21:08:54 I thought that was because people's computers are catching on fire when they clock them up... 21:08:58 Herkamire: That is correct -- you should always optimize afterwards. And, please, measure your code speed before you optimize too. Make sure you're really optimizing what needs to be optimized. 21:09:15 --- join: haroldo (~haroldo@165a.dedicado.com.uy) joined #forth 21:09:32 ft hi haroldo you a forth coder ? 21:09:35 Herkamire: No, 'fraid not. 21:10:29 kc5tja: so now the only thing driving CPU speeds is code cludge? ;) 21:10:38 I440r, hi, I'm curious about forth. I first read about it, fifteen years ago... 21:10:44 cool 21:10:57 :) 21:11:19 well we had the inventor in here earlier 21:11:19 someone said, in lisp code is data, in fort data is code 21:11:30 haroldo: well, you're in the right place I think. we seem to like talking about forth an awefull lot 21:11:35 Herkamire: No. Now it's just a matter of one-upping the competitors. It's a vicious cycle. 21:12:37 kc5tja: hehe :) 21:12:43 Herkamire, look, I'm going to sleep in 45 minutes. Please tell me something coll about forth, which will lead my dreams to new horizons ;-) 21:12:45 With CPUs at 1GHz, computers are so incredibly fast now (we're talking about 1000 times faster than a 1MHz CPU here), that Chuck's assessment that things "should be instantaneous" is dead on the money, in every way. 21:13:30 kc5tja: Tathi and I were starting to wonder if we are Starting to hit the limit of the technology (silicon?). well, mostly tathi was thinking that and I was thinking he might be on to something. 21:14:06 No, we haven't. 21:14:37 Intel and IBM both have transistors that switch at some 18GHz at a minimum, and IBM is researching THz transistors now, using ordinary silicon. 21:14:44 haroldo: sif is a bot in here that executes forth code.... watch: 21:14:56 sif: .( Hello World!) 21:14:57 Herkamire: Hello World! 21:15:07 cool 21:15:15 what else can it do? 21:15:17 sif: cr 21:15:18 kc5tja: 21:15:33 sif: ." Hello World!" 21:15:34 deltab: 21:15:36 sif: : test 0 10 do i . loop ; test 21:15:37 kc5tja: 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21:15:55 I think sif has a bug or do in its do/loop implementation. :) 21:15:56 you need -1 +loop 21:15:59 no 21:16:02 you did it wrong 21:16:04 10 0 do 21:16:07 not 0 10 do 21:16:11 * kc5tja checks ANSI standard... 21:16:12 10 0 do loop or 21:16:20 0 10 do -1 +loop 21:16:25 what's ." ? 21:16:29 I stand corrected. It is 10 0. 21:16:31 prints a string 21:16:31 that has got to be THE most common forth error 21:16:31 --- quit: Herkamire (Read error: 104 (Connection reset by peer)) 21:16:32 * kc5tja almost never uses do. 21:16:43 Herkamire: Precisely why I do not use do/loop. :) 21:16:56 sif: .( hello) 21:16:57 deltab: hello 21:17:00 sif: : test 0 begin dup 10 < while dup . 1+ repeat drop ; test 21:17:01 kc5tja: Word not found: while 21:17:07 sif: ." hello" 21:17:08 deltab: 21:17:11 hmmm.... now that right stinks. 21:17:37 g4 : test 0 begin dup 10 < while dup . 1+ repeat drop ; test 21:17:41 kc5tja: 0 1 2 3 4 5 6 7 8 9 21:17:50 There... :) That's what I wanted. 21:17:55 sif .( my first running forth code) 21:18:01 sif: : fib 0 1 rot 1 do tuck + loop nip . ; 40 fib 21:18:02 I440r: Word not found: tuck 21:18:09 foo! 21:18:15 define TUCK!!! 21:18:20 sif: .( my first running forth code) 21:18:22 haroldo: my first running forth code 21:18:23 I440r: Is it case sensitive? 21:18:30 oh! 21:18:37 dunno 21:18:46 sif: : fib 0 1 rot 1 do swap over + loop nip . ; 40 fib 21:18:48 I440r: 32459 21:18:50 wrong 21:18:58 sif: : test 0 10 do i . loop ; test 21:18:59 haroldo: 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21:19:00 theres a bug in sif for sure 21:19:19 sif: test 0 10 do i . loop ; test 21:19:20 haroldo: Word not found: test 21:19:31 sif: : test 2 10 do i . loop ; test 21:19:32 haroldo: 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21:19:39 sif: : foo 10 0 do i . loop ; foo cr 21:19:41 I440r: 0 1 2 3 4 5 6 7 8 9 21:19:58 end start do stuff loop 21:20:07 not start end do stuff loop 21:20:40 sif: : foo 0 10 do i . -1 +loop ; foo 21:20:42 I440r: Word not found: -1 21:20:46 ugh 21:20:53 sif: 'a . 21:20:54 deltab: stack underflow 21:20:58 sif has a lot of bugs in it... :) 21:21:27 no - just ommissions :) 21:21:39 but the fib result was definatly wrong :) 21:21:40 sif: : pepe 5 15 do .( pepe) -1 loop ; pepe 21:21:41 haroldo: pepe Word not found: -1 21:21:46 sif: : pepe 5 15 do .( pepe) loop ; pepe 21:21:47 haroldo: pepe 21:21:55 sif: : pepe 5 15 do . loop ; pepe 21:21:56 haroldo: stack underflow 21:22:00 sif: : fib 0 1 rot 1 do swap over + loop nip . ; 40 fib 21:22:02 I440r: 32459 21:22:09 sif: : pepe 51 5 do . loop ; pepe 21:22:10 haroldo: stack underflow 21:22:12 aha must be 16 bit ? 21:22:39 sif: 5 i . 21:22:41 deltab: 0 21:22:50 haroldo you need to say i inside tha tloop 21:22:55 well, i think I have to read forth sintax before playing with the bot 21:22:59 : pepe 51 5 do I . loop ; 21:23:09 i is the loop index? 21:23:11 sif: : pepe 51 5 do i . loop ; pepe 21:23:12 haroldo: 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 2 21:23:17 i is the INNER loop index 21:23:21 sif: : pepe 20 5 do i . loop ; pepe 21:23:22 haroldo: 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 21:23:26 you also have outer indicies 21:23:28 i j and k 21:23:36 ah 21:23:49 sif: 5 10 + . 21:23:50 deltab: 15 21:23:56 sif: : 5t 50 0 do i . 5 +loop ; 5t 21:23:57 I440r: Word not found: +loop 21:24:01 dammit :P 21:24:37 oh, I see, everithing is polish notation 21:24:41 Use g4; it's at least complete. :) 21:24:51 no, reverse polish 21:24:54 ok 21:25:07 postfix, isn't it? 21:25:10 yes 21:25:18 Polish notation is essentially what Lisp uses -- + 1 2 3 * 4 5, for example. 21:25:25 Yes, postfix. 21:25:30 and how do I have to think about a loop in that notation? 21:26:26 the word DO takes 2 parametrs. the end index and the start index 21:26:34 Well, for starters, end start is how you'd lay out your boundary conditions, instead of start end. :) 21:26:34 LOOP loops back to the do 21:27:18 one of the merits about lisp, is the great semantics that lies behind its naive syntax, and the ability to parse itself and to generate lisp code. what would you say it is great about forth? 21:27:32 --- join: Herkamire (~Jason_@AC9B308C.ipt.aol.com) joined #forth 21:27:32 sif: 5 0 do i 2 * . loop 21:27:34 deltab: 0 21:28:14 blody &%$*&^ AOL 21:28:41 I440r, and how do you understand the i index after de do word? 21:28:41 sif: 5 2 * . 21:28:42 deltab: 10 21:28:55 deltab: You can only use do/loop inside a :-definition, since it's a compiler word, not an interpretter word. 21:28:59 hang on brb 21:29:06 oh 21:29:43 While some Forths do allow DO/LOOP (and even IF/ELSE/THEN) in interpret mode, they are exceedingly rare. 21:29:59 haroldo: What do you mean by your last question? 21:30:38 I think I've seen logo execute code while defining a term 21:31:22 kc5tja, I don't undertand whose parameter is _i_ in : 5t 50 0 do i . 5 +loop ; 5 for instance 21:31:50 har its easy 21:31:55 har when you say somethign like 21:31:57 5 0 do 21:32:09 the word DO puts the 2 parameters onto the return stack 21:32:16 the word I gets the current index 21:32:31 the word loop increnemts the current index and if its not equal to the end index it lops back 21:33:08 so in a way the i is a post parameter, and the 5 and the 0 are pre parameters... 21:33:15 Think of 'i' as in the BASIC example: FOR i=1 TO 10: PRINT i: NEXT i 21:33:26 That's what 10 0 DO I . LOOP does. 21:33:29 ok 21:33:44 . is the number-print word in Forth. 21:33:56 and ... what about compilers and interpreters? what's available out there? 21:34:11 haroldo: Are you a Windows or Linux user? 21:34:16 I440r likes to explied how the forth system works, rather than what it does in the big picture. 21:34:21 Both, but I prefer Linux 21:34:35 yay linux 21:34:55 In Linux, there is gforth, a GNU Forth implementation that is pretty much fully ANSI Forth compatible. 21:35:13 Just curious, I'm from Uruguay, Latin AMerica, from where are you? 21:35:26 kc5tja, ok 21:35:33 Performance is decent, but I find its documentation to be somewhat lacking. You might want ot have a copy of the ANSI Forth standard lying around for reference. 21:35:47 kc5tja, ok 21:35:55 haroldo: I'm from California, USA. 21:36:08 And no, I do not speak Spanish, and am not Mexican. :-) 21:36:17 Though I live about an hour away from Mexico. 21:36:19 kc5tja :-) 21:36:45 kc5tja you could speak spanish not being a mexican either :-) 21:36:58 I know, but I can't learn the language. It's too foreign a tongue for me. 21:37:18 kc5tja like forth for a pascal guy :-) 21:37:49 ? 21:37:55 Ohh...sorry. 21:37:59 :-) 21:38:06 I didn't catch it right away. I'm getting tired. :D 21:38:22 funny thing is, I can learn most any computer programming language reasonably quickly. 21:38:29 But a spoken language, however, I *just* can't get. 21:38:41 rrrrr did you read my question about lisp and forth strengths? 21:38:58 * kc5tja didn't even see it. Sorry. Can you repost the question? 21:39:05 yes 21:39:25 one of the merits about lisp, is the great semantics that lies behind its naive syntax, and the ability to 21:39:27 parse itself and to generate lisp code. what would you say it is great about forth? 21:39:44 There are several things, actually. 21:39:50 yes ... 21:40:00 Forth can manipulate its own code in source form, as Lisp can manipulate its own code in tree/list form. 21:40:16 Forth exposes the compiler's parser (and even a large portion of the compiler itself) to the programmer. 21:40:31 ahah 21:40:47 In fact, BEGIN/WHILE/REPEAT and even IF and DO/LOOP are all implemented in high-level Forth. :) 21:41:21 However, because they are compiler words, they execute only at compile time, and they compile code right into the word's definition right away (immediately; hence, they're also referred to as "immediate" words). 21:41:29 They're Forth's equivalent to Lisp macros. 21:41:34 so in a way they are similar languages 21:41:57 Forth and Lisp are very closely related actually. Chuck Moore, inventor of Forth, admits that Lisp is a hhuuggee influence to Forth. 21:42:13 --- quit: Herkamire (Read error: 104 (Connection reset by peer)) 21:42:15 However, Chuck has always been a minimalist to the extreme. 21:42:30 oh 21:42:31 While Lisp has garbage collection, and all sorts of other features, Forth manages to make due without them. 21:42:36 yeah...he was a user of Lisp at one point 21:43:07 Storage management is "automatic" in Forth in that it's central data structure is the "stack." You put something onto the stack, and it stays there. As words consume stack elements, these elements are "destroyed." 21:43:29 Traditional Forth actually has very little need for dynamic memory management, interestingly enough. 21:43:33 I see 21:43:39 However, when it does need it, it is often managed in one of two fashions. 21:44:08 The first, you have a queue structure that maintains object references. Items dropped into the queue are known to be referenced, and hence, are not deallocated. 21:44:17 As objects fall out the other end of the queue, they get deallocated. 21:44:30 As you can imagine, the queue is a finite length, and hence, there are a couple of problems with this approach. 21:44:40 (especially if your working set exceeds the queue's size). 21:44:43 yes 21:45:01 The other solution is manual memory management, which is about as difficult to d in Forth as it is in C. 21:45:18 yes the same thing 21:45:30 --- join: Herkamire (~Jason@ACA4E22F.ipt.aol.com) joined #forth 21:45:37 However, I personally have yet to encounter a need where I *need* dynamic memory management. 21:45:56 s/need where/problem where/ 21:46:10 ok 21:46:29 interesting enough 21:46:58 * kc5tja isn't saying it's possible to avoid it in all situations; but I'm just saying that the structure of the Forth language encourages a more static approach to programming, while still yielding incredibly similar results to equivalent Forth programs. :) 21:47:10 dammit -- I'm really tired. 21:47:14 equivalent Lisp programs. 21:47:15 Sorry 21:47:42 Last week Perl's Larry Wall did his Perl 6 presentation in Porto ALegre Brazil 21:47:53 I was there 21:48:03 * kc5tja thinks Larry Wall is a joke. 21:48:24 he said "distrust claimed minimalism" (or something very similar) 21:48:43 Here is a man who has violated every basic tenet of well structured software engineering, every good programming practice in the world, and hten has the nerve to accuse Python and its programmership of being inferior to him. 21:49:15 in a way yoy're right 21:49:33 And he has no foundation for saying that. I'm willing to bet, if you review his speech, he gives zero hard evidence in support of his claims. 21:49:37 but his approch proved somewhat successful 21:50:00 haroldo: And everyone I know of, who uses Perl in a commercial environment, hates every minute of it. 21:50:00 yes, it is as you say 21:50:19 There are forths with dynamic memory management built in. and I'm sure you could find code to do it on any ANS forth. I believe it says something very positive about forth that you usually don't need it. 21:50:48 Herkamire: No need to; ANS Forth has the ALLOCATE/FREE words. But it's still manual memory management. Lisp is automatic, complete with garbage collection. 21:51:20 kc5tja, I dont think is that bad, neither think it's an ideal. 21:51:23 You could implement your own garbage collector in Forth, I suppose; there's nothing preventing that from happening at all. But, still, it's worthwhile only if it's really critical. 21:52:15 haroldo: With respect to what is that comment directed? Manual memory management, or Larry Wall's success with Perl? 21:52:29 LW succwss with Perl ... 21:52:53 * kc5tja thinks that Perl is an absolute abomination. 21:53:04 * kc5tja hasn't touched it since Python 1.3. 21:53:16 And I continue to use Python in all my projects to this day. 21:54:17 people keep talking about liking Python. I can't decide if 21:54:33 kc5tja, yes python must be superior, I think. Perl is good for small scripts, at least it is what I felt using it. Other people whom I kno use Perl heavily without much pain. 21:54:34 If I had suitable permission to, I'd give you the e-mail address of someone I know in our Methodologies group here at work. He can recount numerous horror stories of Perl. He's rewriting Bugzilla in Python, and finding it to take 1/10th the number of program lines. 21:54:48 I should learn all these cool languages people are raving about, or if I should ignore the world for a year or two and make my own OS in my own colorforth. 21:55:45 Herkamire: Python is nice, but it's no Forth. Likewise, Forth is nice, but it's no Python. It's a matter of using the right tool for the right job. I'd learn both. It can never hurt, and it can only help. 21:56:29 Although Dolphin will be written in Forth, and Forth will likely be its first shell environment, I still intend on porting Python to Dolphin, because it's too darn useful not to. :) 21:56:31 kc5tja, but are these horror stories responsibility of Perl itself? 21:56:41 well, now's a good a time as any. I don't have anything to do and this computer dosen't have a compiler on it :) 21:57:05 haroldo: Yes, absolutely and in full. Things where the language semantics itself is ambiguous. Lack of exceptions support. Inability to treat file handles as first-class programmable objects, etc. 21:57:53 kc: what's dolphin? 21:57:54 kc5tja, yes all these things are the ones addressed in Perl 6 (yet to be seen) 21:57:57 haroldo: He admits that Perl 6 is a vast improvement over Perl 5; however, it will not solve all of his problems, and actually introduces some new ones. 21:58:28 kc5tja: what would you use Python for that you wouldn't use Forth for? 21:58:32 haroldo: For this reason, all of his new projects are being done in Python as well, while using Perl 5 only for maintenance issues. He has no intentions of upgrading to Perl 6. 21:58:42 AlephNull: Absolutely. 21:59:05 yes 21:59:32 kc5tja, yes tell us how you do decide between python and forth 22:00:03 That's pretty complicated actually. :) 22:00:10 why? 22:00:24 I normally use Python where i need to do heavy string manipulation, or where having garbage collection is an overwhelming bonus to me. 22:00:26 kc5tja: what is dolphin 22:01:11 is there a page you recomend for getting me started with python? 22:01:17 Herkamire: It's an operating system, with heritage in AmigaOS Classic and VMS, that I'm writing in my spare time. 22:01:42 Herkamire: http://www.python.org has everything pertaining to Python, including library references, tutorials, and language specification. 22:01:58 kc5tja, are you open sourcing it? 22:02:35 haroldo: Yes. 22:02:45 Along with the dialect of Forth I'll be writing it in as well. 22:02:49 kc5tja, do you have an url? 22:03:20 haroldo: No, not at this time. I used to many years ago, back when Dolphin was written in assembly and in C, but due to lack of interest in the project, it's become more of a personal item, rather than a public item. 22:03:34 and language specification. 22:03:43 sorry 22:03:46 I see 22:04:33 I would spend the energy for site maintenance if I could get a beneficial result from it, but without anyone who is interested or willing to help, the time I spend maintaining the site is better devoted towards programming and creating printable documentation. 22:05:01 yes good pointç 22:05:17 * kc5tja is constructing a DOS-based prototype of the FS/Forth environment that I'll be using to write Dolphin in. 22:05:30 let's see if the bot is not yet asleep 22:06:01 I'll be using my FS/Forth System 1 to then target compile FS/Forth System 2, which is the 32-bit native Forth that I'll write Dolphin in. 22:06:05 sif, : . (are you asleep?) 22:06:22 sif: : . (are you asleep?) 22:06:23 haroldo: Word not found: (are 22:06:29 sif: : . ( are you asleep?) 22:06:30 haroldo: 22:06:37 sif: : . ( are you asleep) 22:06:39 haroldo: 22:06:40 You just redefined . to a comment. ;) 22:06:50 sif: : .( are you asleep) 22:06:51 haroldo: Word not found: are 22:06:56 how? 22:07:23 The : word creates a new word in the Forth dictionary. Since . (or .( in your last exxample) is the next word in the input stream, it creates a new word called . or .(. :) 22:07:38 : is also a high-level Forth word that enters the Forth compiler. 22:07:59 sif: . 22:08:00 mmmh 22:08:01 AlephNull: stack underflow 22:08:21 sif: .( are you asleep) 22:08:22 sif: .( are you asleep?) 22:08:22 sif: : NewWord ." This is a new word." ; NewWord 22:08:22 haroldo: are you asleep 22:08:24 Herkamire: are you asleep? 22:08:25 kc5tja: This is a new word. 22:08:50 ." is compile-time only? 22:08:53 As you might imagine, .( is one word in Forth. :) 22:08:56 deltab: Yes. 22:09:01 why? 22:09:02 deltab: According to the ANSI standard. 22:09:09 Because it's a compiler word, like IF or BEGIN. 22:09:20 but there's .( 22:09:31 sif: : new-word 9 . ; new-word new-word new-word 22:09:32 Herkamire: 9 9 9 22:09:41 .( doesn't compile code. Watch: 22:10:15 sif: : NewWord .( Will this print when I execute NewWord?) ; 1 . 2 . 3 . NewWord CR 22:10:16 kc5tja: Will this print when I execute NewWord?1 2 3 Word 22:10:26 ah 22:10:28 Notice when it printed? :) 22:10:41 .( is an "immediate" word -- it executes at compile time. However .( doesn't actually compile anything. 22:10:47 sif: 0 . 22:10:48 AlephNull: 0 22:11:39 sif: 0 . 0 . 0 0 . . 0 . 0 . 0 0 . . 0 . 0 . 22:11:40 AlephNull: 0 0 0 0 0 0 0 0 0 0 22:12:04 sif: .( AlephNull is having way too much fun. ) 22:12:05 kc5tja: AlephNull is having way too much fun. 22:12:10 lol :p 22:12:11 sif: : foo 5 0 do i 2 * . loop ; foo 22:12:12 deltab: 0 2 4 6 8 22:12:52 sif: .( haroldo's eyes are shutting down ) 22:12:53 haroldo: haroldo's eyes are shutting down 22:13:12 sif : square dup * ; 22:13:19 sif: : square dup * ; 22:13:20 deltab: 22:13:28 sif: 5 square 22:13:29 deltab: Word not found: square 22:13:33 * kc5tja considers creating a Forth bot called "sif lord." :D 22:13:51 deltab: The dictionary resets after it executes your one-line request. Security and safety measure. 22:13:55 gforth does this as well. 22:14:13 * kc5tja was expecting gforth to retort with an error on that one... :) 22:14:33 kc5tja, I thought about it, it's a natural thing 22:15:12 Also, Forth exposes its return stack (well, in any non-ANSI compliant system and in a large subset of ANSI Forth systems) as a first-class programming construct. 22:15:12 --- quit: Herkamire (Read error: 104 (Connection reset by peer)) 22:15:31 sif: . 22:15:32 AlephNull: stack underflow 22:15:39 where did the 5 go? 22:15:40 Hence, implementing things like linear continuations take, maybe, around 5 to 16 lines of Forth source code, depending on how trivial or complete you want it to be. 22:16:10 AlephNull: After each line you send to it, the bot will reset its state. 22:16:12 sif: .S 22:16:13 kc5tja: Word not found: .S 22:16:17 g4 .S 22:16:22 kc5tja: <0> 22:16:26 g4 42 22:16:28 ah. thanks. 22:16:32 g4 .S 22:16:37 kc5tja: <0> 22:16:57 what's .S ? 22:17:03 haroldo do you use linux ? 22:17:05 deltab: Print the current contents of the data stack. :) 22:17:13 I440r, yes I do. 22:17:25 thought so... oh, g4 also clears its stack? 22:17:25 x86 based ? 22:17:32 I440r, yes. 22:17:41 haroldo isforth.clss.net 22:17:42 g4 1 2 3 4 .S 22:17:44 :) 22:17:46 deltab: <4> 1 2 3 4 22:17:54 I440r ????? 22:18:06 my linux only x86 only direct threaded forth compiler 22:18:18 give you something to play with 22:18:28 deltab: Yes. 22:18:32 not very much in the way of doccumentation there uyet tho 22:18:33 g4 1 2 3 4 .S 22:18:34 g4 .S 22:18:35 I440r ok, thnkyou 22:18:38 kc5tja: <4> 1 2 3 4 22:18:41 kc5tja: <0> 22:19:26 Note that if you're using GForth at the console, it doesn't clear the stacks unless you get into an error situation. But the bot is actively resetting its own state to ensure no weirdness occurs (e.g., running out of dictionary space, etc) 22:19:30 So I can't do things like: 22:19:44 g4 16777216 ALLOT .( HAHA! Sucked up all your RAM!) 22:19:49 kc5tja: in file included from *the terminal*:-1 22:19:49 kc5tja: /tmp/fsock-sh-server.request.tmp:46: Dictionary overflow 22:19:49 kc5tja: 16777216 ALLOT .( HAHA! Sucked up all your RAM!) 22:19:49 kc5tja: ^^^^^ 22:19:49 kc5tja: Backtrace: 22:19:52 kc5tja, can you access PostgreSQL from forth? 22:20:49 haroldo: To the best of my knowledge, no, since nobody has ever bothered to write a set of C interface words for Forth under Linux. However, under Windows NT, you could, since Win32Forth, a Forth-83 dialect of Forth for Windows, does have such words in place. 22:21:07 However, I do not know how to use such a facility. 22:21:20 Supposedly, Anton was going to write such a feature for GForth, but I don't know if it's been implemented yet or not. 22:21:40 kc5tja, do you prefer Windows NT to Linux? 22:21:49 I do know that, when I write my FS/Forth System 2 for Linux, I definately will include such a feature as well. 22:22:10 haroldo: No way. :) Frankly, I'd rather be using AmigaOS over anything else, but Linux is currently my PC operating system of choice. 22:22:19 [02:11am][haroldo(+iw)] [#forth(+lnt)] 22:22:31 (As soon as Dolphin gets done, that, obviously, will change. ;D) 22:22:41 kc5tja, ahh, better that way :-) 22:22:50 :-P 22:23:09 Even Linux is too big, too bloated, and too error-prone for my needs. 22:23:22 --- join: Herkamire (~Jason@ACA6CCB3.ipt.aol.com) joined #forth 22:23:37 can I start swearing about AOL again? 22:23:45 And when asked, "Why not just contribute to Linux?", the answer is always the same: "Because my changes are so fundamental and far reaching, that it'd take less time to write my own, than to convince the Linux community my changes are preferred." 22:23:47 kc5tja, bye, thanks for your attention ! 22:23:53 bye everyone 22:23:58 73 haroldo!! :) 22:24:02 bye haroldo 22:24:05 73? 22:24:20 It's am amateur radio prosign, meaning, "Nice talking to you, and hope to see you again!" 22:24:33 KC5TJA/6 is my ham radio callsign. :) 22:24:34 not 88? =) 22:24:34 oh!! I missed that one!! 73! 22:24:52 goshawk`: No, 88 is "Hugs and kisses" ... Good grief man, are you trying to label me or something? ;D 22:25:02 I know...I was kidding =) 22:25:14 * kc5tja laughs 22:25:18 hahahaha 22:25:31 well im outa here. got stuph to do tomorrow 22:25:37 and wanna goce before i zzz :) 22:25:44 goce? 22:25:49 well my girlfrind will garbage collect me up if I don't go to sleep, so good bye 22:25:51 I think you need some zzzs... :) 22:26:02 nonono coders NEVER sleep! 22:26:03 heh 22:26:05 Later haroldo -- thanks for stopping by. 22:26:18 haroldo does she have a sister? 22:26:21 :) 22:26:45 nevermind :) 22:26:48 outa here.... 22:26:48 license or licence, not 'lisence' :-) 22:27:01 I440r yes, her sister is 24, just married, and left to Barcelona!! She's a cute girl. 22:27:02 --- quit: I440r (": sleep bed go tuck light off ; immediate") 22:28:24 --- quit: haroldo ("BitchX: don't leave home without it!") 22:29:24 Well, I think I'm probably going to take off myself. 22:29:38 I have a Forth kernel to write, and a ham radio to listen to. :) 22:32:22 have a good one kc5tja 22:33:02 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 22:33:27 --- quit: AlephNull () 22:49:02 --- join: sbk_ (~500@dsl-65-184-98-221.telocity.com) joined #forth 22:52:24 --- quit: sbk_ (Client Quit) 22:55:02 --- join: gforth0 (guest@adsl52013.vnet.hu) joined #forth 22:56:48 --- join: sbk_ (~500@dsl-65-184-98-221.telocity.com) joined #forth 23:06:57 --- quit: gforth (Read error: 110 (Connection timed out)) 23:06:57 --- nick: gforth0 -> gforth 23:37:31 --- quit: Soap` () 23:38:34 --- quit: Herkamire (Remote closed the connection) 23:59:59 --- log: ended forth/02.05.18