00:00:00 --- log: started forth/04.03.03
00:14:46 --- join: networm (~networm@L0662P03.dipool.highway.telekom.at) joined #forth
00:32:57 --- quit: Herkamire ("off to bed")
00:52:40 * hovil looks at prolog
00:53:51 <MarkT> ok, very beginners question soo to follow..
00:54:04 <MarkT> (I'm reading a tutorial..)
00:56:00 <MarkT> in the definition of a word called countdown..  BEGIN  CR   DUP  .  1 -   DUP   0  =   UNTIL  DROP  ;    I understand most of what I'm looking at, except the 0 test.  I'm assuming that it implicitly tests against TOS, thats a no-brainer.. but after the test, whats left on TOS?
00:56:01 <hovil> ok
00:57:13 <hovil> the = test consumes both variables and leaves a true of false flag on the TOS
00:58:02 <MarkT> ok, then until consumes that correct?
00:58:32 <MarkT> (the result of the test that is)
00:59:43 <networm> [x] BEGIN [x] CR [x] DUP [x] [x] . [x] 1 [x 1] - [x-1] DUP [x-1 x-1] 0 [x-1 x-1 0] = [x-1 flag] UNTIL [x-1]
01:00:10 <networm> so what is on the stack afterwards is the last value of the loop variable.. which should be 0
01:00:53 <networm> [] means what is in the stack
01:01:05 <MarkT> ok, thats what I was just thinking after re-reading.  thanks much.
01:01:26 <MarkT> this is getting scary, it actually makes sence to me.. lol
01:01:30 <MarkT> er
01:01:41 <MarkT> s/sence/sense/
01:03:19 --- quit: networm (Read error: 54 (Connection reset by peer))
01:03:41 --- join: networm (~networm@L0662P07.dipool.highway.telekom.at) joined #forth
01:15:08 --- quit: hovil (Remote closed the connection)
01:50:13 <MarkT> i like this
01:53:34 <networm> Forth?
01:53:40 <MarkT> yes
01:53:55 <networm> did what I said before I got disconnected, about FOrth being addictive, get through?
01:54:11 <MarkT> no it didn't, but its very true
01:54:19 <networm> yes, it is :)
01:54:22 <MarkT> it's so damn elegant
01:54:44 <networm> i really started to look into it like 2 weeks ago.. and since then it has been eating up most of my free coding time
01:55:12 <networm> although i wasn't very fond at all about things like RPN
01:55:29 <networm> but now i see, that's just a consequence of the stack
01:55:57 <MarkT> i see why, I looked at it long ago before I did much coding at all and didnt understand much, now that I have a considreable amount of experience I find it very appealing.
01:58:00 <networm> exactly the same here. madgarden was showing me code snippets, and i couldn't see what was the big difference to other languages or just plain asm. now i would take many concepts of it into my self designed language :)
01:59:52 <networm> actually. the stack concept, the way how you can easily define new words, and how everything is tried to keep simple
02:02:08 --- join: hovil (~hovil@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
02:06:10 --- quit: AldoBR (Read error: 113 (No route to host))
02:20:00 <MarkT> how can I delete a word that I created?
02:20:20 <networm> FORGET
02:20:36 <networm> it also will delete all the words created afterwards
02:20:47 <networm> in case there is some reference to that word
02:21:06 <MarkT> hmm, ok
02:21:34 <networm> i guess, some Forthes also have a DELETE, don't really know
02:37:20 <ASau> MarkT, if you don't want redefine all the words you FORGET, you can redefine only necessary.
02:37:50 <ASau> See:  : A1 .... ; : A2 ... ; : A3 ... ;  : A1 .... ;
02:38:18 <MarkT> hmm, ok.
02:38:20 <ASau> The latter will issue a warning, but the word will be redefined.
02:38:44 <ASau> Although!
02:39:01 <MarkT> ok
02:39:02 <ASau> You need to redefine all words referencing A1
02:39:08 <MarkT> right
02:39:25 <networm> DEFER can be used in that case
02:39:47 <MarkT> I took the 'not unique' warning as an error that meant 'sorry, you cant do that'
02:39:48 <networm> i forgot the syntax.. but the defering word could be changed to be any other, later defined, word
02:39:58 <MarkT> right
02:40:53 <MarkT> i see it's just a warning..
02:40:56 <MarkT> cool
02:41:29 <ASau> I use this more frequently to hide lower level words than I use DEFER analogue.
02:41:46 <MarkT> i see.
02:57:45 <ASau> Ma-a-a-at'!
02:58:04 <ASau> God damn GNU!
02:58:10 <ASau> And FSF in addition.
02:58:22 <ASau> /* This file contains a bit of information about the stack allocation
02:58:23 <ASau>    of the processor.  Since there is no general truth we can't say
02:58:23 <ASau>    anything here.  */
02:58:52 <ASau> What is this half K bytes for?
02:59:09 <ASau> One more copy of " Copyright (C) 1999 Free Software Foundation, Inc."?
03:00:33 <networm> heh. someone who complains about GNU *not* because of their license but lack of documentation :)
03:01:06 <ASau> networm, that's their _source_.
03:01:12 <networm> hm
03:01:31 <networm> source of what?
03:02:03 <ASau> Maybe about one third of Glibc is this text:
03:02:16 <ASau>    Copyright (C) 1996,97,98,99 Free Software Foundation, Inc.
03:02:17 <ASau>    This file is part of the GNU C Library.
03:02:17 <ASau>    The GNU C Library is free software; you can redistribute it and/or
03:02:18 <ASau>    modify it under the terms of the GNU Lesser General Public
03:02:18 <ASau>    License as published by the Free Software Foundation; either
03:02:19 <ASau>    version 2.1 of the License, or (at your option) any later version.
03:02:19 <ASau>    The GNU C Library is distributed in the hope that it will be useful,
03:02:21 <ASau>    but WITHOUT ANY WARRANTY; without even the implied warranty of
03:02:23 <ASau>    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
03:02:25 <ASau>    Lesser General Public License for more details.
03:02:27 <ASau>    You should have received a copy of the GNU Lesser General Public
03:02:29 <ASau>    License along with the GNU C Library; if not, write to the Free
03:02:31 <ASau>    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
03:02:33 <ASau>    02111-1307 USA.  */
03:03:24 <networm> heh
03:03:29 <ASau> I frequently see files, where above part takes the same volume as meaningful part.
03:04:19 <ASau> So what should I think of this 17 MB crap?
03:04:24 <networm> yes. another OS project i'm hacking on has a big ASCII logo in every file.. but at least, below that is just the line "see readme.txt for license information"
03:06:39 <ASau> One more file: the note and two more lines.
03:06:45 <ASau> Both are comments.
03:06:59 <ASau> Calculate ratio.
03:18:27 <hovil> hoho
03:27:17 --- quit: hovil (Remote closed the connection)
03:47:04 --- quit: MarkT ("Leaving")
03:56:41 --- join: hovil (~hovil@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
04:00:06 --- join: MarkT (mark@tom.nss.udel.edu) joined #forth
04:00:54 <MarkT> how would I compile to get a stand alone app? or is that implementation specific?
04:01:08 <Robert> Yes.
04:01:54 <ASau> I've done stand alone configuration for my environment.
04:02:13 <ASau> It's easier when you use assembly-written one.
04:03:03 <MarkT> I'm using Win32Forth
04:03:14 <ASau> Take a look at Albert Horst's (IIRC, http://home.hccnet.nl/a.w.m.van.der.horst/ )
04:03:28 <ASau> ciforth.
04:03:35 <MarkT> ok
04:03:36 * hovil finds ASau's rant on osnews.com
04:04:42 <hovil> =)
04:06:12 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
04:15:44 --- quit: MarkT ("Leaving")
04:29:45 <ASau> Want to feel proud of DOES> ? See Glibc source!
04:29:48 <ASau> Slogan.
04:58:48 --- quit: hovil ("Leaving")
05:10:41 --- join: Serg (~z@212.34.52.140) joined #forth
05:12:09 <ASau> Kak zhizn'?
05:13:46 <Serg> xepoBato
05:14:44 <ASau> Nichjo, u menja kuda khuzhe.
05:27:07 <Robert> Hi, russians. ;)
05:28:50 <ASau> Privet!
05:29:02 <ASau> Shwed.
05:29:05 <ASau> :)
05:33:31 <Robert> Privet :)
05:34:10 <Robert> Time to go home!
05:46:57 --- quit: arke (Read error: 104 (Connection reset by peer))
05:47:10 --- join: arke (~Chris@wbar8.lax1-4-11-099-104.dsl-verizon.net) joined #forth
05:54:32 <mur> mmoi
05:54:53 <ASau> Privet!
05:55:32 <mur> privet russkii schwedskij
05:55:53 <mur> i potomy nje amerikanskij toshe
05:56:01 <ASau> A finskijj?
05:56:09 <mur> i can't remember the endings :P
05:56:54 <mur> if you dont believe that i tell then that light shimmers right against the screen and i cant see how i write ;)
05:58:11 <ASau> Why not move screen in shadow?
05:58:38 <mur> because i want to enjoy spring sun without getting wet 
05:58:39 <mur> :)
05:59:10 <ASau> Ah. You've got sunny day there.
05:59:14 <ASau> That's good.
05:59:43 <ASau> It snowed here today.
05:59:52 <mur> here too a bit in the morning
06:00:15 --- join: etrangepulpe_ (~EP@neptune.esial.uhp-nancy.fr) joined #forth
06:00:29 <ASau> Good evening, etrangepulpe_!
06:00:31 <etrangepulpe_> hi
06:00:39 <mur> bonjour
06:01:05 <etrangepulpe_> I'm a beginner in forth
06:01:10 <ASau> mur, here it snowed in afternoon.
06:01:27 <ASau> mur, I hope in 3 days we'll get sun.
06:01:44 <mur> spring sun with no rain is best
06:01:51 <mur> the snow could just disappear without raining
06:02:02 <ASau> etrangepulpe_, and what are you writing in Forth?
06:02:31 <ASau> mur, of course this is the best.
06:03:02 * mur is admiring how good job he did last night
06:03:03 <ASau> mur, but it never rains here.
06:03:04 <mur> in few minutes
06:03:08 <mur> hmm
06:03:12 <ASau> mur, it's just pouring down.
06:03:13 <mur> this looks very good
06:03:23 <mur> and its going to be printed at printhouse
06:03:45 <mur> well used 5 minutes!
06:04:29 <etrangepulpe_> ASau: nothing special, at this moment I'm reading some tutorials, I'm far from knowing everything yet
06:04:57 <ASau> etrangepulpe_, what do you plan to use Forth for?
06:06:37 <etrangepulpe_> I don't know yet :) I already program in asm (on TI-z80) so I thought learning forth would be nice, but I have no project right now
06:07:41 <etrangepulpe_> in fact I like minimalist things
06:07:47 <ASau> Do you use Forth for Z80?
06:08:00 <etrangepulpe_> I plan to.
06:08:44 <ASau> You may find several on Taygeta ( ftp://ftp.taygeta.com/pub/Forth/ )
06:09:20 <Serg> zx forths on taygeta ?
06:09:39 <Serg> i got 2 for Spectrum, but...
06:09:51 <Serg> speccy "os" is BAD THING
06:15:00 <ASau> Serg, what OS do you mean?
06:15:08 <ASau> IIRC, there were several ones.
06:15:16 <etrangepulpe_> ASau: thx, too bad it is not under GPL
06:15:46 <ASau> etrangepulpe_, what do you have against public domain?
06:16:25 <ASau> There exists p.d. Forth.
06:16:44 <etrangepulpe_> oops
06:16:55 <etrangepulpe_> I didn't read clearly :)
06:20:17 <Serg> ASau: TR-DOS in ROM is trash, IS-DOS is overly complex and not widespread
06:25:41 <ASau> I never used both of them. I used ZX Basic. It worked well.
06:26:08 <ASau> ...niether of them.
06:26:26 <Serg> ASau: had your Speccy floppy disk, or tape only ?
06:28:44 <ASau> I used to tape.
06:29:38 <ASau> I had enough RAM.
06:29:58 <Serg> nowdays all use floppy, some - even HDD and CD
06:30:30 <Serg> http://zx.da.ru
06:30:40 <Serg> i bought their CD w/ warez
06:31:42 <ASau> Today, I have no ZX.
06:32:36 <ASau> And it seems, I won't own even in future.
06:33:35 <Serg> if only KGB regime will forbid computers as "copying devices", and people will make 'em covertly, like radio recievers in East Germany
06:36:24 <ASau> This regime will probably leave you without money rather than forbid anything.
06:37:05 <Serg> hell no - my income grows since 1999 w/o drop
06:37:07 <ASau> Without money hardly you will think about computers.
06:37:26 <Serg> it depends on personal ability ;)
06:37:29 <arke> mmm.....ed .... 
06:37:36 <ASau> Don't say "hop".
06:37:49 <arke> I wonder if I could create a line editor which packs much more functionality
06:38:13 <ASau> arke, yet another "ed"?
06:38:32 <ASau> Or "ex"?
06:38:43 <warpzero> line editors :(
06:38:53 <ASau> Take a look on TECO.
06:38:54 <warpzero> needs more syntax expansion
06:39:09 <Serg> arke: why u need one ?
06:39:52 <ASau> Serg, I don't need y.a. "ed".
06:40:05 <ASau> I've got TECO. :)
06:40:05 <Serg> sure
06:42:01 --- join: segr_ghost (~z@212.34.52.140) joined #forth
06:42:30 <segr_ghost> re
06:43:16 <segr_ghost> damn i forgot nickserv paswd
06:44:00 --- quit: Serg (saberhagen.freenode.net irc.freenode.net)
06:44:01 --- quit: Robert (saberhagen.freenode.net irc.freenode.net)
06:44:01 --- quit: warpzero (saberhagen.freenode.net irc.freenode.net)
06:44:44 --- join: Robert (~snofs@c-775a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
06:44:44 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth
06:45:43 --- quit: Robert (Remote closed the connection)
06:45:54 --- join: Robert (~snofs@c-775a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
06:49:14 --- quit: ChanServ (saberhagen.freenode.net irc.freenode.net)
06:51:41 <madgarden> Hello!
06:52:03 <madgarden> Heh, I just have to pass networm's Forth question along here...
06:52:06 <madgarden> "<networm> i wonder what CLIT does"
06:52:58 <ASau> LIT for Character?
06:53:26 <madgarden> Hehe. Yea, probably. But it translates to funny English. :P
06:54:58 --- join: ChanServ (ChanServ@services.) joined #forth
06:54:58 --- mode: irc.freenode.net set +o ChanServ
06:58:12 --- quit: warpzero (Excess Flood)
06:58:59 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth
07:03:55 --- quit: qFox (Read error: 104 (Connection reset by peer))
07:04:07 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
07:06:46 --- quit: etrangepulpe_ ("Client Exiting")
07:07:01 <arke> madgarden: LOL
07:07:14 <arke> about ed: Yes, I'm using Ed. So what?
07:07:24 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
07:07:52 <arke> tathi: o.O
07:07:56 <arke> tathi: you're back!
07:08:20 --- quit: qFox (Read error: 54 (Connection reset by peer))
07:08:27 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
07:09:46 <mur> hmm
07:10:22 <arke> dobroe ubro mur :)
07:10:24 <arke> terve mur :)
07:10:31 <tathi> arke: :)
07:10:32 <arke> ASau: how close was that? :P
07:17:45 <mur> hello arke
07:17:59 <arke> gotta go to school, later
07:18:52 <ASau> Almost right: "utro".
07:20:16 <mur> dobrij utro?
07:20:38 <mur> transliteration is black magic you never can do it correctly
07:22:22 <ASau> mur, see http://www.orwell.ru/info/img/tlit_2.png
07:22:52 <ASau> "dobroe" as "utro" is not masculinum.
07:23:15 <mur> neuter?
07:23:21 <ASau> yes
07:23:26 <mur> hmm
07:23:54 <mur> what's GOST?
07:24:44 <ASau> It means "Gosudarstvennyjj obshhesojuznyjj standart."
07:25:01 <ASau> It used to mean.
07:25:30 <ASau> Now it means "Gosudarstvennyjj standart." (National standard.)
07:26:04 <mur> it reads: "russian] cyrillic alphabets to latin alphabets in ispol.. something's way latin letters
07:26:55 <mur> how they are written with latin letters?
07:27:06 <mur> does that reach the meaning of the end of sentence?
07:28:40 <mur> can you read latin alphabets, the ones with special expansions on them (like ÿ, ä) or do you see them cyrillic?
07:29:04 <ASau> It's standard for one-to-one representation of Cyrillic alphabet in Latin.
07:29:22 <mur> well but they still vary. finnish translitteration uses phonetic rules
07:29:23 <ASau> I see them Cyrillic.
07:29:42 <ASau> You wrote: ja, d (Cyr.)
07:29:58 <mur> that (bi) letter "y" is representated in finnish like 'y'
07:30:06 <mur> y with 2 dots on it
07:30:22 <segr_ghost> e w/ ears ;))
07:30:41 <mur> dutch ij means the same, but they use that because that letter is rarely available
07:30:43 <ASau> e "umlaut"?
07:30:45 --- nick: segr_ghost -> Serg
07:30:47 <mur> y umlaut
07:31:25 <ASau> Ah. You mean short "I".
07:31:55 <mur> tha one that looks like "bi"
07:32:00 <mur> visually
07:32:05 <ASau> Yes, that should not be rare.
07:32:12 <ASau> "bI"?
07:32:25 <mur> i mean umlaut-y is rare in keyboards and charsets
07:32:55 <ASau> Maybe.
07:33:21 <madgarden> What is this doing?
07:33:21 <madgarden> : (;code) r> latest name> ! ;
07:33:27 <ASau> I never have meet usage of "y:"
07:33:46 <mur> ASau: it's the bI letter's approciate translitteration
07:33:49 <mur> phonetically
07:33:51 <ASau> madgarden, this changes CF of latest defined word.
07:33:56 <mur> also used by dutch.. in theory
07:34:23 <madgarden> Oh, right! Thanks.
07:34:26 <ASau> We use "y" without ":".
07:35:57 <ASau> Phonetic transliteration is not possible as we have soft "sh" and  "j"+vowels.
07:36:25 <mur> there are letters for each sound in phonetic char-set
07:36:36 <mur> although they are not rarely seen, full phonetic writing
07:37:11 <ASau> It is not useful as we have variations in pronounsation.
07:38:00 <ASau> There're rather distinct northern, southern and eastern branches.
07:38:04 <mur> phonetic texts do not appear often anyhow
07:38:08 <mur> unless you are linguistic
07:39:35 <ASau> BTW, you can see how Latin is used by Czecks.
07:40:01 <mur> and i think you can see how chinese write :)
07:40:04 <mur> that reminds me
07:40:24 <mur> that there was chinese newspaper at my uni. the address read that subscriber was "Mr. Library"
07:40:27 <mur> :)
07:40:32 <ASau> Yes, but they have reasons for it.
07:40:49 <mur> same historical reasons than any other character set
07:41:15 <mur> korea modernised their set. korean is written like latin, but they have just alphabet that look like chinese in a way
07:41:30 <mur> but they have phoneme written charcterset
07:41:32 <ASau> AFAIR, they have very distink dialects.
07:41:40 <ASau> China is too big.
07:41:57 <mur> i dont talk about possibilities now
07:42:04 <mur> changes go on little by little all the time
07:42:12 <ASau> Yes. I know.
07:42:28 <ASau> Russian changes rather fast.
07:43:08 <ASau> Recent major reform was held in 1918 or 1920.
07:43:40 <mur> what kind of major reform?
07:44:05 <ASau> Some of excess characters were excluded.
07:44:21 <ASau> There were significant changes in grammar.
07:44:26 <mur> like "I" kind of I that is in some cyrilic written languages?
07:44:33 <mur> and "yuk" character?
07:44:46 <ASau> Yes.
07:44:57 <mur> or is yuk character only for other than russian
07:45:12 <ASau> Hard sign usage was limited.
07:45:48 <ASau> Are you linguist?
07:46:02 <ASau> I'm not.
07:46:02 <mur> not really
07:46:24 <ASau> So I may mistake.
07:54:08 <ianp> how is latin used by czechs?
07:58:37 --- quit: Serg ()
08:01:59 <ASau> http://www.main.cz/knihovna/Capek/rur_0.htm
08:03:53 <ASau> Use corresponding font.
08:19:05 <ASau> BTW, mur, arke. See: http://www.orwell.ru/info/tools/scon_02.php
08:24:02 <mur> that'd require inputting cyrillic somehow :)
08:24:17 <mur> i'll bookmark that one too
08:24:27 <Robert> orwell.ru sounds interesting.
08:28:44 <ASau> mur, that depends on your OS.
08:29:37 <mur> i coudl set cyrillic keyboard but i woudnlt know how to use it anywas
08:29:48 <mur> or use character palette and double click each letter 
08:29:58 <Robert> Trial and error, mur. :)
08:30:41 <mur> ;)
08:32:08 <ASau> mur, JCUKENGShShhZKh"
08:32:24 <ASau> mur, FYVAPROLDZhEh
08:32:44 <ASau> mur, JaChSMIT'BJu
08:33:30 <ASau> "Jo" is on <`~> or on </?> key.
08:33:51 <mur> hehee
08:33:56 <mur> i was trying to read that as russian
08:33:58 <ASau> That's the only layout.
08:34:00 <mur> i failed ;)
08:34:30 <mur> what are the commands required to command in gui?
08:34:31 <ASau> Why?
08:34:39 <mur> general short commands that are required
08:34:47 <ASau> It is rather easy read in Russian.
08:35:10 <mur> no i mean i thought that the charmap was actaully some writing
08:35:13 <mur> but of course it is now
08:35:15 <mur> not
08:35:27 <ASau> I don't know your GUI.
08:35:47 <mur> i was thinking of creating asm kind of command set (why not forth too!)
08:35:58 <mur> that you could have simple commands (few) to command everything'
08:36:03 <mur> like risc language for gui
08:36:53 <ASau> If you're running Linux, console is this: setfont alt-8x16; loadkeys ru* ; mapscrn koi2alt
08:37:14 <mur> i run macintosh 
08:37:16 <ASau> echo -ne "\033(K"
08:37:19 <mur> i know how to change the charset
08:37:47 <ASau> Hmm. I never run McIntosh.
08:38:37 <ASau> I don't believe there's no Russian keyboard layout.
08:39:09 <mur> non o
08:39:11 <mur> ther is
08:39:13 <mur> of course
08:39:24 <mur> i just said i dont nkow how to use it because i am not used to it
08:40:05 <ASau> That's not so hard. All you need is switching back.
08:40:38 <ASau> BTW, http://www.trigeminal.com/samples/provincial.html
08:46:24 --- join: Herkamire (stjohns@h000094d30ba2.ne.client2.attbi.com) joined #forth
08:47:56 <ASau> Dobryjj vecher!
08:51:53 --- quit: networm ("Client exiting")
08:54:48 <mur> :)
09:03:12 <ASau> Well, maybe it's clear why I ask not to use umlauts etc.
09:04:05 <ASau> I can write this way: Ìîãó ïèñàòü è êèðèëëèöåé, ýòî ìíå íå òðóäíî.
09:09:30 <ASau> BTW, this made a background of one story.
09:11:00 <ASau> One French had been in Russia, here he found friend.
09:11:53 <ASau> After a half year he returned to France he was to send a box into Russia.
09:12:34 <ASau> His Russian friend sent his address in Cyrillic.
09:13:09 <ASau> He didn't know that French do not usually have fonts needed.
09:13:31 <ASau> So address looked approx. as above.
09:14:47 <ASau> French man signed his box with a strange lines of accented vowels interleaved with digits.
09:16:35 <ASau> It was very impressive to get box by post with unreadable address and subscript after deciphering.
09:18:53 <ASau> The only word that made trasfer possible was "voluntarily" added to the address: "Russia."
09:24:17 --- quit: tathi ("leaving")
09:27:35 <mur> heh
09:27:38 <ianp> ASau - lol :)
09:31:47 <mur> postmasters are awesome sometimes
09:32:37 <ASau> P. O. index is a great thing really.
09:34:04 <mur> Rost Office
09:34:05 <mur> :)
09:43:27 --- join: AldoBR (Aldo@200.141.173.71) joined #forth
09:44:26 <ASau> Gamarjoba!
09:45:03 <ASau> Hello!
10:33:53 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth
10:34:15 --- mode: ChanServ set +o kc5tja
10:38:04 <madgarden> kc5tja, get Blazin' Forth running? :)
10:38:26 <kc5tja> madgarden: No.
10:38:31 <kc5tja> I have more important things to work on at the moment.
10:39:18 <madgarden> Bah!
10:39:26 <madgarden> Priorities...
10:39:46 * kc5tja designed a synchronous-to-asynchronous bus interface circuit today for the 65816, for use in the Kestrel.
10:40:01 <kc5tja> The bus interface is important, because the CPU can address slow-running I/O devices at any time.
10:40:19 <kc5tja> (the VIA chips will run at 5MHz)
10:40:32 <kc5tja> (now that I have this circuit designed; before I had them running at 3.15MHz).
10:40:41 <kc5tja> This means the PIB will run faster too -- 1.25Mbps.  :D
10:43:16 * madgarden goes to look up acronyms...
10:44:11 --- join: Nutssh (~Foo@gh-1063.gh.rice.edu) joined #forth
10:44:37 <madgarden> What is a PIB?
10:45:41 <kc5tja> It's the bit-serial, peripheral interconnect bus that is used to link all the PIC chips together under main CPU control.
10:45:53 <kc5tja> Think of it as a bit-serial IEEE-488 interface.
10:46:04 <kc5tja> Inspired by Commodore's IEC bus, but a *heck* of a lot faster.  :)
10:46:17 <madgarden> Ahh...
10:47:00 <kc5tja> I think the only critical difference is that, to handle the asynchrony of the interface, it'll be a serial loop interface (e.g., it'll need two serial ports; one for sending, one for reading).
10:47:00 <madgarden> What are you using to lay out the board?
10:47:20 <kc5tja> madgarden: I'm probably going to go with ExpressPCB.com's offering.  Free software, cost estimation built right in.
10:47:40 * kc5tja will try to get it working under Wine first.  If that fails, I'll just load it into my WIndows XP harddrive.
10:48:03 <kc5tja> Eventually, I'll probably make my own PCB software.
10:48:08 <madgarden> Hey, that's a cool business model.
10:48:16 <kc5tja> Yep.
10:50:45 <kc5tja> There are several places that offer business models like that.  But most everyone on the 6502.org community use ExpressPCB, so I figure they're pretty good to work with, even if they're not the cheapest.
10:53:36 <madgarden> Just installed it.
10:58:01 <kc5tja> I'm now off to college.
10:58:05 <kc5tja> Back in about 3 hours or so.
11:01:16 --- quit: Nutssh ("Client exiting")
11:06:38 --- join: Nutssh (~Foo@gh-1177.gh.rice.edu) joined #forth
12:24:20 <madgarden> ASau, Dobryjj vecher!
13:17:09 --- quit: qFox ("if at first you dont succeed, quit again")
13:18:13 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
14:08:44 --- quit: Nutssh ("Client exiting")
14:09:12 <kc5tja> Back, for now.  I have to run some errands in a few hours though.
14:09:22 <Robert> Hi
14:09:34 <kc5tja> Howdy
14:14:18 <ianp> Hello.
14:25:34 <kc5tja> You know, I'm kind of half wondering whether I should just load the Kestrel's OS directly into RAM from serial EEPROM.  That would greatly simplify address decoding logic, but would also require a few more ICs.
14:28:20 <madgarden> You were going to run it out of EEPROM?
14:30:10 --- join: Nutssh (~Foo@gh-1177.gh.rice.edu) joined #forth
14:30:25 <kc5tja> Yes.
14:30:37 <kc5tja> The EEPROM would contain code to copy itself into RAM, then bank out the EEPROM.
14:30:59 <kc5tja> While the EEPROM was banked in, the CPU effectively ran at 3MHz.
14:31:47 <kc5tja> But if I have a PIC chip that performed the boot process by loading, say, 1KB of code from a serial EEPROM into RAM, then booted the 65816 into that code, then I could use that boot-strap to read the remainder of the code off the serial EEPROM.
14:32:32 <kc5tja> While I would need a few more chips to do this with, it would eliminate the whole fast/slow CPU thing, which might be a net win.
14:32:51 <kc5tja> Especially since I'm going to have two PICs in the system already (one to generate VGA timings, and one to generate audio).
14:34:00 <kc5tja> The particularly nice thing about doing it this way is I can fake a serial EEPROM via the PC's parallel port.
14:34:16 <kc5tja> So I never have to burn an EEPROM chip, especially while in development.
14:43:53 <kc5tja> Just evaluating some options right now...
14:46:06 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
14:58:59 <TreyB> kc5tja: you do electronic design, yes?
15:02:00 <TreyB> Ah, away.  I'll try again later.
15:02:03 <madgarden> I like the idea of extra CPU speed.
15:03:28 --- join: TheRealIanP (ian@inpuj.net) joined #forth
15:03:36 <madgarden> Plus, it would give the user more control over the OS, non?
15:05:09 --- quit: ianp (Read error: 60 (Operation timed out))
15:05:14 --- nick: TheRealIanP -> ianp
15:13:28 --- join: dasbear (das@adsl-64-219-100-35.dsl.lgvwtx.swbell.net) joined #forth
15:17:54 * warpzero is away: Sunny day, sweeping the clouds away, on my way to where the air is sweet!  Can you tell me how to get, how to get to Sesame Street?
15:18:25 * warpzero is back (gone 00:00:24)
15:18:47 <warpzero> grr
15:20:08 <kc5tja> TreyB: Back
15:20:18 <kc5tja> TreyB: I can do it, yes.
15:20:57 <kc5tja> madgarden: You make it out like the CPU will run at 3MHz all the time.  It doesn't.  It only has those extra wait-states when accessing the EEPROM.
15:21:19 <madgarden> And how often will it be executing code from EEPROM?
15:21:29 <kc5tja> madgarden: What the boot-copy hardware does is it provides a simpler address decoder.
15:21:36 <kc5tja> madgarden: Except for the code to copy ROM to RAM, *never*.
15:21:52 <warpzero> is mister isforth around
15:22:02 <madgarden> Oh, well OK then. :) I though the system portion of your Forth would be running out of EEPROM.
15:22:07 <kc5tja> That's the whole point of being able to bank the EEPROM in and out of memory space.
15:22:25 <kc5tja> madgarden: It still needs some kind of ROM.  It can't just boot into thin air.
15:22:32 <madgarden> Of course.
15:22:43 <warpzero> isnt chandler the one with isforth
15:22:54 <kc5tja> warpzero: I440r is the one who wrote isForth.
15:23:05 <warpzero> whats his other nick
15:23:25 <kc5tja> madgarden: Actually, I might just stick with the regular EEPROM.  Supporting the boot-PIC requires resources from the VIA chip that I don't want to invest.
15:24:55 <kc5tja> Oh, and even though BoxOS exists in EEPROM, the computer won't have the ability to reflash itself.  Forth will inevitably overwrite itself under those circumstances.  Any write to the ROM space will actually pass through to the RAM underneath the ROM.
15:25:02 <kc5tja> (as it does with the Commodore 64 and Commodore 128).
15:26:28 --- nick: scope_ -> scope
15:29:06 <warpzero> kc5tja: whats i440r's other nick?
15:30:00 <kc5tja> mmanning
15:30:24 <warpzero> right!
15:30:40 <warpzero> so how is life kc?
15:31:15 <kc5tja> It'd be going a lot better if I had some cash to my name.
15:31:26 <warpzero> :(
15:31:27 <kc5tja> After rent, I'll have only $500 in the bank.
15:31:35 <kc5tja> And car insurance is due -- conveniently, it's $550.
15:31:46 <warpzero> :( damn dude that sucks
15:31:53 <kc5tja> Yup.
15:32:07 <warpzero> one of the many reasons i dont have a car
15:32:16 <kc5tja> I *need* a car.
15:32:29 <kc5tja> Though I've been riding my bike more often, my client is some 40 miles away from me.
15:32:34 <warpzero> I kinda do too, there isn't much else in montana.
15:32:34 <kc5tja> No way in hell I'm going to bike to his location.
15:32:41 <kc5tja> (not in any reasonable time frame at least.)
15:32:45 <warpzero> yeah
15:32:58 <warpzero> aren't you making some uber forth box thing too?
15:33:00 <dasbear> 40 miles is too far for every day
15:33:10 <warpzero> yeah
15:33:20 * kc5tja doesn't go to his location every day.
15:33:22 <kc5tja> Once a week.
15:33:34 <warpzero> i don't understand how kc5 is so smart but so poor :(
15:33:40 <kc5tja> But 40 miles is also not so far.  Plenty of people in California commute 30 to 50 miles daily.
15:33:53 <kc5tja> warpzero: easy.  The economy doesn't like smart people.
15:33:58 <warpzero> guess not
15:33:58 <kc5tja> If you're a yes-man, you rise to the top.
15:34:02 <kc5tja> If you're smart, you're an outcast.
15:34:10 <kc5tja> "Worse Is Better."
15:34:19 <warpzero> :(
15:34:22 <warpzero> stupid america
15:35:27 --- quit: tathi ("leaving")
15:35:33 <kc5tja> And, as far as the ForthBox Kestrel is concerned, it's hardly uber.  :)
15:35:56 <warpzero> well how much is one?
15:36:10 <kc5tja> (Raven, on the other hand, will be rather a nice system.  Still won't compete against PCs, but, well, it'll probably make Amiga computers fear...)
15:36:23 <kc5tja> I'm going to go out on a limb and guess somewhere in the $65 to $80 range.
15:36:27 <kc5tja> But:
15:36:35 <kc5tja> 1) It's just a PC board with a bunch of parts you put together yourself.
15:36:45 <warpzero> oh okay
15:36:47 <warpzero> sign me up
15:36:50 <kc5tja> 2) As such, it needs instruction manuals on HOW to put it together.  Therefore, that adds to the price.
15:37:26 <dasbear> anyone know when i440r is usually around?
15:37:26 <kc5tja> 3) These prices are preliminary.  I'm making every effort I can to reduce the cost on this thing as much as possible, while still being halfway economically viable for me.
15:37:41 <kc5tja> dasbear: He's on a contract position right now, so he rarely pops in.
15:37:54 * kc5tja realizes that if the price is too high, nobody will buy one.
15:38:05 <kc5tja> If the price is too low, I can't recoup costs, much less strive for a living making these things.
15:38:08 <warpzero> dasbear: we are both looking for him hehe
15:38:22 <dasbear> kc5tja: thanks... I'll email him then
15:38:48 <kc5tja> I wonder how much of a market there is for homebrew kit computers.
15:39:00 <kc5tja> According to ZX81kits.com, they sold over 1800 Sinclair ZX-81 kits.
15:39:26 <kc5tja> That's rather appreciable.  At 99 pounds a pop, that's close to 100,000 pounds in revenue.
15:39:34 <warpzero> so what are these thigns gunna have?
15:39:47 <kc5tja> warpzero: http://www.falvotech.com/cgi/fsforth/ForthBoxKestrel
15:40:39 <arke> hiya
15:40:45 <warpzero> ugh, polled io?
15:40:55 <kc5tja> Love it or leave it.
15:41:08 <kc5tja> Unless you have a better solution for a kitable computer that is under $100.
15:41:17 <warpzero> flash?
15:41:23 <kc5tja> What about it?
15:41:32 <warpzero> maybe you could use a flash memory controller?
15:41:40 <kc5tja> Yes, they're called IDE controllers.
15:41:45 <warpzero> oh
15:41:47 <warpzero> nm then
15:41:52 <kc5tja> And the CPU doesn't have DMA channels; ergo, it's polled.
15:42:38 <kc5tja> Note the paragraph that reads, "Note that 'high performance,' . . . and other awe-inspiring traits are not in the above list."
15:42:42 <kc5tja> There is a reason for it.
15:42:43 <kc5tja> $$$$ is that reason.
15:43:01 <warpzero> um
15:43:06 <warpzero> what about memory?
15:43:12 <kc5tja> It costs ME money to put those in (R&D and fabrication costs), costs the customers money because I have to pass on that cost, and it adds to the customer's frustration because anything else these days is surface-mount technology.
15:43:14 <warpzero> i dont see a spec for it
15:43:23 <kc5tja> It'll ship with around 512KB of static RAM standard.
15:43:30 <warpzero> hmm
15:43:31 <kc5tja> It'll probably be expandable to 2MB on the motherboard.
15:43:31 <arke> yeah, theres no mem spec for the kestrel
15:43:41 --- join: Sonarman (~matt@adsl-64-160-166-93.dsl.snfc21.pacbell.net) joined #forth
15:44:19 <dasbear> Why not use a game console as a homebrew kit
15:45:11 <kc5tja> There.
15:45:21 <kc5tja> dasbear: Because it's not homebrew -- it's already built.
15:45:25 <kc5tja> Thus, defeating the whole purpose.
15:45:31 <warpzero> i dont think im gunna be able ot build one of these thigns
15:46:07 <kc5tja> warpzero: You'll never know.
15:46:08 <dasbear> true, atmel avr boards are simple to make, even on a solderless breadboard
15:46:19 <warpzero> kc5tja: ive never soldered before :(
15:46:44 <dasbear> "solderless", on that you just plug in the parts and plug in wires
15:47:04 <kc5tja> dasbear: Those breadboards are next to useless with any signal above 4MHz.
15:47:08 <kc5tja> Too much capacitance.
15:47:54 <dasbear> true. But the avr has most of what you need built in, the external items can be slow. I've done it before.
15:47:55 <warpzero> is there open source vhdl or verilog simulator software stuff?
15:48:13 <kc5tja> warpzero: None to my immediate knowledge.
15:48:17 <warpzero> fuck
15:48:19 <dasbear> freehdl
15:48:26 <dasbear> and ghdl
15:48:30 <warpzero> oh ho!
15:48:36 * kc5tja wasn't aware of those.
15:48:43 --- nick: arke -> chris-xp
15:48:45 <kc5tja> Now the question is, how complete are they?
15:48:51 <dasbear> both simulate, but don't synthesis
15:48:59 <kc5tja> That's not what I asked.  :)
15:49:07 <dasbear> I played with both.
15:49:26 <dasbear> for what I was doing they were quite complete
15:49:28 <kc5tja> Convenient, because the ISE software from Xilinx synthesizes, but doesn't simulate.
15:49:43 <chris-xp> by the way, my editor of choice is ed
15:49:56 <warpzero> what do you mean by synthesize?
15:50:08 <kc5tja> warpzero: Compile is to software as synthesize is to hardware.
15:50:17 <dasbear> create a netlist for downloading to a pld
15:50:48 <warpzero> okay
15:50:53 <Sonarman> pld? :)
15:51:03 <dasbear> cpld or fpga  :)
15:51:09 <warpzero> programmable logic device
15:51:22 <Sonarman> thanks
15:51:32 <kc5tja> dasbear: Now if only vendors would release the specs to the netlist formats their chips understands....
15:51:47 <dasbear> hehehe... yep. :)
15:58:01 <warpzero> dammit i really want an fpga
15:58:09 <warpzero> but fuck if i have that much money :(
15:58:12 <kc5tja> warpzero: http://www.fpga4fun.com
15:58:23 <kc5tja> Look at their Pluto board.
15:59:52 <warpzero> 500 logic elements eh
16:00:18 <kc5tja> Hey, well, it's at least a start.  :)
16:00:38 <warpzero> aren't cpld's more dense than that
16:00:53 <kc5tja> FPGAs can go up to hundreds of thousands of logic blocks.
16:00:58 <kc5tja> They're both about equally 'dense.'
16:01:05 <kc5tja> It's just how they are organized internally.
16:01:08 <dasbear> only 500 logic elements force you to really unit test your design. :)
16:01:20 <warpzero> define unit test
16:01:23 <kc5tja> CPLDs have fewer logic blocks, but they are BIG logic blocks.  Hence, they are well suited for lots of combinatorial logic.
16:01:43 <kc5tja> FPGAs, however, have lots of *little* logic blocks, so they're good for sequential logic like CPUs and DMA controllers and such.
16:01:54 <kc5tja> dasbear: Unit test?
16:02:05 <dasbear> test each design unit, in other words, make small modules
16:02:16 <dasbear> and test each one individually
16:02:22 <kc5tja> Sure, but why can't you do that with a huge, unweildly FPGA?
16:02:33 <dasbear> you can
16:02:33 <warpzero> kc5tja: i would like something i could maybe implement a simple simple cpu on or something
16:03:13 <kc5tja> warpzero: You'll probably want something along the lines of a mid-class Spartan IIe from Xilinx then.
16:03:20 <dasbear> I do that at work, via a processor running forth, I test subdevice in an fpga
16:03:20 <kc5tja> Lots of pins on those things though.
16:03:32 <kc5tja> Most will undoubtedly go unused for a CPU.
16:03:40 <warpzero> er uh
16:03:45 <warpzero> those are expensive :(
16:03:58 <kc5tja> warpzero: An unprogrammed Spartan IIe is reported to cost $13.50 for one chip.
16:04:08 <kc5tja> Yes, it's "expensive" when compared to, say, a $6 16MHz 65816.
16:04:10 <warpzero> kc5tja: yeah but you have to uh program it
16:04:36 <kc5tja> But considering you can probably get a 33MHz MISC core on it easily, and have room to spare, well, it suddenly looks really attractive compared to, say, a $150 25MHz 68020.  (Yes, they cost that much.)
16:05:07 <warpzero> are you fucking with me? thats more than i pay for my athlons!
16:05:09 <kc5tja> warpzero: If you have Windows, you can download Xilinx's ISE software for free.  You won't be able to simulate, but it will synthesize.
16:05:17 <warpzero> i dont have windows
16:05:43 <kc5tja> Yep.  I'm with you.  I was *shocked* when I saw the price of a 68020 now-a-days.  Simply insane.
16:05:55 <kc5tja> It's Motorola, trying to deliberately kill off that processor line.
16:06:04 <kc5tja> PowerPCs are way faster and way cheaper.
16:06:17 <dasbear> the coldfire line is the replacement for 68xxx
16:06:19 * kc5tja can get a PowerPC 403ga for something like $44 from IBM.  It'll do 66MHz at least.
16:06:45 <warpzero> wait here
16:06:49 <kc5tja> dasbear: Yeah, but they're not true 68K processors.  They're reduced functionality, and many don't even have distinct user/supervisor modes.
16:07:02 <warpzero> http://www.fpga4fun.com/board_dragon.html would this work?
16:07:18 <warpzero> TI still uses 68K
16:07:28 <dasbear> kc5tja: true, but a lot of things in embedded systems, that does not matter
16:08:11 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
16:09:04 <kc5tja> dasbear: If I'm designing a real computer, I want it to be real, doggone it.  :)
16:09:35 <kc5tja> warpzero: Well, yes; you can still get a 68000 for pretty cheap in embedded forms, like the 68332 Dragonball processor (used in Palm Pilot and compatible devices)
16:10:02 <warpzero> kc5tja: spartan ii e you say?
16:10:18 <kc5tja> That's what I was told.
16:10:42 <kc5tja> I'm not sure where he got his prices from; I'm sure he ordered parts from a distributor; I can't think of any other way he got them so low.
16:10:47 <dasbear> kc5tja: I know what you mean. protected memory you can handle with an fpga, but user/supervisor you can't. (to prevent overwriting interrupt vectors and such) 
16:11:14 <warpzero> kc5tja: i mean as far as gatecount goes, the iie is enough for a cpu right?
16:11:46 <kc5tja> warpzero: Well, Spartan II is a whole family of FPGAs.  They come in various sizes.
16:12:02 <kc5tja> That's why I said 'mid-class,' because the smallest of them probably won't work.
16:12:10 <dasbear> well, I gotta go. I'll be back later.
16:12:15 <kc5tja> But you don't want to go too big either, lest you waste the chip's capabilities.
16:12:19 <kc5tja> 73 dasbear 
16:12:36 <warpzero> kc5tja: where do i look up which has how many?
16:12:38 <dasbear> bye
16:12:43 --- quit: dasbear ("leaving")
16:12:50 <kc5tja> warpzero: Xilinx has data sheets on their website.
16:13:05 <kc5tja> You have to browse their online documentation.
16:14:03 <kc5tja> If your CPU comes to something like 4000 gates, look for a chip that has around 6000 gates at least; this is because the synthesizer tool will never be able to pack everything in super tight.  On average, you'll get 80% or so density usage in the chip.
16:14:17 <warpzero> pdfs
16:14:31 <warpzero> okay
16:14:46 <warpzero> whats the average gate > transistor conversion just wondering
16:15:34 <kc5tja> That depends on the manufacturer and the process technology.  My guess of '1 gate is 6 transistors' is just a total, utter wild guess (I'm making the assumption that the design will "average out" to a lot of 3-input gates).
16:15:49 <warpzero> hmm
16:16:51 <warpzero> so are there really 4k gate cpus?
16:16:57 <kc5tja> Well, I have to get back to work here.  I have lots of homework to complete, and then I need to spend about 4 hours coding for my customer.
16:17:06 <kc5tja> warpzero: The 6502 is a 4K gate CPU, yes.
16:17:17 <warpzero> jesus thats tiny
16:18:19 <warpzero> one more question
16:18:19 <kc5tja> That's partly why it's so damn fast for an 8-bit CPU.
16:18:30 <warpzero> how do i read this table :(
16:18:48 <warpzero> Logic Cells is the number of groups of gates right?
16:19:17 <kc5tja> Yes.  Each logic cell (aka Logic Block) constitutes precisely *one* programmable "chunk" of the chip.
16:20:05 <warpzero> adn so when it says 23K-50K "Typical System Gate Range"? this is the number of usable gates about?
16:20:12 <AldoBR> kc5tja you know a cheap ic to connect a small processor (like the 68000) to a 10mbit ethernet ?
16:20:49 <kc5tja> AldoBR: No.  You have to read the Ethernet controller chip's data sheets.  They'll tell you how to get data in and out of the chip.
16:21:01 <AldoBR> right
16:21:04 <kc5tja> warpzero: Roughly, based on some nebulous concept of an "average design," yes.
16:21:12 <AldoBR> but i need a chip name to search for pdfs :P
16:21:25 <warpzero> kc5tja: thanks for the help duders
16:21:29 <kc5tja> AldoBR: RealTek supposedly makes discrete Ethernet chips.
16:21:32 <kc5tja> warpzero: n/p
16:21:35 <warpzero> kc5tja: send me a forth computer kit plz
16:21:35 <AldoBR> hm
16:21:40 <AldoBR> ok, thanks...
16:21:46 <AldoBR> i´ll search for realtek
16:21:59 <AldoBR> hope they sell this thing here in brazil...
16:22:01 --- join: blockhead (default@dialin-704-tnt.nyc.bestweb.net) joined #forth
16:22:01 <kc5tja> warpzero: Depends on whether I receive payment for it.  :-)
16:22:19 <blockhead> kc5tja: seeling drugs?  :D
16:22:35 <kc5tja> blockhead: http://www.falvotech.com/cgi/fsforth/ForthBoxKestrel
16:23:03 <blockhead> cool!
16:23:16 <warpzero> kc5tja: hey how fast do you think you could get me one?
16:23:25 <blockhead> and you're calling it kestrel!  So very very cool!
16:24:22 <kc5tja> warpzero: I have to design, build, and debug the prototype first.
16:24:28 <kc5tja> I don't know how long that can/will take.
16:24:35 <kc5tja> blockhead: :)
16:25:33 <warpzero> go!
16:25:39 <blockhead> mmmmmm, wiki
16:25:49 <warpzero> you have 10 minutes, and you must not break combo!
16:26:07 <blockhead> "break combo"?
16:27:14 <warpzero> thats right
16:27:58 * blockhead doesn't understand.  what combination is being broken?  huh?
16:28:20 <warpzero> itt
16:28:31 <warpzero> this 300k max gate thing would rox
16:29:00 <chris-xp> Rush - Vital Signs
16:29:01 <chris-xp> kickass!
16:29:23 <chris-xp> http://yro.slashdot.org/yro/04/03/03/1322244.shtml <-- o.O
16:33:53 <blockhead> kc5tja: raven is an even cooler name for the computer ... but eh case will have to be black  :D
16:34:02 <chris-xp> :P
16:34:49 <kc5tja> Raven is also a way cooler design, too.
16:36:13 <warpzero> kc5tja: um one last question XD
16:36:16 <kc5tja> Where does Mozilla keep its lock files?
16:36:32 * kc5tja can't launch Mozilla 'cos it thinks it's still running somewhere.
16:36:37 <warpzero> kc5tja: most of these development boards dont have any built in I/O right so you would have wire them to something
16:36:55 <warpzero> kc5tja: .mozilla/default/random numbers and letters/lock
16:36:57 <chris-xp> kc5tja: su -c init 6 should fix it :P
16:37:43 <kc5tja> warpzero: Mine doesn't have one.
16:37:53 <kc5tja> That's where I looked originally.
16:38:02 <warpzero> ???
16:38:09 <warpzero> thats messed up
16:38:58 <warpzero> $ ls .mozilla/default/nafbwom2.slt/lock
16:38:58 <warpzero> .mozilla/default/nafbwom2.slt/lock
16:41:23 <kc5tja> We'll see what updatedb and 'locate' has to say about this.
16:42:08 <warpzero> i really want an fpga kit but i dont know if i have the balls to make it do anything
16:43:40 <kc5tja> *nothing*
16:43:53 <warpzero> kc5tja: time to reboot
16:44:20 <chris-xp> kc5tja: find /tmp -name moz
16:44:27 <kc5tja> OK, fixed it.
16:44:31 <chris-xp> :)
16:44:35 <warpzero> oh cool
16:44:49 <kc5tja> I found a zombie shell process that had a <defunct> MozillaFirebird-bin sitting under it.
16:44:52 <kc5tja> Somehow.
16:44:53 * kc5tja kill -9'ed it.
16:45:29 <warpzero> wow
16:45:39 <warpzero> firebird/firefox both leak memory hardcore on my box
16:45:47 <tathi> Gotta love mozilla...
16:46:14 <chris-xp> tathi: you're back :)
16:46:19 <ianp> At least it doesnt fuck up your whole OS like IE does!
16:46:55 <warpzero> ianp: malloc(1000000000) doesnt fuck up your box?
16:47:14 <ianp> warpzero: I guess I never notice compared to IE freezing up all the time at runtime
16:47:28 <warpzero> yeah thats lovely i hate ie
16:47:30 <ianp> isnt it just 100000000
16:47:38 <ianp> :P
16:47:41 <warpzero> i also like how ie doesn't draw anything in the window untill it has data
16:47:48 <ianp> yeah
16:47:54 <ianp> that one is esepcially competent
16:48:03 <ianp> but i like the freezing explorer.exe feature better
16:48:13 <warpzero> and how it fails to understand css
16:48:18 <warpzero> oh yeah thats cool
16:48:25 <warpzero> i should get a replacement shell for my laptop
16:48:29 <ianp> warpzero: Doing server side programming open a whole new view of IE
16:48:38 <warpzero> ne way
16:48:46 <ianp> warpzero: e.g. if you want to serve binary data with application/pdf MIME type...
16:48:51 <ianp> IE makes THREE requests
16:49:06 <chris-xp> ianp: !!
16:49:06 * warpzero is away: This is the end my one and only friend, the end.
16:49:12 <chris-xp> ianp: ......WHY!?
16:49:15 <ianp> you have to add .pdf at the end of the URL to skip over one
16:49:22 <warpzero> omfg
16:49:32 <warpzero> what a poorly programmed browser
16:49:41 <warpzero> anyway im out
16:49:47 <chris-xp> warpzero: but we knew that already!
16:49:50 <chris-xp> warpzero: lates
16:50:02 <ianp> it works fine if it actually works
16:50:02 <chris-xp> although, firefox and konqueror aren't optimal either.
16:50:12 <chris-xp> but it STILL doesn't have tabs
16:50:19 <ianp> chris-xp: look all over the net, anywhere such code exists there's hooks checking for IE's dumb ass...
16:50:24 <ianp> and it's still there in IE6, yes.
16:50:28 <ianp> *shakes head*
16:50:37 <chris-xp> *sigh*
16:50:50 <chris-xp> did you hear, they're trying to patent the concept of a virtual desktop
16:50:57 <ianp> haha.
16:50:59 <ianp> .newp
16:51:07 <ianp> they patented part of XML somehow too right, wasnt it?
16:51:13 <chris-xp> yeah
16:51:18 <chris-xp> who made XML btw?
16:51:26 <ianp>  Note that Internet Explorer incorrectly ignores server-provided MIME types, so it sometimes "does the right thing" when the server is misconfigured. Other browsers correctly heed the server-provided MIME types, so they will reveal server misconfigurations.
16:51:29 <ianp> I dunno.
16:51:46 <chris-xp> I think it was in part them, not sure tho
16:54:37 <ianp> oh yeah, the other cool thing about IE... on error pages if the content is <5k it doesnt show it :)
16:54:58 <ianp> it's mind boggling not really that these 'features' exist, but that they were never fixed.
16:56:04 <kc5tja> ianp: It's not in their economic best interests to finish them.
17:17:06 <ianp> kc5tja: well I was thinking, it might not be so simple, i can't imagine what the source code looks like. but how do you mean?
17:18:25 <Sonarman> well, maybe now you can see what the source code looks like yourself :)
17:19:29 <ianp> true, I want to see that stuff
17:19:36 <ianp> too lazy though
17:27:17 <ianp> not like i'll gain anything from it :)
17:30:53 <kc5tja> ianp: Microsoft prefers to introduce "new features" or subjugate control from third-party standards organizations, than fix bugs.
17:36:37 <ianp> kc5tja: Makes sense from a marketing POV
17:36:42 <ianp> they already got all the web's balls in a vice
17:36:45 <ianp> :>
17:37:00 <slava> hi
17:42:50 <ianp> lo
17:53:09 --- join: arke (~Chris@wbar8.lax1-4-11-099-104.dsl-verizon.net) joined #forth
17:53:19 --- nick: arke -> FBI
17:53:35 --- quit: FBI (Client Quit)
17:55:34 --- nick: blockhead -> CIA
17:55:37 --- quit: CIA ("Client Exiting")
18:37:12 --- quit: Nutssh ("Client exiting")
18:45:15 --- join: jdrake (~uniq1@CPE00045afdd0e8-CM014410113717.cpe.net.cable.rogers.com) joined #forth
18:45:44 --- quit: tathi (Read error: 60 (Operation timed out))
19:02:02 --- quit: ChanServ (Shutting Down)
19:02:37 --- join: ChanServ (ChanServ@services.) joined #forth
19:02:37 --- mode: irc.freenode.net set +o ChanServ
19:38:29 --- quit: qFox ("if at first you dont succeed, quit again")
20:00:16 <jdrake> !
20:02:53 <slava> hi jdrake
20:03:40 <jdrake> bonjour!
20:03:50 <jdrake> i have been looking into ficl
20:04:11 <jdrake> and my god it looks easy to implement the exact way I was thinking, i just need something to work in allegro :-)
20:04:27 <jdrake> the  ficl app is so simple 
20:11:00 <Sonarman> what are you wanting to do, again?
20:12:26 <jdrake> think C64 booting into graphics mode basic
20:12:39 <jdrake> now change it to an allegro graphics application running ficl
20:23:06 <madgarden> jdrake, need help with Allegro?
20:25:42 <jdrake> i am just getting into it
20:25:52 <jdrake> my intent is to basically recreate a simple console 
20:26:02 <jdrake> then to add run ficl inside of it
20:26:15 <jdrake> then create graphics bindings for ficl
20:26:17 <madgarden> Yep, sounds fun. ;)
20:26:42 <jdrake> i came up with some things that the api might have
20:26:48 <jdrake> like pset ( c x y -- )
20:26:59 <jdrake> i think that is the right way of expressing it forthy right?
20:27:15 <madgarden> is that like putpixel?
20:27:19 <jdrake> yes
20:27:27 <jdrake> pset is the name of it in gwbasic
20:28:19 <madgarden> Ahh. 
20:29:12 <slava> jdrake, cool
20:29:15 <jdrake> ten things like line, box, fbox, etc
20:29:17 <madgarden> You may want the colour to go on the stack last.
20:29:25 <slava> i've wanted to make opengl bindings for my language.
20:32:57 <ASau> Dobre jitro!
20:33:28 <ASau> jdrake, better pset is ( x y c -- )
20:34:13 <jdrake> ASau, why
20:34:51 <ASau> If you have configured colours as constants, then you can optimize access.
20:35:02 <ASau> Otherwise you have ROT -ROT
20:35:13 <ASau> ...have to...
20:37:07 <jdrake> one thing I was going to be doing with this graphical environment was configure a retargetable output
20:37:14 <ASau> When you want to show _any_ colour, you'll be able to use 0 PSET
20:37:38 <jdrake> so that you could set the printing to go to drawing, screen, or log file
20:37:56 <jdrake> so that there is not special draw text command, it is just done via regular output
20:38:13 <ASau> ForthScript?
20:39:00 <ASau> The idea of such a kind floats around.
20:39:20 <jdrake> i do not know what you refer
20:39:24 <madgarden> 10 10 50 50 20 30 pget line
20:39:54 <ASau> jdrake, you may take a look at PostScript.
20:40:58 <jdrake> no thank you ;p
20:41:06 <ASau> Also, there is a MS Research's PAN.
20:41:40 <ASau> That is Haskell based, but I know how make it in Forth.
20:42:13 <ASau> The problem is my laziness and lack of time.
20:43:03 <ASau> jdrake, why? PostScript is simple enough.
20:43:10 <ASau> It's a kind of Forth.
20:43:12 <jdrake> 20 30 pget 10 10 50 50 line
20:43:27 <jdrake> i think that is more simple
20:43:45 <TreyB> And minimizes stack depth.
20:44:21 <jdrake> stack depth is of no concern
20:44:36 <TreyB> How un forthy of you :-)
20:45:03 <jdrake> it really doesn't matter with how small the typical function will be
20:45:05 <ASau> TreyB, use relative to current position.
20:45:42 <ASau> You'll get rid of two more cells on stack.
20:45:44 <kc5tja> Turtle graphics!!
20:46:14 <ASau> Also, you can pack X and Y in one cell.
20:46:28 <jdrake> ASau, certain things are too much paint
20:46:29 <jdrake> pain
20:46:40 <ASau> Especially if you're 32 bit guy.
20:47:36 <ASau> jdrake, that's a demonstration that the stack depth is to be of no concern.
20:47:39 <madgarden> kc5tja, hmm, a Logo in Forth would be pretty fun and easy to make!
20:47:57 <slava> yes!
20:48:22 <Herkamire> jdrake: I reccomend turtle graphics too
20:48:43 <jdrake> turtle graphics aren't my specialty
20:48:52 <Herkamire> hmmm... I'm not sure I have the term right, but having line-to move-to and set-color
20:48:54 <jdrake> my favourite api for ease was basics
20:49:02 <jdrake> that could work too
20:49:07 <TreyB> Herkamire: path-based
20:49:10 <ASau> jdrake, that's ZX Basic.
20:49:19 <TreyB> a la PostScript
20:49:34 <kc5tja> ASau: Was ZX-Basic path-based or absolute-based?
20:49:35 <TreyB> (or pen plotters, etc.)
20:49:39 <jdrake> gwbasic 
20:49:46 <chris-xp> jdrake: FUN!
20:49:48 <chris-xp> jdrake: :)
20:50:00 <chris-xp> jdrake: whats going on in #hprog right now, anything worth seeing?
20:50:13 <slava> hi chris-xp
20:50:15 <ASau> kc5tja, it's more path-based.
20:50:17 <jdrake> no
20:50:23 <kc5tja> ASau: I see.
20:50:37 <Herkamire> what's absolute-based?
20:50:39 <jdrake> #hprog is a slime-channel with religious unproductive arguments mostly
20:51:14 <madgarden> Logo manual: ftp://anarres.cs.berkeley.edu/pub/ucblogo/usermanual
20:51:22 <ASau> Herkamire, line(x1,y1,x2,y2,colour) is "abs-based",  lineto(dx,dy,colour) is "path-based".
20:52:40 <ASau> MSX Basic has both kinds of instructions, IIRC.
20:53:51 --- join: hovil (~hovil@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
20:54:01 <ASau> Dobre jitro!
20:55:39 <kc5tja> madgarden: Neat thing about Logo -- it's basically Lisp.
20:55:57 <kc5tja> It can do pretty much everything that a typical Lisp system can (though not necessarily as succinctly).
20:55:59 <slava> kc5tja, didn't think you'd consider that 'neat'
20:56:06 <kc5tja> slava: I love Lisp.
20:56:09 <slava> kc5tja, really?
20:56:12 <kc5tja> Of course I'd consider it neat.  :)
20:56:16 <kc5tja> I just like Forth better.
20:56:17 <slava> kc5tja, common lisp especially is rather "large"
20:56:23 <slava> didn't think you'd like that.
20:56:27 <kc5tja> Now, where did I write Common Lisp above?
20:56:29 <Herkamire> once you define move-to line-to and set-color you can do:   : line ( x y x2 y2 color -- ) set-color move-to line-to ;
20:56:40 <kc5tja> I said it was "a Lisp."  Not "a Common Lisp."
20:56:42 <madgarden> kc5tja, yes, I'd read that. It is rather neat. Neato, even.
20:56:58 <ASau> slava, Lisp is not only Common, also Forth is not only Gforth.
20:57:10 <slava> i know.
20:57:15 <slava> scheme is nice and simple.
20:57:33 <ASau> Yes, see also Arc.
20:57:52 <ASau> It has some good features.
20:58:35 <kc5tja> I'm not familiar with Arc.
20:58:54 <ASau> IIRC, http://www.paulgraham.com
20:58:55 <kc5tja> Scheme is really cool too.  (As long as you don't get into that really esoteric stuff.)
20:59:06 <slava> call/cc is cool
20:59:30 <kc5tja> Call/cc is cool, but it has a lot of run-time overhead, and really confuses the code.
20:59:49 * kc5tja finds manipulating a return stack directly is much more to-the-point, and runs a HECK of a lot faster.
21:00:02 <slava> a continuation is basically this though
21:00:18 <kc5tja> Partial continuations is what forth supports.
21:00:21 <slava> how would you do co-operative multitasking in forth for example.
21:00:27 <kc5tja> Complete continuations is what Scheme supports.
21:00:32 <slava> you'd switch the stack pointers right?
21:00:39 <kc5tja> Yes.
21:00:42 <kc5tja> I know what continuations are.
21:00:52 <slava> a good continuation implementation is not necessarily slow.
21:00:54 <kc5tja> Continuations are a lot more than just switching the stacks though.
21:01:07 <kc5tja> slava: Find me one that can compete with Forth, and I'll agree.  :)
21:01:08 <slava> in scheme, yes, because of the more complex nature of the language.
21:01:25 <slava> in a postfix language continuations are very simple to conceptualize.
21:01:31 <kc5tja> Sure.
21:01:38 <kc5tja> I don't recall ever objecting to this.
21:01:49 <slava> you said "manipulating a return stack directly"
21:01:57 <slava> a continuation is higher level.
21:02:03 <kc5tja> Ummm.....no.
21:02:09 <kc5tja> A partial continuation is just a return address.
21:02:24 <slava> yes
21:02:31 <kc5tja> A complete continuation (hence the call/cc construct) is the whole return stack, the parameters, and other state.
21:03:38 <ASau> UP, SP and RP addresses.
21:04:06 <ASau> Every other thing is global by defn.
21:04:08 <slava> ASau, *copies* of the stacks are needed though -- you can return to a continuation more than once and the exact same state must be restored.
21:04:32 <ASau> Hmm. Agree.
21:04:32 <slava> ASau, and some way of passing cells from the 'restorer' of the continuation to the continuation, to implement co-routines and such.
21:05:41 <ASau> You have global var-s, user var-s and stacks.
21:11:00 <ASau> Hm. I thought about writing multi-threaded Forth.
21:11:24 <ASau> I think these continunations should be very easy to implement.
21:12:04 <ASau> It's like swapping out thread state.
21:12:08 <slava> yes
21:12:13 <ASau> You only give it a name.
21:12:22 <slava> the hard part about continuations is not implementation, but understanding their full potential.
21:12:34 <ASau> It's very easy to implement.
21:12:38 <slava> exceptions, co-operative multitasking, co-routines and other things i've forgotten are all specific instances of continuations.
21:13:17 <ASau> You only have to save user area, stacks and local dictionary.
21:13:23 <ASau> Ha!
21:14:10 <ASau> That's the same as I've done already. It's saving part of memory.
21:14:18 <jdrake> Dr. Evil Speaking in Forth: here bone a me throw
21:16:56 --- join: Smerdyakov (~adam@206.169.168.190) joined #forth
21:17:04 <ASau> Dobre jitro!
21:17:12 --- quit: hovil (Remote closed the connection)
21:34:59 --- quit: Sonarman (Read error: 60 (Operation timed out))
21:36:28 * warpzero is back (gone 04:47:25)
21:36:51 <Smerdyakov> warpzero, did you choose your name to be an anagram of warezpro?
21:36:53 --- join: Sonarman (~matt@adsl-66-124-254-243.dsl.snfc21.pacbell.net) joined #forth
21:37:19 <ASau> Doboe utro, Sonarman!
21:37:50 <warpzero> Smerdyakov: nope
21:40:38 <jdrake> what was that channel with the forth interpreter?
21:48:26 <ASau> Ask Herkamire for his bot.
21:52:38 <warpzero> Herkamire: bot plz
21:52:51 --- join: Nutssh (~Foo@gh-1177.gh.rice.edu) joined #forth
21:56:21 <Sonarman> ASau: what is the UP?  I assume it stands for "User Pointer," but what does that mean?
22:00:16 <Herkamire> warpzero: http://herkamire.com/jason/sifbot
22:01:03 <warpzero> its written in bash?
22:02:23 --- join: Serg (Serg_Pengu@212.34.52.140) joined #forth
22:04:29 <Herkamire> warpzero: irc protocol in bash. forth in C
22:04:37 <slava> why bash?
22:04:54 <jdrake> slava, so you would ask him why :-)
22:05:03 <ASau> Dobroe utro, Serg!
22:05:07 <Herkamire> I didn't know perl at the time :)
22:05:16 <ASau> Sonarman, read about user variables.
22:05:29 <ASau> UP is pointer to user variables area.
22:05:29 <Herkamire> I didn't know a language with regular expression stuff
22:06:43 <Herkamire> and where I knew how to redirect stdin/stdout/7/6
22:06:57 <ASau> Herkamire, isn't it easier to use simple substring search.
22:06:58 <ASau> ?
22:08:19 <Herkamire> don't see why it would be easier to program.
22:08:52 <Herkamire> it was just a quick hack. I just used what I knew how to use already
22:09:08 <Herkamire> If I had known perl as well as I do now, I would have used that.
22:09:30 <Herkamire> asuming I could get it to read/write from descriptors 6 and 7
22:09:46 <ASau> Herkamire, there're ready to use substring search routines.
22:09:55 <ASau> Even in "C."
22:10:11 <Herkamire> I didn't want to spend time learning new APIs
22:10:19 <Smerdyakov> ASau, why did you quote it?
22:10:59 --- quit: jdrake ("Oops. This machine just fell asleep")
22:11:05 <Herkamire> My bot is really crappy. If you're looking for anything more than a very simple example of the IRC protocol, then you should probably look elsewhere.
22:11:21 <Herkamire> I reccomend the RFC. it was very simple and clear, and not even very long.
22:11:39 <ASau> I mix different styles. It's a drawback of different kinds of work done in parallel.
22:12:02 <Herkamire> the only snag I ran into, was that I didn't have it replying to PINGs at first, so it got booted off after an hour or 3
22:12:38 <Herkamire> I only have sifbot up on the web because people have expressed interest in it.
22:14:39 <Smerdyakov> Anyone know of an automated theorem prover implemented in Forth?
22:15:28 <ASau> Smerdyakov, those problems are not usually solved in Forth.
22:15:37 <Smerdyakov> Heh. I guessed as much.
22:15:39 <ASau> Smerdyakov, Forth is not ML.
22:15:50 <Smerdyakov> Oh, OK. That's good to know.
22:16:37 <ASau> Of course, noone forbids you writing your own prover.
22:16:44 <slava> Smerdyakov, there is one written in joy.
22:16:55 <Smerdyakov> slava, URL?
22:17:47 <ASau> Smerdyakov, I know Mikhail Gassanenko worked on back tracking in Forth.
22:18:23 <Smerdyakov> ASau, what does that mean, exactly?
22:18:26 <ASau> Maybe he could work on something like theorem proving.
22:18:46 <slava> Smerdyakov, you don't know what back tracking is?
22:18:59 <ASau> That's a Prolog style of program control.
22:19:10 <Smerdyakov> "Bactracking" is a pretty general term.
22:19:19 <slava> Smerdyakov, i can't seem to locate the joy theorem prover anymore. i recall it being mentioned on the concatenative list once...
22:19:25 <Smerdyakov> But if you mean something like a Prolog interpreter, then I understand.
22:19:41 <ASau> Better refer to M.L.G. site.
22:19:47 <slava> ASau, backtracking can be done with continuations too.
22:19:53 <Smerdyakov> M.L.G.?
22:19:59 <ASau> He maintains large bibliography.
22:20:47 <Smerdyakov> Well, I don't think I'd be interested in the details.
22:20:48 <ASau> http://forth.org.ru/~mlg
22:21:42 <Smerdyakov> My question was just more of a pondering on the limitations on views of computation that programming in Forth will encourage.
22:22:04 <slava> Smerdyakov, automated theorem provers are not an important use of a language.
22:22:47 <ASau> It seems noone did it before.
22:23:19 <ASau> Prolog-like Forth extension exists.
22:23:25 <Smerdyakov> Well, I think it would be fairer to say that you must look at more than theorem proving to evaluate a general purpose language.
22:24:00 <Smerdyakov> This is just an example of a task that is thought of in a highly structured way.
22:25:22 <ASau> Also there're several Lisp-like extensions.
22:25:42 <kc5tja> Well, I am now going to bed.
22:25:52 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
22:27:02 <Smerdyakov> ASau, why do you mention that?
22:27:34 --- join: hovil (~hovil@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
22:27:44 <ASau> That's somehow connected with you TP theme.
22:28:04 <ASau> I never have heard of T.P. in Forth.
22:29:02 <Smerdyakov> It's connected because Lisp is popular for ATP development?
22:29:57 <ASau> I know about expert systems (Feucht et al.) and grammars (Paysan?).
22:30:09 --- join: MarkT (~mark@tom.nss.udel.edu) joined #forth
22:30:20 <ASau> Dobroe utro, MarkT!
22:30:31 <MarkT> hello
22:30:33 <MarkT> :)
22:30:38 <Herkamire> Smerdyakov: well put :)  ("pondering on the limitations on views of computation that programming in Forth will encourage"
22:30:46 <MarkT> got gforth for my linux box  :)
22:31:06 <ASau> Lisp uses complex enough structures.
22:31:28 <ASau> Bare Forth is the bariest thing you can have.
22:32:21 <Smerdyakov> I'm skeptical that Lisp is suited for effective development of programs that use complex data structures, though.
22:32:22 --- quit: Sonarman ("leaving")
22:32:31 <Herkamire> in my thinking you can do just about anything in forth, but you often have to code a bunch of stuff before you can really get started.
22:32:54 <slava> Smerdyakov, heard of CLOS? beats the pants off anything ML has.
22:33:14 <Herkamire> MarkT: welcome :)
22:33:19 <MarkT> ty :)
22:33:28 <Smerdyakov> Herkamire, I don't agree. Human information processing limitations will stop you from doing things when you need to remember too much of the complex structure you have to set up.
22:33:30 <Herkamire> ASau: I wouldn't say that. I'd say ASM is as bare as you can get.
22:33:48 <Smerdyakov> slava, I have yet to be impressed by any "object system." Yes, I have read about CLOS at length a few years back.
22:33:52 <slava> Smerdyakov, that's why you have abstraction and information hiding.
22:34:00 <slava> Smerdyakov, to manage compexity.
22:34:07 <Smerdyakov> slava, which I think Forth doesn't provide very well.
22:34:22 <Herkamire> Smerdyakov: I don't follow what that has to do with forth's limitations.
22:34:35 <ASau> Smerdyakov, you're wrong.
22:34:44 <ASau> Forth can hide information.
22:34:48 <Smerdyakov> Herkamire, Forth requires more work to set up foundations, so less of your brain is left for the conceptual stuff.
22:34:49 <ASau> I've tested.
22:34:56 <slava> Smerdyakov, have you ever written "real" programs or just theorem provers and calculating pi to a million digits?
22:35:26 <Smerdyakov> ASau, yes, but from what I know of Forth, it does not support abstraction _well_ compared to alternatives.
22:35:31 <slava> Smerdyakov, you seem to argue mundane points at great length purely for the sake of argument.
22:35:35 <Smerdyakov> slava, what are you trying to suggest?
22:35:41 <Herkamire> Smerdyakov: ok, I understand that, but I figure it goes more that you have to code the foundation first. Then you are free to code/think the conceptual stuff
22:35:59 <Smerdyakov> slava, you're quoting "real" like I did, so I must ask you for a definition like you did of me.
22:36:14 <ASau> Smerdyakov, no one ever thought of this.
22:36:27 <slava> Smerdyakov, by "real" I mean real.
22:36:28 <ASau> Smerdyakov, it contradicts main philosophy.
22:36:31 <Smerdyakov> ASau, what?
22:36:34 <slava> Smerdyakov, as in, industry.
22:36:42 <ASau> Smerdyakov, that's not a Forth style.
22:36:52 <Smerdyakov> slava, interesting view of reality. You must be American. :D
22:37:12 <Herkamire> hehe :)
22:37:13 <Smerdyakov> ASau, and I claim that this stylistic choice inhibits effective development.
22:37:22 <Herkamire> some of us are a little cranky tonight :)
22:37:30 <slava> Smerdyakov, see? you just come to #forth to try and convince us we're wrong.
22:37:36 <slava> Smerdyakov, its not going to work.
22:37:45 <Smerdyakov> slava, I've done large amounts of coding in all of personal projects, industry, classes, and academic research.
22:37:48 <slava> Smerdyakov, take your theorem provers and ML coding and PhD attitude elsewhere.
22:37:52 <ASau> If no one did it before that does not mean it's hard.
22:38:25 <Smerdyakov> slava, perhaps you could benefit from viewing things objectively instead of as a chance for you to protect your turf.
22:38:47 <Smerdyakov> ASau, OK, and your point is?
22:38:54 <slava> Smerdyakov, i'm not interested in hearing your mundane arguments that seem to all hinge on the precise semantic definition of words.
22:39:07 <ASau> Smerdyakov, just try to do it yourself.
22:39:09 <Herkamire> Smerdyakov: I agree with you that some projects will be more work in forth because you will have to code a foundation that some languages have already.
22:39:36 <Herkamire> Smerdyakov: what I think is cool though, is that often when you code the foundation yourself it is much better suited to your project.
22:39:37 <ASau> Smerdyakov, I'm not interested in theorem proving. 
22:39:45 <Smerdyakov> slava, you can't have a meaningful argument when you can't understand what the other party is saying.
22:39:56 <slava> Smerdyakov, so stop trying to argue.
22:39:58 <ASau> Smerdyakov, all I can do is providing links.
22:40:19 <Smerdyakov> ASau, as I've already said, I'm not interested in having or writing a Forth theorem prover. I only used that example to highlight what I view as a limitation of Forth.
22:40:38 <ASau> Smerdyakov, that's not a limitation.
22:41:02 <Smerdyakov> slava, I'm not "trying to argue." I'm trying to participate in mutual discussion and learning of software development methodology.
22:41:03 <ASau> Smerdyakov, do you have tons of ML based OSes?
22:41:14 <Smerdyakov> ASau, no, only a few.
22:41:22 <ASau> Smerdyakov, are they self-hosted?
22:41:44 <slava> Smerdyakov, you're basically saying "I am correct, you are wrong, I won't listen to your reasons, nyah nyah nyah". then when you feel like you've lost a point, you pull out some word out of your ass and ask us to clarify what we mean by that word.
22:41:46 <ASau> It seems they require C?
22:41:48 <Smerdyakov> C runtime systems.
22:42:05 <Smerdyakov> slava, no, I don't think I've ever said that. I think that, for some reason, you are just terribly insecure.
22:42:17 <ASau> Smerdyakov, so is that a limitation of ML or its run time support?
22:42:32 <Smerdyakov> ASau, certainly a limitation of ML.
22:42:32 <Herkamire> slava: easy does it. I think Smerdyakov is being perfectly reasonable
22:43:18 <ASau> Smerdyakov, so your point is ML can't produce machine code directly. Should I understand this way?
22:43:33 <ASau> Smerdyakov, I've thought ML is Turing complete.
22:44:07 <Smerdyakov> ASau, ML can't produce memory management code for bare hardware directly, for example.
22:44:08 <slava> ASau, it can, but its like java, it needs a hiary runtime.
22:44:14 <ASau> Smerdyakov, you say it's a limitation because nobody tried before.
22:44:35 <Smerdyakov> No, I say it's a limitation because ML purposely limits the range of possible programs.
22:45:20 <ASau> I've thought ML is Turing complete.
22:45:34 <Smerdyakov> Turing completeness is not relevant to this sort of thing.
22:45:35 <ASau> So I'm mistaken.
22:45:37 <ASau> Sorry.
22:45:58 <slava> ASau, javascript is turing complete for instance, but it runs in a browser and its I/O is very limited.
22:45:59 <Smerdyakov> Very few languages are really Turing complete, since they don't run on Turing machines!
22:46:02 <ASau> What do you call "possible" then?
22:46:32 <Smerdyakov> Say your hardware requires you to control screen output by writing to a particular memory address.
22:46:35 <Smerdyakov> This is not possible in ML.
22:46:39 <Smerdyakov> ML has no concept of memory.
22:47:00 <MarkT> *cough*
22:47:30 <Smerdyakov> ?
22:47:30 <ASau> Smerdyakov, don't ML support numeric functions?
22:47:45 <ASau> ...doesn't...
22:47:48 <Smerdyakov> ASau, if you mean functions that take and return integers, sure.
22:48:20 <ASau> So what makes IO impossible?
22:48:44 <Smerdyakov> The fact that you need to generate special opcodes that the ML compiler never generates, for example.
22:48:59 <Smerdyakov> MarkT, why "*cough*"?
22:49:10 <ASau> MS Research made screen I/O control language Haskell-based.
22:49:26 <Smerdyakov> Right, and ghc uses a C runtime system.
22:49:43 <MarkT> the view that ml has no concept of memory has no bearing on whether or not you can store values to an address that mey have been set up somewhere...  
22:50:07 <ASau> Why can't you eliminate C?
22:50:13 <Smerdyakov> MarkT, I don't follow. How can you write to memory when the language doesn't let you know that memory exists?
22:50:23 <ASau> Don't you know op. codes?
22:50:32 <slava> MarkT, exactly, consider jnode's: class Unsafe { native int peek(int address); native void poke(int address, int value); }
22:50:38 <MarkT> I know for a fact that on the Commodore VIC20 (which has a flat memory model) that you can indeed manipulate the screen ram because I've done it several times
22:50:50 <Smerdyakov> ASau, you know the op codes, but the ML language without modification has no way to generate arbitrary opcodes.
22:51:04 <slava> Smerdyakov, sure it does. write them to a file.
22:51:12 <slava> write an appropraite header first.
22:51:13 <MarkT> Smerdyakov: language does not define memory, hardware does
22:51:17 <slava> make the file executable, according to OS conventions.
22:51:46 <Smerdyakov> slava, now you're not using ML/whatever other language. You're generating executable code directly or almost directly
22:51:53 <ASau> I never thought you can't add necessary functions to your languages.
22:52:05 <slava> Smerdyakov, but certainly its possible to write a native ML compiler in ML.
22:52:08 <Smerdyakov> ASau, if you add them, what language will you implement them in?
22:52:18 <Smerdyakov> slava, yes.
22:52:20 <ASau> Smerdyakov, ML.
22:52:33 <Smerdyakov> ASau, how do you implement "write to screen memory" in ML?
22:52:48 <ASau> See "meta-compilation".
22:52:49 <slava> Smerdyakov, is there a concept of a foreign function interface?
22:52:56 <slava> Smerdyakov, write these functions in C.
22:53:02 <Smerdyakov> MarkT, surely you agree that it's impossible to send a packet over the network in toy Brainfuck implementations, yes?
22:53:04 <MarkT> Smerdyakov: that depends on how the screen is set up and the memory model of the archetecture
22:53:19 <Smerdyakov> slava, there is, but then *YOU*ARE*USING*C*.
22:53:24 <slava> Smerdyakov, or even, extend the language and have your compiler generate the appropraite code.
22:53:29 <Smerdyakov> slava, ASau is claiming you can do this with only ML code.
22:53:33 <ASau> Smerdyakov, write it in *ML*!
22:53:48 <Smerdyakov> ASau, it is impossible./
22:54:01 <ASau> *Noone* ever forbid writing HEX integers!
22:54:07 <Smerdyakov> Untrue.
22:54:12 <Smerdyakov> The Standard ML Definition forbids it.
22:54:22 <Smerdyakov> If you could do it, you would no longer have Standard ML.
22:54:42 <ASau> Then send your standard to the hell.
22:54:50 <slava> Smerdyakov, your language limits you.
22:55:01 <Smerdyakov> You would also be violating the spirit of ML if you allowed memory writes.
22:55:08 <ASau> You have limited yourself, not a language.
22:55:23 <Smerdyakov> Of course the language limits the programmer! That's why we like to use it so much!
22:55:27 <slava> what if you wanted to write an OS in ML? is a large and complex undertaking like this against the "spirit" of the language?
22:55:35 <MarkT> lets put it this way, there isn't *anything* that you cannot do in ML. the question is "do you really want to?"  part of the basis for this view is 1) originally there was 'only' ML and 2) the languages eventually get compiled or interpreted into ML anyway.  so by definition it is possible because it is being done.
22:55:45 <Smerdyakov> No, and there _are_ OS's written in ML, which was what started this discussion.
22:56:04 <Smerdyakov> MarkT, that didn't make any sense.
22:56:15 <MarkT> Smerdyakov: in what way?
22:56:21 --- quit: Nutssh (Read error: 110 (Connection timed out))
22:56:23 <ASau> Smerdyakov, then there're *NO* OS written in ML.
22:56:44 <Smerdyakov> ASau, right. They are written in ML and a comparatively small bit of C and assembly.
22:56:59 <ASau> Smerdyakov, so ML is not complete.
22:57:06 <Smerdyakov> ASau, whatever that means. :P
22:57:23 <ASau> You may think it over.
22:57:24 <Smerdyakov> MarkT, you cannot write to video memory in ML, for instance.
22:57:35 <ASau> Forth *is* complete.
22:57:38 <MarkT> sure you can
22:57:42 <MarkT> I've done it
22:57:46 <Herkamire> ASau: what do you mean complete?
22:57:47 <Smerdyakov> ASau, are you claiming that your "completeness" is useful in a general purpose programming language?
22:57:56 <ASau> I have seen written in Forth OSes.
22:58:01 <Smerdyakov> MarkT, how?
22:58:20 <ASau> So I can have ATP in Forth, you can't.
22:58:33 <ASau> That's your own limitation.
22:58:34 <MarkT> Smerdyakov: look at the VIC20  screen ram is set up in hardware.  just write to it and BOOM its there.
22:58:35 <Smerdyakov> ASau, that doesn't bother me.
22:58:41 <Smerdyakov> MarkT, do you know what ML is?
22:58:48 <MarkT> *sigh*
22:59:03 <ASau> Smerdyakov, me too.
22:59:06 <Smerdyakov> MarkT, I _don't_ mean "machine language."
22:59:26 <Smerdyakov> MarkT, I mean the family of high level programming languages.
22:59:56 <ASau> "There's no abstract truth. Truth is always concrete."
23:00:01 <Smerdyakov> ASau, it doesn't bother me because your "completeness" doesn't bring practical benefits.
23:00:06 <MarkT> Smerdyakov: hmm. lest I looked ML was Machine Language.. oh well. 
23:00:07 <Smerdyakov> MarkT, did you think I meant "machine language"?
23:00:12 <MarkT> sure did
23:00:22 <Smerdyakov> MarkT, it depends on what community you're in. :)
23:00:30 <MarkT> apparently  :)
23:00:42 <slava> come on, everybody knows that ML == functional language.
23:00:52 <Smerdyakov> However, ML as Meta Language is much more popular overall than ML as machine language.
23:00:55 <slava> who says "ML" to mean machine language?
23:01:10 <Herkamire> slava: nobody I've heard
23:01:11 <Smerdyakov> Witness Google search results, the contents of the programming language ML category in the Open Directory, etc..
23:01:25 <ASau> Smerdyakov, your ATP does not bring practical benefits at all. ;) The only science that matters is chemical technology.
23:01:50 <Smerdyakov> ASau, on the contrary, automated theorem provers can help you find problems in your chemical technology software.
23:02:38 <slava> this discussion is going nowhere.
23:02:59 <ASau> Smerdyakov, I never have heard of this usage. So it does not exist. It's a limitation of ATP. Hence it can't be done. Conclusion: ATP is useless.
23:02:59 <Smerdyakov> I don't know about that. I see some preliminary confusion dissipating slowly. :)
23:03:29 <slava> what do you hope to achieve?
23:03:39 <Smerdyakov> ASau, you have misunderstood my argument. I claim that Forth makes it difficult to write ATP's. It is this difficulty that matters, not the current lack of Forth ATP's.
23:03:42 <Herkamire> what's ATP?
23:03:57 <Smerdyakov> slava, learn more about programming methodology.
23:04:03 <Smerdyakov> Herkamire, automated theorem prover
23:04:07 <Herkamire> ahhh
23:04:28 <slava> Smerdyakov, you seem to have your mind firmly set on debunking anything that is claimed as a benefit of forth.
23:05:02 <ASau> Smerdyakov, learn more about science philosophy.
23:05:07 <Smerdyakov> slava, sure. The current information that I have makes it seem like that is the best course for any given claimed benefit of Forth. However, whether that is right or wrong, someone is going to learn something, and hopefully everyone will.
23:05:21 <Smerdyakov> ASau, learn more about software engineering. :P
23:05:37 <ASau> Smerdyakov. and logic also. Mathematical logic.
23:05:44 <Smerdyakov> ASau, Forth requires too much focus on unimportant details, limiting the scope of a design the programmer can hold in his head.
23:05:51 <slava> Smerdyakov, do you know this?
23:05:55 <slava> Smerdyakov, or is it conjecture?
23:06:09 <Smerdyakov> ASau, I claim the level of an ATP is beyond what most programmers could manage to keep in their heaps to develop in Forth.
23:06:17 <Herkamire> Smerdyakov: are you saying that forth doesn't have any good sides?
23:06:26 <Smerdyakov> slava, it's a reasonable inference from my knowledge of programming in general.
23:06:30 <slava> Smerdyakov, do you develop ATPs all day?
23:06:33 <Smerdyakov> Herkamire, no.
23:06:44 <Smerdyakov> slava, no. I don't think I've ever written one, really. :P
23:06:53 <slava> Smerdyakov, so everything is just hand-waving and conjecture?
23:07:00 <slava> Smerdyakov, there is a technical term for this kind of behavior
23:07:02 <ASau> Smerdyakov, I claim that ATP is useless thing, hence it even does not needed to implement. What d'you have against?
23:07:04 <slava> Smerdyakov, intellectual masturbation
23:07:28 <Smerdyakov> slava, I don't follow. I am an expert on programming, so I have knowledge on many different areas, and I may combine what I know to make reasonable inferences.
23:07:41 <slava> Smerdyakov, so why don't you go write some code/
23:07:49 <Smerdyakov> Herkamire, I have yet to see an argument for Forth used in non-resource-limited platforms.
23:08:04 <Smerdyakov> slava, because I don't want to do that right now.
23:08:51 <Herkamire> Smerdyakov: I think that's because those arguments go against your programming methodologies
23:09:11 <Smerdyakov> Herkamire, OK. Could you give me an example of an objectively valid argument that you think I disregard inappropriately?
23:09:30 <ASau> Smerdyakov, simplicity. You have to write the code you need only.
23:09:41 <Smerdyakov> I find ML quite simple.
23:09:44 <ASau> Smerdyakov, there's no resource limited platforms.
23:10:06 <ASau> Smerdyakov, RAM is never enough.
23:10:22 <Smerdyakov> RAM is cheap these days.
23:10:38 <ASau> If you have enough RAM it means you do not work.
23:10:45 <slava> pointless language wars, they never end!
23:10:52 <Smerdyakov> And the difference between roughly equivalent compiled Forth and ML programs in terms of resource usage is small.l
23:10:53 <slava> Smerdyakov, try pulling this kind of stunt in #java, you'll get banned.
23:10:57 <Herkamire> Smerdyakov: (let me know if this is too abstract) building the foundation yourself often produces an environment much better suited to your project. [pro forth] forth encourages and is well suited to doing this. [con forth] you sometimes have to.
23:11:22 <Smerdyakov> slava, ah, you speak like a true programming novice, unversed in the hard realities of software development, where tool choice really does matter.
23:11:46 <slava> Smerdyakov, i realize tool choice matters but arguing on irc about "my language is better than yours nyah nyah" is fruitless.
23:11:55 <slava> Smerdyakov, do you hope we will all convert to ML and write ATPs?
23:11:57 <Smerdyakov> Herkamire, and why would I choose Forth over Lisp for that?
23:12:05 <ASau> Smerdyakov, I know pretty well, what is domain spec. S.D.
23:12:13 <Smerdyakov> slava, I don't hope anything from you, since it's clear that you're not too interested in effective programming.
23:12:22 <AldoBR> lol
23:12:22 <ASau> Smerdyakov, your S.D. is not enough.
23:12:32 <AldoBR> he talks like a kind of guru
23:12:33 <AldoBR> ;)
23:12:37 <Herkamire> Smerdyakov: you won't. you like more abstraction, and more things done for you already.
23:12:44 <AldoBR> "youre not interested in the true force"
23:12:55 <slava> Smerdyakov, you just dismiss me because i'm interested in practical programming, unlike you.
23:12:59 <ASau> I've never seen any S.D. expert that could write useful software.
23:13:00 <Smerdyakov> Herkamire, I don't understand. Lisp also allows you to build up all your own control flow operations, syntax, etc..
23:13:09 <Herkamire> Smerdyakov: I will choose forth because it's simpler, it encourages better factoring, and I can write bug free applications quicker, and end up with much more code that will be useful in other projects
23:13:11 <ASau> It has major drawbacks.
23:13:31 <Herkamire> and because I find lisp very hard to read
23:13:41 <Smerdyakov> ASau, well, I can write useful software, just so you know. ;)
23:13:46 <slava> Smerdyakov, sometimes when developing software, you have to get something *done*, and not just write code for the sake of it.
23:14:04 <Smerdyakov> slava, right. It's funny how we are both accusing each other of the same thing. :)
23:14:10 <slava> you just seem to enjoy doing ASCII art in ML.
23:14:15 <AldoBR> this is the old "lisp x forth" flame war
23:14:17 <AldoBR> hauhuahuau
23:14:27 <Smerdyakov> slava, what leads you to conclude that?
23:14:28 <ASau> Smerdyakov, OK. Phase diagrams calculations. :)
23:14:33 <AldoBR> he is a lisp priest, at least, looks like one :)
23:14:43 <slava> AldoBR, who is "he"?
23:14:53 <AldoBR> Smerdyakov
23:15:03 <AldoBR> i saw this kind of discussion on and on and on
23:15:04 <Smerdyakov> Herkamire, what you say could be convincing if it were backed up.
23:15:16 <slava> backed up with what?
23:15:17 <AldoBR> there are plenty of texts about it around the net
23:15:19 <Herkamire> Smerdyakov: OKAD II
23:15:29 <slava> Herkamire, it doesn't do ATP! :)
23:15:31 <Smerdyakov> slava, line beginning with "I will choose forth"
23:15:33 <slava> Herkamire, its no good! :)
23:15:35 <Herkamire> slava: lol
23:15:37 <Smerdyakov> Herkamire, what is OKAD II?
23:15:45 <Smerdyakov> slava, can you tell me how old you are?
23:16:03 <AldoBR> Smerdyakov you think you´ll trully convert us to another language ?
23:16:07 <Herkamire> Smerdyakov: VSLI CAD program written in forth
23:16:12 <slava> Smerdyakov, what is it to you?
23:16:23 <slava> Smerdyakov, you think you're right and i'm wrong because you're younger/older?
23:16:26 <Smerdyakov> slava, well, you act like an impetuous teenager. :)
23:16:30 <AldoBR> or are you just trying to find someone to go flame war with you ?
23:16:41 <Smerdyakov> Herkamire, OK. How does that support what you said?
23:16:42 <Herkamire> Smerdyakov: maybe he is. let him be
23:16:47 <slava> Smerdyakov, and you act like a lonely depressed man looking for flamewars on IRC.
23:17:00 <Herkamire> Smerdyakov: it was written by one man, and I don't think it took him that long
23:17:16 <Smerdyakov> Herkamire, OK, but what about a comparison of a one-man effort using other tools?
23:17:39 <Herkamire> Smerdyakov: I'm not going to prove it to you. I'll just give you my opinions
23:17:58 <slava> Herkamire, give up. show him an example of one man forth > one man ml, and he'll next say the ml programmer was incompetent.
23:17:58 <Smerdyakov> Herkamire, yeah, I was just hoping for some non-opinion arguments.
23:18:01 <Herkamire> Smerdyakov: I think the way you find out how well forth works is to use it.
23:18:18 <Herkamire> Smerdyakov: you could try things in forth and see if you can achieve the benifits others claim to get
23:18:35 <ASau> Smerdyakov, have you ever seen real examples of CAD in ML?
23:18:37 <slava> Herkamire, show him some C code and forth code written by the same person, where the forth code is better, he'll say the guy was on crack when he wrote the C code.
23:18:39 <Herkamire> Smerdyakov: then get off IRC
23:18:56 <Smerdyakov> I could, but I think it's reasonable to want some evidence that I can expect experiments with Forth to be worthwhile.
23:19:02 <Smerdyakov> Since I only have finite time available. :)
23:19:05 <ASau> Smerdyakov. So, you have N yrs of Forth programmer < \infty of ML programmer.
23:19:10 <Smerdyakov> ASau, I haven't.
23:19:11 <slava> Smerdyakov, then stop wasting your precious finite time on IRC.
23:19:20 <Herkamire> Smerdyakov: you can't prove one language is better than another
23:19:27 <ASau> Smerdyakov, that's all. You've got example.
23:19:30 <Smerdyakov> Herkamire, I don't agree with that.
23:19:38 <Herkamire> Smerdyakov: how?
23:19:41 <AldoBR> he talks like a bot :)
23:19:55 <slava> AldoBR, ML is good for eliza-type things -- maybe here is one
23:19:58 <Smerdyakov> Herkamire, it's possible to draw very likely conclusions based on studies of how people use a tool to reach some goal.
23:20:09 <AldoBR> hope he program as good as he is on rethorics ;)
23:20:28 <Herkamire> Smerdyakov: even if you can agree on your results there, that only shows that it's good for that goal
23:20:30 <slava> i'll write an IRC bot in my forthish language that does eliza, then i'll put it in #hprog, and Smerdyakov can argue with it about ML all day long.
23:20:39 <AldoBR> i´m trying to find a wiki where all these flame wars where resumed...
23:20:52 <Herkamire> slava: lol
23:20:56 <ASau> slava, there is Eliza in Forth. :)
23:21:07 <slava> ASau, aha :)
23:21:09 <Smerdyakov> Herkamire, right. It doesn't make sense to talk about tools apart from goals.
23:21:20 <ASau> slava, you can find it in the net. 
23:21:24 <Herkamire> Smerdyakov: we're talking apples and oranges here.
23:21:30 <Smerdyakov> AldoBR, if you see any flames except from slava repeatedly insulting me, do let me know!
23:21:32 <AldoBR> but the server is written in C++ and is crashing due to memory leakages
23:21:41 <AldoBR> damn... 
23:21:43 <AldoBR> :P
23:21:56 <slava> Smerdyakov, i only insult you because of your excruciatingly irritating and dry manner.
23:21:57 <Herkamire> Smerdyakov: you can't prove which one is better. You might be able to prove that one is better for a particular purpose.
23:22:04 <AldoBR> Smerdyakov this kind of discussion is pointless
23:22:08 <Smerdyakov> Herkamire, right. That is what I meant by "better."
23:22:09 <ASau> Smerdyakov, wouldn't your majesty answer my question above?
23:22:12 <slava> Smerdyakov, if you try to behave like a real human and not a bot with built-in ATP algorithms, i'll take you seriously.
23:22:22 <Smerdyakov> ASau, which question?
23:22:48 <AldoBR> i use, C/C++, pascal, forth, clipper (huuu)....
23:22:56 <Herkamire> Smerdyakov: what particular purpose are you wondering how good forth is for?
23:23:03 <AldoBR> there is no sense in saying one language is better then another
23:23:10 <AldoBR> its just forms of expression
23:23:13 <slava> AldoBR, exactly, the pragmatic programmer uses a range of tools and doesn't worry themselves with showing which one is "better"
23:23:17 <AldoBR> like human language
23:23:20 <Smerdyakov> Herkamire, I don't have anything in mind now. I'd like to learn of a purpose for which it is a good choice.
23:23:27 <slava> AldoBR, english > french :-)
23:23:35 <slava> AldoBR, lets start a new flamewar :)
23:23:43 <AldoBR> i could not say that english is better than french
23:23:44 <Smerdyakov> AldoBR, again, this attitude is common among people who do not know how to develop software effectively.
23:23:45 <AldoBR> but
23:23:46 <ASau> Smerdyakov, you have an example of  N years of Forth programmer for CAD system. That's less than 20 years of ML existance with no CAD system. That's example that ML is the worst P.L. ever.
23:23:59 <ASau> Smerdyakov, what do you have against?
23:24:01 <AldoBR> i could say that french is better than english in diplomatic texts, for example
23:24:02 <slava> Smerdyakov, there you go again, with your "I'm the best programmer" talk
23:24:10 <Herkamire> Smerdyakov: I think forth is great for writing operating systems
23:24:12 <Smerdyakov> ASau, it doesn't make sense. You use invalid inference rules. :)
23:24:13 <slava> AldoBR, please stop! i was joking :)
23:24:23 <Smerdyakov> Herkamire, and there I think I might even agree with you.
23:24:24 <AldoBR> slava huauhuahua
23:24:42 <Smerdyakov> Herkamire, but OS's should only be a small portion of all software implemented.
23:24:50 <ASau> Smerdyakov, that does make sense. You have no one real application of your language.
23:25:11 <AldoBR> Smerdyakov about efficience, i only talk to you when you show some kind of work done efficiently
23:25:14 <ASau> So, what do you have against?
23:25:20 <Smerdyakov> ASau, not sure what you mean. Do you really need me to give you a link to "functional programming in the real world"?
23:25:31 <AldoBR> efficience thats only present in your mind will not change the world
23:25:50 <AldoBR> but, if you show me some examples...
23:25:52 <ASau> Smerdyakov. Yes, I want a real world CAD example from you.
23:25:52 <Smerdyakov> AldoBR, well, my web site is at http://www.schizomaniac.net/ . You may be able to deduce something from poking around. :)
23:26:22 <Smerdyakov> ASau, I'm not familiar with CAD, so probably you could hunt for examples as easily as I could.
23:26:24 <ASau> Smerdyakov, prove your ML has a _real_ application.
23:26:38 <Smerdyakov> ASau, mlDonkey and unison are popular examples.
23:26:56 <ASau> Smerdyakov, doesn't it show ML is much worse than Forth?
23:27:01 <Herkamire> ASau: what do you mean by "real"?
23:27:06 <slava> summary of Smerdyakov's home page: "I love ML"
23:27:09 <Smerdyakov> Of course, I view logical frameworks and theorem provers as real applications, which yields a slew of real ML applications.
23:27:14 --- quit: AldoBR (Read error: 104 (Connection reset by peer))
23:27:15 <ASau> Herkamire, I mean "existent".
23:27:31 <ASau> Smerdyakov, theorems are not application.
23:27:43 --- join: Amante_Profissio (Aldo@200.141.173.71) joined #forth
23:27:44 --- nick: Amante_Profissio -> Aldo
23:27:53 <Herkamire> ASau: mlDonkey exists. I've heard of it
23:27:53 <ASau> Smerdyakov, theorems are "spheric horses in vacuum".
23:27:55 <Smerdyakov> ASau, ah, luckily a few smart people saw beyond that viewpoint and changed the future of computer security forever. :)
23:27:59 <Aldo> your site is infected 
23:28:13 <ASau> Herkamire, what's mlDonkey?
23:28:26 <Herkamire> ASau: an eDonkey P2P client
23:28:42 <Aldo> http://c2.com/cgi/wiki?ForthVsLisp <- here, all this pointless flamewar resumed...
23:28:43 <ASau> Herkamire, it's all 'bout words.
23:29:07 <ASau> Herkamire, show me integrated circuits it produce.
23:29:12 <Smerdyakov> Aldo, again, all flames here are one-sided.
23:29:28 <Smerdyakov> Aldo, and, recent definitions excluded, I believe wars must be multi-sided.
23:29:41 <ASau> Herkamire, or any another info, I could set and experiment and test it.
23:29:44 <Aldo> Smerdyakov i think you dont have a point
23:29:54 <Herkamire> Smerdyakov: don't act so innocent. you came into the #forth channel repeatedly saying that ML is great and that you see no value in forth
23:29:57 <Aldo> you just keep say "oh you dont want to be a effective programmer"
23:30:04 <Herkamire> Smerdyakov: that's called TROLLING
23:30:07 <Aldo> this is all bullshit for me
23:30:28 <ASau> Smerdyakov, you may believe in whatever god. "Truth is always concrete."
23:30:33 <Aldo> my professor of introductory programming said that java was the most effective programming language around
23:30:35 <Smerdyakov> Herkamire, I don't follow. I've listed some areas for which Forth is well suited just in the last few minutes.
23:30:43 <Aldo> what i discovered later ? 
23:31:00 <Aldo> that he owned a private venture whose majority of products where java based...
23:31:04 <ASau> Smerdyakov, you see Forth is useful in real world.
23:31:13 <Smerdyakov> ASau, yes. I didn't say it wasn't.
23:31:14 <ASau> Smerdyakov, where're examples for ML?
23:31:23 <Aldo> this kind of "dogma" are generated by economic interests...
23:31:23 <Smerdyakov> ASau, I just listed some.
23:31:34 <Herkamire> Smerdyakov: just in the last few minutes. Sorry I left that out. You still seem to be trying to prove to a bunch of forth enthousiasts that ML is better.
23:31:36 <ASau> Smerdyakov, what? mlDonkey?
23:31:45 <Smerdyakov> ASau, that was one example.
23:31:48 <ASau> Smerdyakov, it is not a real-world.
23:31:56 <Smerdyakov> Herkamire, yes, I am. What's the problem with that?
23:32:02 <Smerdyakov> ASau, what is your definition of "real-world"?
23:32:21 <Aldo> http://c2.com/cgi/wiki?ForthIsDeadFlames 
23:32:22 <Aldo> oh
23:32:23 <Aldo> :)
23:32:26 <ASau> Smerdyakov, anything that produces a kind of information I can test in experiment.
23:32:39 <Herkamire> Smerdyakov: You're TROLLING
23:32:45 <Smerdyakov> Herkamire, define "trolling."
23:32:53 <Aldo> lot
23:33:06 <ASau> Smerdyakov, CAD is of such a kind.
23:33:06 <Aldo> his database engine didnt found this word ;)
23:33:15 <Herkamire> Smerdyakov: stative views that you know will agitate the crowd you're in
23:33:18 <ASau> Smerdyakov, mlDonkey is not.
23:33:38 <Smerdyakov> Herkamire, OK, then I don't mind the attribution, since that doesn't sound like a bad thing to me.
23:33:48 <Herkamire> ASau: what's unreal about mldonkey. it does networking. it transfers data from one computer to another. without networking we would not be talking here.
23:33:56 <Smerdyakov> Herkamire, since my purpose isn't to agitate anyone... that only comes as a side effect.
23:34:13 <Aldo> he uses sofismas...
23:34:15 <ASau> Herkamire, it does not produce anything.
23:34:19 <Herkamire> Smerdyakov: problem is that all you're doing is agitating people
23:34:34 <ASau> Herkamire, it only transfers information. No reality.
23:35:02 <Aldo> [04:32] <Herkamire> ASau: what's unreal about mldonkey. it does networking. it transfers data from one computer to another. without networking we would not be talking here. <- there goes implicit the concept that we dont want networking (at least, he tries to make appear that only his solution is able to do netwrk...)
23:35:04 <Herkamire> ASau: you could download something with mlDonkey, and then print it.
23:35:09 <ASau> Herkamire, it's done by computer geeks for computer geeks.
23:35:18 <Smerdyakov> Herkamire, then that is a flaw of the audience, and one I would have hoped not to find.
23:36:10 <Herkamire> Smerdyakov: you've chosen the wrong audience. that's what's wrong with it.
23:36:33 <Smerdyakov> Maybe, but only if Forth programmers can be expected to be illogical.
23:36:53 <Herkamire> Smerdyakov: we're not illogical. we just have different views than you.
23:37:04 <Smerdyakov> In that case, then I didn't pick the wrong audience.
23:37:10 <Herkamire> Smerdyakov: different programming methodologies, techniques, and priorities
23:37:11 <Aldo> Smerdyakov guy, my college study is psycology, and, sure, you need help...
23:37:34 <Smerdyakov> Herkamire, that shouldn't matter here.
23:37:38 <Aldo> youre clearly trying to find someone to fight with... 
23:37:44 <Smerdyakov> Herkamire, we should be able to discuss objectively the reasons for our preferred methods.
23:38:12 <Aldo> Smerdyakov your motivations are subjetive ones...
23:38:27 <Smerdyakov> Aldo, no.
23:38:27 <Herkamire> Smerdyakov: we could if you were interested. but you keep batting them down and saying that you want prove.
23:38:34 <Herkamire> s/prove/proof/
23:38:43 <Smerdyakov> Herkamire, how else can you have a pragmatic discussion?
23:38:58 <Aldo> Smerdyakov you are trying to argue
23:39:03 <Aldo> so YOU need to prove...
23:39:08 <Herkamire> Smerdyakov: reasons for liking things are not objective
23:39:10 <Aldo> got it ?
23:39:22 <Smerdyakov> Herkamire, they _should_ be objective for engineers.
23:39:25 * Herkamire looks up pragmatic
23:39:26 <hovil> no
23:39:39 * Aldo away
23:39:40 <hovil> programming is far to much a personal thing to be objective
23:39:49 --- nick: Aldo -> AldoBR
23:40:00 <Smerdyakov> I'm not interested in that personal side here. I implicitly refer only to what the best engineering decisions are to reach specific goals.
23:40:18 <AldoBR> dicothomy Mind X Body ;)
23:40:20 <ASau> Herkamire, I can download anything without it and it won't differ. So it doesn't produce and has no real-world ground.
23:40:40 <Herkamire> Smerdyakov: what specific goals? what engineers?
23:40:55 <Smerdyakov> Herkamire, right. We have to fix these for each discussion.
23:41:35 <Herkamire> Smerdyakov: I've asked you for specific goals. you have not given them.
23:42:07 <ASau> Smerdyakov. Well, you have said it yourself. So what do you have against CAD?
23:42:23 <Smerdyakov> The goals fall into the category of trying to produce software meeting a specification with minimal time and effort.
23:42:38 <Smerdyakov> ASau, I don't have anything against CAD. Some of my best friends are CAD
23:43:07 <Herkamire> Smerdyakov: ok, that's a good start. now unfortunately we will have to get a lot more specific to be able to say if forth or ML is better
23:43:29 <Smerdyakov> Herkamire, right, and we can get as specific as necessary to get information that helps us make software engineering decisions.
23:43:30 <ASau> Well, ML community has not produced any for 20 years of existance. Forth has at least two.
23:43:48 <Herkamire> Smerdyakov: it will depend on what sort of software we are trying to produce, what quality/effeciency is nessesary, and what programmer(s) are coding it.
23:44:18 <Smerdyakov> Herkamire, sure, though we may determine that some factors have little effect on tool choice.
23:44:32 <ASau> Sorry, three known to me. And I'm not E.E.
23:45:35 <Herkamire> Smerdyakov: ok, so let's say hypothetically, that we choose a programmer, to make a specific piece of software, and we have her code it in ML and forth.
23:45:44 <Herkamire> Smerdyakov: would that prove anything?
23:46:18 <Herkamire> Smerdyakov: Is there some way we could alter the test to be more objective?
23:46:36 <Smerdyakov> Herkamire, yes. We'll need to look at more people.
23:46:52 <Herkamire> Smerdyakov: it's so contextual though
23:46:58 --- quit: Serg ()
23:46:59 <Smerdyakov> Herkamire, we can also deduce things from cognitive properties shared by all healthy adults.
23:47:11 --- join: OrngeTide (orange@rm-f.net) joined #forth
23:47:47 <Smerdyakov> I take it as obvious that there are factors about human abilities, based on physical properties of brains, that lead to clear meanings of assertions that some programming languages are flat out better for programmer efficiency than others.
23:48:00 <Herkamire> even if you have 10000 people do the test, and you have have do the forth version first... that still doesn't tell us that one is better than another. perhaps the difference in quality is a function of the different learning curves for the two languages.
23:48:01 <Smerdyakov> It's another issue altogether how we can determine which such assertions are true and which are false.
23:49:04 <Smerdyakov> Every "programming case study" provides useful evidence.
23:49:29 --- quit: OrngeTide (Client Quit)
23:49:33 <Herkamire> it's circumstatial
23:49:35 <ASau> Smerdyakov, how many people do you want to test?  How reliable is your deduction model? What's it based upon? Statistics?
23:50:07 <Herkamire> it only shows you that with those sorts of people doing that sort of project (or maybe even just that project specifically) one is better
23:50:09 <Smerdyakov> ASau, I'm not proposing a deducation model now. I'm just asserting that "language X is better than language Y" has a clear meaning, though it may be hard to determine whether the statement is true.
23:50:26 <Smerdyakov> Herkamire, the human nervous system is a finite thing.
23:50:39 <Herkamire> Smerdyakov: that does not have a clear meaning
23:50:55 <Herkamire> Smerdyakov: would you say that "fruit x is better than fruit y" hase a clear meaning?
23:51:04 <ASau> Smerdyakov, do you really think that you know nervous system enough for this investigation?
23:51:30 <Smerdyakov> Herkamire, it does. The meaning is that, given human capabilities, we can predict how efficiently humans will perform tasks using different tools.
23:51:34 <Smerdyakov> Herkamire, on average.
23:51:51 <Herkamire> Smerdyakov: what is the meaning?
23:51:57 <Smerdyakov> ASau, no, but I think that this is a justification for drawing inferences from limited evidence: there are common underlying factors.
23:52:31 <Smerdyakov> Herkamire, quantify over all possible programming projects and all possible programmers. If X is better than Y, the average time to complete a project is lower in X than Y.
23:53:27 <Herkamire> Smerdyakov: that is not possible
23:53:36 <Smerdyakov> What is not possible?
23:53:51 <Herkamire> quantifying all possible programming projects and all possible programmers.
23:53:52 <AldoBR> lol
23:54:01 <Herkamire> Smerdyakov: what are you smoking?
23:54:05 <AldoBR> this is the perfect capitalist world
23:54:11 <ASau> Smerdyakov, "S" curve vs. "Z" curve. How do you want to compare if it happens?
23:54:27 <AldoBR> where programmers are machines
23:54:29 <AldoBR> ;)
23:54:41 <Herkamire> Smerdyakov: you can't even begin to compaire programming languages.
23:54:43 <Smerdyakov> Herkamire, it's somewhat ill-defined, but we can get more specific by quantifying over all people currently employed as programmers, for example.
23:54:47 <ASau> AldoBR, are you socialist?
23:54:53 <AldoBR> no
23:54:57 <Smerdyakov> Herkamire, and we can consider for each the next project he will be asked to complete.
23:55:03 <AldoBR> but not capitalism sympathizer either...
23:55:12 <ASau> Anarchist?
23:55:12 <Smerdyakov> Herkamire, then we can look at how long he would take at it in language X vs. language Y.
23:55:26 <Herkamire> Smerdyakov: no we can't
23:55:43 <Smerdyakov> Herkamire, we can't physically look, but the proposition is well enough defined.
23:55:45 <AldoBR> nop
23:56:06 <AldoBR> i dont have these labels inside my mind :P
23:56:36 <Herkamire> Smerdyakov: how could we look at how long someone would take would take at it in language X vs. language Y?
23:56:49 <AldoBR> now lets start a anti-usa flamewar
23:56:49 <Smerdyakov> Herkamire, we have to consider "alternate timelines."
23:56:58 <AldoBR> they tend to get VERY hot ;)
23:57:11 <AldoBR> these days...
23:57:28 <Herkamire> Smerdyakov: you can't predict the future. we're trying to look at evidence, because we don't know which is better
23:57:43 <Herkamire> AldoBR: battling fire with fire eh?
23:58:01 <Smerdyakov> Herkamire, if we consider human brains as formal machines, it's easy to say what it means for the machines to be able to create a working program using one language more quickly than in another language.
23:58:06 <ASau> Herkamire, that's what I'm trying to do.
23:58:22 <Herkamire> Smerdyakov: we have no working model of the human brain
23:58:32 <ASau> So what do you think on USA president elections?
23:58:34 <AldoBR> lol
23:58:36 <Herkamire> Smerdyakov: noone can predict brains thaht well
23:58:39 <AldoBR> how much bullshit
23:58:40 <AldoBR> huahuauhauhauhuah
23:58:42 <MarkT> they suck
23:58:54 <AldoBR> Smerdyakov like i said, dichotomy mind X body
23:58:56 <AldoBR> :)
23:58:57 --- join: OrngeTide (orange@rm-f.net) joined #forth
23:59:01 <Smerdyakov> That's OK. I think it's safe to posit that there is a model good enough for these purposes. We don't need to demonstrate to rely on its existence for the definedness of propositions.
23:59:12 <Herkamire> I donno if we have any pro USA government people here
23:59:15 <Smerdyakov> s/demonstrate/demonstrate it
23:59:49 <Herkamire> Smerdyakov: no way there is a good enough model
23:59:49 <ASau> Smerdyakov, do you thing that human brains can be well described as formal machines???
23:59:54 <Smerdyakov> ASau, yes.
23:59:54 <AldoBR> Smerdyakov and if your perfect programmer gets some vodka ?
23:59:59 <ASau> Smerdyakov, have you tried it?
23:59:59 --- log: ended forth/04.03.03