00:00:00 --- log: started forth/03.08.04
00:00:54 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
00:01:00 <Serg_Penguin> hi !
00:03:02 <kc5tja> hello
00:03:40 <Serg_Penguin> would u need an article on my control structures 4 one-liners ?
00:03:55 <Serg_Penguin> they encourage better factoring ^)
00:04:07 <Serg_Penguin> and less noise in source ^)
00:05:37 <Serg_Penguin> ( n -- ) rep
00:05:42 <kc5tja> Sure, but the site is down at the moment.  At least it was when I last checked.
00:05:59 <Serg_Penguin> ( flag -- ) ?? ?; ?re
00:06:00 <kc5tja> Speaking of which, . . .
00:06:05 * kc5tja notes that the topic for this channel is gone.  :(
00:06:34 <Serg_Penguin> ( -- ) re
00:07:23 <Serg_Penguin> they should be rewritten in ASM ;((
00:09:55 * kc5tja nods
00:10:18 <kc5tja> Probably as immediate words that compiles assembler.
00:10:20 <Serg_Penguin> are u interested in them ?
00:12:10 <kc5tja> Maybe in rep.
00:12:18 <kc5tja> Not sure about the others.
00:12:49 <Serg_Penguin> .. 25 rep line. ..
00:13:05 <Serg_Penguin> instead of .. 25 0 do line loop ..
00:13:47 * kc5tja nods
00:28:13 --- quit: Serg_Penguin ()
00:32:13 <kc5tja> Off to bed.
00:32:25 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
00:34:19 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
00:34:33 --- quit: Serg_Penguin (Client Quit)
02:08:29 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
02:08:33 --- quit: Serg_Penguin (Client Quit)
02:24:18 --- join: rk_ (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
02:27:33 --- quit: rk (Read error: 113 (No route to host))
04:28:35 --- join: krish (KRISHNAKUM@61.1.220.183) joined #forth
04:29:36 --- part: krish left #forth
04:30:09 --- join: mur (murr@baana-62-165-189-186.phnet.fi) joined #forth
04:46:54 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
04:47:04 <Serg_Penguin> hi mur !
04:47:12 <Serg_Penguin> i was missing you ^)
04:47:27 <Serg_Penguin> i undigged my Sea Legends, and now play like wild :)
04:47:50 <Serg_Penguin> it runs OK under XP, even sound works
04:49:57 <mur> hey Serg_Penguin :)
05:11:31 --- quit: Serg_Penguin ()
08:33:05 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
08:53:08 --- quit: Herkamire ("leaving")
08:54:31 --- join: wossname (wossname@HSE-QuebecCity-ppp81007.qc.sympatico.ca) joined #forth
09:17:07 --- join: gilbertdeb (gilbert@fl-nken-u2-c3b-118.miamfl.adelphia.net) joined #forth
09:33:38 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
10:01:27 --- nick: rk_ -> rk
10:44:51 --- join: jma (~jma@h-64-105-21-62.DNVTCO56.covad.net) joined #forth
10:45:02 <jma> 'lo all
10:45:09 <gilbertdeb> lo
10:45:55 <jma> any grand discussions at the moment?
10:46:20 <gilbertdeb> no, we were waiting for you to start one.
10:46:53 <jma> well then (no pressure)...
10:47:06 <jma> dunno if anyone here remembers me, havent been here in a while
10:47:17 <jma> i was the one who made a Forth for the GameBoy Advance...
10:47:18 <mur> terve gilbertdebi
10:47:35 <gilbertdeb> terve mur
10:47:36 <jma> its rockin' now with a solid userbase :)
10:47:51 <gilbertdeb> a forth for the game boy advance???
10:47:54 <gilbertdeb> do you have a url?
10:48:06 <gilbertdeb> (or at least submit one to http://forth.bespin.org )
10:48:11 <jma> www.simforth.com
10:48:31 <jma> it is pretty neat... the Forth part is called TIN
10:48:32 <gilbertdeb> what chip does the gameboy advance use?
10:48:37 <jma> ATM7TDMI
10:48:39 <jma> ARM even
10:48:55 <jma> I have a BASIC compiler built over it, which is compiled to Forth
10:49:00 <jma> then compiled to native code
10:49:22 <jma> Got lots of kids excited about learning Forth (they can make their own API functions in Forth for use in BASIC)
10:49:52 <gilbertdeb> hmmm.
10:49:56 <gilbertdeb> i don't see the forth part.
10:50:10 <jma> only a few users right now have access to it
10:50:20 <gilbertdeb> it is a proprietary forth?
10:50:23 <jma> ... for now... later i will release it completely
10:50:38 <jma> yes. and not ANSI either
10:50:44 <jma> I got plans for it :)
10:50:47 <gilbertdeb> :)
10:50:53 <gilbertdeb> how did you come to write a forth for the GBA?
10:51:25 <jma> Wrote a simple forth for it initially (easiest compiler to write)
10:51:35 <jma> then wrote BASIC compiler since no one used the Forth one
10:51:56 <jma> then decided to revamp the whole thing and combine them... BASIC is easier for everyone else, Forth easier for me
10:52:04 <gilbertdeb> :)
10:52:09 <gilbertdeb> so BASIC generates forth?
10:52:17 <jma> the compiler is command line based. passing "-tin" will assume that the source file is Forth
10:52:24 <jma> yes BASIC->Forth->ARM
10:52:31 <jma> totally transparent
10:52:46 <gilbertdeb> interesting. will you document your techniques?
10:52:57 <jma> I could, haven't thought about it much
10:53:10 <gilbertdeb> <hint><hint>
10:53:14 <jma> :)
10:53:39 <jma> there were a couple difficulties in the translation that may be interesting to others
10:53:45 <gilbertdeb> well?
10:53:49 <gilbertdeb> write them up :)
10:53:55 <jma> ok then :P
10:54:22 <jma> I know the guy writing Retro Forth OS is interested in how it was done
10:54:53 <jma> im in the process of taking it all a step further
10:54:53 <gilbertdeb> from the coiner of "You. Forth. the machine. simplicity"?
10:55:13 <jma> currently the whole compiler is written in ANSI C
10:55:23 <gilbertdeb> ah okay.
10:55:24 <jma> I'm working on making TIN a C library (like FICL)
10:55:35 <jma> then re-coding much of the compiler in TIN
10:55:54 <jma> so that others can then re-write much of the compiler if they want
10:56:13 <jma> they will have access to the ARM assembler, etc.
10:56:29 <jma> so that the compiler will be interpreted (if that makes sense)
10:56:33 <gilbertdeb> Hmmm. I thought someone else here was writing a forth for the arm. but I don't recall.
10:56:58 <jma> the arm is (simply put) the most bad-ass processor there is :)
10:57:15 <jma> I can't even look at x86 code any more without dry-heaving
10:57:15 <gilbertdeb> is it now?
10:57:29 <gilbertdeb> what other ones follow?
10:57:50 <jma> my personal favorites?
10:57:54 <gilbertdeb> sure.
10:58:04 <jma> Z80 and 6502
10:58:21 <wossname> 6502? :l
10:58:33 <wossname> surely that's far worse than x86
10:58:37 <jma> it was the first processor i worked on -- has a special place in my heart
10:58:43 <jma> :)
10:58:46 <wossname> ah, well then :p
10:58:55 <gilbertdeb> on an apple II or something?
10:59:01 <jma> atari 2600
10:59:18 <gilbertdeb> the 6502 gets quite a bit of air time here btw.
10:59:26 <jma> nice to hear :)
10:59:50 <jma> what projects y'all workin on?
11:00:16 <gilbertdeb> none at the moment for me.
11:00:30 <gilbertdeb> I'm mostly musing at other languages.
11:00:34 <gilbertdeb> like k (from kx)
11:00:46 <gilbertdeb> and I started reading the maude manual yesterday.
11:00:51 <jma> interesting, my current "toy" language is Haskell
11:00:53 <gilbertdeb> a little too much math for me.
11:01:05 <gilbertdeb> what do you think of haskell so far?
11:01:15 <jma> has some very cool features
11:01:24 <jma> but making it purely functional really bites
11:01:25 <gilbertdeb> incoporable into forth?
11:01:34 <jma> i would doubt it
11:01:50 <jma> one of Haskell's best features is lazy evaluation
11:02:01 <jma> that just plain rocks :)
11:02:06 <gilbertdeb> really?
11:02:11 <gilbertdeb> what's the advantage?
11:02:15 <jma> oh yeah... 
11:02:18 <gilbertdeb> it looks to see what its going to do before doing it?
11:02:24 <XeF4> you can lazy evaluate in forth too, you know..
11:02:31 <gilbertdeb> do tell XeF4.
11:02:32 <jma> you can, Xe?
11:02:52 <jma> gilbert, no, lazy evaluation means it doesnt do something until it needs to
11:03:10 <gilbertdeb> ack.
11:03:10 <XeF4> not quite as automagically as in Haskell, but yes
11:03:16 <jma> for example, let's say you have an array of numbers
11:03:27 <jma> now you want to +1 to every element in the array
11:03:31 <jma> then grab the 10th index
11:03:42 <jma> Haskell won't actually do the +1 to each element of the array
11:03:46 <jma> only then 10th one
11:03:51 <gilbertdeb> cool.
11:03:55 <gilbertdeb> 'clever' even.
11:04:18 <jma> with lazy eval, you can have "infinite" arrays, etc.
11:04:35 <jma> XeF4, can you give an example?
11:05:28 <XeF4> jma: you can eg, make array access words that build up a list of XTs to execute next time the value of some array becomes critical (usually =is read)
11:05:45 <mur> eval is the coolest function there is!
11:06:20 <jma> interesting
11:10:53 <XeF4> I should learn haskell sometime
11:10:57 <XeF4> some time, even
11:11:20 <jma> its difficult to get past the fact that its purely functional
11:11:23 <jma> monads suck
11:11:47 <jma> I/O is very frustrating
11:13:37 <gilbertdeb> but haskell is supposed to be functional isn't it?
11:13:59 <gilbertdeb> they warned you before telling you about it didn't they? ;)
11:14:17 <jma> that they did :)
11:14:33 <jma> lisp is functional, too... but side-effects are nice :)
11:16:00 <XeF4> lisp is also *simple* which is very nice
11:17:01 <jma> true, too
11:17:14 <jma> () get annoying, but you can do so much with them
11:17:27 <jma> another nice thing about Haskell is (like Forth) Currying
11:19:39 <gilbertdeb> what do you mean?
11:19:44 <gilbertdeb> btw have you seen joy?
11:19:55 <gilbertdeb> since you're into all this functional 'stuph'
11:20:22 <jma> hehe
11:20:33 <jma> by currying i mean, say function FOO takes 3 arguments
11:20:45 <jma> and you only pass 1, it will return a function that expects the other 2
11:20:51 <jma> so you can pass them later
11:21:05 <jma> Joy is interesting. Lisp meets Forth
11:21:12 <jma> afk for a min
11:21:29 <gilbertdeb> lisp meets forth?
11:21:37 <gilbertdeb> I thought that was pop-11 for some reason.
11:21:44 <gilbertdeb> or rather the HP calculator language.
11:22:02 <gilbertdeb> ?
11:22:29 <jma> haha
11:22:40 <jma> work calls... gotta cruise
11:22:44 <jma> be back l8r all :)
11:22:48 <jma> nice to be back...
11:23:00 --- quit: jma ()
11:36:08 --- quit: rk (Read error: 54 (Connection reset by peer))
11:36:47 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
11:44:29 --- quit: mur ("MURR! save the http://rainforest.care2.com/")
11:47:29 --- join: mur (jukka@baana-62-165-187-147.phnet.fi) joined #forth
11:50:11 <Herkamire> I remember jma. too bad I missed him
11:50:32 <gilbertdeb> pretty neat thing with the game boy, aye?
11:50:44 <gilbertdeb> <jma> be back l8r all :)
11:50:44 <gilbertdeb> <jma> nice to be back...
11:50:50 <gilbertdeb> he'll be back
11:55:22 <mur> Terve taas :)
11:56:42 <Herkamire> I don't know about BASIC, or the why he won't share the source
11:57:25 <Herkamire> but that's very cool to hear that he got it to work. IIRC last I heard from him he was asking about how to get the GBA connected to his computer
11:58:33 <gilbertdeb> heheh.
11:58:42 <gilbertdeb> sounds like a mighty leap from that to finished product.
11:59:24 <Herkamire> yeah :) haven't seen him in a while :)
11:59:37 <Herkamire> very cool to hear that he's got it to the point where others are using and interested in it
12:00:02 <gilbertdeb> I like the fact that basic generates forth.
12:00:24 <Herkamire> I've been working on my forth for several months and I've only showed it to one person
12:01:00 <gilbertdeb> yours is colorforth slanted isn't it?
12:01:20 <Herkamire> yeah. it's got a few slants I haven't done before
12:01:31 <Herkamire> such as asm, colorforth, pre-parsed source
12:01:59 <Herkamire> that's why it's taking me so long. never worked with asm before. or CF or ...
12:03:33 <Herkamire> I'm now in the middle of writing my first editor (which is 95% usable)
12:04:03 <Herkamire> and when that's done, I'm going to start playing with the BIOS for the first time
12:04:22 <gilbertdeb> but Herkamire very very few people have ever played with CF :)
12:04:33 <Herkamire> I know
12:04:46 <Herkamire> there's not much to read about it
12:05:32 <gilbertdeb> there is a #color4th channel.
12:05:36 <Herkamire> I read some e-mail conversation between tcn (or was it sean?) and Jeff Fox about it (flux I think)
12:17:53 <gilbertdeb> Hmmm I don't remember where it was anymore.
12:32:58 <Herkamire> here's what I was reffering to: http://www.ultratechnology.com/enthflux.htm
12:34:28 <Herkamire> where's the color4th channel?
12:35:08 <gilbertdeb> I can't seem to find it...
12:35:11 <gilbertdeb> lemme try agai.
12:37:17 <gilbertdeb> bah.
12:44:47 <Herkamire> maybe it ceased to exist
12:45:00 <gilbertdeb> i forgot the name.
12:45:03 <gilbertdeb> it is #c4th
12:45:10 <gilbertdeb> ... which server??
12:45:12 <gilbertdeb> I dunno yet.
12:48:33 <gilbertdeb> I'll have to ask r|o when he comes back.
12:53:08 <Herkamire> irc.stealth.net
12:54:58 <gilbertdeb> that helps.
12:54:59 <gilbertdeb> thanks.
13:39:55 --- quit: wossname ("fishznit")
14:25:28 --- quit: rk (Read error: 54 (Connection reset by peer))
14:25:51 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
14:37:57 --- quit: gilbertdeb ("Client Exiting")
16:42:44 --- join: suprdupr (CrowKilr@Ottawa-HSE-ppp3654067.sympatico.ca) joined #forth
16:44:00 <suprdupr> forth.bespin.org works for me now
16:44:25 <XeF4> suprdupr: I was looking through my homedir for an sdl stub and stumbled on another tiny tokenforth thing I had written and then somehow forgotten about =)
16:44:57 <suprdupr> can you send it to me, or explain how it works?
16:45:50 <XeF4> moment
16:46:17 <XeF4> scping the tarball to the usual place
16:46:39 <mur> XeF4,  :)
16:46:44 * mur goes sleep soon
16:46:49 <mur> 3 am already, XeF4 :)
16:47:22 <suprdupr> are you still doing the asm demo competition?
16:47:58 <XeF4> who knows.. hopefully but only 3.5 days left
16:48:35 <mur> hurry up! ;)
16:51:49 <XeF4> yeesyes
16:51:54 <mur> :)
16:51:57 <mur> hyvää yötä
16:52:04 <XeF4> öitä
16:52:45 <XeF4> supr: iirc, there was a version after this one with only newop and copybytes in the initial state
16:52:55 <XeF4> and some useful codewords but I can't recall where I put it
16:53:01 --- quit: mur ("MURRR!")
16:54:14 <suprdupr> =(
16:57:25 <suprdupr> what do you mean by newop, and what was the encoding of those tokens?
17:01:30 --- quit: suprdupr (orwell.freenode.net irc.freenode.net)
17:01:31 --- quit: XeF4 (orwell.freenode.net irc.freenode.net)
17:02:35 --- join: suprdupr (CrowKilr@Ottawa-HSE-ppp3654067.sympatico.ca) joined #forth
17:02:35 --- join: XeF4 (xef4@lowfidelity.org) joined #forth
17:04:55 <suprdupr> [b]*Last update: August 2, 2003*[/b]
17:04:55 <suprdupr> This example is the compiler compiling a
17:04:55 <suprdupr> program having the same functionality.
17:04:55 <suprdupr> This code could be rewritten
17:04:55 <suprdupr> to structure dependencies in a more
17:04:55 <suprdupr> logical way, like putting 4, 2, and 1,
17:04:55 --- quit: suprdupr (Excess Flood)
17:05:01 --- join: suprdupr (CrowKilr@Ottawa-HSE-ppp3654067.sympatico.ca) joined #forth
17:05:30 * suprdupr saves a note to himslef: when in doubt about the content of the clipboard, open the clipboard viewer
17:08:07 <XeF4> actually. this seems to be the wrong file given that the 2 compiler words newop and , are missing 
17:08:39 <XeF4> time to start using some sort of version control, I guess
17:35:57 --- join: TheBlueWizard (TheBlueWiz@ip-207-198-223-133.nyc.ny.fcc.net) joined #forth
17:35:57 --- mode: ChanServ set +o TheBlueWizard
17:36:03 <TheBlueWizard> hiya all
17:37:04 <suprdupr> hi
17:37:23 <TheBlueWizard> hiya suprdupr
17:37:57 --- quit: suprdupr ("gtg sorry, hope you're doing well! bye !")
17:40:58 <TheBlueWizard> gotta go...bye all!
17:41:19 --- part: TheBlueWizard left #forth
17:51:44 --- quit: ian_p (orwell.freenode.net irc.freenode.net)
17:51:45 --- quit: TreyB (orwell.freenode.net irc.freenode.net)
17:51:45 --- quit: skylan (orwell.freenode.net irc.freenode.net)
17:51:46 --- quit: XeF4 (orwell.freenode.net irc.freenode.net)
17:51:46 --- quit: ChanServ (orwell.freenode.net irc.freenode.net)
17:51:46 --- quit: rk (orwell.freenode.net irc.freenode.net)
17:51:46 --- quit: Herkamire (orwell.freenode.net irc.freenode.net)
17:52:07 --- join: ChanServ (ChanServ@services.) joined #forth
17:52:07 --- join: XeF4 (xef4@lowfidelity.org) joined #forth
17:52:07 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
17:52:07 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
17:52:07 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth
17:52:07 --- join: skylan (sjh@vickesh01-4919.tbaytel.net) joined #forth
17:52:07 --- join: ian_p (ian@inpuj.net) joined #forth
17:52:07 --- mode: orwell.freenode.net set +o ChanServ
18:39:37 --- join: Taaus (~T@cpe.atm0-0-0-196156.0x3ef21ebe.boanxx5.customer.tele.dk) joined #forth
20:18:23 --- quit: rk (Read error: 54 (Connection reset by peer))
20:23:32 --- part: Taaus left #forth
20:30:45 --- join: gilbertdeb (gilbert@fl-nken-u2-c3b-118.miamfl.adelphia.net) joined #forth
21:01:07 --- join: kc5tja (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
21:01:07 --- mode: ChanServ set +o kc5tja
21:01:21 <gilbertdeb> .
21:01:30 * kc5tja has money in his account for the first time in ..... almost a year.
21:01:40 <kc5tja> Too bad it'll all be spent pretty quick though.
21:01:42 <gilbertdeb> did the freedos project succeed?
21:01:45 * kc5tja needs more customers like this last one.
21:01:50 <kc5tja> For the most part, yes.
21:02:00 <kc5tja> I can't get GEOS to print to the printer -- no driver support.  >:/
21:02:00 * gilbertdeb applauds wildly.
21:02:05 <kc5tja> Otherwise, everything else is working perfectly.
21:02:09 <gilbertdeb> what printer is it?
21:02:21 <kc5tja> HP OfficeJet G85
21:02:36 <gilbertdeb> the client can setup a sneakernet no?
21:02:53 <kc5tja> Of course.
21:03:41 <gilbertdeb> excellent.
21:04:14 <kc5tja> That won't solve his problems though.
21:04:30 <kc5tja> But at least he can access critical business-sensitive data.
21:04:35 <XeF4> kc5: doesn't GEOS predate the Officejet series by about.. erhm.. 15 years?
21:04:44 <kc5tja> Yep!
21:04:58 <kc5tja> But GEOS does have DeskJet support.
21:05:21 <kc5tja> Unfortunately, the OfficeJet doesn't speak PCL3.
21:05:34 <XeF4> don't the DeskJet and Laserjet both speak PCL though?
21:05:46 <kc5tja> But he doesn't have any of those.
21:05:58 <kc5tja> He has an OfficeJet.
21:06:32 <gilbertdeb> is an officejet a laser printer as well?
21:06:38 <kc5tja> No, ink-jet.
21:07:03 <gilbertdeb> don't people realize how costly it is to run inkjet printers as opposed to laser printers?
21:07:08 <kc5tja> I really don't see why HP decided to not make the OfficeJet look like a normal ink-jet printer to the computer.
21:07:27 <kc5tja> Yes.  My laser printer is about twice as expensive to run than an ink-jet.
21:07:39 <kc5tja> Ink jets are virtually disposable printers now-a-days.
21:07:54 <kc5tja> And ink is on par with toner cartridges for lifespans.
21:07:56 <XeF4> but the ink
21:08:00 <gilbertdeb> kc5tja: your laser printer is more expensive to run?
21:08:05 <kc5tja> Yes.
21:08:14 <gilbertdeb> really?? this is news.
21:08:19 <kc5tja> XeF4: Volume -- more ink sold makes for cheaper ink.
21:08:25 <kc5tja> Where've you guys been?  :)
21:09:00 <XeF4> don't HP's ink carts still retail for $50 or somesuch?
21:09:10 <kc5tja> ?!?!
21:09:12 <kc5tja> No way
21:09:14 <kc5tja> Not here at least.
21:09:15 <gilbertdeb> 35
21:09:24 <kc5tja> I've seen them as cheap as $15 for black.
21:09:30 <gilbertdeb> and for color?
21:09:33 <kc5tja> Color maybe, but black is dirt cheap.
21:09:49 <kc5tja> I think maybe $35 for color -- maybe $40 to $45 for all four inks combined.
21:09:50 <gilbertdeb> how many pages do you get per cartridge?
21:09:58 * XeF4 has trouble believing a toner cartridge would run out after 1000 or so pages.. maybe they don't make consumer toner cartridges like the cartriges of yore, or..?
21:10:18 <kc5tja> XeF4: My toner cartridges are rated for 500 pages, and that's it.
21:10:19 <gilbertdeb> XeF4: it is supposed to run much much much longer.
21:10:30 <gilbertdeb> kc5tja: what sort of laser printer do you have?
21:10:40 <kc5tja> Brother CL-1250
21:11:26 <kc5tja> I think we got about 750 pages out of the last toner cartridge.  Still, it's rated pretty low.
21:11:36 <gilbertdeb> kc5tja: what about re-filling them?
21:11:51 <gilbertdeb> might it be that hp lasers have the higher rating?
21:12:18 <kc5tja> gilbertdeb: Perhaps, but HP LaserJet printers are *damn* expensive printers.
21:12:33 <gilbertdeb> a used one cost about $150
21:12:46 * kc5tja doesn't buy used.
21:13:09 <gilbertdeb> do you refill the cartridges?
21:13:09 <XeF4> at least the canon lbp-8 cartriges last for aeons
21:13:16 <kc5tja> gilbertdeb: Refill toner?  No.
21:13:22 <gilbertdeb> why not?
21:13:32 <gilbertdeb> I know they sound horrible after recycling them but ...
21:13:57 <kc5tja> gilbertdeb: It takes too long, it is way too messy (you get black powder ink all over the place, and it's basically like soot: impossible to clean)
21:14:40 <gilbertdeb> no not  yourself.
21:14:41 <kc5tja> Then you have to clean the cartridges after you refill them to make sure the rollers don't get clogged, etc. -- it is just a mess to do.  Not worth it.
21:14:46 <gilbertdeb> have a service do it for you.
21:15:15 <kc5tja> Oh, so pay a service the $20 for new ink, plus an hourly wage?  That's even more expensive than a new ink cartridge for it.
21:17:12 <gilbertdeb> hmmmm.
21:17:12 <XeF4> I would just use some (already stained and grase-covered) work clothes and refill the cartridge outside
21:17:20 <gilbertdeb> hahaha
21:17:49 --- join: KarlO (~hko@famties.dsl.xmission.com) joined #forth
21:18:20 <KarlO> Hi-all
21:18:25 <gilbertdeb> .
21:19:10 <KarlO> 1st forth I programmed was for a PDP-11 back in 1980.
21:19:22 <gilbertdeb> :)
21:19:33 <gilbertdeb> all hail the pdp-11.
21:19:37 <gilbertdeb> how was your experience?
21:19:38 <kc5tja> re
21:19:44 <kc5tja> brb phone
21:20:15 * Herkamire hates Internet Exploder
21:20:20 <KarlO> yea, it was for a seismic data gathering system for the Nevada Test Site.
21:20:37 <Herkamire> can't bloody display pngs properly
21:20:44 <KarlO> Was working for about a yrar befor I left the company
21:20:58 <gilbertdeb> did you write a forth?
21:21:36 <KarlO> No, but I modified the 1 I was using to use multitasking
21:21:51 <gilbertdeb> really?
21:21:53 <gilbertdeb> was it hard?
21:22:04 <KarlO> about 200 screens that generated about 32k of code
21:22:23 <gilbertdeb> sounds like a lot!
21:22:28 <gilbertdeb> was the machine running unix btw?
21:22:42 <KarlO> total application, the multitasking was about 10 screens if I remember correctly
21:22:58 <KarlO> no, RT11
21:23:40 <gilbertdeb> have you touched forth since then?
21:24:24 <KarlO> not in any sig. way, all my other employers had the language specified for all.
21:25:00 <gilbertdeb> it isn't too late then, you can write a forth too :)
21:25:11 <gilbertdeb> someone's written one for the Game boy advance.
21:25:24 <gilbertdeb> he came in today and mentioned it.
21:25:30 <Herkamire> KarlO: it's hard to get a job doing forth these days
21:25:37 <KarlO> Went to Asilamar 1 year for a conference, meet most of the sig. forth people for the time.
21:26:01 <gilbertdeb> did you meet chuck moore or E. rather?
21:26:08 <KarlO> both
21:26:12 <gilbertdeb> really?
21:26:21 <gilbertdeb> what was e. rather like?
21:27:06 <KarlO> yea, boy was that place plush, by the ocean in cabins. really nice. I was paid by the gov. for a 5 day vacation.
21:28:00 <KarlO> to tell the truth, I don't remember what she was like. It was a long time age.
21:28:10 <KarlO> ago
21:29:01 <KarlO> yea, Ting, ragsdal & others
21:30:15 <gilbertdeb> what was the conference about though?
21:30:19 <KarlO> Hard to get a job, period, esp. in any type of software, when you're 60
21:30:20 <gilbertdeb> just forth in general?
21:30:52 <gilbertdeb> you gotta start some sort of venture capital fueled company :)
21:30:57 <KarlO> yea, A little about forth79, & general forth
21:31:04 <gilbertdeb> it seems that money has run out too.
21:33:25 <KarlO> It'l get better, I remember what it was like in the early 70's. Man I that that the "stagflation" was never going to quit.
21:34:28 <KarlO> & price & wage controls. the result of that was runaway inflation & NO raises.
21:34:53 <gilbertdeb> what did you think of forth when you used it?
21:35:05 <gilbertdeb> in comparison to other languages you've used..
21:36:06 <KarlO> It was fun because you are so close to the hardware. The documentation was tough though. I bet they just recoded it in C or something after I left.
21:36:37 <gilbertdeb> the documentation for the seismic application in forth or the documentation for forth?
21:37:15 <KarlO> the forth. I was the only forth program in my section.
21:37:31 <gilbertdeb> hmmm.
21:37:39 <KarlO> programmer
21:37:57 <gilbertdeb> how did you come to learn forth to begin with?
21:38:12 <gilbertdeb> or did you simply adapt because of forth?
21:38:49 <gilbertdeb> I meant adapt because of the job.
21:39:01 <KarlO> I guess from s/w mag's is where I learned of it. I taught myself the language
21:39:18 <gilbertdeb> cool!
21:39:40 <KarlO> used Fig-forth
21:40:37 <gilbertdeb> there still might be incarnations of that somewhere on the web.
21:40:43 <gilbertdeb> have you seen colorforth yet?
21:41:13 <Herkamire> I hear very good things about fig-forth. I think there are a bunch of forthes based on it.
21:41:20 <KarlO> I tried it on some of my PC's, never got any of them to even display anything
21:41:33 <Herkamire> I think fig-forth runs under DOS
21:42:27 <Herkamire> Chuck's colorforth is very picky about hardware. I hear others have been hacking on it and have made versions that work on more PCs
21:42:31 * Herkamire searches for a link
21:43:26 <KarlO> It could be used under a operating system, but mainly it was for a bunch of micros, 6502, 6800, 6809,tms9900,...
21:44:11 <KarlO> the 1st was prob. a PDP-11
21:44:19 <Herkamire> http://www.oakland.edu/~maslicke/colorforth/
21:45:00 <Herkamire> the above link has a version of colorforth for linux, and a few for PCs that are much more likely to have the graphics work.
21:46:16 <KarlO> When I tried  colorforth on my main PC I had a devel of a time getting windows back!
21:47:00 <gilbertdeb> thats odd!
21:47:08 <gilbertdeb> how many reboots ;)
21:47:23 <gilbertdeb> Herkamire: do you have it working on the mac yet?
21:47:45 <KarlO> I finally had to redo some of the ini's to clear it up
21:48:10 <gilbertdeb> did you run it off the floppy?
21:48:19 <KarlO> Hey!, I'd love to have forth working on a 68000 MAC
21:48:39 <Herkamire> KarlO: try MOPS
21:48:56 <KarlO> I don't have a floppy, I use a LS120 as my floppy
21:49:18 <Herkamire> gilbertdeb: I just downloaded the version for bochs, but it needs an x86 assembler for the makefile...
21:49:51 <gilbertdeb> isn't mops a hybrid language Herkamire?
21:50:52 <KarlO> the assembler needs to be the MS MACRO assembler, also
21:54:32 <gilbertdeb> KarlO: do you remember much about the pdp-11?
21:57:31 <Herkamire> MOPS is a forth for 68K macs
21:57:53 <Herkamire> or older versions were at least
21:58:07 <KarlO> a little, I was an assembler for that machine for ten years(systems programmer)
21:58:26 <gilbertdeb> hehehe. 'a little' is an understatement!
21:59:09 <KarlO> that was almost 20 years ago
21:59:19 <gilbertdeb> I'm considering getting a copy of lion's commentary on unix, just to know what all the noise is about...
21:59:42 <gilbertdeb> i believe the asm language parts in his book were for the pdp-11.
22:00:23 <KarlO> I believe that unix was written on a PDP-11
22:00:49 <gilbertdeb> how fast/slow were those machines anyway?
22:00:53 <gilbertdeb> in mhz.
22:02:25 <KarlO> oh, shoot, it's hard to translate. those were the days when everyone thought a pipeline carried oil or water, but I think 10 to 20 mhz for the fastest
22:02:45 <gilbertdeb> Incredible.
22:02:48 <gilbertdeb> what about ram?
22:04:31 <KarlO> the best PDP-11 that I worked on had 256K of ram, the winchesters had maybe 128 meg. those were BIG things, 2.5'x2.5'x4' high
22:05:22 <gilbertdeb> the winchesters?
22:05:30 <gilbertdeb> the harddisk?
22:05:33 <KarlO> old speak for hard drive
22:05:34 <TreyB> Yes.
22:05:45 <gilbertdeb> they went out of business :)
22:06:35 <KarlO> winchesters stood for 30-30 because the 1st had 30 meg non-removeable/30 meg removable
22:06:45 <KarlO> like the rifle
22:06:52 <gilbertdeb> ack.
22:07:10 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
22:07:11 <gilbertdeb> so 20mhz, 256k ram, 128meg.
22:07:12 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
22:07:16 <gilbertdeb> thats a SMALL machine!
22:07:21 <Serg_Penguin> hi
22:07:25 <gilbertdeb> aloha
22:07:38 <KarlO> yea, took a room 30'x30'
22:07:53 <gilbertdeb> hehehe.
22:07:57 <KarlO> with serious air conditioning
22:08:12 <Serg_Penguin> i just fixed a bug in gp-forth ;)
22:08:22 <gilbertdeb> gp?
22:08:23 <KarlO> hey! cool
22:08:35 <Serg_Penguin> www.forth.ru
22:08:51 <Serg_Penguin> RU DOS .com 64k 16-bit forth
22:09:06 <gilbertdeb> I see square text.
22:09:13 <Serg_Penguin> makes kool tiny proggies
22:09:22 <Serg_Penguin> gilbertdeb: u see russian text ;)
22:09:39 <gilbertdeb> ??????
22:09:42 <gilbertdeb> thats what I get :)
22:09:46 <KarlO> ".ru"
22:09:52 <gilbertdeb> and i'm even using apl fonts!
22:10:04 <gilbertdeb> lemme see something.
22:10:11 <KarlO> U ever see a APL program?
22:10:32 <gilbertdeb> I'm reading the gilman and rose book.
22:10:37 <gilbertdeb> yes i've seen apl programs ;)
22:10:42 <gilbertdeb> I'm learning to write them too.
22:10:55 <KarlO> I think russian makes more sense
22:10:58 <KarlO> to me
22:11:00 <gilbertdeb> hahaha
22:12:05 <KarlO> I wrote some very small prog's in school back in '73 in APL.
22:13:27 <KarlO> It's cool for vector math
22:19:10 <KarlO> so long everybody
22:19:13 <Herkamire> gilbertdeb: have you gotten cf to work under bochs?
22:19:18 --- part: KarlO left #forth
22:19:18 <Herkamire> bye KarlO 
22:19:18 <gilbertdeb> Herkamire: no.
22:19:25 <Herkamire> gilbertdeb: tried?
22:19:27 <gilbertdeb> bochs doesn't like me.
22:23:31 <Herkamire> this is all I know: # vesa-chuck-8bit - A version by Alex Thiel intended for use under Bochs 
22:23:43 <Herkamire> don't know what settings Bochs should have
22:31:42 --- join: Fractal (bron@we.brute.forced.your.pgp.key.at.hcsw.org) joined #forth
22:50:04 --- quit: Herkamire ("goodnight")
22:51:55 <kc5tja> whew -- back
22:58:11 <kc5tja> What Herkamire is about to find out is that Bochs does not support VESA video modes.
22:58:27 <gilbertdeb> </narrator>
22:58:28 <gilbertdeb> hahaha
22:59:11 <kc5tja> Sorry I took so long; some of the things Karl0 said were not 100% true, due partly to failing memory.
22:59:27 <kc5tja> Unix was actually written on a PDP-7, and was later ported to a PDP-11.
22:59:39 <kc5tja> fig-forth was originally written for the 6502, not the PDP-11.
22:59:56 <gilbertdeb> heheh.
23:02:09 * kc5tja has done more research on how to handle event-driven programming.
23:02:31 <kc5tja> My conclusion is that it's probably going to be well worth the expense of implementing a simple task switcher in the Forth environment.
23:02:57 <kc5tja> With one task handling one specific event type, the software ends up being much smaller, easier to write, and easier to debug.
23:04:02 * kc5tja notes that ColorForth uses concurrently running tasks in the same manner, though the task allotment is static.  There is one task dedicated to updating the screen, one (foreground) task for managing keyboard input, and optionally, one extra "miscellanious" background task that is available, but rarely used.
23:04:32 <kc5tja> In my Forth, I'm planning on supporting more than three tasks.
23:04:58 <kc5tja> Hence, I'll need a slightly more general purpose multitasker.
23:05:30 <gilbertdeb> how are you finding out about this stuff?
23:05:35 <gilbertdeb> is there a book you're using?
23:05:42 <kc5tja> As far as inter-task synchronization, I'll borrow from the Concurrent Sequential Processes and use fully synchronous message reads and writes.  This eliminates the need for any other kind of synchronization method.
23:05:49 <kc5tja> Mostly Google.
23:05:57 <kc5tja> the CSP thing is a book, but I'd have to find my reference to it again.
23:06:15 <gilbertdeb> Google: the ultimate url.
23:06:25 * kc5tja couldn't understand most of what was in the book.  I was reminded about it when I was reading a research paper written by Rob Pike on concurrent GUI programming.
23:06:47 <gilbertdeb> heheh, someone is sniffing around plan9 afterall :)
23:08:09 --- quit: Serg_Penguin (Read error: 110 (Connection timed out))
23:08:34 <kc5tja> http://citeseer.nj.nec.com/rd/13761483%2C136608%2C1%2C0.25%2CDownload/http://citeseer.nj.nec.com/cache/papers/cs/761/http:zSzzSzcm.bell-labs.comzSzcmzSzcszSzdoczSz89zSz1-a.pdf/pike89concurrent.pdf
23:08:43 <gilbertdeb> hahahah.
23:08:48 <kc5tja> Pike's research into GUIs predates even AmigaOS.
23:08:55 <gilbertdeb> tinyurl.com <hint><hint>
23:08:59 * kc5tja notes that AmigaOS's layers.library is based 100% on Pike's work with layers.
23:09:04 <gilbertdeb> really?
23:09:07 <kc5tja> Yes.
23:09:18 <gilbertdeb> an unpealing of the amiga onion at last!
23:09:19 <kc5tja> AmigaOS was one of the first commercial OSes to utilize his work on layers.
23:09:26 <kc5tja> ???
23:09:41 <kc5tja> There's nothing to unpeal.
23:09:45 <gilbertdeb> :)
23:10:00 <gilbertdeb> for those who know the amiga back and forth. sure.
23:10:43 * kc5tja has mentioned how the Amiga's user interface system is structured a number of times here.
23:11:01 <kc5tja> intuition.library, graphics.library, and layers.library all cooperate to form the Amiga's user interface.
23:11:03 <gilbertdeb> yes yes, but I can't read this anywhere else.
23:11:12 <kc5tja> And it's one of the fastest user interfaces I know.
23:12:35 <kc5tja> To quote from Pike:
23:13:49 <kc5tja> "The channels that carry messages between processes in Newsqueak, and hence within the window system, are synchronous, bufferless, channels as in Communicating Sequential Processes (CSP) or Squeak, but carry objects of a specific type, not just integers . . . ."
23:14:22 <kc5tja> (note: his references to Squeak and Newsqueak are to window management languages, NOT to the Smalltalk clone.
23:14:41 <gilbertdeb> yes, I noticed.
23:16:38 <gilbertdeb> I like this bit:
23:17:12 <gilbertdeb> Window systems are not inherently complex. they seem complex because we traditionally write them, and their applications, as single processes ... etc
23:17:44 * kc5tja nods
23:17:57 <kc5tja> You know, I stumbled across this paper entirely by accident.
23:18:02 <kc5tja> I think the paper is sorely misnamed.
23:18:06 <gilbertdeb> awww. 
23:18:14 <gilbertdeb> I was hoping you'd say something about plan9 :)
23:18:34 <kc5tja> It isn't so much about concurrent window systems, as it is about how to use concurrency in handling events to make the software simpler.
23:18:52 <gilbertdeb> kc5tja: I was gonna say that it suggests a methodology for other things as well.
23:18:53 <kc5tja> How could I?  This paper predates Plan/9.
23:19:10 * kc5tja nods
23:19:54 <gilbertdeb> dare i say a lot of the quality methods for writing small, fast applications are not of this era?
23:20:25 <gilbertdeb> they had physical limitations that we might never see again.
23:20:54 <kc5tja> I think that's largely a valid statement.
23:21:02 <kc5tja> The use of concurrency is from this era though; 1989 isn't that long ago.
23:21:12 <kc5tja> Well, I suppose it kinda sorta is.
23:21:40 <kc5tja> But still, I'm not aware of a single operating system that implements its GUI architecture using such heavy use of multithreading.
23:21:57 <gilbertdeb> I'm beginning to think my 195mhz sgi box isn't quite slow enough to do interesting things with :)
23:21:57 <kc5tja> I think QNX comes the closest, but it's still not 100% there.
23:22:38 <gilbertdeb> Maybe a MAC se/30 or older ...
23:22:55 <gilbertdeb> or a 286 box. or even an apple ][ gs.
23:23:37 <kc5tja> You'll be amazed at what the IIgs (BTW, ][ isn't used for the IIgs -- Apple chooses either IIgs or //gs, because it has a whole "new image").
23:23:39 <kc5tja> err.
23:23:47 <gilbertdeb> :)
23:23:58 <kc5tja> You'll be amazed at what the IIgs can do; heck, even the Commodore 64 with GEOS/64 rocked.  It was slow, but it was amazingly capable.
23:24:09 <gilbertdeb> ][ does look like an I doesn't it?
23:25:35 <kc5tja> That's the basic idea.
23:25:47 * kc5tja prefers IIgs though -- it makes more sense to me.
23:25:56 <gilbertdeb> the machine?
23:26:02 <kc5tja> The name moniker.
23:26:23 <gilbertdeb> // looks slick.
23:27:39 <kc5tja> Yep.
23:27:46 <kc5tja> The IIgs was a slick machine.
23:27:59 <gilbertdeb> theres one I can get for about 30 bucks.
23:28:06 <gilbertdeb> I wonder if I should have it picked up or not.
23:28:17 <kc5tja> Certainly no comparison to the Amiga, which it was competing against (fools -- they didn't even increase the clock speed; it was running at a monsterously fast speed of 2.8MHz).
23:28:27 <gilbertdeb> hahaha
23:28:31 <kc5tja> But for an 8/16-bit computer, it definitely wasn't a bad design.
23:30:29 <gilbertdeb> I wonder what other relics from the past people are refusing to learn from...
23:31:49 <gilbertdeb> another thought: a windowing system based on pike's paper for the *nix console.
23:31:49 <kc5tja> Well, I think, despite the Atari 800's horridly ugly graphics, the Atari 800's chipset is a great model to learn from.  The Amiga is its direct descendent, actually.
23:32:44 <kc5tja> gilbertdeb: I thought about that, and I can't really see a way to make that happen easily.
23:33:05 <gilbertdeb> based on the pike paper?
23:33:11 <kc5tja> One could use MGR as a conceptual basis, where the application's sole connection to the window manager is through stdin and stdout.
23:33:13 * kc5tja nods
23:33:24 <gilbertdeb> hmmm.
23:33:24 <kc5tja> Hence, the window manager doubles as a terminal emulator.
23:33:50 <gilbertdeb> the TurboVision UI didn't look too bad to me when I first saw it.
23:33:55 <kc5tja> But events and GUI commands are multiplexed on the same I/O channel.
23:34:11 <kc5tja> To take advantage of CSP concepts, you'd need a demultiplexer library of some kind.
23:34:21 <kc5tja> (which shouldn't be hard to write, but still, it needs to be written)
23:34:40 <gilbertdeb> where is that assistant you ordered again? ;)
23:34:47 <gilbertdeb> ... or did you forget to order one?>
23:34:53 <kc5tja> gilbertdeb: MGR is fully graphical, and it uses stdio to affect everything.  It's even backward compatible with pure-textual applications.
23:36:01 <gilbertdeb> how on earth did you come upon this paper btw?
23:38:07 <kc5tja> Like I said, it was entirely on accident.
23:38:22 <kc5tja> I was looking for papers on how various GUIs handle events.
23:38:44 <gilbertdeb> have you seen the other things he's written?
23:38:48 <kc5tja> This paper was listed in the bibliography of an otherwise useless paper.
23:38:53 <kc5tja> Some of them.
23:38:57 <kc5tja> He's written a lot fo stuff.
23:39:14 * kc5tja found his "Systems Reserach Is Irrelavent" slideshow to be spot on the money.
23:39:32 <gilbertdeb> I was going to point you to this one: http://tinyurl.com/j0xz
23:40:06 <gilbertdeb> it refers to that particular paper and others.
23:40:13 <gilbertdeb> but I think you might have seen that list already.
23:43:59 * kc5tja nods
23:44:36 * kc5tja refers you to "Graphics in Overlapping Bitmap Layers" -- the seminal paper on layers.
23:44:56 * gilbertdeb googles.
23:44:56 <kc5tja> Unfortunately, layers is actually *slow* on the PC, because video memory bandwidth is so doggone slow.
23:45:03 * kc5tja has a (LONG!) URL for you.  :)
23:45:06 <kc5tja> If you want
23:45:08 <gilbertdeb> heheh
23:45:13 <gilbertdeb> tinyurl.com it .
23:45:16 <gilbertdeb> I'll look at it.
23:45:59 <kc5tja> OK, just a second
23:46:20 <gilbertdeb> I thought there was a tiny url bot somewhere abouts...
23:46:24 <kc5tja> http://citeseer.nj.nec.com/144447.html
23:46:28 <kc5tja> actually no need too.
23:46:44 <kc5tja> That'll bring up the root page -- just click on the PDF link in the upper right-hand corner of the page.
23:46:54 <gilbertdeb> thats better.
23:48:26 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
23:48:45 <Serg_Penguin> re
23:48:57 <gilbertdeb> :
23:49:21 <Serg_Penguin> power failure - changed auto-off'er
23:50:09 <kc5tja> :/
23:50:49 <Serg_Penguin> kc5tja: yesterday i wrote GREAT piece of ASM code ;)
23:51:00 <Serg_Penguin> in gp-forth what i use...
23:51:09 <Serg_Penguin> ?KEY was mega buggy
23:51:21 <Serg_Penguin> i wrote new one ;)
23:51:59 <Serg_Penguin> the code:
23:52:18 <Serg_Penguin> : cat begin ?key ifnot bye then emit again ;
23:52:27 <Serg_Penguin> was missing every second letter ;((
23:52:32 <Serg_Penguin> now it's OK !
23:52:34 <gilbertdeb> how???
23:53:04 <kc5tja> Sounds like ?key consumed the first key trying to wait on a key, while delivering the second.
23:53:11 <Serg_Penguin> yes !
23:53:18 <Serg_Penguin> brb, user calls...
23:53:55 <kc5tja> Ahh, the joys of system administration.  :D
23:59:59 --- log: ended forth/03.08.04