00:00:00 --- log: started forth/02.10.08
00:47:49 --- quit: Serg_Penguin (Killed (NickServ (Nickname Enforcement)))
00:47:53 --- join: Serg_p (~Z@nat-ch1.nat.comex.ru) joined #forth
01:54:27 <onetom> ianni: well, im not a professional programmer, but i was 6yrs old when ive learnt basic on a zx81. now im 25yrs old
02:15:47 --- nick: Serg_p -> Serg_penguin
02:56:34 --- quit: Serg_penguin (card.freenode.net irc.freenode.net)
02:56:34 --- quit: Soap` (card.freenode.net irc.freenode.net)
02:56:38 --- quit: ianni (card.freenode.net irc.freenode.net)
02:56:38 --- quit: TreyB (card.freenode.net irc.freenode.net)
02:56:38 --- quit: skylan (card.freenode.net irc.freenode.net)
02:57:15 --- join: Serg_penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
02:57:15 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth
02:57:15 --- join: skylan (sjh@Sprint8102.tbaytel.net) joined #forth
02:57:15 --- join: ianni (ian@inpuj.net) joined #forth
02:57:15 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth
03:09:21 --- quit: Serg_penguin (Read error: 110 (Connection timed out))
03:19:30 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
03:19:48 --- quit: Serg_Penguin (Client Quit)
03:30:45 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
03:57:47 --- join: jacereda (~crc@VA1-1G-u-0007.mc.onolab.com) joined #forth
04:01:05 <XeF4> i440r: which fpc version do you use?
04:05:11 <Serg_Penguin> hi
04:05:23 <XeF4> hi
04:28:55 <Robert> Good afternoon :)
04:40:04 --- quit: jacereda (Remote closed the connection)
04:43:47 --- join: jacereda (~crc@VA1-1G-u-0007.mc.onolab.com) joined #forth
05:02:09 <XeF4> afternoon
05:02:19 <XeF4> fpc seems very cool
05:02:37 <XeF4> a bit chubby, but still nice
05:03:36 * Serg_Penguin does use two russian forths - smal32 (DOS DPMI 32 and linux) and gpforth (dos 64k)
05:03:53 <Serg_Penguin> mostly second - no big tasks
05:04:51 <XeF4> has either one a good editor?
05:05:27 <Serg_Penguin> no, both like gcc - compiler only
05:05:48 <Serg_Penguin> u can compile special source (included) to get dialog system
05:06:00 <XeF4> aha.
05:06:06 <Serg_Penguin> err - its gpforth
05:06:20 <Serg_Penguin> smal32 has both in one .exe
05:09:04 <XeF4> fetching smal32..
05:10:15 <Serg_Penguin> it has RU docs
05:10:22 <Serg_Penguin> be warned
05:10:38 <XeF4> is smal_eng EN docs?
05:10:42 <XeF4> er.. smal_eng.zip
05:11:29 <Serg_Penguin> doG knows
05:15:55 <Serg_Penguin> -> XeF4 does it work ?
05:16:41 <XeF4> fetching unrar now
05:26:53 <Serg_Penguin> -> XeF4 started ?
05:42:15 --- quit: Serg_Penguin ()
05:49:03 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
06:15:16 --- join: Herkamire (~jason@wsip68-15-54-54.ri.ri.cox.net) joined #forth
06:15:38 <XeF4> started, no time to test it
06:16:51 <XeF4> Serg: only RU docs found so far.
06:18:48 --- join: LuckyPhil (~phowlett@CPE-203-45-248-201.qld.bigpond.net.au) joined #forth
06:20:21 --- quit: LuckyPhil (Client Quit)
06:20:43 --- quit: Serg_Penguin (Read error: 104 (Connection reset by peer))
06:22:17 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
06:28:37 --- join: tathi (~josh@wsip68-15-54-54.ri.ri.cox.net) joined #forth
06:28:46 <XeF4> tahti
06:31:29 <tathi> hi hi
07:19:59 --- quit: Serg_Penguin ()
07:36:23 --- join: Stepan (~stepan@Charybdis.suse.de) joined #forth
07:37:21 <Stepan> hi
07:37:50 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
07:37:54 <Stepan> is there any good information about the implementation of vocabularies available?
07:42:11 <Serg_Penguin> testmsg
07:43:00 <Stepan> Serg_Penguin: ack.
07:45:26 <Serg_Penguin> thanx, it needn't ack, i just tryed what happens if i /msg to a channel, not man
07:45:39 * ianni likes the look of isforth
07:46:11 <Serg_Penguin> but my nasm dislikes its %xdefine :(
07:57:10 * ianni wonders if he should learn ASM before learning any more forth
07:57:58 <tathi> heh
07:58:07 * Serg_Penguin considers lame if compiler is written in asm and cannot compile itself
08:02:42 * ianni decides to do some NASM 
08:04:58 <Robert> Heh.
08:05:11 <Robert> Serg_Penguin: It's not lame :)
08:06:52 <Serg_Penguin> maybe i was too crude, but it's no good
08:09:57 <Robert> Hehe.
08:10:01 <Robert> Why not?
08:10:11 <Robert> An x86 assembler is a quite complicated thing.
08:11:59 <tathi> besides, how many forths are there that can compile themselves, anyway?
08:12:18 * Robert wonders.
08:12:20 <tathi> I can only think of a couple
08:12:35 <Serg_Penguin> spf4 is compiled by spf374, dunno tryed by itself, but why not ?
08:12:54 <Serg_Penguin> -> tathi pse gimme names of that couple
08:15:26 <tathi> Serg_Penguin: I'm pretty sure MOPS does, but that's for MacOS (think < 10)
08:15:48 <tathi> I was thinking of another one a minute ago
08:16:03 <tathi> but now I'm drawing a complete blank
08:16:08 <tathi> hmm...
08:17:11 * Serg_Penguin will try to compile spf4 by itself, if wont forget
08:21:21 --- quit: Serg_Penguin ("time to go home...")
08:33:55 --- join: I440r (~mark4@1Cust169.tnt3.bloomington.in.da.uu.net) joined #forth
08:34:24 <ianni> lo I440r 
08:34:28 <I440r> hi
08:34:54 <I440r> wasnt feelin too good yesterday, but i went to bed at 7 and didnt get up till 9 this morning heh
08:35:00 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
08:35:13 <I440r> serg! hi dood
08:35:30 <I440r> serg did you get isforth compiled ?
08:36:56 <ianni> I440r: I'm trying to learn some NASM before I move on
08:37:01 <ianni> so i might understand things better
08:37:13 <ianni> I440r: too much sleep makes me tired :D
08:37:24 <Robert> Hello, I440r :)
08:37:25 <ianni> I'm having problems compiling the an nasm hellow world
08:37:31 <ianni> s/an //
08:37:54 <ianni> [~/nasm]$ nasm -f bin hello.asm -o hello
08:37:54 <ianni> [~/nasm]$ ./hello
08:37:54 <ianni> tcsh: ./hello: cannot execute binary file
08:38:01 <ianni> perms are ok
08:38:17 <I440r> erm dont do -f bin. do =f elf
08:38:37 <I440r> then you need to run the hello.o thru the linker (ghey)
08:38:43 <Serg_Penguin> -> I440r no, i forgto the right version of nasm
08:38:51 <I440r> serg hang on
08:39:29 <I440r> serg ftp://67.241.61.169
08:39:38 <I440r> there is a nasm gzip in there. use that one
08:40:05 <Serg_Penguin> better put binary on web
08:40:06 <ianni> I440r: ack - ghey indeed
08:40:11 <I440r> ianni you do
08:40:17 <I440r> nasm -f elf somefile.asm
08:40:19 <I440r> then you do
08:40:38 <I440r> ld -o somefile simefile.o
08:40:41 <I440r> ./somefile
08:40:49 <ianni> hello.asm:1: error: parser: instruction expected
08:40:59 <ianni> org 100h
08:40:59 <ianni> mov dx,msg
08:41:13 <I440r> even though your not actually LINKING anything to your hello.asm object you STILL need to link
08:41:16 <I440r> LAME
08:41:23 <I440r> oh
08:41:27 <ianni> haha
08:41:30 <I440r> is this a DOS executable ?
08:41:43 <ianni> I440r: oh, yeah, heh.... I guess thats not portable huh
08:41:50 <ianni> haha
08:42:07 <I440r> if its a DOS executable then -f elf is wrong hehe
08:42:10 <I440r> -f bin is correct
08:42:36 <I440r> with a .bin you need to specify the ORG 
08:42:44 <I440r> if its an elf file you shouldnt specify ORG
08:42:59 <Serg_Penguin> -> 12I440r0 tnx, noticed version, dunno wanna waste traffic now
08:43:04 <I440r> yea ?
08:43:05 <ianni> It's a DOS hellow world, I440r 
08:43:07 <ianni> -w
08:43:13 <Serg_Penguin> -> I440 tnx, noticed version, dunno wanna waste traffic now
08:43:19 <Serg_Penguin> ( colors glitched )
08:43:48 <I440r> ianni dont need to ld -o it
08:44:06 <Robert> Eeew...
08:44:07 <Robert> +c :P
08:44:11 <I440r> serg you can grab that one off of me or just get the latest from nasm.sourceforge.net
08:45:01 <ianni> should I learn on NASM? seems appropriate
08:45:09 <XeF4> you reopened your sourceforge account?
08:46:01 <I440r> no
08:46:04 <I440r> nasm did tho
08:46:25 <I440r> i dont like having to babysit a shell account like sf.net
08:46:33 <I440r> way too much management is needed to use it
08:46:33 <XeF4> bwah! stress bad, decent eyesight good
08:46:39 <XeF4> I saw 'isforth.sourceforge.net' for some reason
08:46:44 <I440r> heh
08:47:17 <Robert> I440r: Be proud! I IsForth to demonstrate Forth code, and the IsForth page to demonstrate how ugly you can make a website.
08:47:23 <Robert> I use Isforth*
08:47:26 --- quit: XeF4 ("sauna")
08:47:55 <I440r> lol
08:48:04 <ianni> hhaha
08:48:08 <I440r> isforths web page isnt ugly, ist MINIMAL :)
08:48:24 <Robert> Minimal pages aren't in pink.
08:48:36 <Robert> They're black & white/grey.
08:48:44 <I440r> its NOT pink
08:48:45 <Robert> NOT pink and purple.
08:48:46 <I440r> its PURPLE
08:48:48 <Robert> Whatever...
08:48:52 <I440r> your color-blind :P
08:49:00 <Robert> I am color-blind, yeah..
08:49:06 <I440r> i know heh
08:49:26 <Robert> Yup.
08:49:38 <Robert> Damn hard to sort resistors :P
08:49:46 <I440r> lol
08:49:56 <I440r> its hard to do that even if your NOT colorblind :P
08:50:42 <Robert> Well, s/hard/fucking impossible/
08:51:19 <I440r> ianni did you get that hello world to assemble ?
08:51:36 <I440r> rename the resultant .bin file to hello.com
08:51:40 <I440r> and try execute it
08:52:00 <ianni> No
08:52:22 <I440r> erm paste the asm here
08:52:48 <ianni> I can't run it on a linux box
08:52:53 <ianni> right? 
08:53:10 <ianni> int 21h is gonna be diff on a linux box than a dos box even tho nasm is portable, righT?
08:53:49 <ianni> hehe
08:53:50 <I440r> erm your doing it as a dos executable in linux ?
08:53:59 <ianni> heh
08:54:00 <ianni> right...
08:54:04 <ianni> looking for source code now
08:54:10 <I440r> ok. take the ORG out
08:54:14 <I440r> and do -f elf
08:54:16 <ianni> ok
08:54:21 <I440r> then ld -o hello hello.o
08:55:04 <I440r> oh
08:55:10 <tathi> I440r: he has source to a hello world program for DOS, but he's on a linux box I think
08:55:10 <I440r> erm it STILL wont work heh
08:55:16 <I440r> yea 
08:55:21 <I440r> you need to do it different
08:55:31 <I440r> you cant use interrupt 21 in linux
08:55:35 <ianni> ok good
08:55:39 <ianni> I understand completely
08:55:45 <ianni> I feel like a retard for picking this
08:55:48 <I440r> you need to do a sys write to standard out
08:55:51 <ianni> but this is my first real try at asm :)
08:55:59 * ianni finds some more source to steal
08:56:06 <I440r> heh
08:56:48 <ianni> woot - 
08:56:48 <ianni> [~/nasm]$ ./hello
08:56:48 <ianni> Hello, world!
08:56:49 <ianni> :)
08:57:08 <ianni> now where are the docs I should be reading ?
08:57:33 <ianni> ill start with a linux tutorial i guess 
08:57:57 <I440r> well if you want to do asm for linux then www.linuxassembly.org is good
08:57:57 <ianni> this rules..
08:58:00 <Robert> The Linux kernel source!
08:58:04 <ianni> aoready therem I440r 
08:58:08 <ianni> already there,    rather
08:58:13 <I440r> but most of that assumes you know asm
08:58:21 <I440r> you need a good book on x86 assembler
08:58:26 <I440r> unfortunatly theres only one
08:58:31 <tathi> what's that?
08:58:42 <I440r> the 8086 book (includes 8088) by intel
08:58:47 <I440r> i can get you an sibn if you want
08:58:50 <I440r> isbn i mean
08:58:54 <I440r> its an OLD book 
08:59:09 <I440r> but its the ONLY fucking book worth reading on x86 assembly
08:59:20 <I440r> getr a more modern book and you will have a 5 inch book
08:59:23 <ianni> there wasa decent pdf on HLA I read last year
08:59:32 <I440r> the first 2 inches are on the differences between masn, tasm and spazzum switches
08:59:33 <ianni> I440r: word...
08:59:46 <I440r> the next 1.5 inces is on binary, octal and decimal
08:59:52 <tathi> yup yup :(
08:59:57 <I440r> etc, etc, etc
09:00:08 <I440r> next to fucking useless for learning assembler
09:00:13 <I440r> do you have an actual dos machine ?
09:00:14 <tathi> and the last few pages are a crappy reference :)
09:00:17 * Serg_Penguin tosses x86 out a window
09:00:37 <Serg_Penguin> what commercial CPU has fewest commands set ?
09:00:40 <ianni> I don't use DOS
09:00:44 <ianni> I use unix
09:00:51 <I440r> your sane heh
09:00:55 <I440r> thats not acceptable :P
09:01:34 <I440r> serg dont know. do you mean processor or controller.
09:01:43 <ianni> I440r: lol really..
09:01:44 <I440r> the 8051 instruction set isnt HUGE
09:01:54 <ianni> Fuck DOS. I dont care about it at all.
09:01:59 <ianni> x86 - *maybe*
09:02:06 <I440r> also the msp-430 instruction set is REALY cool
09:02:09 <ianni> but ppc is what I use mainly for the last 2 years
09:02:12 <Serg_Penguin> processor for all-purpose compueter
09:02:22 <Serg_Penguin> ppc - mac or amiga ?
09:02:33 <I440r> amiga never used ppc
09:02:38 <I440r> it used 68k
09:02:47 <ianni> Mac
09:03:23 <Serg_Penguin> but amiga used PPC 'accelerator' - plug-in card what replaces CPU and part of chipset
09:03:43 <Serg_Penguin> where is even linux for it
09:04:21 <ianni> 68k?
09:04:26 <ianni> can use netbsd i think
09:05:02 <Serg_Penguin> amiga rulez, i even think of buying one, but two computers are too cumbersome, and Amiga has difficulties w/ Svga CRT
09:05:47 <I440r> serg i have THREE computers 
09:06:01 <I440r> my server is an athlon 700 mhz
09:06:05 <Serg_Penguin> it's native mode is TV, some progs glitch on svga
09:06:10 <ianni> i have 2 at home
09:06:15 <I440r> my workstation (currently booted to windows 98) is a 900 mhz athlon
09:06:22 <I440r> my laptop is a 550 hmz k6-3 :)
09:06:28 <ianni> 1.6ghz athlon
09:06:32 <ianni> 400mhz ppc
09:06:56 <ianni> and enough parts to make a  500 kk6/2
09:06:59 <ianni> k6/2
09:07:13 <Serg_Penguin> but i gotta only 13 m^2 of lifespase, w/ angry and dirty mother
09:08:25 <Serg_Penguin> just no place to store all boxes...
09:09:06 <Serg_Penguin> 4swallowed u'r tongues ? :(
09:09:58 <ianni> haha
09:10:31 <I440r> yea but it didnt taste good so i spat it out :P
09:10:47 <Serg_Penguin> :)
09:11:47 <Serg_Penguin> and i was thinking about becoming ham radio man...
09:11:50 <Serg_Penguin> :)
09:12:44 <I440r> ham radio would be cool
09:13:39 <Serg_Penguin> but hams were first 'spies' at Stalin time, so i'd better not say Big Brother what i can do such things...
09:15:58 <Serg_Penguin> building cool recievers to compete w/ soviet jammers was popular at time of USSR
09:16:42 <I440r> :)
09:17:49 <Serg_Penguin> thanx to Germany, USSR didn't seized all Europe and was finally beaten down in Cold War
09:18:55 <ianni> \ forth can use any damned base it wants to
09:18:55 <ianni> hehehe
09:19:16 <Serg_Penguin> XXI U-boten w/ soviet crews could cut USA's throat, coz USA heavily depends on 'third world' resources
09:20:08 <Serg_Penguin> ( only at 80-s USSR built a submarine outclassing German XXI series )
09:20:33 <Serg_Penguin> but germans built whem at 43
09:20:36 <ianni> if someone cut USAs throat im sure we'd shoot lots of missles at em
09:20:41 <ianni> thatd be nasty
09:22:24 <Herkamire> heh :) I've never seen the background color on the isforth site. I have bgcolor turned off in my user-friendly browsers (dillo, lynx)
09:22:35 <I440r> heh
09:22:57 <Serg_Penguin> missiles ? at WWII ?
09:23:15 --- quit: Herkamire ("lunchtime")
09:23:33 <ianni> I like links
09:23:48 <Stepan> re
09:24:30 --- part: Serg_Penguin left #forth
09:27:24 <ianni> re
09:37:21 <ianni> getenv Segmentation fault (core dumped)
09:37:21 <ianni> [~/isforth-1.09b]$
09:37:33 <ianni> how do I do strings in isforth?
09:40:19 <ianni> ."   " doesnt print?
09:41:17 <I440r> you cant use ." blah balh" in interactive mode
09:41:26 <I440r> the word ." is a compiling word. 
09:41:29 <I440r> you have to do
09:41:36 <I440r> : foo ." blah blah" cr ;
09:41:37 <I440r> foo
09:41:40 <I440r> you CAN do
09:41:44 <I440r> .( blah blah blah)
09:42:53 <ianni> Got it.
09:43:05 <ianni> what's ," ?
09:44:48 <I440r> that compiles a string at HERE
09:44:52 <I440r> a counted string
09:44:56 <I440r> in the form
09:45:04 <I440r>  db length,'xyzzy'
09:47:11 <ianni> k
09:47:37 <I440r> you could do
09:47:45 <I440r> create foo ," xyzzy"
09:47:49 <I440r> foo count type
09:47:49 <ianni> im not worried about string
09:48:01 <ianni> I need to learn file i/o so I can make a useful forth pogram
09:48:05 <ianni> or stdio at least
09:48:08 <ianni> same thing anyway im sure
09:48:47 <ianni> what's dot ?
09:49:22 <ianni> I dont understand how these lines work either
09:49:23 <ianni> fload forthsrc/comment.f  dot   \ + commenting words
09:49:40 <ianni> I'd expect something like ." forthsrc/comment.f" dot fload
09:49:41 <ianni> ?
09:52:32 <I440r> ok
09:52:35 <I440r> \ is a comment
09:52:36 <I440r> but
09:52:44 <I440r> its NOT defined within the kernel
09:53:00 <I440r> its in the extensions. what you are looking at is isforth.f which includes all the extensions
09:53:08 <I440r> comment.f defines \
09:53:19 <ianni> everythign to the left of \ I was looking at  :)
09:53:24 <ianni> I understand \, as you mean it..
09:53:36 <I440r> it also defines a word called DOT that i invoke on every included file
09:53:46 <I440r> fload path/to/file.f dot
09:53:54 <I440r> executes the fload then executes DOT
09:54:19 <ianni> but how does the kernel know the word path/to/file.f
09:54:23 <ianni> thats what I dont get atm
09:54:45 <I440r> ok
09:55:00 <I440r> forths interpreter extracts ONE space delimited string at a time
09:55:02 <I440r> FLOAD
09:55:16 <I440r> it then searches the dictionary for fload and executes it
09:55:23 <ianni> interesting
09:55:39 <I440r> fload takes the next space delimited string (the file name) 
09:55:47 <I440r> and INTERPRETS that file
09:55:55 <ianni> awesome
09:57:53 <ianni> so is fload written in  kernel?
09:58:00 <ianni> is it in the kernel
09:58:03 <ianni> (ie not forth)
10:12:52 <I440r> fload is written into the kernel
10:12:57 <I440r> had to be
10:13:04 <ianni> makes sense
10:13:26 <ianni> all I know is basic math
10:13:31 <ianni> I need to start doing something useful
10:13:40 <ianni> maybe I should start learning struct-like things
10:15:48 <I440r> no
10:15:59 <I440r> just learn stack manipulation and basic stuff first :)
10:16:01 <I440r> swap
10:16:02 <I440r> dup
10:16:05 <I440r> drop
10:16:05 <I440r> nip
10:16:07 <I440r> over
10:16:16 <I440r> + - * / . 
10:16:46 --- join: Herkamire (~jason@wsip68-15-54-54.ri.ri.cox.net) joined #forth
10:18:22 <ianni> I440r: I got all that
10:18:24 <Robert> Hello.
10:18:29 <Robert> ianni: Most important: bye
10:18:55 <ianni> I440r: now I'm at a point where I need to start learning to use stdio and other such things so that my forth can actually become useful in computing outside of doing mathematical computation
10:19:12 <ianni> or to solve higher level problems
10:19:14 <ianni> i ngeneral
10:19:28 <I440r> well you can read fronm stdin with key and ?key
10:19:38 <I440r> and write to stdout using emit or ." xyzzy"
10:19:44 <ianni> simple.
10:19:47 <ianni> I should o that now..
10:19:52 <ianni> doing
10:20:03 <ianni> oh, what do I do to get a standalone app
10:20:11 <ianni> so I can pipe to stdin for ex
10:20:24 <ianni> run isforth with a .f with a bye in it ?
10:20:43 <ianni> :)
10:21:12 <I440r> you CAN pipe to stdin
10:21:19 <I440r> or
10:21:25 <I440r> look at test.f
10:21:25 <I440r> thats +x
10:21:30 <ianni> I440r right but im used to running/writing forth interactively
10:21:30 <I440r> it has 
10:21:34 <ianni> I440r: great, thanks
10:21:40 <I440r> #! ./isforth - fload
10:21:40 <I440r> as the first line
10:21:40 <I440r> you can 
10:21:40 <I440r> ./test.f
10:21:43 <I440r> and execute it!
10:21:58 <ianni> yeah
10:22:00 <ianni> Got it..
10:22:00 <ianni> :)
10:22:01 <ianni> Loading fortran...done
10:22:04 <ianni> damn emacs!
10:22:13 <I440r> lol
10:22:36 <ianni> beautiful :)
10:22:50 <I440r> what is ?
10:22:51 * ianni likes isforth
10:23:04 <I440r> its a work in progress still
10:23:07 <ianni> I440r: i dunno, i had never been able to do even use my own file before. I am a total n00b
10:23:12 <I440r> need a debugger stioll
10:23:27 <I440r> heh so was i once :)
10:24:18 <ianni> my terminal is getting messed up
10:24:20 <ianni> heh
10:24:37 <I440r> with what ?
10:24:52 <ianni> LF and or CR's
10:26:17 --- quit: Stepan ("Do you think it is air you are breathing? Hmm?")
10:26:46 <tathi> ianni: I've never been able to convince emacs that .f doesn't mean Fortran.
10:27:04 <tathi> finally gave up and switched over to vim :)
10:28:12 <I440r> ugh
10:28:15 <I440r> use joe
10:28:19 <I440r> vi and emacs are evil
10:28:24 <I440r> i mean REALY fucked up
10:28:35 <I440r> actually so is joe, joe is NOT a good editor
10:28:47 <I440r> its just the lesser of 284975623789654237849 evils
10:29:22 <tathi> sorry
10:29:23 <ianni> tathi: Hahah, really? it's like one line in your .emacs
10:29:27 <I440r> heh
10:29:27 <tathi> I'm thoroughly enjoying vim
10:29:29 <ianni> lemme see how to do it
10:29:46 <tathi> hmm. I never figured it out
10:30:59 <ianni> tathi, emacs is very configurable
10:31:22 <tathi> I know it is. But after a while I got tired of it taking me 3 hours to figure out how to configure it.
10:31:36 <tathi> (every time I wanted to change one little thing)
10:32:00 <tathi> with the .f thing, I got as far as finding an array that appeared to map file extensions to modes
10:32:18 <ianni> uck
10:32:23 <ianni> I dont understand all that crap
10:32:27 <tathi> and managing to set it local to one buffer, but couldn't manage to set it globally
10:32:27 <ianni> I just use the internet to tell me what to do
10:33:00 <tathi> ah, maybe that was my problem -- I was just using the help and searcing through the info docs
10:33:04 <tathi> searching
10:33:06 <ianni> yeah
10:33:12 <ianni> I dont ever expect to know that crap
10:33:19 <ianni> I just like that once I do figure it out, I just put it in .emacs
10:33:25 <ianni> btw, did you try the online configuration?
10:33:42 <ianni> that can do stuff like that many rimes
10:33:46 <ianni> M-x customize
10:33:49 <tathi> I found that rather frightening
10:33:51 <ianni> haha ok
10:33:57 <ianni> well I'll let ya know when I find it
10:33:58 <ianni> hehe
10:34:14 <tathi> and it was too deep of a heirarchy, always took a while to figure out what section I wanted
10:34:45 <tathi> ok, I'd be interested to see
10:36:10 <ianni> yeah
10:36:13 <ianni> it's a big confusing mess
10:36:21 <ianni> but it works ...
10:36:44 <ianni> if vim could be configured as well as emacs maybe i'd use it then
10:37:08 <Robert> emacs is confusing.
10:37:17 <I440r> vi has a fucked up user interface
10:37:23 * Robert uses vim when he has to, otherwise normal editors.
10:37:24 <I440r> so does joe, but vi is MUCH worse
10:37:32 <Robert> Hehe.
10:37:39 <I440r> escape colon shift azlt backspace backspace f10
10:37:41 <Robert> DOS edit.com forever!
10:37:48 <I440r> now your in insert mode
10:37:57 <Robert> More like... "i" :P
10:38:03 <I440r> shift alt control f5 delete enter
10:38:06 <I440r> now your in overwrite mode
10:38:13 <I440r> ok im exaggerating SLIGHTLY for effect
10:38:14 <Robert> More like "insert" ;)
10:38:16 <Robert> Hehe.
10:38:21 <Robert> Slightly.. Hah :)
10:38:32 <I440r> but. any editor that takes more than TWO fucking keypresses for ANY fucking operation is a pile of shit
10:38:38 <tathi> I like the modal thing, personally
10:38:42 <I440r> and that includes the use of shift, alt or control
10:38:45 <tathi> heh
10:38:45 <I440r> control X
10:38:47 <I440r> alt Y
10:39:05 <Herkamire> simple programs for simple minds
10:39:06 <tathi> I don't think I have enough modifier keys to fit everything I want to do into two keypresses
10:39:23 <Herkamire> ditto
10:39:34 <tathi> no, actually, that's not true
10:39:36 <Herkamire> pluss my terminal only supports two modifiers
10:39:53 <tathi> I just want a simple way to record longer sequences easily and then repeat them
10:39:58 <tathi> which vim has
10:40:04 <I440r> thats all that should EVER be used
10:40:14 <I440r> the 'command mode' is fucked up
10:40:24 <Herkamire> you're fucked up
10:40:30 <I440r> joe is bad, you have to hit control K something-else
10:40:34 <Herkamire> command mode is what lets you do almost anything in two characters!!!
10:40:36 <I440r> every time you want to do something
10:40:40 <I440r> control k b
10:40:42 <I440r> starts a block
10:40:49 <I440r> control k k ends it
10:40:54 <I440r> you should be able to do 
10:41:00 <I440r> move cursor to start of block
10:41:01 <I440r> hit shift
10:41:04 <tathi> see, I really like the command mode. It's nice to be able to use single-letter commands when I just want to make edits as opposed to entering text
10:41:06 <I440r> scroll down to end of block
10:41:08 <I440r> now its marked
10:41:27 <I440r> tathi how do you DELETE a line in vi ?
10:41:32 <tathi> dd
10:41:38 <Herkamire> I440r: or in vim, you  could hit v, and scroll down to the end of the block. then it's highlited.
10:41:40 <I440r> escape colon [what goes here?}
10:42:02 <Herkamire> I440r: how about you learn how to use it!!
10:42:03 <tathi> I440r: why are you acting like insert mode is the "normal" mode?
10:42:11 <I440r> herk but if your not in the right mode you have to fucking switch to it. that counds as extra keypresses for me
10:42:24 <ianni> I hate command mode - it's niec to be able to have the same flexibility in a stateless environment.
10:42:24 <I440r> herk the point is. there should be ONE mode.
10:42:26 <I440r> EDIT mode
10:42:30 <ianni> I440r++ !
10:42:35 <I440r> and ALL functions should be available there
10:42:42 <I440r> insert, delete, mark, move, copy
10:42:46 <I440r> etc, etc,et c
10:42:47 <tathi> I440r: other thing is, vim makes it supremely easy to remap the keyboard
10:42:53 <Herkamire> I440r: you can't have both. if there is only one mode, then all your key sequences will be longer.
10:43:06 <I440r> tathi obviously has never used multiedit in dos
10:43:12 <tathi> bleh
10:43:19 <tathi> just kidding
10:43:25 <tathi> multiedit is pretty nice
10:43:26 <ianni> tathi:
10:43:26 <ianni> (setq auto-mode-alist
10:43:27 <ianni>       (append
10:43:27 <ianni>         '(("\\.f$"     . fundamental-mode  )
10:43:27 <ianni>                      ) auto-mode-alist))
10:43:31 <ianni> there ya go
10:43:40 * ianni knows ZERO lisp
10:43:41 <ianni> btw\
10:43:42 <ianni> :)
10:43:54 <tathi> ianni: in my emacs that added the .f to the end. it still thought .f was fortran because that came _way_ earlier in the list
10:44:02 <I440r> i use multi edit version 3.01
10:44:13 <ianni> tathi: this works for me
10:44:21 <tathi> ianni: cool cool
10:45:10 <tathi> I440r: I don't remember -- does multi-edit have some kind of a scripting language?
10:45:11 <Herkamire> I440r: what is this should/shouldn't about modes in editors. I think an editor _should_ be easy to use, configure and get things done fast. I don't care about modes, and to be honest, I didn't like the idea at first. but I get things done much quicker now with vim.
10:45:48 * ianni likes that M-x customize will configure his .emacs automatically, and it's a no-brainer to use the UI to M-x customize
10:45:57 * ianni likes not having to remember modes while editing
10:46:03 <Herkamire> and editor _should_ be fast. vim achieves this by using modes. I say the more power to us.
10:46:07 <ianni> the latters why I use emacs mainly 
10:46:23 <tathi> ianni: yeah, modes _definitely_ aren't for everybody
10:46:38 <ianni> emacs doesnt have to use modes to do it though.. I see that as superior design, personally
10:46:41 <ianni> but
10:46:49 <ianni> I am not one of these people who have used vi+regexp for 10 years
10:46:50 <ianni> :)
10:47:15 <Herkamire> I don't see a big difference between hitting esc in vim and hitting M-x in emacs.
10:47:23 <Herkamire> or C-k in joe
10:47:43 <ianni> me neither
10:47:43 <Herkamire> both are a mode.
10:47:52 <Herkamire> just that vim it stays for more than one command
10:47:55 <ianni> in emacs it's decoupled from the editing though
10:48:10 <Herkamire> who cares?
10:48:16 <Herkamire> I don't want to do anything but edit with my editor
10:48:29 <Herkamire> (well, and scroll and search)
10:48:30 <ianni> exactly 
10:49:11 <Herkamire> I'm sure you could bind a key in vim to work just like M-x in emacs
10:49:11 <tathi> ianni: I've been using vim for just about a month :)
10:49:32 <tathi> but anyway, personal preference. I didn't mean to start a religious war about editors :)
10:49:36 <ianni> I guess it's just preference. I can't wriet regexp well, so I use emacs :)
10:49:48 <ianni> but if i want to use regexp in emacs I can
10:49:53 <ianni> yeah, this is the wrong place for the emacs vs vi fight
10:49:56 <ianni> sorry :)
10:50:15 <tathi> and it's really the lisp in emacs that I object to, not the modes thing
10:50:22 <Herkamire> I switched from emacs (which I had been using intencely for about a year) to vim, and after about a week just as fast as I was with emacs, and after two weeks I was quite a bit faster.
10:50:45 <ianni> tathi: I do not use or know lisp. I just use emacs cause it works
10:51:09 <Herkamire> ianni: heh :) that's why I used emacs too.
10:51:15 <Herkamire> untill I found something that worked better :)
10:52:36 <ianni> right on
10:52:40 <ianni> I'm just piseed that..
10:52:53 <ianni> I can't do en masse edits in a mode-ful editor w/o using regexp
10:53:00 <ianni> or macros
10:53:25 <ianni> until now it seems
10:53:34 <ianni> misfire
10:54:13 <ianni> like I had to convert "a b c d e .........." to "a:b:c:d:e:f ...." the other day
10:54:17 <ianni> and I just wanted to freakin do it normally
10:54:24 <ianni> but i couldnt
10:54:33 <ianni> what vim needs is some kind of emacs-like mode
10:54:42 <ianni> heh
10:54:45 <ianni> pico like mode
10:54:57 <tathi> yeah, why not? I mean, emacs has ViPER or whatever it's called...
10:55:09 <ianni> yeh
10:55:12 <ianni> exactly
11:25:01 <ianni> I440r: like if I run test.f it doesnt do CR or LF at all
11:25:05 <ianni> do i need term=linux i wonder
11:25:35 <ianni> ahhh. twas my Term client :)
11:25:38 <ianni> sorry.
11:29:45 <ianni> there's no page up is there?
11:32:54 <Herkamire> funny how 4.6MB of the emacs 21.1 tar.gz file is elc files... when the source for those is in there too.
11:34:04 <ianni> heh
11:34:11 <ianni> nie to not haver to compile i guess ...
11:41:41 <Herkamire> if you don't want to compile, you don't download the source...
11:46:16 <ianni> good point...
11:46:19 <ianni> there may be a reason for it
11:50:11 <Herkamire> possible
11:51:17 <Herkamire> but I don't think it's worth it. between removeing those and switching to bzip2, it can take it from 19.8MB to about 12MB
11:51:41 <Herkamire> that may make it small enough that dial-up people would consider downloading it.
11:53:01 <ianni> haha
11:53:03 <ianni> really
11:54:53 <Herkamire> some :) some dialup people
12:01:29 --- join: XeF4 (xef4@lowfidelity.org) joined #forth
12:08:29 <ianni> no im agreeing "really."
12:15:05 <Herkamire> ok, time to write block support for fpos
12:18:47 <ianni> ok so why doesnt this work, giving stdin:
12:18:47 <ianni> while ?key
12:18:48 <ianni>         key emit
12:18:48 <ianni> repeat
12:18:56 <ianni> god i know thats porbably so wrong
12:19:10 <ianni> but i need guidance
12:19:11 <ianni> haha
12:24:19 <tathi> ok, 1) you have to put it in a word, the control-flow structures in forth usually don't work at interpret time
12:25:02 <tathi> 2) the condition would go before the loop thingy
12:25:35 <ianni> ahhhhhhhh
12:25:39 <ianni> (1)..
12:25:50 <ianni> stupid stupid :)
12:25:59 <ianni> right on
12:26:22 <tathi> does that work then?
12:26:25 <tathi> what forth are you using?
12:26:56 <ianni> isforth
12:27:08 <Herkamire> : echokeys repeat key emit ?key while ;
12:27:42 <Herkamire> something like that should work
12:27:47 <ianni> usgh
12:27:54 <ianni> I have while    THEN repeat
12:28:00 <I440r> no
12:28:03 <I440r> no then
12:28:06 <I440r> begin
12:28:07 <I440r>   do test
12:28:15 <I440r>  while (while test is true)
12:28:18 <I440r>   do all this
12:28:20 <I440r>  repeat
12:28:37 <ianni> ok
12:28:53 <I440r> or
12:28:58 <I440r> begin
12:29:07 <I440r>  do usefull stuff
12:29:10 <I440r>   do test
12:29:15 <ianni>  echokeys begin do ?key while key emit repeat ;
12:29:19 <ianni> whats wrong ?
12:29:20 <I440r> unitl  \ repate until test is true
12:29:38 <Herkamire> ianni: remove "do"
12:29:54 <ianni> hmm nope
12:30:12 <Herkamire> what happens?
12:30:16 <I440r> betgin
12:30:17 <I440r>   key?
12:30:18 <ianni> oh "do test"
12:30:20 <I440r> while
12:30:23 <ianni> misread :)
12:30:24 <I440r>   key emit 
12:30:25 <I440r> repeat
12:30:37 <I440r> but that will exit IMMEDIATLY
12:30:43 <I440r> being
12:30:47 <ianni> : echokeys begin ?key while key emit repeat ;
12:30:47 <ianni> is what I have..
12:30:49 <ianni> and 
12:30:49 <ianni> [~/isforth-1.09b]$ echo "hell world" | ./key.f
12:30:49 <ianni>  ?key ?
12:30:57 <I440r>   key dup $1b <>
12:31:00 <I440r> erm
12:31:01 <I440r> begin
12:31:08 <I440r>   key dup $1b <>
12:31:10 <tathi> oh
12:31:10 <I440r> while
12:31:13 <I440r>   emit
12:31:15 <I440r> repeat 
12:31:16 <I440r> drop
12:31:24 <I440r> 1b is an escape
12:31:28 <ianni> wtf :)
12:31:31 <I440r> you could change that to $0d if you like
12:31:59 <tathi> I440r: key? or ?key
12:32:10 <ianni> oh, right
12:32:13 <ianni> it's complaining on that word
12:32:15 <ianni> I jsut realized :P
12:32:30 <ianni> there we g
12:32:30 <ianni> o
12:32:52 <ianni> infinite loop yay
12:33:01 <ianni> What is the escape and why am I using it?
12:33:10 <ianni> what are $1b and $0d?
12:33:11 <I440r> er i forget which is is heh
12:33:14 <ianni> especially?
12:33:15 <I440r> ?key i think
12:33:16 <ianni> specially I mean
12:33:19 <ianni> specifically I meant
12:33:23 <I440r> $1b is an escape
12:33:26 <I440r> $0d is a cr
12:33:46 <ianni> oh the key.
12:33:51 <ianni> why do i need it
12:33:57 <tathi> print out keys until you get an escape
12:34:01 <ianni> and whats <> ?
12:34:01 <I440r> you need to have a way OUT of the loop
12:34:06 <Herkamire> what good is cr?
12:34:10 <ianni> er
12:34:11 <Herkamire> doesn't linux use lf?
12:34:11 <I440r> <> is not equals
12:34:15 <ianni> I44...
12:34:21 <ianni> isnt that what "key?" is for?
12:34:28 <I440r> no
12:34:30 <ianni> (guess not :)
12:34:38 <I440r> key? just tells you if ANY key has been presssed
12:34:42 <ianni> ok well
12:34:43 <Herkamire> ianni: you're right
12:34:44 <I440r> key READS the keyboard
12:34:47 <I440r> key? is non blocking
12:34:50 <I440r> key is blocking
12:34:57 <ianni> I want to echo "foo foo" | ./key.f
12:35:05 <I440r> : key begin key? until (key) ;
12:35:07 <ianni> w/ an indweterminate string end
12:35:08 <Herkamire> I440r: ianni is reading from the keyboard untill he get's it all. when key? fails he exits
12:35:12 <I440r> is a BAD way to do key in linux
12:35:17 <I440r> where (key) reads stdin
12:35:28 <I440r> thats wrong
12:35:29 <ianni> I440r: that def. of key? makes sense, ok,
12:35:33 <I440r> key? will fail immediatly
12:35:49 <I440r> try this
12:35:55 <ianni> I440r: so there is no clean way to write echo "any string w/o special termination" | myfile
12:36:00 <I440r> : foo begin key? not until ;
12:36:01 <I440r> foo
12:36:02 <ianni> ? :)
12:36:04 <ianni> ok
12:36:08 <I440r> foo will wait for a NO key condition
12:36:13 <I440r> i.e. it will exit immediatly
12:36:31 <ianni> it looks to never exit to me
12:36:42 <ianni> if I run like this:  echo "hell world" | ./key.f
12:36:42 <I440r> try this
12:36:43 <I440r> : foo
12:36:49 <I440r> begin key? until ;
12:36:56 <I440r> THAt wont exit until you press a key
12:37:03 <ianni> it's happenign the other way around :)
12:37:04 <I440r> begin key? while blah repeat 
12:37:07 <ianni> as you've just said
12:37:08 <I440r> will exit IMMEDIATLY
12:37:12 <ianni> the one with not until hangs
12:37:41 <ianni> but acts as you say if I run w/o piping 
12:41:58 <ianni> whgeres moonphases
12:42:42 <I440r> shouldnt hang
12:42:48 <I440r> should exit on first keypress
12:43:05 <I440r> erm or am i getting confused now heh
12:43:53 <ianni> dude, I got confused a long time ago
12:43:54 <ianni> :D
12:46:09 <Fractal> I440r : I put the terminal in CANONical mode, how do you do it?
12:46:20 <ianni> i dunno
12:46:26 <ianni> normal rh linux bash shell
12:46:51 <Fractal> Er, I'm talking about the word key
12:47:23 <Fractal> BTW, frugal has a powerful debugger now. It's l33t.
12:47:47 <I440r> cool
12:47:52 <Fractal> Breakpoints, stepping over instructions, stepping into instructions. It's fun stuff.
12:48:00 <ianni> oh misread
12:48:02 <I440r> yes
12:48:09 <I440r> im going to do one for isforth
12:48:29 <I440r> with ability to single step coded defs too
12:48:30 <Fractal> You can take a look at mine soon. I'm just writing the documentation for it, then I'll release it.
12:48:48 <I440r> :)
12:50:02 <Fractal> What's really great is that the debugger uses query and interpret for the interactive modes, so you can do anything you can at the forth interpreter inside the debugger. Very useful if you want to find out what a variable's value is and such.
12:51:51 <Fractal> I440r : BTW, I implemented locals if you want to take a look at them too.
12:52:37 --- quit: XeF4 ("Z")
12:52:40 <ianni> arent locals Bad?
12:52:48 <I440r> erm no thanx, im sorta against them, i could implement them np
12:52:48 <Fractal> Well, it's debatable.
12:53:10 <I440r> its being debated heh
12:53:21 <Fractal> I think it's a good idea to have them even if you don't use them though.
12:53:46 <Fractal> They can simplify a lot of stack kung-foolery in certain instances.
12:54:07 <Fractal> The best part of them though is the dynamic namespace.
12:55:37 <I440r> but if you need them to sort out the stack manipulations you need to rethink your code anyway so...
12:57:31 <Fractal> Not necessarily. They're great if, say, you have to be passed 6 arguments in order to be reentrant. You can then fork out many small words for handling those variables.
12:58:13 <Fractal> I use kind of an interesting approach. When you start locals, it creates a new word ";", which when called, closes the definition, and does "forget ;"
12:58:35 <Fractal> Maybe that's how other people do it, I dunno.
12:59:17 <Fractal> Maybe they vector ";"... Hm...
13:00:12 <Herkamire> factor factor factor
13:00:18 <I440r> most forths just give you a way to NAME the varilus stack items
13:00:33 <I440r> | local0 local1 local2 |
13:00:47 <I440r> you can now do local0 @ or local0 ! etc
13:01:00 <I440r> and your manipulating the param stack items
13:01:12 <I440r> not sure how good it works if you do a rot or a drop heh
13:01:38 <Fractal> Hm. I do mine on the return stack, so the param stack is clear.
13:01:57 <I440r> :)
13:02:07 <I440r> might be better to use a dedicated locals stack
13:02:08 <Fractal> Mine might look like this:  : my_word l{ x y z } x + y - ;
13:02:14 <I440r> that way doing >r and r> wont mess u up
13:02:21 <Fractal> Meh. :)
13:03:59 <Fractal> I just say "do not access locals in a do loop" and "do not access locals if you alter the return state until you restore it"
13:04:14 <Fractal> s/state/stack
13:04:15 <ianni> ah
13:04:52 <Fractal> If you wanna use a do loop in a locals word, you can, just don't access locals inside it. Press em on the stack first or something.
13:06:29 <ianni> how do I do a character literal in isforth
13:06:46 <Fractal> I think it's 'c'
13:06:53 <I440r> yes
13:06:59 <I440r> works with every character except space
13:07:09 <ianni> great
13:07:11 <I440r> ' ' is interpreted as tick tick
13:07:16 <ianni> so wait
13:07:17 --- quit: jacereda (Read error: 110 (Connection timed out))
13:07:18 <Fractal> Use bl for space
13:07:19 <I440r> which returns the cfa of the word '
13:07:24 <ianni> n/m
13:07:26 <I440r> yes
13:07:27 <ianni> fuck it
13:07:33 <I440r> 'x' 'y'
13:07:37 <I440r> for every other char
13:07:45 <I440r> no [char] x
13:07:51 <I440r> thats BUTT UGLY
13:07:52 <ianni> nah i was gonna ask how t o get wqhat wer wer talking about earlier to work
13:07:55 <ianni> im gonna try some more
13:07:57 <I440r> no ascii y
13:07:57 <Fractal> I440r : Does that work for both interpretation and compilation?
13:08:02 <ianni> yeah, 'x' is just how i would want
13:08:10 <I440r> fractal what do you mean
13:08:12 <ianni> perfect
13:08:13 <Fractal> 'x'
13:08:17 <I440r> oh
13:08:22 <I440r> 'x' emit
13:08:35 <I440r> in interactive mode emits the x
13:08:36 <I440r> : foo 'x' emit ;
13:08:40 <I440r> compiles 'x' as a literal
13:08:47 <I440r> because 'x' is like $45  
13:08:55 <ianni> ok so I cant use this: : foo begin emit key? not until ;
13:08:56 <I440r> handled by NUMBER
13:09:00 <I440r> number looks for 
13:09:01 <ianni> because key? will be 0 first?
13:09:06 <I440r> \xxx  as an octal number
13:09:15 <I440r> $yyy as a hex number
13:09:23 <I440r> and 'z' as a character literal
13:09:37 <I440r> no. you have nothing to emit
13:09:44 <Fractal> Hm. Don't you use base?
13:09:57 <I440r> yes
13:10:00 <I440r> base exists too
13:10:17 <I440r> but. you can be in ANY base and insert octal, binary and hex numbers
13:10:23 <I440r> and character literals
13:10:29 <I440r> 3 base !
13:10:35 <I440r> #20
13:10:35 <I440r> erm
13:10:39 <I440r> $20  <-- hex
13:11:19 <Herkamire> ianni: does this work?      : cat begin key? while key emit repeat ; cat bye
13:11:54 <Fractal> Yeah, I only use base. 2-36 is the range.
13:12:10 <Fractal> Don't want to complicate "number" any more than I have to.
13:12:12 <ianni> Herkamire: yeah, except for nevr exiting.
13:12:20 <I440r> i allow you to set base to anything but 2 to 36 is the only realy useful range
13:12:26 <ianni> there's gotta be a way unix programs know that stdin ended
13:12:28 <ianni> hmmmmmmmmmmmmmmmmmmmmmmm
13:12:37 <Fractal> Yeah, me too. The only displayable range, anyways.
13:12:44 <I440r> i like being able to use $xx for a hex number because its less ambiguous
13:12:58 <Fractal> Fair enough.
13:13:11 <I440r> and base is only useful for inputting and outputting nubmers so the digigs HAVE to be displayable heh
13:13:20 <I440r> so 92 base !
13:13:24 <I440r> is useless heh
13:13:28 <I440r> i jsut dont check the range
13:13:30 <Fractal> Exactly.
13:13:31 <Herkamire> ianni
13:13:34 <Fractal> Ya, me neither.
13:13:44 <Herkamire> ianni: that's probably because isforth does not support EOF
13:14:13 <Herkamire> I440r: what does "key?" return on EOF?
13:14:22 <ianni> wh0rd
13:14:30 <I440r> \s
13:14:35 <I440r> oh
13:14:45 <I440r> no. isforth doesnt knnow its being piped to yet
13:14:49 <I440r> i have to fix that
13:14:53 <I440r> it just reads stdin heh
13:15:12 <I440r> id doesnt know that stdin is a pipe yet and to quit when the data terminates
13:15:28 <I440r> but if you put a bye at the end of the piped data it will be a SORT of fix heh
13:15:40 <I440r> \s is a comment. everything else in the file is a comment
13:15:44 <I440r> usefull for debugging
13:15:59 <I440r> other than that i dont support multi line comments
13:16:01 <Herkamire> It doesn't have to quit, but key? should definately return false.
13:16:07 <I440r> \ on every line
13:16:13 <I440r> not 
13:16:13 <I440r> comment:
13:16:13 <I440r>  blah blah
13:16:13 <I440r>  blah blah
13:16:13 <I440r>  blah blah
13:16:22 <I440r> ;comment
13:16:29 <I440r> key? is none blocking and returns either true or false
13:18:09 <Fractal> I don't use key?, because what it does depends on your terminal's state. In mine you would use poll and read, which would handle EOF.
13:18:47 <I440r> key? uses poll
13:19:05 <I440r> key? uses poll and key uses read
13:19:07 <Fractal> I440r : So isforth's terminal is always in CANONical mode?
13:19:28 <I440r> but isforth doesnt knnow how to distinguish between keyboard and piped data
13:19:55 <I440r> canonical is where its in one keypress = one key returned
13:19:56 <I440r> yes ?
13:20:07 <I440r> not. read entire string then get them one at a time till the cr
13:20:33 <Fractal> Right. I think. I might have it backward.
13:20:53 <I440r> usually when you call read on a file it reads up to the next cr 
13:21:07 <I440r> then each successive call to read returns one of the already buffered characters
13:21:34 <I440r> canonical mode makes it return each character at the time the key was pressed
13:21:42 <I440r> and 'I' buffer it
13:21:47 <Fractal> No, when you call read() on a file it will read until EOF, or the max number of bytes is read.
13:22:07 <I440r> oh yea
13:22:12 <Fractal> Same with a socket. It's a terminal that will behave that way if it's in non-canonical mode.
13:22:13 <I440r> well
13:22:13 <I440r> stdin has no eof
13:22:17 <I440r> i dont believe
13:22:42 <I440r> i specifically dont want it in non canonical mode
13:23:12 <Fractal> stdin does have an EOF. For instance, when you type ^D to a piped application like cat, the terminal devices goes into EOF state.
13:23:56 <I440r> oh yea
13:24:12 <Fractal> People say everything in unix is a file, but they're wrong. Everything in unix is a device. :)
13:24:26 <I440r> not true
13:24:38 <I440r> oh heh
13:24:39 <ianni> everything in unix is a FILE i think not a file
13:24:40 <Fractal> Well, no, of course not. :)
13:24:42 <I440r> yea
13:24:47 <ianni> hehe
13:24:59 <Herkamire> I440r: fix "key?"
13:25:10 <I440r> fix it how ?
13:25:12 <ianni> nah it's ok
13:25:16 <Herkamire> your polling stuff must not be working. after you reach EOF it keeps returning true.
13:25:18 <I440r> whats wrong with it
13:25:27 --- join: Soap- (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth
13:25:34 --- quit: Soap` (Read error: 104 (Connection reset by peer))
13:25:34 <ianni> I440r: well, there IS an end to stdin
13:25:37 <I440r> all key? returns is "is there a key pressed"
13:25:38 <ianni> there has to be
13:25:46 <ianni> hmmmmmmmm
13:25:50 <I440r> there are all sorts of flags returned in the poll-fd structure that im totally ignoring
13:25:57 <Herkamire> at EOF there is NOT a key pressed
13:26:06 <Fractal> Herkamire : That's because the descriptor is NOT blocking, it's reached EOF.
13:26:34 <Fractal> The behaviour is correct, it's just that key? is not well suited to a unix environment, which is why I don't use it.
13:27:03 <ianni> Makes sense to me.
13:27:11 <Fractal> :)
13:27:15 <ianni> is key my only opetion?
13:27:19 <ianni> in isforth
13:27:34 <ianni> if so, it's ok, I'll move on - im not accomplishing anything here
13:27:36 <Fractal> I440r : You provide direct access to poll() and read(), right?
13:27:45 <I440r> no, you can use <open> <read> <write> and <close>\
13:27:50 <I440r> those are the syscalls
13:27:59 <ianni> I see
13:28:04 <I440r> there is a <poll> syscall yes
13:28:31 <I440r> plus
13:28:34 <Herkamire> Fractal: what do you think key? should return when you reach EOF?
13:28:37 <Fractal> ianni : Instead of key?, use poll(). Instead of key, use read(). When read() returns 0, you've reached EOF.
13:28:42 <I440r> if there is a syscall that you need but isnt there you ca use the word syscall to create it
13:28:55 <I440r> #parameters syscall-number syscall <syscall-name>
13:29:05 <I440r> i wrap all syscall names in < > chars
13:29:21 <Fractal> Herkamire : Both T and F are wrong in that case. There is no character waiting on the device, but the device is NOT waiting for a character.
13:30:20 --- join: tcn (tcn@tc3-login22.megatrondata.com) joined #forth
13:30:38 <ianni> Fractal: kl
13:30:40 <ianni> Fractal: k
13:30:50 <Fractal> ianni : Seriously, I wouldn't worry about it for now.
13:31:21 <Fractal> Somebody needs to come up with a better I/O API for forth in a unix environment, IMO.
13:31:25 <I440r> tcn!
13:31:35 <tcn> hey
13:32:49 <tcn> me and i440r came up with a good I/O system..
13:33:25 <I440r> its not perfect tho
13:33:34 <I440r> it needs some work heh
13:33:39 <I440r> and i changed it a little :)
13:33:47 <tcn> you just use KEY ACCEPT TYPE . EMIT etc.. and redirect stdin/out
13:33:48 <I440r> key? uses a pollfd
13:33:52 <Herkamire> Fractal: so..... which makes more sense, true of false?
13:33:55 <I440r> i dont have accept
13:34:00 <I440r> im using query and expect
13:34:04 <tcn> expect, whatever :)
13:34:08 <I440r> hehe
13:34:29 <I440r> is "accept" just expect renamed ?
13:34:46 <tcn> no.. you have to give an address & count where to store the string
13:34:54 <tcn> and preallocate it
13:35:23 <Fractal> Herkamire : Well, that's an interesting question. The unix behaviour is T, because that descriptor is considered not-blocking, so you can safely use the descriptor without blocking your process. Like I say, though, neither of them are optimal behaviours for forth.
13:35:54 <Fractal> Well, let me rephrase: Neither of them are optimal behaviours for key?
13:36:20 <I440r> aha
13:36:31 <I440r> for expect you mean
13:36:39 <I440r> expect expects n characters to address a
13:36:48 <I440r> which is usually TIB :)
13:36:56 <tcn> oh, then accept IS the same
13:36:58 <Fractal> Actually, I think expect is okay. It's practically a direct call to read() :)
13:37:08 <Fractal> Oh, sorry, wrong conversation.
13:37:28 <I440r> fractal expect cant be a direct call to read because of line editing
13:37:33 <tcn> i440, you messed with the select() call?
13:37:35 <I440r> which isforth doesnt realy have yet
13:37:44 <I440r> im not using select
13:37:52 <I440r> if you need a select syscall add it :)
13:38:03 <Fractal> I440r : Yeah, I don't use it either. I don't personally think it's necessary.
13:38:08 <I440r> #parameters syscall-number syscall <select>
13:38:27 <Fractal> Line editing, that is.
13:38:28 <I440r> its being depricated by poll
13:38:33 <tcn> oh
13:38:48 <I440r> oh. i will be implementing both that and a command history :)
13:39:20 <Fractal> That's one problem I have with emacs though... It tries to do everything...
13:39:26 <tcn> heh
13:40:08 <tcn> well I expect command history/editing.. i don't give a damn if it has a newsreader written in its own programming language
13:40:19 <Fractal> I like a nice simple editor, a separate interpreter, a speparate shell, etc.
13:40:25 <Fractal> tcn : Heh.
13:40:38 <tcn> exactly.. everything separate
13:43:04 <Fractal> Honestly, I don't spend much time actually writing code in my forth environment. I almost always load the code from files.
13:43:21 <ianni> Fractal: I would too, if I could ever get aprogam started
13:43:31 <Fractal> ianni : What are you having trouble with?
13:43:57 <ianni> Fractal: nothing - just starting!
13:44:03 <Fractal> :)
13:44:09 <ianni> Fractal: really, havbing trouble finding a good problem to solve!
13:44:11 <ianni> :)
13:44:23 <Fractal> Well, try to write a program that will solve factorials
13:44:38 <Fractal> factorial ( n -- !n ) 
13:44:41 --- join: OrngeTide (orange@65.19.141.250) joined #forth
13:44:44 <ianni> sounds like a good suggestion.
13:44:46 <ianni> Hi OT.
13:44:50 <OrngeTide> hiya!
13:44:57 <Fractal> Where !n = 1 * 2 * ... * n-1 * n
13:44:58 <I440r> hi orange!
13:45:00 <I440r> ltns :)
13:45:05 <OrngeTide> what's that one page that explains why forth isn't mainstream even though it's technically superior?
13:45:08 <tcn> yeah fractal, I'd rather have an RPN calculator program and do my programming in an editor
13:45:11 <ianni> Fractal yes :)
13:45:26 <I440r> OrngeTide i cant remember :)
13:45:29 <Fractal> tcn : Agreed.
13:46:06 <OrngeTide> d'oh!
13:46:20 <Fractal> Oh, hey all, I had an idea: You guys heard of the IFPC?
13:46:29 <Fractal> International Functional Programming Contest?
13:46:32 <tcn> fractal: and it should work like a calculator, you know, single keystrokes instead of words, 1-letter variables, and graphics..
13:46:34 <ianni> wait
13:46:39 <ianni> Fractal: that's really too easy
13:46:40 <ianni> ehehe
13:46:44 <ianni> il do it anyway
13:46:56 <Fractal> ianni : You still might learn a thing or two. :)
13:47:27 <ianni> you're right.
13:47:31 <Fractal> tcn : Well, that wouldn't be too tough to code in forth.
13:48:32 <tcn> frac: i'm gonna use C/svgalib
13:49:07 <OrngeTide> don't use svgalib. it's junk.
13:49:09 <Fractal> tcn : Oh yeah? Cool. I had a really nice RPN calculator for macos a while ago.
13:49:25 <tcn> what do you use orangetide?
13:49:43 <OrngeTide> use direct framebuffer. just mmap it in. or use SDL (which supports MacOS, Framebuffer, X11, QNX, EPOC, BeOS, Win32 DirectX, ...
13:49:45 <tathi> svgalib doesn't compile on PPCs...grr...
13:50:01 <tathi> (at least on mine running Linux)
13:50:07 <OrngeTide> SDL is really simple. once i finish my forth it's going to have SDL bindings.
13:50:22 <OrngeTide> tathi, i think there is an svgalib wrapper floating around that will interface to framebuffer.
13:50:29 <tcn> i thought SDL just uses X, svgalib, fb, etc..
13:50:36 <OrngeTide> svgalib doesn't support ANY of my videocards. and it doesn't work on my iBook/Linux box
13:50:40 --- quit: tathi ("Lost terminal")
13:50:47 <OrngeTide> tcn, no. SDL supports like *everything*
13:51:09 <tcn> natively?
13:51:20 --- join: tathi (~josh@wsip68-15-54-54.ri.ri.cox.net) joined #forth
13:51:26 <OrngeTide> SDL framebuffer is what I use for my MAME/SNES/NES box. it has input support so i can use the joysticks to select my roms.
13:51:29 <OrngeTide> tcn, define native.
13:51:47 <OrngeTide> it does native sound, input and framebuffer support. for X it uses XDGA or X MIT SHM
13:51:56 <OrngeTide> for Win32 it goes through DirectX.
13:52:13 <tcn> it has its own card drivers?
13:52:26 <OrngeTide> if by native you mean directly use the hardware. then that's a bad idea.
13:52:34 <OrngeTide> tcn, of course not. because that would be a mistake.
13:52:48 <OrngeTide> it's hard to write drivers for every videocard. especially when compoanies like nvidia won't hand out specs.
13:53:01 <Fractal> tcn : Basically, it's a series of wrappers that provides a common API across different drivers.
13:53:11 <OrngeTide> it's better to centralize the videocard support in the kernel as a driver.
13:53:20 <tathi> oh, and don't forget, SDL supports aalib too :)
13:53:40 <Fractal> Ha! You guys seen Quake2 in aa?
13:53:45 <Fractal> Funny shit.
13:53:52 <tcn> hahah
13:53:56 <tathi> yeah, herk was doing that the other day
13:53:59 <OrngeTide> plus using the system drivers SDL will get accelleration for blitting. as well as RLE compressed sprites for blitting sprites w/ transparency.
13:54:20 <OrngeTide> Fractal: my friend did the original quake1 port to linux then the aalib port. 
13:54:36 <tathi> you know RockDodger, from that SDL competition? That actually works really well under aalib.
13:54:45 <OrngeTide> tathi, neat:)
13:54:54 <Fractal> OrngeTide : Heh. Cool.
13:54:58 <OrngeTide> I need to patch VNC server into SDL. then i can play stuff remotely in full color. :)
13:55:10 <OrngeTide> VNC rules aalib. 
13:55:27 <tcn> oh i actually downloaded SDL.. ain't doing anything with it yet
13:56:29 <tcn> i figure you guys can make my vgalib program work with whatever you use
13:56:30 <OrngeTide> SDL is my favorite library. my graphical "mud" (it's more like a network zelda) is all SDL based. was finishing up my public domain PNG loader for it. (i don't want to have to deal with too much GPL stuff cuz i don't want to have to give out like a dozen .DLL files for the windows version)
13:56:56 <tcn> i hate libraries dammit
13:57:13 <OrngeTide> tcn, linux framebuffer is simpler than vgalib. you just mmap it and start writing to it. can't get any dumber than that:)
13:57:19 <Fractal> OrngeTide : Why don't you GPL your code and statically link?
13:57:26 <OrngeTide> yah. i hate libs too. that's why i'm doing a lot of my own code. so i can statically link it.
13:57:31 <OrngeTide> fractal, because I hate GPL.
13:57:40 <I440r> orange yea exactly
13:57:42 <tcn> well linux framebuffer doesn't have the right driver
13:57:44 <OrngeTide> i want to beable to have actual control over my code. 
13:57:50 <I440r> ill be doing frame buffer stuff soon :)
13:57:58 <OrngeTide> tcn, linux framebuffer always has the right driver.
13:58:07 <tcn> not in 2.2
13:58:15 <OrngeTide> just use the VESA one. it sets up the fb with vesa bios. (you need VESA 2.0 bios)
13:58:39 <OrngeTide> 2.2 has vesa. it doesn't have BIOS32 but if you set the mode in lilo you're fine.
13:59:22 <tcn> ok let's see..
13:59:28 <Fractal> Companies have definitley run a huge campaign against the GPL saying it strips freedoms away, but please don't take this at face value.
13:59:37 <OrngeTide> 2.4 lets you change your mode at runtime through bios32. 2.2 forces you to be stuck with whatever mode you picked at startup. but it works with any PCI vga card and some VLB cards. doesn't seem to work on any of my ISA vga cards though
13:59:52 <tcn> good enough for a laptop :)
13:59:53 <Fractal> There is more than 1 kind of freedom.
14:00:16 <OrngeTide> Fractal: GPL is inferior to BSD and Public Domain. It's my code. i wish to give people the freedom to add propritary modifications to it and sell it.
14:00:31 <OrngeTide> (because I'm going to be adding my own propritary modifications to it once I get it worked out)
14:00:34 <Fractal> If it wasn't for the GPL, the free software/open source would have been dead over a decade ago.
14:00:45 <OrngeTide> well GPL gives you safety, not freedom.
14:01:13 <OrngeTide> GPL makes it safe to hand out code. knowning that nobody can ever take it away from you and that it will always be available to you and to the public.
14:01:15 <Fractal> Not true. It gives you an essential freedom: To see how others improve and modify your code.
14:01:56 <OrngeTide> Fractal: that's not a freedom. that's restricting the freedom of others so your code cannot be hidden away from you.
14:02:04 <Fractal> MS/adobe/apple/others are trying to get people to believe that this is not a freedom, and are doing a decent job of it, sadly.
14:02:23 <OrngeTide> some people were burnt BAD by writing something cool. giving it out and then some company takes it. extends the crap out of it and then charge like $1000 to use it.
14:03:31 <OrngeTide> I had a coworker who wrote a major part of psos. he gave that part out for free. Then 5 years later he had to actually buy psos which at that time was mostly based on code he wrote. he was kinda pissed. he even asked for a free copy. but they were like "ahhahaha"
14:03:38 <I440r> they will lose. 
14:03:47 <OrngeTide> that kind of crap really sucks. you do a favor for the community and then someone screws you.
14:03:58 <OrngeTide> if GPL were around then I bet he would've done that. 
14:04:19 <OrngeTide> GPL is great for education too. because you can write something and others will extend it. but nobody can hide the code away from students.
14:04:29 <OrngeTide> my agenda is different though. :)
14:04:45 <tcn> students don't need to see CRAP
14:04:51 <OrngeTide> i *want* to hide the code. and I want parts of the code to be hidden from me.
14:04:55 <Fractal> Definitley. The GPL isn't about protecting your own code, though. It's about protecting the public's freedom to view the source code to all software.
14:05:02 <OrngeTide> tcn, ehhee. :)
14:05:23 <OrngeTide> Fractal, well i guess it's more about personal freedom versus public freedom.
14:05:34 <tcn> the problem is people who want to use a computer without bothering to learn programming
14:05:52 <OrngeTide> tcn, my problem is with students who don't bother to read their textbooks.
14:06:04 <tcn> those damn lusers outnumber us now
14:06:08 <Fractal> Yes, in a sense. The united states is built upon personal freedom: I have the right to enslave black people. However, the rest of the world doesn't work like that.
14:06:09 <OrngeTide> they just hop into my channel. paste like 20 lines of code. and says "This doesn't work! why?"
14:06:23 <I440r> lol
14:06:43 <OrngeTide> Fractal: i think there is a place for GPL. and a place for BSD or PD.
14:06:58 <OrngeTide> I'd GPL a compiler or a kernel. just to keep people from doing propritary drivers and crap. 
14:07:30 <tcn> they'll do it anyway.. companies ignore the GPL when it's convenient
14:07:43 <OrngeTide> because it's so much easier to just do a quick hack to your architecture and make it propritary rather than hand out your hacks to the rest of the community. gcc would not support as much as it does today without gpl. (makes me wonder how well TenDRA, a bsd licensed C compiler will turn out)
14:07:53 <OrngeTide> tcn, FSF has actually busted some people on it.
14:08:05 <tcn> all you need to say is "(c)2002 Your Name - Public Domain"
14:08:23 <Fractal> tcn : Well, it can't be copyrighted if it's public domain.
14:08:33 <OrngeTide> the evil thing about GPL is that when you've violated the license you are actually denied from ever using that product again.
14:08:41 <OrngeTide> which gives GPL some teeth.
14:09:16 <OrngeTide> "oh. sold gcc without source. won't give out source. telling us you 'can't' ? ... well you're not allowed to distribute gcc ever again. even if you do comply with the license."
14:09:21 <tcn> anything you write is copyrighted whether you say it or not.. "Public Domain" just says you're not collecting royalties
14:09:22 <Fractal> ianni : How's that factorial program coming?
14:09:36 <OrngeTide> tcn, public domain is actually not copyright
14:09:44 <OrngeTide> you have to mark stuff *clearly* as being public domain. 
14:09:50 <Fractal> tcn : No, public domain infers no copyright whatsoever.
14:10:08 <Fractal> For example, after copyright expires, the work becomes public domain.
14:10:21 <OrngeTide> if someone takes a public domain program and breaks thier computer. they can't sue me for damages.
14:10:29 <OrngeTide> if someone takes a GPL program copyright by me. they can.
14:10:30 <tcn> and you can voluntarily make it PD
14:10:48 <Fractal> OrngeTide : Not true. The GPL explicitly says no warranty whatsoever.
14:11:00 <OrngeTide> at least in the US. DMCA has made the "no warranty" clause meaningless. software automatically has a warranty.
14:11:03 <tcn> GPL ain't the law
14:11:13 <OrngeTide> afaik.
14:11:28 <Fractal> OrngeTide : Well, I don't consider the US laws. It's simply not a free country. :)
14:11:30 <tcn> hell, DMCA might not be the law too long either :)
14:11:45 <OrngeTide> yah. there is a big push to overturn DMCA.
14:11:54 <OrngeTide> i mean do you really need your copyright to last like 190years? 
14:11:55 <tcn> courts don't seem to like the DMCA
14:12:04 <OrngeTide> the purpose was to protect your creative works for your own lifetime.
14:12:05 <tathi> does copyright _ever_ expire under the DMCA?
14:12:15 <tcn> yeah.. dmca extends it 20 years
14:12:16 <OrngeTide> but sonny bono wanted his great great great grandchildren to profit from his horrible singing.
14:12:31 <tcn> heh
14:12:38 <Fractal> OrngeTide : The problem is that in the US, the corporations write the laws. So if they want Mickey Mouse to be their IP for another 20 years, so be it.
14:13:04 <tcn> well.. now everyone's out to get "greedy corporations"
14:13:28 <OrngeTide> well corperations were not ment to outlast the original creater's lifetime. it was ment to be a way to create a pact between partners and distribute the fiscal and legal responsibility accrossed partners.
14:13:38 <tcn> anyway, the law is just the law. nobody says you gotta follow it.
14:13:42 <OrngeTide> but some court cases have turned corperations into little entities.
14:14:38 <OrngeTide> tcn, well if you want society to function you have to follow laws.
14:14:45 <OrngeTide> at least most of them. :)
14:14:57 <tcn> the ones that make sense.. you gotta sift 'em
14:15:13 <OrngeTide> like don't kill people, or kidnap children and sell them into slavery, or rob banks, etc.
14:15:40 <OrngeTide> tcn, well our problem is we "hire" representives full-time. if they were part time they wouldn't have so much free time to make up all these bullshit laws:)
14:16:13 <Fractal> tcn : Agreed. Except the problem is that firms don't sift them based on ethics like individuals, they sift them based on risk-of-getting-caught to profit ratios.
14:16:15 <OrngeTide> like i'd be a congressman for 1 day a week. it could be like the army reserve. :)
14:16:53 <tcn> yeah, damn career politicians
14:17:12 <tcn> Massachussetts is like a fucking monarchy it's so bad
14:21:34 <tcn> well, i guess this vesa fb is ok.. let's see
14:24:56 <OrngeTide> california is like a despot. it's horrible.
14:25:26 <ianni> seen opensecrets.org
14:25:34 <OrngeTide> vesafb is fun because you can draw in text mode. if you watch where the cursor is you can actually have like a graphical cat program if you wanted.
14:25:47 <OrngeTide> what's that?
14:25:49 <Fractal> ianni : How's the factorial word coming along?
14:26:09 <ianni> Not working on it atm
14:26:12 <OrngeTide> ohh. this is neat. there is a california voters association website that does this for local government.
14:26:15 <ianni> but i did realize it was much harder than I expected :0
14:26:19 <Fractal> :)
14:26:21 <ianni> :)
14:26:31 <Fractal> Want to see a really neat implementation of it?
14:26:33 <Fractal> 1-liner?
14:26:35 <ianni> yes!
14:26:39 <OrngeTide> i worry more about my state government than my national government. california's economy is the 5th largest in the world. 
14:26:49 <ianni> hmm
14:26:53 <ianni> I was thinking to use return stack
14:26:57 <ianni> I spose you dont need to at all
14:26:59 <OrngeTide> is it a recursive factorial ?
14:27:12 <Fractal> : factorial ( n -- !n ) dup 1 > if dup 1- recurse * then ;
14:27:14 <Fractal> OrngeTide : Yes.
14:27:38 <OrngeTide> i only have 32 cells of data stack.. how deep does it have to go usually?
14:27:40 <tcn> orngtide: i bet it's not the 5th largest economy during the port lockout!
14:27:48 <OrngeTide> tcn, ahaha:)
14:27:59 <OrngeTide> the port lockout is a bitch. i dunno where i'm going to buy my green tea.
14:28:03 <Fractal> OrngeTide : n-1 cells.
14:28:09 <tcn> hahah
14:28:12 <OrngeTide> hopefully none of the cool asian stores near my go out of business.
14:28:19 <OrngeTide> s/my/me
14:28:19 <Fractal> Plus n return stack cells.
14:28:33 <tcn> forget Iraq, GW should send the army into California
14:28:37 <OrngeTide> Fractal: oh. i only have 16 return stack cells. hrm.
14:29:05 <OrngeTide> nah. they should just pay the port workers decent wages.
14:29:16 <tcn> $100,000 a year ain't decent?
14:29:18 <OrngeTide> and pass the dollar onto the consumer.
14:29:39 <OrngeTide> tcn, is that the avergae salary? i thought that was a high for like a supervisor?
14:29:46 <ianni> Fractal: nice
14:29:53 <tcn> not sure. they're unionized, they must be rich.
14:30:00 <OrngeTide> toll workers in CA get $65k
14:30:05 <Fractal> OrngeTide : That's okay !13 will overflow a 32 bit signed integer, so... :)
14:30:16 <tcn> I bring home $15k
14:30:43 <tcn> laborers are lucky to get $10k here
14:30:47 <Fractal> ianni : Thanks. What you've got to remember is there's always more than one way of doing something in forth. :)
14:30:56 <OrngeTide> Fractal: oh. my forth has 64bit cells.
14:30:59 <OrngeTide> my forth is a bit ... queer
14:31:16 <Fractal> OrngeTide : Cool.
14:31:19 <OrngeTide> my friend wants to port it to S/390 :) 
14:31:32 <OrngeTide> or 360 or whatever that mainframe is called
14:31:34 <ianni> Fractal; for sure
14:31:41 <Fractal> OrngeTide : 360/370
14:31:42 <tcn> are you gonna call it FAGforth?
14:31:56 <OrngeTide> tcn, no.. not that kind of queer:P
14:32:01 <tcn> rofl
14:32:11 <tcn> you know.. FIGforth, FAGforth
14:32:16 <Fractal> Heh. Ya.
14:32:18 <OrngeTide> yes, yes. :P
14:32:21 <tcn> for those who live in hollywood
14:32:29 <OrngeTide> hollywood?
14:32:43 <tcn> oh you ain't been on #osdev
14:32:43 <OrngeTide> i'm in san francisco bay area...
14:33:17 <OrngeTide> no. all the OSes i code are closed source.
14:33:32 <tcn> oh. what do you use?
14:33:41 <OrngeTide> QNX right now. was using vxWorks (bleh!)
14:33:47 <OrngeTide> vxWorks is okay. but the tools suck.
14:34:02 <OrngeTide> QNX is weird/cool and it's tools are reasonable. 
14:34:13 <OrngeTide> mostly because it's tools are a unix rip off. :)
14:35:08 <OrngeTide> man. it was only $3 for this 4.23oz bag of loose-leaf green tea. those little asian stores are awesome. i got a tastey vietnamese sandwich there for $1.50 too
14:35:32 <tcn> hm
14:36:52 <ianni> loev green tea
14:36:52 <OrngeTide> i'm not have much luck pressuring my company to use forth for the command-line interface on our NAS/SAN product:)
14:36:59 <OrngeTide> green tea makes you pee! :)
14:37:13 <ianni> haha
14:37:22 <OrngeTide> s/i'm/i've
14:37:44 <OrngeTide> man. i make so many errors in my typing you'd think i'd just came over on a boat from china.
14:37:47 <tcn> i didnt' have much luck pushing forth either
14:38:38 <OrngeTide> tcn, they f33r forth. they are going to use some unix shell they got from qnx. they didn't even want to use pdksh which is less buggy than the qnx 'fesh'
14:38:57 <tcn> hehe
14:39:50 <OrngeTide> well i'm going to go back to work. seeya. if anyone remembers that link to the paper on why forth isn't mainstream please msg it to me. 
14:39:56 <tcn> thing is, forth influences your syntax.. makes it less than ideal, just for a little more flexibility. what if you don't need flexibility? you know
14:40:15 <tcn> nice chatting with ya
14:40:46 <Fractal> Later, man.l
14:41:48 --- quit: Herkamire ("leaving")
14:47:52 --- quit: tathi ("leaving")
14:48:24 --- quit: tcn ()
14:54:04 <ianni> 
got to
14:55:47 <Fractal> Later.
14:55:57 <Fractal> Everybody's ditching here....
15:00:51 <Robert> :)
15:00:53 <Robert> Bye.
15:10:03 <ianni> 
n/m
15:10:08 <ianni> sorry
15:10:09 <ianni> bye
15:21:14 <Fractal> Hey Robert, how's it going?
15:31:22 --- join: Herkamire (~jason@ip68-9-58-75.ri.ri.cox.net) joined #forth
15:41:22 <Robert> Fine :)
15:59:57 <Fractal> Good, good.
16:01:26 <Fractal> Any cool forth projects?
16:02:43 <Robert> Hmm... Yes. Once I get the TinyVM running, I'll implement a Forth for it.
16:02:48 <Robert> (Hopefully tomorrow)
16:03:00 <Fractal> Nice.
16:03:15 <Fractal> I've just finished documenting my powerful forth debugger.
16:03:29 <Fractal> I'm gonna release frugal in a minute if you wanna check it out.
16:03:53 <Fractal> It's kind of neat. You can step into instructions, skip instructions, put breakpoints in, etc.
16:05:01 <Robert> It's 1am here, school tomorrow :)
16:05:05 <Robert> I'll soon go to bed.
16:05:57 <Fractal> Heh. OK. I've got tea.fs in the progs/ directory now.
16:07:38 <Fractal> OK, all the moment you've all been waiting for: www.hcsw.org/frugal/   
16:07:45 <Fractal> 0.9.7 :)
16:08:11 <Robert> Heh, nice.
16:08:20 <Robert> Well, good night :)
16:08:24 <Fractal> Night.
16:18:48 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth
16:19:53 --- quit: galexand ("brb")
16:20:35 --- join: galexand (~galexand@adsl-80-153-166.rdu.bellsouth.net) joined #forth
17:03:02 --- quit: proteusguy (Read error: 110 (Connection timed out))
17:03:11 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth
17:14:07 --- join: tathi (~josh@ip68-9-68-50.ri.ri.cox.net) joined #forth
17:33:00 --- quit: I440r (Read error: 113 (No route to host))
18:02:09 --- join: I440r (~mark4@1Cust174.tnt3.bloomington.in.da.uu.net) joined #forth
18:02:20 --- quit: proteusguy (Read error: 110 (Connection timed out))
18:02:55 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth
18:55:50 --- quit: air (Read error: 110 (Connection timed out))
19:03:25 --- quit: proteusguy (Read error: 110 (Connection timed out))
19:03:57 --- join: proteusguy (~proteusgu@24-197-147-197.charterga.net) joined #forth
19:08:02 --- quit: Robert (Remote closed the connection)
19:08:08 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #forth
19:37:02 --- quit: proteusguy ("Client Exiting")
20:03:12 <onetom> c u guys
20:03:36 <onetom> im gonna leave 4 2 days 1h l8r ;)
20:03:49 <onetom> thatis: b back on friday
20:03:59 <Herkamire> have fun
20:04:32 <onetom> Fractal: keep on writing furgal! thats nice & very educational
20:05:02 <onetom> Robert: good luck 4 the tinyvm
20:05:12 <Fractal> onetom : Thanks. :)
20:05:19 <ianni> re
20:05:23 <onetom> Chef: :/ he isnt here.
20:05:35 <ianni> later ontom
20:05:41 <ianni> e +
20:06:07 <onetom> ianni: hey, i have ~half an hour left until departure
20:06:11 <Fractal> onetom : When you get back, you should check out frugal's forth debugger.
20:06:29 <Fractal> Wait a sec if you're going to test it now. There's a bug I gotta fix.
20:06:31 <onetom> Fractal: of course i will!
20:06:33 <Herkamire> for those who care, I have fixed my gforth httpd (which uses tcpserver for the networking). it's at http://herkamire.homeip.net:9000/ (it's serving an intro page and it's own source.)
20:08:33 <onetom> all: we should invite abi here, so we can stop repeating the often typed urls, like the 1 4 frugal, isforth, various important papers & concentrator sites, like wikis & also our PERSONAL 4th homepages
20:09:23 <onetom> Herkamire: oh, man, u didnt know how miserable was my 1st intro of ur web server :)
20:09:53 <onetom> Herkamire: the final conclusion was: 4th is bullshit, it doesnt work @ all ...
20:10:17 <Herkamire> onetom: heh :) I'm so glad I figured it out finally
20:10:44 <onetom> Herkamire: tho, i tried 2 defense meself like saying: i didnt play much w strings, i dont know the http protocol, im not the 1 who wrote the program ....
20:10:52 <Herkamire> what's abi?
20:10:56 <Fractal> onetom : www.hcsw.org/frugal/
20:11:05 <onetom> Herkamire: but i couldnt fix 4 them :(
20:11:23 <onetom> Fractal: i know the url very well. also bookmarked it, 
20:11:30 <Fractal> onetom : The debugger is sweet. debug> my_word
20:11:33 <Fractal> onetom : Heh. Cool.
20:11:48 <onetom> Fractal: & i even see it on the top of my screen, in the irc scrollback buffer
20:11:50 <Fractal> Then you can step, bwd, fwd, run, into
20:11:58 <onetom> just disappeared ;)
20:11:59 <Fractal> Ha! :)
20:12:25 <Fractal> Set breakpoints in the code, etc.
20:12:31 <onetom> Fractal: u mean, u dont have a documentation 4 it yet! grrr!! ;)
20:12:43 <Fractal> It's got a comprehensive manual: docs/DEBUG
20:12:44 <onetom> Herkamire: plz explain the solutoin
20:12:56 <Fractal> All the docs are kept in docs/
20:13:20 <onetom> Herkamire: dont let me die so dumb :)
20:13:24 <Herkamire> onetom: you have to read in from the socket everything that the client sends you. most clients do not look for data From the server until they finish sending the request To the server
20:14:25 <onetom> Herkamire: hmm.. doesnt that content get buffered?
20:14:49 <onetom> Herkamire: the 1 what is sent as a reply during the request from the client
20:14:49 <Herkamire> I just added:   : eatall begin ?key while key drop repeat ;   and called it before I output it before I output my responce to the client.
20:15:11 <Herkamire> onetom: I thought buffering would take care of it, but apparently not.
20:15:38 <onetom> eatall <- loll
20:16:45 <Herkamire> :)
20:16:57 <Fractal> onetom : Check the debugger out?
20:17:11 <onetom> Herkamire: abi is a bot on most opn hosted channels
20:17:31 <onetom> Herkamire: its a simple interactive knowledge base
20:18:09 <onetom> Fractal: im busy w packin 4 the trip
20:18:18 <Fractal> Heh. Ok. Where you going?
20:18:32 <onetom> Fractal: but im gonna test it just right after packin if i have some time left
20:19:01 <onetom> Fractal: im going 2 help some chicks diggin plants
20:19:04 <Fractal> Cool. I think it's really powerful.
20:19:10 <Fractal> Heh. Sweet.
20:19:23 <onetom> Fractal: they work 4 the local botanical institute
20:20:19 <Fractal> Ha! Hitting on gardener chicks. Sweet.
20:20:22 <onetom> Fractal: & probably i will also drive a car if the owner of that old LADA gets tired
20:21:19 <Fractal> Cool.
20:21:49 <onetom> Fractal: :) they r a lil bit more than gardener chicks. 2of them r teacher, 1of the 2 is writing her dissertation & the 3rd is a phd student
20:22:09 <Fractal> Heh. Smart gardener chicks. Even better.
20:22:56 <onetom> do u mind if i still dont refer 2 them as gardener chicks? :D
20:23:04 <Fractal> Heh. Whatever.
20:23:37 <onetom> tho, none of them is a gf of mine
20:23:40 <onetom> ...
20:23:43 <onetom> yet ;)
20:23:51 <Fractal> Heh.
20:24:14 <onetom> Fractal: have u seen my keyboard routine?
20:24:22 <Fractal> Nah, what is it?
20:24:39 <onetom> Fractal: ive rewritten a real world example 4 Chef's picforth
20:25:08 <onetom> its an optimizing native code compiler 4 pic micros
20:25:14 <Fractal> Oh. Neat.
20:25:57 <onetom> hajaj, how neat is it! it generates such a code as an asm programmer
20:26:04 <onetom> look:
20:26:14 <onetom> macro
20:26:14 <onetom> : col:  portb swap bit-set ;
20:26:14 <onetom> : r:    portb swap bit-set? if ;
20:26:14 <onetom> : |     >w then ;
20:26:14 <onetom> target
20:26:16 <onetom> : key fe ( nokey code) >w
20:26:19 <onetom>   7 col: 1 r: 03 | 2 r: 06 | 3 r: 09 | 4 r: 0B |
20:26:21 <onetom>   6 col: 1 r: 02 | 2 r: 05 | 3 r: 08 | 4 r: 00 |
20:26:24 <onetom>   5 col: 1 r: 01 | 2 r: 04 | 3 r: 07 | 4 r: 0A | w> ;
20:26:31 <onetom> it generates 30 instructions from it!
20:26:55 --- quit: Robert (Remote closed the connection)
20:27:00 <onetom> its a decoder routine 4 a 3x4 keys keypad
20:27:00 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #forth
20:27:30 <Fractal> onetom : Oh, I see. That's neat. Using the parallel port, I assume.
20:27:47 <onetom> that is 12keys... & the compiled code is hardly longer than the twice of the number of keys
20:28:15 <onetom> Fractal: no, it runs on a PIC microcontroller
20:28:35 <Fractal> Oh.
20:28:46 <onetom> 0x0003  30FE    movlw   0xFE
20:28:46 <onetom> 0x0004  1786    bsf     0x06,7
20:28:46 <onetom> 0x0005  1886    btfsc   0x06,1 
20:28:46 <onetom> 0x0006  3003    movlw   0x03 
20:28:46 <onetom> 0x0007  1906    btfsc   0x06,2 
20:28:48 <onetom> 0x0008  3006    movlw   0x06
20:28:51 <onetom> 0x0009  1986    btfsc   0x06,3
20:28:53 <onetom> 0x000A  3009    movlw   0x09
20:28:56 <onetom> 0x000B  1A06    btfsc   0x06,4
20:28:58 <onetom> 0x000C  300B    movlw   0x0B
20:29:01 <onetom> ...
20:29:09 <Herkamire> easy there killer
20:29:11 <onetom> this is the code 4 the 1st 2rows of the word key
20:29:38 <Fractal> Alright...
20:29:53 <onetom> bsf = Bit Set Fileregister
20:30:19 <onetom> btfsc = Bit Test Fileregister Skip if Clear
20:30:22 <Fractal> Ok, I'm not familiar with PIC Microcontrollers, but I can kinda see what it does.
20:30:48 <onetom> w = work register (the only 1)
20:30:51 <Fractal> The forth code anyways.
20:31:24 <onetom> i think its very useful 2 learn the pic asm
20:31:30 <onetom> coz its so simple
20:31:53 <onetom> & u can experiment easily w compilation techniques on it
20:32:02 <Fractal> Ya.
20:32:28 <onetom> & the gathered experiences can b utilized writing such stuff 4 other processors
20:32:43 <onetom> a pic only has ~33 instructions
20:33:02 <onetom> absolutely straightforward instructions..
20:33:21 <Fractal> Nice.
20:33:29 <onetom> Fractal: im, gonna present u similar tests 4 frugal, i hope
20:33:41 <onetom> Fractal: dont u plan a macro feature?
20:33:50 <onetom> ahh! stop!
20:34:03 <onetom> uve told me frugal know tail recursion
20:34:05 <onetom> how?
20:34:20 <Fractal> Maybe. Macros are just immediate words that compile stuff, so... It'd be easy to add.
20:34:22 <Fractal> Ok.
20:34:26 <onetom> &what is tail recursion in ur opinion?
20:35:00 <Fractal> Well, tail recursion is a particular instance of recursion where the return value of a function is not needed.
20:35:26 <Fractal> Quicksort is a good example of it. It calls itself twice, once of which can be tail recursed.
20:35:45 --- quit: tathi ("'night all...")
20:35:48 <Fractal> SO, for a tail recursive function, there is no need to store the previous stack frame: You can write over it.
20:36:09 <I440r> i.e. make it a jmp, not a call
20:36:11 <Fractal> So to tail recurse in frugal, simply recurse r> drop
20:36:27 <Fractal> Er, r> drop recurse
20:36:47 <Fractal> I440r : Yes. There's another tactic for it.
20:37:17 <Fractal> Just put a goto to the top of the word. :)
20:39:10 <I440r> or you could do a : foo begin dup X > if .... recurse .... again then .... ;
20:39:10 <I440r> heh
20:39:25 <Fractal> I440r : Yup. Definitley.
20:39:47 <I440r> begin if again then in that order is kinda bad code tho :)
20:39:55 <I440r> you would prolly have to do
20:40:14 <I440r> begin if [ swap ] again then 
20:40:35 <Fractal> Yeah, I would personally do something like begin ... if exit then again
20:41:00 <onetom> Fractal: /me prepares meanwhile...
20:41:27 <I440r> dont need if exit then
20:41:32 <I440r> use ?exit or not ?exit
20:41:50 <I440r> unless you have if dothis-first exit then
20:42:00 <Herkamire> well, my gforth httpd hooked in through tcpserver is 3.5 times slower than apache2 (wich apache creating an index of a directory, and my script just outputing static html) according to my tests with apache benchmark (ab) wich 1000 hits, 20 at once.
20:42:12 <Fractal> Well, I don't have ?exit in mine, so...
20:42:23 <onetom> hm, nice tricks :)
20:43:02 <onetom> but id prefer the : word ~~~ word ; ~~~ ; notation 4 it ;)
20:43:35 <Fractal> The what?
20:43:54 <onetom> multiply word exit points..
20:43:57 <onetom> like in color4th
20:44:01 <Herkamire> recurse is a branch back to the begining of the function
20:44:30 <Fractal> Herkamire : Not exactly. The IP is pushed onto the return stack as well.
20:44:49 <onetom> Herkamire: it isnt , but if it is, than its a tail recursion :)
20:44:52 <Fractal> A TAIL recurse would simply be a branch.
20:45:06 <Herkamire> sorry :) I meant that's what tail recursion is.
20:45:10 <Fractal> :)
20:45:24 <Herkamire> sorry sorry
20:45:39 <onetom> Fractal: so u can simply create a word named "recurse;"
20:45:44 <onetom> Fractal: hm, what about it?
20:45:58 <Herkamire> you can do tail recursion with a function that returns a value. but only when you want to return whatever value the recursion returns
20:46:36 <Fractal> Herkamire : True. What I meant is that you can't back propogate the return value to get the answer.
20:46:56 <Herkamire> nope :) you have to calculate it as you go
20:46:57 <Fractal> onetom : Yeah, sure. Think I should?
20:48:06 <Fractal> Has anyone here checked out my debugger yet?
20:49:13 <Herkamire> heh :) no
20:50:02 <Fractal> :)
20:50:05 <Fractal> Heh. Damn.
20:50:33 * Herkamire tries it
20:50:57 <Fractal> onetom : recurse; VM_BRANCH , l @ HEADER_LEN cells + , ; immediate compile-only
20:51:04 <I440r> fractal not yet hehe im trying to get some USEFUL info on the get/set_sockopt syscalls
20:51:11 <onetom> Fractal: but dont bear up! we will ;)
20:51:18 <Fractal> Heh. Ok. :)
20:51:42 <I440r> isforth has compile-only words but doesnt check that your using them interactivly
20:51:47 <I440r> not sure if it should or not
20:52:00 <onetom> Fractal: hmmm, ur branch is realtive? & what is that l @?
20:52:29 <Fractal> onetom : "l" is a variable the returns the address to the dictionary pointer.
20:52:43 <Fractal> "l @" is the address of the last defined word.
20:54:02 <onetom> l as Last?
20:54:06 <onetom> ah, c
20:54:12 <Fractal> Ya, lastdictp :)
20:54:35 <Herkamire> lol: *** DUDE! HOW CAN I STEP INTO THAT INSTRUCTION!
20:54:35 <Fractal> And no, the branch is absolute.
20:54:40 <Fractal> Herkamire : :)
20:54:59 <Fractal> When you try to step into a branch or literal or something. :)
20:56:17 <onetom> gotta got. bye
20:56:18 <Herkamire> cool :)
20:56:21 <Fractal> Later.
20:57:49 <Herkamire> nice debugger
20:57:58 <Fractal> Thanks.
20:58:48 <Fractal> I don't think it'll port easily to isforth...
20:59:04 <Fractal> :(
21:00:23 <I440r> maybe, maybe not :)
21:00:31 <I440r> ill check it out eventually
21:00:40 <Fractal> Cool.
21:00:42 <I440r> i need to implement SEE but there are issues iwth the current kernel
21:01:01 <Fractal> You don't have see? I thought I saw a see.f?
21:01:07 <I440r> i took shortcuts with my hand-made colon definitns in the kernel
21:01:10 <I440r> its a work in progress
21:01:29 <I440r> and is also incorrect
21:01:40 <Herkamire> Day changed to 09 Oct 2002
21:01:42 <I440r> ive defered SEE till after i have completed the meta compiler
21:01:44 <Fractal> I440r : BTW, how do you handle compiled strings in the current definition?
21:01:48 <Herkamire> bedtime for me :)
21:01:54 <I440r> compiled strings where
21:01:55 <Fractal> Night, man.
21:01:56 <I440r> in ." ?
21:02:03 <I440r> or inside colon defs ?
21:02:06 <I440r> i dont have s"
21:02:08 <Fractal> I440r : Yeah, s"
21:02:12 <I440r> i wont support s"
21:02:12 <Fractal> Or ,"
21:02:14 <I440r> its DUMB
21:02:21 <I440r> i say do this
21:02:26 <I440r> here ," xyzzy"
21:02:33 <I440r> : foo ..... literal ..... ;
21:02:45 <I440r> factor the string OUT of your : definition
21:02:53 <I440r> s" is an abomination
21:02:57 <I440r> i also wont define evaluate
21:03:12 <Herkamire> I'm trying to decide if I'm going to put strings in my source file at all
21:03:12 <I440r> so theres no lame ass "10 0 do i . loop" evaluate
21:03:30 <Herkamire> (for our colorforth)
21:03:31 <Herkamire> anyway
21:03:34 <Fractal> I see.
21:03:36 --- quit: Herkamire ("goodnight")
21:03:55 <Fractal> I440r : I don't see how you handle ." though.
21:04:03 <Fractal> : foo ." Yo!" ;
21:04:21 <I440r> ." compiles a (.") inside the : definition. thats totally different to s"
21:04:37 <I440r> which compiles a branch arround the data that you injected into the execution stream
21:04:55 <I440r> (.") actually USES the string, not just returns its address and branches arround it
21:04:56 <Fractal> Oh really? So this won't work: : foo ." not-a-word" ;
21:05:01 <I440r> yes
21:05:13 <I440r> it WILL work
21:05:19 <I440r> ." is an immediate word
21:05:22 <Fractal> So, what, do you have a separate heap?
21:05:30 <I440r> it compiles (.") and then compiles the string to here
21:05:36 <I440r> thats DIFFERENT
21:05:48 <onetom> thats UNUSUAL ;p
21:06:10 <Fractal> Um. That's the same as s", effectively.
21:06:11 <I440r> : (s") r> dup count + >r ;
21:06:15 <I440r> NO 
21:06:16 <I440r> its not
21:06:20 <I440r> s" doesnt USE the data
21:06:23 <I440r> ." DOES
21:06:30 <I440r> i.e. for ." the data is NOT in the way
21:06:43 <I440r> s" injects DATA into the execution stream and DOESNT USE IT
21:06:46 <I440r> evil
21:07:13 <Fractal> Oh I see. Well, here's my ." :   : ." postpone s" compile type compile flush ; immediate compil-only
21:07:21 <I440r> ugh
21:07:27 <I440r> postpone is ANOTHER evil word
21:07:42 <Fractal> postpone is only used for overriding immediate.
21:07:44 <I440r> i would say that postpone is my absolute #1 hated NON_FORTH word
21:07:49 <I440r> ugh 
21:07:50 <I440r> no
21:07:52 <Fractal> : postpone ' , ; immediate compile-only
21:08:05 <I440r> postpone is used as a single replacement for both compile and [compile]
21:08:10 <I440r> do you have the isforth sources ?
21:08:24 <I440r> look in asmsrc/compile.1 and search on the word compile
21:08:26 <I440r> read my comment there
21:08:28 <Fractal> Yeah, somewhere.
21:10:09 <Fractal> Well, I don't have [compile], I have postpone instead.
21:11:51 <Fractal> I think postpone is a perfectly good name for a single purpose word: It postpones the execution of immediate words.
21:12:31 <I440r> i think its a bad name
21:12:36 <I440r> its totally non descriptive
21:12:58 <I440r> it doesnt postpone execution of immediate words, it COMPILES immediate words
21:13:13 <I440r> saying it postpones execution is looking at it backwards
21:13:18 <I440r> totally fucking ans :)
21:14:16 <Fractal> Well, we'll have to agree to disagree then. I think postpone is an excellent name for it.
21:14:50 <Fractal> It "postpones" the execution of this immediate word. What could be more descriptive?
21:15:06 <I440r> ugh
21:15:21 <I440r> it compiles the immediate word.... thats a better way of thinking about it
21:15:23 <I440r> imho
21:15:25 <Fractal> Just as : blah 1 2 + . ; postpones the execution of "1 2 + ."
21:15:41 <I440r> no it doesnt
21:15:42 <I440r> ugh
21:15:45 <I440r> it COMPILES it!!!!!!!!!!!
21:15:47 <I440r> lol
21:15:53 <I440r> your thinking about it ALL WRONG!!!!!!!!!!!
21:15:55 <I440r> lol
21:16:09 <Fractal> Um. You're thinking about it all wrong!!!!!
21:16:10 <I440r> thats like saying the main funciton of a forth compiler is realy as an interpreter
21:16:17 <I440r> and thats NOT correct
21:17:14 <Fractal> Ok, look at it this way: When you compile sources using a compiler, like gcc, you compile it into a executable file for later execution.
21:17:27 <Fractal> The word "compile" implies later usage.
21:17:47 <Fractal> Compiling an encyclopedia is not used instantly, it's postponed for later usage.
21:18:52 <Fractal> And if you speak reasonable english, the word postpone should be obvious. :)
21:19:26 <I440r> english is my 3'rd or 4'th language :)
21:19:55 <Fractal> Well, no offence, but maybe that's the problem. :)
21:20:05 <Fractal> Parlez vous fracais?
21:20:11 <Fractal> Francais?
21:22:31 <I440r> hehe
21:22:33 <I440r> no
21:22:40 <Fractal> D'accord. :)
21:22:41 <I440r> i speek fluent gobbledegook
21:22:42 <I440r> assembler
21:22:44 <I440r> and forth
21:22:45 <I440r> heh
21:22:52 <Fractal> Oh, I see.
21:27:42 <Fractal> Well, that's the great thing about forth, theres so much room for improvement/customization/evolution.
21:28:06 <I440r> and arguement
21:28:07 <I440r> heh
21:28:12 <Fractal> Also, forth is the only language I can think of that has an active community who, by choice, doesn't accept the ANS standard. :)
21:28:21 <I440r> :)
21:28:35 <I440r> i have alot of issues with ans
21:28:39 <Fractal> Don't get me wrong: There are some stupid things about ANSI, but I think postpone is good.
21:28:41 <I440r> mostly that its totally ghey
21:28:52 <I440r> i dont even like the word "cell"
21:28:53 <I440r> ugh
21:29:06 <I440r> cell+ is WRONG
21:29:10 <I440r> you have 2+ and 4+
21:29:15 <I440r> YOU decide which to use
21:29:31 <Fractal> Well, the idea is to easily incorporate multiple celled machines.
21:29:37 <I440r> dont be lame and try write code that works on 8 bit, 16 bit, 32 bit.... machines
21:29:42 <I440r> bad idea
21:29:49 <I440r> rewrite the code for the specific machines
21:29:59 <I440r> DONT try make jack-of-all code
21:30:09 <I440r> portability is a myth anyway
21:30:35 <Fractal> Not true. Portability is a concept, not a concrete attribute.
21:31:34 <I440r> the overhead in doing an absolute 100% from scratch rewrite of any forth project is NOTHING!!!!!!!!!!!!!1
21:31:43 <I440r> specially if you use the existing sources as a base
21:31:51 <Fractal> A perfectly, flawlessly portable prorgam is probably a myth, but adapting portable design methodologies can make a world of difference.
21:32:00 <I440r> so writing code to fit everyone is a totally fucked up idea
21:32:07 <I440r> imho heh
21:32:33 <Fractal> I440r : What are you talking about? There's a very large overhead: programmer time, computing time, etc.
21:32:41 <I440r> no.
21:33:07 <I440r> the ammount of time you spend sifting through 40 gigs of #ifdefine condition compilation outweighs it
21:33:19 <I440r> i gota go walk the doggie brb :)
21:33:30 <Fractal> Alright...
21:40:29 --- quit: Soap- (Read error: 104 (Connection reset by peer))
21:52:35 <I440r> back
21:53:45 <Fractal> Yo.
21:59:07 <ianni> what does recurse do in isforth?
21:59:46 <Fractal> It compiles a reference to the last entry defined to 'here'.
22:00:06 <Fractal> At least, that's what it does in any sensible forth. :)
22:00:21 <ianni> i dont understand
22:00:30 <ianni> I'm looking at the factorial word you showed me
22:00:37 <Fractal> Ok, these are the exact same:
22:00:41 <ianni> recurse does what to the top of the stack?
22:00:50 <I440r> nothing
22:01:01 <ianni> well, i know that
22:01:03 <ianni> hehe
22:01:09 <ianni> now
22:01:23 <Fractal> : factorial dup 1 > if dup 1- recurse * then ;  is the exact same as....
22:01:32 <Fractal> : factorial dup 1 > if dup 1- factorial * then ;
22:01:39 <ianni> oh :)
22:01:40 <I440r> : factorial dup 0 do dup 1+ * loop ;  \this is just a guess
22:01:46 <I440r> oh right
22:01:51 <I440r> thats proilly better heh
22:01:58 <I440r> erm no
22:02:00 <I440r> its recursive
22:02:04 <I440r> recursion is always evil :)
22:02:23 <ianni> evil as in good?
22:02:30 <I440r> ill have to test my itterative method
22:02:38 <ianni> :P
22:02:42 <I440r> no. evil as in to be avoided at all costs
22:02:45 <ianni> I didnt know i could do that
22:02:47 <ianni> ah
22:03:08 <ianni> thought maybe so
22:03:08 <Fractal> No, recursion is a very useful tool.
22:03:40 <I440r> i dislike it
22:03:44 <Fractal> It takes a particular kind of mindset to "think recursively" though, which is why some people tend to avoid it.
22:03:58 <I440r> ive never seen a REAL case where recursion couldnt be replaced wiht a simpler itterative method
22:04:16 <Fractal> I440r : How about the quicksort algorithm?
22:04:40 <I440r> Fractal i wouldnt use quick. i would use heapsort or shell sort or fastest of all
22:04:44 <I440r> radix counting sourt
22:05:00 <Fractal> Radix sort is only useful in very specific instances.
22:05:01 <I440r> i.e. i would use a counting sort that implemented the radix huristic
22:05:18 <I440r> fractal the radix sort is absolutely the fastest sort method ever devised.
22:05:21 --- join: Serg_Penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
22:05:21 <I440r> well
22:05:24 <I440r> the counting sort is
22:05:36 <I440r> the radix huristic just makes it practical
22:05:56 <I440r> counting sort makes absolutely NO comparisons between items AT ALL
22:06:13 <Fractal> Radix sort is good if you have a number of equally length, direct digit corresponding numbers. It simply can't be applied to all problems.
22:06:25 <I440r> it just counts how many of each item there are and figures out each items place based on how many items there are equal or less than itself
22:06:29 <Serg_Penguin> hi
22:06:31 <I440r> (or equal to it or greater than)
22:08:15 <Serg_Penguin> I440r, can u put bin of isForth on web ?
22:08:28 <I440r> a binary? sure
22:08:28 <Fractal> Because of it's inflexibility, quicksort is probably the most useful sorting algorithm invented.
22:08:34 <I440r> linux x86 elf
22:09:09 <Fractal> Er, because of RADIX's and other linear time sorting algorithms, quicksort is probably the most useful sorting algo.
22:09:47 <I440r> if you /dns me serg and look in my ftp you will see a kernel.com and an isforth
22:09:53 <I440r> download them and make them both +x
22:09:59 <I440r> ./isforth is the extended kernel
22:10:04 <I440r> you could try ./kernel.com
22:10:07 <I440r> fload isforth.f
22:10:19 <I440r> kernel.com is what nasm produces
22:10:50 <Fractal> One reason is that you have to know the largest number to be sorted, so often you have another O(n) search algorithm to implement before you use radix or counting sort.
22:11:09 <I440r> therse no such thing as a radix sort
22:11:14 <I440r> the "Radix" is a huristic
22:11:23 <I440r> you can use the radix huristic with ANY stable sort
22:11:36 <I440r> including shell
22:11:52 <I440r> shell sorts are stable. QUICK sorts are not
22:11:54 <Fractal> Well, the point is that it's a linear time sort.
22:12:17 <Fractal> Whatever stable sort you use.
22:12:37 <I440r> who cares. given the same data the radix/counting is going to out-perform a quick sort every time
22:13:11 <I440r> unless its only got 1 or 2 items unsorted and then a shuttle sort is going to be best
22:13:19 <I440r> FAR FAR faster than a quick sort
22:13:28 <I440r> and shuttle sort is only one step up from a damned bubble sort
22:13:53 <Fractal> However, there are simply too many sets of data that a radix sort can sort.
22:14:20 <I440r> err radix isnt a sort method! its a huristic
22:14:38 <I440r> given 16 bit data items to sort you first sort on the high order byte
22:14:41 <I440r> then on the low oder byte
22:15:05 * Serg_Penguin remembers bubble sort of 33x33 numbers on i386 25mhz, on TurboBasic
22:15:10 <I440r> as long as you use a stable sort the sort on the low order bytes wont jumble up the high order bytes
22:15:12 <Fractal> A radix sort does exist. It uses another stable sorting algorithm to sort a different set of data.
22:15:24 <I440r> fractal NO!!!!!!!!!!!!!!1
22:15:26 <I440r> its NOT a sort
22:15:35 <I440r> its a HURISTIC you apply to a sort method
22:15:48 <I440r> you use a XYZZY sort to sort all HIGH order digits
22:16:03 <I440r> then you use the same XYZZY sort to sort on each successivly lower digit
22:16:06 <Fractal> Ok, first off, you're misspelling "heuristic", and radix is a method of ordering entities, hence a "sort".
22:16:19 <I440r> ugh no its not a sort heh
22:16:39 <Fractal> It relies on other algorithms, but that doesn't mean it isn't a sort.
22:16:42 <I440r> its a HEURISTIC <- i do know this spelling, i just slippoed
22:16:48 <Fractal> Heh.
22:16:58 <I440r> yes it does, a heuristic by iteslf doesnt do anything
22:17:22 <I440r> the radix heuristic would make absolutely NO sense exceopt when applied to a counting sort
22:17:31 <Fractal> Also, radix sort often has large overheads associated with it, such as byte ordering and word alignment.
22:17:32 <I440r> but it WILL work with a shell sort still
22:17:39 <I440r> no
22:17:42 <I440r> look
22:17:50 <I440r> hang on
22:18:32 <I440r> http://67.241.61.174/html/sort.html read that
22:18:43 <I440r> then read the link to counting sort
22:18:47 <Fractal> Look, I know how a radix sort works: It works by applying counting (or another stable sort) one position at a time to a set of data.
22:19:31 <I440r> lets look at 32 bit data sorted with an xyzzy sort using a radix heuristic
22:19:41 <I440r> you dont sort the data ONCE, you sort it FOUR times
22:19:43 <I440r> once per digit
22:20:03 <I440r> the radix part just decides which digit. thats NOT a sort, its a decision. the xyzzy is the sort
22:20:15 <I440r> radix applied to ANYTHING except a count sort is BOGUS
22:20:20 <I440r> it will just slow you down
22:20:31 <Fractal> Look, I'm not denying that radix sort is computationally less complex than other sorts.
22:20:43 <I440r> radix applied to a count sort makes it possible to do the count sort without needing a 256 meg buffer just for the counts
22:20:56 <I440r> its NOT a radix sort
22:21:01 <I440r> its a COUNT sort 
22:21:03 <I440r> ugh!
22:21:22 * Serg_Penguin dunno understand anydamnthing better than bubble sort
22:21:38 <Fractal> Look, a radix sort is a series of count (or other) sorts applied to a series of digits.
22:21:46 <I440r> no
22:21:50 <I440r> absolutely NOT
22:21:56 <I440r> the radix is NOT a sorting method
22:22:11 <I440r> else you would be able to do NOTHING except the radix part and still end up with sorted data
22:22:14 <I440r> you CANT
22:22:19 <I440r> therefore its not a sort method
22:22:35 <I440r> its a heuristic that you apply to your inner sort method for what ever reason
22:22:48 <Fractal> Look, we're arguing semantics here. If you want to insist that an algorithm that orders data is not a sorting algorithm, well, whatever.
22:23:03 <I440r> only place it makes sense is with a counting sort because a radix heuristic is GUARANTEED to slow down ANY sort method its applied to
22:23:16 <I440r> radix isnt an algorithm
22:23:25 <I440r> its a heuristic. BIG difference!
22:24:06 <I440r> http://67.241.61.174/html/count.html   <-- just read that
22:24:08 <I440r> heh
22:25:38 <Fractal> Well, from what I understand, a heuristic is a less than optimal solution that comes up with an *approximate* answer.
22:25:54 <Fractal> I don't see radix sort as an heuristic.
22:26:14 <I440r> it is. it does make your sort less than optimal
22:26:30 <I440r> but if you were sorting word items you would need 512k just for the sort
22:26:52 <Fractal> No, the result is still exact.
22:26:53 <I440r> if you were soring 32 bit items you would need about 4 megs JUST FOR THE COUNTS!!!!!!!!!!!!!!
22:27:13 <Fractal> The method is definitley not resource optimal, that's not what we're arguing about.
22:27:34 <I440r> so. by sorting the data on each digit one at a time you are making the sort method practical
22:27:39 <Fractal> Newton's method for calculating differentials, is a pretty fucking good approximation, BUT it's not exact. So it's an heuristic.
22:28:22 <I440r> a heuristic is a method of achieving a solution to a problem by getting you there in small steps
22:28:29 * Serg_Penguin just invented new method of sorting
22:28:31 <I440r> a heuristic is not guaranteed to find a solution
22:28:46 <I440r> but its almost guaranteed to find the solution faster than a pure algorithm 
22:28:59 <I440r> a heuristic is a rule of thumb
22:29:06 <Fractal> However, the radix sort IS guaranteed to find a solution.
22:29:07 <I440r> manhattan distance is a heuristic
22:29:11 <I440r> no
22:29:18 <I440r> dont use anything EXCEPT radix
22:29:21 <Fractal> If your data is set up correctly.
22:29:24 <I440r> dont use radix/shell
22:29:28 <I440r> dont use radix/quick
22:29:29 <Fractal> Er, is acceptable.
22:29:30 <I440r> just use radix
22:29:37 <I440r> ui can guarantee you NEVER find a solution
22:29:46 <Serg_Penguin> 1. bubble sort w/ interval of half the array size
22:29:53 <Fractal> Well, obviously. You need a stable sort to carry out a radix sort.
22:29:56 <I440r> you are going to be doing NOTHING on each successive digit
22:29:58 <Serg_Penguin> 2. w/ interval 1/3 of array size....
22:30:05 <Serg_Penguin> 3. ... down to 1
22:30:37 <I440r> if radix doesnt sort anything by iteslf how can you claim its a sort method
22:30:43 <Serg_Penguin> ( normal bubble sort )
22:30:48 <I440r> even without radix the shell IS a sort methoid
22:30:52 <I440r> quick IS a sort method
22:31:01 <I440r> (quick is unstable btw)
22:31:38 <Fractal> I440r : Because, it's a technique that, when applied to data, produces sorted data. The fact that it uses other algorithms to achieve these results is irrelevant.
22:31:40 <I440r> apply radix to your count sort or shell sort and you dont have TWO paralell sort methods working together
22:31:45 <Fractal> Yes, I am aware that qsort is unstable.
22:31:54 <I440r> its a heuristic
22:32:06 <I440r> i challenbed some friends of mine to make a stable quick sort heh
22:32:09 <I440r> they gave up :P
22:32:27 <Fractal> Heh. How the fuck are you going to make a stable qsort? :)
22:32:50 <Fractal> But seriously, stable sorts rarely matter.
22:32:56 <I440r> anyway, the count sort is the fastest sort there is. so i wouldnt use quick (the start of teh conversation)
22:33:08 <I440r> i.e. my itterative sort method out-performs your recursive method
22:33:28 <Fractal> In certain instances. Often you simply can't use your counting sort.
22:33:38 <I440r> i say for every recursive algorithm in existance there is a better itterative method of achieving the same gial
22:33:45 <I440r> its an out of place sort
22:33:47 <I440r> yes
22:34:03 <I440r> so i use shell sort or a heap sort
22:34:09 <I440r> or some other sort
22:34:27 <I440r> but i DONT use ANY recursive methods.
22:34:48 <Fractal> Well, thinking recursively CAN be difficult. I don't blame you.
22:34:55 <I440r> region flood fills - i use my itterative method. outperforms every single recursive method i ever saw
22:35:21 <I440r> the only case i say where an itterative method wont work is with things like ackerman
22:35:31 <I440r> which has absolutely NO fucking practical use what so ever
22:35:38 <I440r> so who gives a fuck :)
22:35:53 <Fractal> Often, though, you're forced to implement a separate stack to keep track of things. Recursion is a better way to produce "correctness" than iteration, generally.
22:36:23 <I440r> no. i just used the parameter stack
22:36:45 <Fractal> Yes, but with algorithms that have 2 recursion calls.
22:36:55 <Fractal> (Or more).
22:37:00 <I440r> well. ask yourself
22:37:04 <I440r> what IS recursion for
22:37:08 <I440r> its a way to do LOOPS
22:37:17 <I440r> a recursive call is TWO loops
22:37:29 <Fractal> It is a way to do LOOPS and keep track metadata with no programmer overhead.
22:37:33 <I440r> one loop from the top to the call. the next one from the call to the exit
22:37:39 <Fractal> keep track OF metadata
22:38:08 <I440r> show me a USEFULL recursive function
22:38:18 <I440r> something that actually does a job (not quicksort)
22:39:18 <Fractal> Ok...
22:40:11 <Fractal> Calculating harmonic series.
22:40:21 <Fractal> For instance, 1+ 1/2 + 1/3 + ...
22:40:39 <I440r> i.e. genrates a square wave heh
22:41:02 <I440r> maths usually does fit better with recursion. CODE doesnt
22:41:05 <Fractal> Not only waves, but computing vital constants like e and I think Pi.
22:41:12 <I440r> unless its code to do something mathematical
22:41:19 <Fractal> Which is everything.
22:41:25 <I440r> no
22:41:28 <I440r> not everything
22:41:41 <I440r> why calculate pi
22:41:48 <I440r> 3.1415926535
22:42:00 <I440r> if you think you need it to more places than that you REALY need a labotomy
22:42:41 <Fractal> Pi is not an excellent example, but e definitley is. Often you need very precise differential calculations that use e.
22:43:10 <I440r> well, what is the value of E ?
22:43:11 <I440r> erm e
22:43:12 <Fractal> Ok, how about computing prime factors?
22:43:19 <I440r> are they usefull 
22:43:20 <I440r> ?
22:43:22 <I440r> ANYWHERE ?
22:43:23 <Fractal> 2.17 something... I dunno.
22:43:35 <Fractal> Um, cryptography, compression...
22:43:46 <I440r> theres a number thats 1 less than its own reciprocal
22:43:57 <I440r> but im not sure if THAt has any practical use either heh
22:44:02 <Fractal> Heh.
22:44:03 <I440r> compression ?
22:44:14 <I440r> 1.61803499.....
22:44:25 <Fractal> No, no.
22:44:31 <Fractal> Prime factors.
22:45:03 <I440r> heh
22:45:10 <I440r> why calculate :P
22:45:15 <Fractal> There are a lot of real world problems that are naturally expressed as the same problem, just a little smaller.
22:45:20 <I440r> just use a very large database :)
22:46:38 <Fractal> Heh.
22:46:42 <I440r> heh
22:46:50 <I440r> that WAS humor btw
22:46:50 <Fractal> Anyways, I don't know what the fuck this has to do with forth. :)
22:46:57 <I440r> me either :)
22:46:59 <Fractal> Ok, good. :)
22:47:28 <I440r> if you used a database all those public key encryption methods would be way too easy to decrypt :)
22:47:36 <I440r> unless you kept the database itself secret
22:48:13 <Fractal> Well, the encryption methods don't need to be secret; just the keys.
22:48:26 <I440r> i know
22:48:35 <Fractal> Ok, I don't know what you're talking about then.
22:48:41 <I440r> did the guy that released PGP go to jail ?
22:48:50 <Fractal> Phil Zimmerman?
22:48:52 <I440r> yea
22:49:02 <Fractal> Not exactly. Charges were dropped.
22:49:08 <I440r> good
22:49:11 <Fractal> He may have spent some time there, but he was not convicted.
22:49:33 <Fractal> Yeah, that is good.
22:49:42 <I440r> i heared a story about a frenchy guy that gave talks on rot13 encryption IN FRANCE. he was then invited to usa to talk on a totally different subject and was arrested on arrival
22:49:46 <Fractal> Good ol "land of the free" :)
22:49:48 <I440r> spent 3 years in prison or something
22:49:55 <Fractal> On rot13? Ha!
22:49:59 <I440r> yes
22:50:06 <I440r> its used by adobe i believe
22:50:46 <I440r> i thinkk even debian is going to be releasing strong encryption from the main site 
22:50:54 <I440r> i.e. no more non-us sites
22:51:14 <Fractal> Well, I'm not a US citizen, so it doesn't concern me.
22:51:29 <Fractal> But that will be good for the people in the US, I guess.
22:51:34 <I440r> if you were french it would,
22:51:43 <I440r> any use of encryption there and they arrest you
22:52:03 <Fractal> Are you sure? I thought GPG was developed in france?
22:52:23 <I440r> im sorta sure, i recall one of my linux books mentioning it
22:52:37 <Fractal> Hm. Interesting.
22:53:00 <I440r> neway its definatly time to go zzz
22:53:09 <Fractal> Ya, man. I'm out for a smoke. Later.
22:53:09 <I440r> and radix is a HEURISTIC! still
22:53:11 <I440r> dammit :)
22:53:15 <Fractal> Heh. Whatever.
22:53:22 --- quit: I440r ("Reality Strikes Again!")
22:53:55 <Serg_Penguin> 3here in damn russia even codepage difference was named 'criminal encryption to shut down wrong-thinker
22:54:59 <Serg_Penguin> 3 we do have 3 codepages and guy using less popular one was blamed in criminal use of cryptography
22:55:44 <Fractal> Ah yes. Are these remenants of Stalinist Russia, or new restrictions imposed by the US government?
22:56:38 <Serg_Penguin> no US sure , more likely 1st
22:56:43 <Fractal> Heh.
22:57:14 <Fractal> BBIAB
22:57:26 * Serg_Penguin broke many Caesar cryptogram created by mail gates because of codepage mess
22:57:40 <Fractal> Heh.
22:57:47 <Serg_Penguin> i even wrote proggie 4 breaking them
22:58:49 <Serg_Penguin> using 'pattern words' - ones w/ repeating letters
23:02:21 <Serg_Penguin> BBIAB ?
23:05:22 <Fractal> "Be Back I A Bit". I'm back. :)
23:05:36 <Fractal> Be Back In A Bit
23:20:20 <Serg_Penguin> :)
23:22:42 <Serg_Penguin> historical cryptanalysis is one of my hobbies
23:35:42 --- quit: Serg_Penguin ()
23:59:59 --- log: ended forth/02.10.08