00:00:00 --- log: started forth/03.12.07
00:02:59 --- quit: Shain ("ircII+tkirc2")
00:57:04 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
01:37:53 --- join: schihei (~schihei@p5085DD3F.dip.t-dialin.net) joined #forth
03:09:31 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
03:11:24 --- quit: Robert (Read error: 54 (Connection reset by peer))
03:40:27 --- nick: Robert_ -> Robert
05:02:06 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
05:04:55 --- quit: ChanServ (ACK! SIGSEGV!)
05:19:50 --- quit: Robert (Read error: 110 (Connection timed out))
05:24:18 --- nick: Robert_ -> Robert
05:27:03 --- join: haroldo_ (~haroldo@r200-40-166-8-dialup.adsl.anteldata.net.uy) joined #forth
05:39:40 --- quit: haroldo (Read error: 110 (Connection timed out))
05:41:01 --- join: ChanServ (ChanServ@services.) joined #forth
05:41:01 --- mode: tolkien.freenode.net set +o ChanServ
06:45:02 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
06:46:10 --- quit: Robert (Read error: 54 (Connection reset by peer))
07:12:27 --- quit: ChanServ (ACK! SIGSEGV!)
07:14:14 --- join: ChanServ (ChanServ@services.) joined #forth
07:14:14 --- mode: tolkien.freenode.net set +o ChanServ
07:44:47 --- nick: Robert_ -> Robert
07:59:07 --- join: karingo (karingo@73.portland-01-03rs.or.dial-access.att.net) joined #forth
08:02:12 --- join: rO| (~noclue@p508078A2.dip.t-dialin.net) joined #forth
08:02:40 * rO| greets
08:03:37 <mur> hey
08:07:46 <rO|> brb
08:08:32 --- join: rO|_ (~noclue@pD95238B2.dip.t-dialin.net) joined #forth
08:12:35 --- join: rO|back (~noclue@pD9E0AE4D.dip.t-dialin.net) joined #forth
08:14:27 --- join: rO|| (~noclue@pD9523720.dip.t-dialin.net) joined #forth
08:23:51 --- join: dpb9cpu (~dpb@lexx.daves.net) joined #forth
08:27:31 --- quit: rO| (Read error: 110 (Connection timed out))
08:31:18 --- quit: rO|_ (Read error: 110 (Connection timed out))
08:32:46 --- quit: rO|back (Read error: 110 (Connection timed out))
08:35:48 --- quit: schihei (Client Quit)
08:44:51 --- join: sk1p (skip@p5086D8F1.dip.t-dialin.net) joined #forth
08:46:30 --- join: schihei (~schihei@pD9548A24.dip.t-dialin.net) joined #forth
08:55:43 --- quit: karingo ()
08:59:40 --- quit: ChanServ (Shutting Down)
08:59:53 --- join: ChanServ (ChanServ@services.) joined #forth
08:59:53 --- mode: tolkien.freenode.net set +o ChanServ
09:07:06 --- quit: schihei (Client Quit)
09:08:18 --- join: schihei (~schihei@pD9548881.dip.t-dialin.net) joined #forth
09:15:05 --- quit: Robert (tolkien.freenode.net irc.freenode.net)
09:15:05 --- quit: onetom (tolkien.freenode.net irc.freenode.net)
09:15:05 --- quit: Herkamire (tolkien.freenode.net irc.freenode.net)
09:15:06 --- quit: wUoNrFk (tolkien.freenode.net irc.freenode.net)
09:16:30 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
09:16:54 --- join: onetom (~tom@cab.bio.u-szeged.hu) joined #forth
09:17:09 --- join: Robert (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
09:28:05 --- quit: schihei (Client Quit)
09:29:20 --- join: wUoNrFk (~unfy@ip68-99-27-190.om.om.cox.net) joined #forth
09:36:20 <arke> hey
09:36:22 <arke> everybody read this
09:36:25 <arke> http://www.complang.tuwien.ac.at/forth/ftp.dei.isep.ipp.pt/pub/forth/docs/cm891023
09:38:21 <arke> he is just bagging on ANS
09:48:12 --- join: rO| (~noclue@pD9E0AF87.dip.t-dialin.net) joined #forth
09:49:04 <arke> http://www.complang.tuwien.ac.at/forth/ftp.dei.isep.ipp.pt/pub/forth/docs/cm891023
09:49:07 <arke> :)
10:06:14 --- quit: rO|| (Read error: 110 (Connection timed out))
10:10:54 <jc> Languages are interesting.  I've always proselytized Forth, but I've done virtually nothing in it.  CREATE/DOES> is a very powerful word, but it's not one that I know how to use effectively.  In Perl, there was a day when map() just hit me like a brick.  Now I don't use it like some people, but the light came on, and I frequently saw map() as a way to simplify code.  I can write basic Forth, although effective string management still eludes me (
10:10:54 <jc> Perl and even C have, from my perspective at this time, easier to use string functions).
10:12:05 <jc> Basically, I don't recognize opportunities to use CREATE/DOES>, so no doubt part of the code I have written are inefficient.
10:14:16 <arke> yeah, strings in forth are quite hard.
10:15:00 <jc> I assume you've been following the colorForth ranting in c.l.f?
10:16:54 <arke> i did for a while, then gave up :)
10:17:47 <jc> I recently just got back into newsgroup reading for c.l.f and comp.os.cpm.  Anyway, it's pretty amusing some of the opinions on why Forth does or doesn't have the acceptance it does.
10:19:42 <jc> Personally, one of the reasons I think Forth doesn't have a larger acceptance on platforms like Linux is the lack of bindings to libraries that have been around for years.  Not having any resources is (more) acceptable on a embedded processor, but when you need TCP/IP, FTP, UIs, crypto, etc, why re-invent everything in Forth when the libs already exist?
10:24:07 <arke> gforth does have lib.fs, which provides a crutch interface for libraries.
10:25:30 <jc> Yea, but it's one of the few.
10:39:27 --- join: wossname (wossname@HSE-QuebecCity-ppp81239.qc.sympatico.ca) joined #forth
10:39:48 <arke> wossname!
10:39:49 <arke> If you want to talk to fellow forthers on IRC
10:39:49 <arke> Get yourself an IRC client and join us at #forth on irc.freenode.net
10:39:53 <arke> er
10:39:54 <arke> oops
10:39:58 <arke> http://www.complang.tuwien.ac.at/forth/ftp.dei.isep.ipp.pt/pub/forth/docs/cm891023
10:39:58 <wossname> ?~
10:40:01 <wossname> :D
10:40:02 <arke> check that out!
10:40:02 <arke> :)
10:40:08 <arke> wossname: how have you been?
10:41:32 <wossname> i've been bleh
10:42:31 --- quit: sk1p ("/quit")
10:43:08 <wossname> how have you been?
10:44:09 <arke> ive been very7 very very very very good :)
10:44:21 <wossname> that good, huh? good for you :l
10:44:21 <arke> been coding on my environment server
10:44:23 <arke> in forth :
10:44:24 <arke> :)
10:44:27 <arke> ^_^
10:44:28 <wossname> environment server?
10:44:36 <arke> its is hard to explain...
10:45:00 <arke> imagine like an X server, but it can run on console, or an actual X server, or something else
10:45:03 <arke> sorta like that
10:45:52 <wossname> what is it to accomplish, then?
10:48:42 <arke> beat emacs
10:48:46 <arke> :)
10:48:55 <arke> beat emacs without actually being an editor
10:49:15 <arke> ^_^
10:50:29 <jc> emacs isn't an editor, it's an abomination against heaven and nature.
10:52:20 <arke> well, its all of those lol
10:52:22 <arke> be right back
10:52:26 <arke> ill explain more when im backj
11:01:17 <I440r> jc no - thats vi
11:01:26 <I440r> emacs is easy to ignore - vi isnt
11:01:39 <jc> vi is the editor of the gods.
11:01:51 <jc> When the universe was created, the script was written with vi.
11:05:00 <I440r> your evil :P
11:05:22 <I440r> the ONLY thin i ever needed with vi is escape colon q bang
11:05:36 <I440r> and i hated having to learn that
11:05:53 <I440r> the entire interface is an abomination - i understand the reason for it but i still dont like it
11:06:25 <I440r> unfortunatly in linux you cannot guarantee ANY single keypress will return a keycode other than the anlpha numerics
11:06:47 <I440r> every other key can be stolen from you by the operating syste, the terminal or the window manager
11:06:56 <jc> I had to relearn WordMaster the other day.  Now I keep crossing up WM and vi keystrokes.
11:07:30 <I440r> heh
11:14:01 <arke> :)
11:14:17 <arke> I440r: you gotta keep the link for isforth up ... i almost panicked about not being able to gt it!!!
11:14:35 --- join: schihei (~schihei@pD9548395.dip.t-dialin.net) joined #forth
11:16:03 <arke> vi is the best.
11:16:06 <arke> I440r: there?
11:16:16 <arke> I440r: you've got sockets, right??
11:17:13 <I440r> sort of
11:17:21 <I440r> dont have dns queries finished yet
11:17:27 <I440r> they are on the todo list
11:17:38 <I440r> but the irc bot in src/bot does connect
11:18:39 <arke> UNIX sockets?
11:18:43 <arke> thats my main interest.
11:18:49 <arke> I'm sorta not liking gforth...
11:19:01 <arke> I440r: you got deferred boot or deferred entry?
11:19:18 <I440r> defered boot ? you mean a defered init chain?
11:19:22 <I440r> theres default
11:19:30 <I440r> also, quit is defered in isforthg
11:20:16 <arke> er..
11:20:17 <arke> ok
11:20:21 <arke> wow
11:20:31 <arke> if i want (on start up) to run the word foo, what do i do?
11:20:44 <I440r> : foo defers default ...... ;
11:20:49 <I440r> thats ALL you need to do
11:21:01 <I440r> but it will run after everything else
11:21:12 <I440r> if foo displays anything it will happen after hello
11:21:27 <I440r> type hello in isforth to see what hello does hhe
11:21:34 <I440r> do it from a cleared sceen
11:22:00 <I440r> or if you do not want any of the other stuff to run do
11:22:03 <I440r> : foo ..... ;
11:22:08 <I440r> ' foo is default
11:22:14 <I440r> fsave/turnkey fooblah
11:22:36 <I440r> if its turnkey then foo must have a 0 <exit> in it
11:22:45 <I440r> erm wait
11:23:20 <arke> hehe
11:23:23 <I440r> if foo deferes default it should have a bye in it. if you do ' foo IS default and turnkey then foo must have a 0 <exit> in it
11:23:35 <I440r> i need to fix it so that bye can be used in both cases
11:23:50 <arke> OK
11:24:22 <arke> : rkg-start ... 0 ; ' rkg-start is default fsave
11:24:24 <arke> is that it?
11:24:42 <I440r> no .... 0 <exit>  ;
11:25:01 <I440r> <exit> isnt exit - <exit> is the syscall exit
11:25:12 <arke> oh
11:25:14 <arke> so
11:25:22 <arke> <blah> is the syscall <blah> ?
11:25:26 <arke> thats neat.
11:25:36 <arke> : bye 0 <exit> ; ???
11:25:43 <I440r> also. if your going to do a turnkey then i suggest you make a new copy of isforth.f, remove all the shit your app does not need - change the fsave isforth to  fsave foocompiler
11:25:57 <I440r> then have foocompiler compile your application and do
11:26:03 <I440r> ' foo is quit turnkey fooblah
11:26:12 <I440r> i kinda need to document all this
11:28:23 <arke> heh
11:28:23 <arke> ?
11:28:30 <arke> ' foo is quit turnkey fooblah
11:28:53 <arke> basically, int main() { atexit(foo); fooblah(); }
11:28:54 <arke> like that?
11:29:15 <arke> :)
11:29:33 <arke> do you mind if my project ships with a version of isforth?
11:31:00 <arke> (did i mention you need a make install?)   install: cp isforth /usr/local/bin/
11:31:09 <arke> :)
11:32:40 <arke> whats u<<?
11:35:01 <Robert> Unsigned shift left.
11:35:10 <Robert> er
11:35:14 <Robert> Yes.
11:35:37 <Robert> I.e. fill the bottom bit with a zero.
11:35:57 <I440r> no thats just a shift left dummy hhe
11:36:05 <arke> ok
11:36:09 <I440r> thers no difference between an unsigned shift left and a signed shift left :)
11:36:10 <Robert> I440r: Isn't that what it does?
11:36:16 <arke> isforth seemes to not like it.
11:36:21 <Robert> I440r: Sure, but that's what it does, no?
11:36:28 <arke> said u<< and u>> didnt exist
11:36:36 <I440r> thers << u>> and >> but no u<<
11:36:46 <arke> so i added : u<< << ; : u>> >> ; at the top of words.f
11:36:49 <arke> and it segfaulted
11:37:11 <I440r> what dide
11:37:12 <I440r> words ?
11:37:34 <arke> nforth :)
11:37:37 <arke> robert nforth
11:37:41 <arke> robert's*
11:37:44 <arke> runs on isforth
11:37:45 <I440r> oooh heh
11:38:33 <Robert> Heh.
11:38:44 <Robert> It's not meant to be used
11:38:54 <Robert> Just for forth freaks to look at and laugh
11:39:01 <I440r> lol
11:39:20 <arke> lol
11:41:07 <arke> well, im gonna go get to work.
11:41:23 <arke> I440r: how long are you gonna be here? i might need some help setting up my unix sockets..
11:42:50 <arke> (sorry, thats what you get for writing such an awesome forth! ^_^)
11:43:07 <I440r> arke lol i dunno
11:43:12 <I440r> im trying to fix my local network :P
11:43:15 <arke> :P
11:43:36 <arke> IRC works, so who gives a shit about the local network? :)
11:44:08 <Robert> The one who uses it?
11:44:21 <arke> :)
11:45:22 <arke> nip nip nip!?
11:45:29 <arke> nip ( a b -- b ) right?
11:45:47 <Robert> Yes.
11:45:53 <Robert> : nip  swap drop ;
11:46:03 <arke> :)
11:46:30 <arke> 3nip ( a b c d -- d ) :)
11:47:18 <arke> : tuck ( a b -- b a b ) swap over ;
11:47:41 <Robert> 3nip, heh.
11:48:21 <jc> : catnip ( cat -- ) drop ;
11:51:13 <arke> lol
11:52:16 <arke> : xnip [for] compile nip [next] ; immediate
11:52:28 <I440r> wtf is [for] ?
11:52:34 <I440r> heh
11:52:45 <arke> an interpreted for
11:52:46 <arke> :P
11:52:48 <I440r> isforth uses for/nxt  <-- i refuse to call it NEXT
11:52:55 <Robert> jc: :)
11:53:08 <Robert> And I call it "next" just to tease I440r.
11:53:12 <arke> I440r: yeah, i understand that. is quite confusing.
11:53:40 <arke> :)
11:53:43 <arke> ^_^
11:55:04 --- join: Sonarman (~matt@adsl-64-169-95-66.dsl.snfc21.pacbell.net) joined #forth
11:55:30 <arke> Sonarman: ^_^
11:56:03 <Sonarman> are you oninshiko?
11:56:27 <arke> no...lol
11:56:35 <arke> oninshiko is moving right now.
11:56:42 <arke> ^_^
11:56:48 <Sonarman> ^_-
11:57:03 <arke> I440r: could you gimme a hand please?
11:57:05 <arke> ...
11:57:08 <arke> that sounded wrong.
11:57:26 <arke> I440r: can you give me a hand with UNIX sockets in isforth please?
11:58:07 <I440r> maybe
11:58:10 <I440r> what you trying to do
11:59:21 <arke> well, first, im just trying to set up a simple UNIX socket client and server to communicate, and i can build from there.
12:04:06 <I440r> hmm
12:04:13 <I440r> well ive not looked at my sockets code in AGES
12:04:21 <I440r> but that irc bot basically does just that
12:10:19 <arke> but its not UNIX socket :)
12:10:36 <Sonarman> arke: you mean UDP?
12:10:59 <arke> Loading irc.f
12:10:59 <arke>  #inbuff ?
12:11:06 <arke> Sonarman: no, UNIX sockets :)
12:11:13 <arke> AF_UNIX
12:11:46 <I440r> irc.f ?
12:11:49 <I440r> try fload bot/bot.f
12:12:42 <arke> src/bot/io.f open error
12:12:44 <arke> erm
12:12:45 <arke> oh
12:12:46 <arke> lol
12:13:08 <arke> ok, worked this time...
12:15:04 --- join: [Forth] (~Forth@ca-cmrilo-cuda1-c3b-124.vnnyca.adelphia.net) joined #forth
12:16:08 <arke> [Forth]: greetings
12:16:15 <arke> [Forth]: Quit
12:16:18 <arke> ugh
12:16:19 <arke> bah
12:16:21 <arke> .
12:16:30 <arke> alright, i think I got it :)
12:17:42 --- quit: schihei (Client Quit)
12:18:03 <arke> uum
12:18:07 <arke> how do I quit it? :)
12:18:22 <I440r> erm press x in the console hhe
12:18:34 <I440r> it sometimes takes a second or 2 before it quits
12:18:46 --- quit: [Forth] (Client Quit)
12:18:48 <I440r> it doesnt respond to in channel
12:18:53 <I440r> its a DO NOTHING bot
12:18:55 <I440r> as yet
12:18:57 <arke> :)
12:22:50 <arke> I440r: C" for c-strings?
12:23:04 <I440r> err no
12:23:09 <I440r> what are you trying to do
12:23:21 <arke> make a C string
12:23:23 <I440r> do you want a counted string or a null terminated string
12:23:27 <arke> null terminated
12:23:31 <arke> but show me both
12:23:36 <I440r> create counted-string
12:23:42 <I440r>  ," this is a counted string"
12:23:48 <I440r> create null-terminated-string
12:23:57 <I440r>  ,' this is an uncounted string' 0 c,
12:24:06 <I440r> YOU have to put the null terminator
12:24:33 <I440r> you CANNOT embed a string inside a : definition
12:24:41 <I440r> other than a ." or an abort"
12:24:46 <arke> hrm
12:24:47 <arke> dam
12:24:49 <arke> ok
12:24:57 <I440r> its a fucked up idea anyway
12:25:03 <I440r> you cant execute a STRING
12:25:09 <I440r> you can execute a ." tho
12:25:35 <I440r> you embed a string insde a : def and its execution vector is to a word to SKIP past the string
12:25:55 <I440r> just fscking compile the address of the string as a literal and keep the STRING outside the : def
12:26:12 <I440r> iwish the fuck ans people would stop fucking coding gordian knot bullshit
12:26:16 <arke> makes sense.
12:26:18 <arke> heh
12:26:27 <arke> i read a transcript from some ANS meeting earlier
12:26:39 <arke> chuck kept bagging and bagging and bagging on them
12:26:42 <arke> about various things
12:26:55 <arke> like MOVE, IF, EVALUATE, and others
12:27:12 <I440r> the company im working for had one of the guys on the ans std team working for them - from what i can gather he was a good forth coder but spend months there with NOTHING to show for it
12:27:23 <I440r> he produced huge ammounts of NOTHING 
12:27:38 <I440r> move is anothyer fucked up featureitis word
12:27:45 <I440r> use cmove and cmove>
12:27:59 <I440r> move arbitrates which of the above to use
12:28:01 <I440r> bah
12:28:28 <I440r> thats so the dumbass coders dont need to know the language and decide which of the above two he/she needs
12:28:48 <arke> yeah
12:28:58 * I440r wont cater to the moronic "i dont know how to code really" group
12:29:29 <arke> i need to learn strings :P I don't get strings of any kind in any dialect of any sorts
12:33:37 <arke> what number is bind ?
12:37:32 <arke> wtf!?
12:37:48 <arke> i cant find bind!
12:38:59 <Sonarman> syscall number?
12:40:06 <arke> socketcall number'
12:40:43 <Sonarman> oh
12:42:12 <arke> ugh
12:42:16 <arke> i hate this fucking keyboard
12:54:18 --- quit: I440r ("Leaving")
12:55:47 --- quit: wossname (Connection timed out)
13:13:58 <arke> Robert: does isforth use EXIT or RETURN?
13:14:10 <arke> erm, nvm
13:45:39 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth
13:45:39 --- quit: I440r (Client Quit)
13:46:26 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth
13:55:11 --- join: _gps_ (~gps@166.70.196.201) joined #forth
14:30:16 --- quit: ChanServ (Shutting Down)
14:30:28 --- join: ChanServ (ChanServ@services.) joined #forth
14:30:28 --- mode: tolkien.freenode.net set +o ChanServ
14:30:37 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
14:31:06 --- quit: Robert (Nick collision from services.)
14:31:14 --- nick: Robert_ -> Robert
15:11:40 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
15:11:52 --- quit: Robert (Read error: 104 (Connection reset by peer))
15:12:11 --- nick: Robert_ -> Robert
15:29:45 --- quit: rO| (Read error: 110 (Connection timed out))
15:39:40 --- quit: I440r ("Leaving")
16:30:59 <Herkamire> oops. left irc on all weekend while I was away
16:31:05 <Herkamire> I'm back!
16:31:20 <Herkamire> wow somebody actually looked at my website and sent a comment.
16:49:22 <Robert> Hi :)
16:54:14 <Herkamire> hi robert :)
17:13:54 <jc> I think what I need is a good example of using Forth to implement something real-worldish, like taking stdin, replacing every occurence of foo with bar, and writing to stdout.  I don't know this for a fact, but I think Forth would tend to appeal to the unix-y types more than the Windows-y types.  And in Unix, we do everything with files, redirection, and pipes.  And I have no idea in something like isforth or gforth of how to implement reading f
17:13:54 <jc> rom stdin.  Something like "cat foo | gforth myfilter >arf"
17:42:33 --- quit: haroldo_ (Read error: 110 (Connection timed out))
17:43:31 <arke> jc: yeah :)
17:43:35 <arke> jc: you can use isforth for that.
17:43:46 <arke> jc: strings are not very nice in forth, however.
17:44:04 --- join: haroldo_ (~haroldo@r200-40-164-123-dialup.adsl.anteldata.net.uy) joined #forth
17:44:16 <warp0x00> dude
17:44:19 <warp0x00> arke: 
17:44:24 <warp0x00> strings aren't very nice in computers
17:44:29 <arke> jdrake: yeah.
17:44:30 <arke> er
17:44:32 <arke> warp0x00: yeah
17:44:36 <arke> lol
17:45:10 <warp0x00> who invented strings anyway
17:48:17 --- quit: _gps_ (".")
17:53:16 <arke> they invented themselved.
17:53:50 <Herkamire> strings aren't very nice in C
17:53:50 <Sonarman> they were first discovered after they overflowed their buffers, entering the realm of humankind
17:55:04 <arke> :)
18:00:36 <jc> What?!?  Strings in C are trivial.  Heck, only Basic and Perl make strings any easier.
18:02:12 <arke> BASIC!???
18:02:16 <arke> OMG
18:02:24 <arke> you are on some serious fucking crack
18:02:25 <Herkamire> jc: strings are a pain to manipulate in C
18:02:45 <Herkamire> most interpreted languages make strings easier than C
18:03:50 <Herkamire> you can't do this in C:    a = "Current value: " + x "\n";
18:04:05 <Herkamire> you can't do this in C:    a = "Current value: " + x + "\n";
18:04:11 <Herkamire> brb
18:04:14 --- quit: Herkamire ("leaving")
18:14:43 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
18:45:31 <jc> I write thousands of lines of C every year.  Strings are trivial.  And I was using BASIC as a comparison for how simple strings are to use.  I didn't say I liked basic.
18:45:53 <jc> Agreed, you can't use + (not talking C++ here), but concat'ing strings is still easy.
19:12:28 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth
19:16:56 --- quit: ChanServ (Shutting Down)
19:17:10 --- join: ChanServ (ChanServ@services.) joined #forth
19:17:10 --- mode: tolkien.freenode.net set +o ChanServ
19:20:03 --- quit: ChanServ (ACK! SIGSEGV!)
19:24:58 --- join: ChanServ (ChanServ@services.) joined #forth
19:24:58 --- mode: tolkien.freenode.net set +o ChanServ
19:25:12 --- quit: ChanServ (ACK! SIGSEGV!)
19:29:58 --- join: ChanServ (ChanServ@services.) joined #forth
19:29:58 --- mode: tolkien.freenode.net set +o ChanServ
20:44:25 <arke> I440r: there?
20:44:33 <I440r> nope
20:44:35 <I440r> :)
20:44:37 <arke> I440r: :)
20:44:49 <arke> I440r: RECURSE in isforth isnt true tail-recursive
20:44:56 <arke> any way that can be amended?
20:45:12 <I440r> depends lol
20:45:28 <arke> heh
20:45:55 <arke> how does a word look like in memory?
20:46:00 <arke> (i might have an idea
20:46:51 <arke> DTC doesn't allow jumps, but maybe you can declare words as tail-recursive
20:47:19 <arke> and in that case, isforth would jump to the second slot in the code field
20:47:25 <arke> the first slot being rdrop
20:47:29 <I440r> dood use goto
20:47:39 <I440r> : blah ...... goto foo ;
20:47:42 <arke> and RECURSE would go to 1st field (rdop )
20:47:45 <I440r> YOU make the words tail recursive
20:47:47 <arke> GOTO IS EVIL!
20:47:52 <arke> wheres foo?
20:47:54 <I440r> no stop thinking basic
20:47:59 <I440r> goto is NOT evil
20:48:03 <arke> goto still is evil
20:48:09 <I440r> ok whatever
20:48:12 <I440r> thats how you do it
20:48:13 <arke> but anyway
20:48:17 <arke> how does goto work?
20:48:23 <arke> : blah ... goto blah ; ?
20:48:36 <I440r> no
20:48:41 <I440r> : blah ..... recurse ;
20:48:43 <I440r> bad
20:48:57 <I440r> you CANT toto blah because its not revealed
20:49:04 <I440r> but thats not tail recursion
20:49:05 <I440r> use
20:49:06 <I440r> begin
20:49:08 <I440r> again
20:49:12 <I440r> fuck recursion
20:49:35 <I440r> anything you can do WITH recursion can be done FAR fucking better without
20:49:58 <arke> actually, i do like tail recursion very much in certain cases.
20:50:08 <arke> makes some things neater
20:50:14 <arke> which is why i want to use it
20:50:39 <I440r> you tell ME the difference between begin ... again and label: ...... jump label
20:50:43 <I440r> there IS none
20:50:46 <I440r> use begin/again
20:50:55 <I440r> or begin while repeat or begin until
20:51:03 <I440r> recursion is for lamers
20:51:17 <arke> i am lame
20:51:22 <arke> so there.
20:51:41 <arke> : recurse
20:51:41 <arke>   last @ name> , ; immediate
20:52:47 <arke> whats count and lexmask?
20:53:26 <I440r> brband ill tell u
20:53:31 <arke> ok
20:55:14 <I440r> : tail-recurse last @ name> @ compile goto , ;
20:55:47 <arke> lemme test :)
20:57:42 <arke> it doesnt recurse.
20:58:22 <I440r> it jumps back to the start of the definition
20:58:35 <arke> yeah
20:58:40 <arke> it doesnt work...
20:59:01 <arke> : blah 1 + tail-recurse ;  ok
20:59:01 <arke> 0 blah  ok
20:59:07 <arke> should give an infinite loop
20:59:18 <arke> . 1 ok
21:00:05 <I440r> hang on
21:03:28 <arke> it seems that goto is the problem here ...
21:03:52 <arke> unless...
21:04:10 <I440r> ok let me go test
21:04:37 <I440r> ohh lol
21:04:40 <I440r> make it immediate
21:04:41 <arke> : tail-recurse last @ name> @ compile goto , ; immediate  ok
21:04:41 <arke> : blah 1 + dup . tail-recurse ;  ok
21:04:41 <arke> 0  ok
21:04:41 <arke> blah 1 Speicherzugriffsfehler
21:04:42 <arke> chris@chris:~/RkG>
21:04:51 <arke> i was already doing that...
21:04:53 <I440r> : tail-recurse ..... ; immediate
21:04:55 <I440r> sorry lol
21:05:00 <arke> it only seems to print it once though.
21:05:06 <jc> It's very difficult to recurse through directory trees without recursion.  Where as the state can naturally be maintained on the stack, you're forcing the user to allocate space to preserve the state, and know ahead of time how much space he'll need.
21:05:13 <I440r> no its not
21:05:48 <I440r> i do depth first and breadth first tree spanning with itterative methods all the time
21:05:54 <I440r> and tey are MUCH easier to understand
21:05:54 <arke> hrm
21:06:05 <arke> it seems that tail-recurse crashes now
21:06:13 <I440r> ok wait lol
21:06:49 <I440r> oh. yea it will. lol
21:07:09 <I440r> it fils the return stack hang on ill re-write it
21:07:49 <I440r> : trecurse last @ >body compile branch , ; immediate
21:08:45 <I440r> i would put a  dup . cr inside your loop to prove its running
21:09:06 <arke> hrm it crapped out on me after about 4k iterations
21:09:06 <I440r> : test 1+ dup . trecurse ;
21:09:07 <arke> which shouldnt be happening
21:09:08 <arke> heh, thats exactly what i had
21:09:16 <arke> so i think it still fills the rstack somehow...
21:09:24 <arke> do you have a rdepth or such?
21:09:55 <I440r> no. its a stack crash tho let me investigate lol
21:10:03 <arke> :)
21:12:54 <I440r> i dun get it
21:13:03 <I440r> thers a reason its doing this i just dont know what it is lol
21:13:26 <arke> well, it definetely does increase the rstack by 1 each time
21:13:58 <arke> try inserting : rdepemit rp@ rp0 swap - 2/ 2/ . ; in trecurse, and see what it does
21:14:09 <I440r> the return stack is being overflowed
21:14:12 <I440r> and it shouldnt be
21:15:06 <I440r> yup the return stack is filling up
21:17:27 <arke> grr.
21:17:33 <I440r> i know where its going wrong
21:17:37 <arke> you do?
21:17:39 <arke> where?
21:17:41 <I440r> but now to fix it
21:17:50 <arke> whats it doing wrong?
21:18:00 <arke> heh
21:18:04 <arke> is it the branch? :)
21:18:09 <I440r> the branch address is wrong - last @ is not giving the address of the word USING tail-recurse
21:18:35 <arke> er
21:18:39 <arke> shouldnt it?
21:19:17 <I440r> : zzz last @ ; immediate  ok
21:19:17 <I440r> : test zzz literal h. ;  ok
21:19:17 <I440r> test 80D1B53  ok
21:19:17 <I440r> ' test h. 8054846  ok
21:19:34 <I440r> OOOH i think i got it lol
21:19:39 <I440r> wait
21:19:44 <I440r> dumbass grr 
21:19:54 <I440r> last @ gets the NFA not the cfa
21:23:21 <I440r> : tail last @ name> >body compile branch , ; immediate  ok
21:23:21 <I440r> : test 1+ dup . cr tail ;  ok
21:23:22 <I440r> lol
21:23:22 <arke> er? 
21:23:44 <I440r> last @ returns the address of the most recent words name field
21:23:54 <arke> oh
21:23:55 <arke> haha
21:24:02 <arke> and its not in it before ; ?
21:24:10 <I440r> ?
21:24:12 <I440r> that works
21:24:17 <I440r> but its pointless
21:24:23 <I440r> : test begin ...... again ;
21:24:31 <I440r> MUCH more readable
21:26:01 <arke> thanks
21:26:12 <arke> i like the recursive approach better in certain cases :)
21:26:55 <I440r> be warned
21:27:05 <I440r> that definition will NEVER be in the isforth release
21:27:10 <I440r> so if you want it you keep it
21:28:33 <arke> I will.
21:28:43 <arke> thanks alot :)
21:32:28 <I440r> arke can you ssh to my ip and tell me what it does ?
21:32:48 <I440r> does it refuse connection or ask for a login
21:33:23 <arke> ; immediate
21:33:23 <arke> : cstring-len-do ( addr 0 -- addr+len len )
21:33:23 <arke>   over c@ if
21:33:23 <arke>     \ Again...
21:33:26 <arke>     1 + swap 1 + swap
21:33:28 <arke>     tail-recurse
21:33:30 <arke>   then
21:33:33 <arke> ;
21:33:35 <arke> : cstring-len  ( c-addr -- len )
21:33:38 <arke>   0 cstring-len-do nip
21:33:41 <arke> ;
21:34:26 <arke> for some reason, it gives 14 for a size 6 string
21:35:04 <arke> er
21:35:05 <arke> nevermind
21:35:10 <arke> that was my own stupidity
21:35:13 <arke> it works :)
21:35:23 <jc> why 1 + instead of 1+?
21:35:38 <arke> dunno.
21:35:43 <arke> I440r: you need a disassembler.
21:35:46 <arke> :)
21:35:52 <I440r> i need an assembler first
21:36:07 <jc> PLEASE, not a postfix assembler...
21:36:07 <arke> shouldnt be too hard.
21:36:10 <arke> stack baded?
21:36:15 <I440r> can you /dns me ?
21:36:15 <arke> jc: those are fun though.
21:36:18 <I440r> whats my ip ?
21:36:22 <jc> I hate postfix assembly.
21:36:27 <arke> chris@chris:~> ping 12-160.lctv-a5.cablelynx.com
21:36:27 <arke> PING 12-160.lctv-a5.cablelynx.com (24.204.12.160) 56(84) bytes of data.
21:36:27 <arke> --- 12-160.lctv-a5.cablelynx.com ping statistics ---
21:36:27 <arke> 3 packets transmitted, 0 received, 100% packet loss, time 1999ms
21:36:27 <arke> chris@chris:~> ssh 24.204.12.160
21:36:41 <arke> (this went on about 3 minutes silent until i ctrl+c it)
21:36:55 <arke> jc: its not the greatest, but the best on a forth sys.
21:37:13 <jc> best?  Maybe easiest, but I wouldn't say best.
21:37:25 <arke> easiest, fastest, you name it.
21:37:42 <arke> easiest to implement, actually
21:37:47 <arke> not easiest to use...
21:37:48 <I440r> can someone please try sshing to this ip ?
21:37:53 <I440r> 24.204.12.160
21:37:56 <arke> I440r: i did. didnt work. :)
21:38:00 <I440r> i need to know if its refused
21:38:05 <I440r> was it refused or just nothing
21:38:12 <arke> it was doing nothing at all
21:38:25 <jc> I think your firewall is blackholing it.
21:39:09 <arke> mine might.,
21:39:53 <jc>  7:  so-1-0-0.gar1.Atlanta1.Level3.net (67.72.8.5)        asymm  8 117.707ms
21:39:53 <jc>  8:  so-6-0-0.gar2.Atlanta1.Level3.net (209.247.9.161)    asymm  9 117.862ms
21:39:53 <jc>  9:  so-0-3-0.bbr2.Atlanta1.Level3.net (209.247.11.225)   asymm 10 117.309ms
21:39:53 <jc> 10:  so-0-1-0.bbr1.Dallas1.Level3.net (64.159.1.109)      136.647ms
21:39:53 --- quit: I440r (Read error: 104 (Connection reset by peer))
21:39:53 <jc> 11:  no reply
21:40:00 <arke> and once again, i shall wake up at 2am
21:40:34 <jc> no reply on hops 11 through 16
21:44:13 <arke> ugh
21:44:14 <arke> fuck
21:44:15 <arke> he left
21:44:16 <arke> dammit
21:44:48 <jc> Prolly restarting the firewall or somesuch.  That usually drops a connection
21:45:11 <Herkamire> : cstring-len-do 2dup + @ 0= if exit then 1 + ;
21:48:04 <Herkamire> or if you have the handy word 0=; (works in place of "0= if exit then") it's just:   : cstring-len-do 2dup + @ 0= 0; 1+ ;
21:48:39 <Herkamire> I mean:   : cstring-len-do 2dup + @ 0=; 1+ tail-recurse ;
21:49:38 <jc> that should be c@, shouldn't it?
21:49:43 --- join: jdrake (~whore@CPE00045afdd0e8-CM014410113717.cpe.net.cable.rogers.com) joined #forth
21:49:57 <Herkamire> jc: :) yes
21:50:13 <arke> Herkamire: its supposed to be tail-recursive!
21:50:38 <Herkamire> arke: huh?
21:50:41 <arke> thats why you skip the 0= and put tail-recurse instead of exit
21:50:44 <arke> makes i better :)
21:51:07 <Herkamire> you have to exit when you get to the 0 terminator
21:51:10 <Herkamire> not recurse
21:52:52 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth
21:52:52 --- mode: ChanServ set +o kc5tja
21:52:55 <arke> kc5tja!!!!
21:52:56 <arke> yay!
21:53:09 <kc5tja> re, for now.
21:53:11 <arke> kc5tja: i gave up on gforth.
21:53:17 <arke> kc5tja: isforth-time.
21:53:20 <arke> kc5tja: :)
21:53:21 <kc5tja> I have to work tomorrow, so I will be going to bed within an hour or so.
21:53:25 <arke> :)
21:53:57 * kc5tja spent some time reviewing the ELF32 header format this morning.
21:54:22 <kc5tja> Although I think it's quite doable (e.g., isForth does it), I'm not sure I want to use it to generate stand-alone executables in FS/Forth.
21:54:41 <kc5tja> It seems like an awful lot of work to go through, and I'm not sure the gain is all that worth it for me.
21:55:16 <arke> FS/Forth = what type of forth?
21:55:25 <arke> ITC?DTC?STC?TTC?NTC?
21:55:36 <arke> and what the heck is doubly ITC?
21:55:59 <kc5tja> Building and maintaining an ELF header is beneficial in that it enables me to compile faster software (fewer register-indirect addressing modes), but maintaining relocation tables just seems like a pain.  And the string table doesn't seem like it'd help much.
21:56:05 <kc5tja> arke, I already told you.  :)
21:56:15 <kc5tja> FS/Forth is native code compiling.
21:56:20 <arke> so NTC :)
21:56:26 <kc5tja> And doubly indirect threaded code is just what it says: doubly-indirect.
21:56:29 <arke> (which doesnt make mark very happy)
21:56:43 <kc5tja> Fuck Mark.
21:56:48 <arke> heh
21:56:55 <arke> so basically, a double pointer?
21:57:01 <arke> pointer to pointer to code field!?
21:57:01 <kc5tja> I mean, I like him and all, but his opinions have no bearing on what I do.
21:57:03 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth
21:57:05 <arke> whats teh point of that?
21:57:15 <arke> I440r: speaking of whom...
21:57:18 <arke> lol
21:57:18 <arke> haha
21:57:18 <kc5tja> arke: It depends.
21:57:20 <arke> I440r: :)
21:57:43 <kc5tja> FS/Forth is my Forth.  If others find it useful, that's fine.  If not, that's not my problem.  :)
21:57:50 <arke> kc5tja: :)
21:57:59 <arke> kc5tja: I'll port RkG to it :)
21:58:08 <kc5tja> If/when I get it done.
21:58:09 <arke> kc5tja: for now, RkG will be isforth turnkey
21:58:15 <kc5tja> At this point, progress is next to zero.
21:58:20 <jc> Forth interpeters are like assholes.  Everyone has one, no one wants to look at anyone elses...
21:58:41 <arke> jc: that is not true at all.
21:58:43 <kc5tja> jc: I looked at a *lot* of other interpreters before choosing my implementation.
21:58:47 <jc> No, wait.  That's supposed to be "opions", isn't it?
21:58:51 <arke> jc: IsForth is a very beautiful Forth
21:59:12 <jc> er, opinions.  I can type.  Really.
21:59:19 <arke> :)
21:59:38 <kc5tja> I haven't used IsForth, but it seems like a pretty decent system.  It has an impressive compile speed, and it is currently the only Forth for Linux that can turnkey itself easily.
21:59:40 <arke> which reminds me, i need to re-extend isforth to not show the title bar (which allows scrollback!)
21:59:50 <arke> kc5tja: it is soooo easy
22:00:09 <kc5tja> arke: That's nice, but I just don't need that capability.
22:00:13 <arke> ' entry is quit turnkey my-application
22:00:16 <arke> done! :)
22:00:19 <arke> kc5tja: :)
22:00:21 <kc5tja> :(  Arthritis acting up again.  :(
22:00:26 <arke> :(
22:00:36 <arke> gforth can't turnkey
22:01:00 <kc5tja> I would like FS/Forth to turnkey, but I don't see that happening.
22:01:14 <arke> :)
22:01:24 <arke> well, if you're awake at 2am, i will too :)
22:01:25 <Herkamire> kc5tja: does/will fs/forth link to libraries?
22:01:28 <kc5tja> ELF is just too intensive.
22:01:39 <arke> (my newest thing: sleep until 2am, then wake up and code/play freeciv
22:02:09 <kc5tja> Herkamire: FS/Forth doesn't exist for Linux yet.  Yes, it will link to libraries, via libdl.so (e.g., the linkage will be entirely built at run-time by explicitly opening libraries by name, rather than linking against them).
22:02:55 <arke> ill be watching you while i sleep .. literally! keep talking
22:02:56 <arke> night
22:03:03 <Herkamire> kc5tja: cool. I like that. but then why would you need symbol tables and such in your ELF header?
22:03:08 <kc5tja> Herkamire: However, FS/Forth, being native code compiling, will only support x86.  I have no immediate plans to support PowerPC
22:03:49 <Herkamire> FWIW my PPC forth can turnkey easily
22:04:10 <Herkamire> not sure of the exact definition of turnkey, but it spits out an ELF that runs by it's self
22:04:13 <kc5tja> Herkamire: To create the initial bindings to libdl.so, you need to link against it.  Hence, you need a symbol table for those symbols.  You also need the symbol table for any exposed functions too (e.g., if one were to use FS/Forth to write a .so file, for example)
22:04:36 <kc5tja> Herkamire: Your code maintains a complete list of address relocations as it compiles?
22:04:57 <Herkamire> oh, I thought you were dynamically linking to libsdl too.
22:05:07 <kc5tja> That's what I'm concerned about: I already have two tables (the Forth and Compiler wordlists), plus I have a whole stack of USER-variables; I don't want to have to extend that.
22:05:12 <kc5tja> Herkamire: I am.
22:05:23 <kc5tja> libsdl = dlopen( "libsdl.so" );
22:05:26 <kc5tja> if( libsdl != 0 )
22:05:36 <Herkamire> and you need to have a symbol table for that?
22:05:42 <kc5tja>   func = dlsym( libsdl, "sdl_sync" );
22:05:46 <kc5tja>   func();
22:05:52 <kc5tja> No.
22:05:54 <Herkamire> (sorry for stupid questions, I've never done dynamic linking)
22:06:11 <kc5tja> You need a symbol table to link against libdl, so that I can use dlsym, dlopen, and dlclose.
22:06:27 <kc5tja> Linux has no intrinsic API for manipulating, or even using, shared libraries.
22:06:33 <kc5tja> Neither do most Unices.
22:07:11 <Herkamire> perhaps what I do wouldn't count as turnkey. when you press shift-S it compiles a new kernel (elf header and all) and then saves that along with source code in an ELF
22:07:17 <kc5tja> Shared libraries works under Unix only because of a really convoluted process involving dynamic loaders and such.  This is why ld.so must be used as the process 'loader' -- it understands how to load a binary that links against a dynamic library.
22:07:30 <kc5tja> Herkamire: No, that's turnkeying.
22:07:32 <Herkamire> when the ELF starts up (unless you change something) it starts running the source saved in itsself
22:07:42 <kc5tja> That is the very definition of turnkey.
22:07:49 <kc5tja> But...
22:08:00 <kc5tja> I cannot assume my code loads at a fixed address.
22:08:10 <kc5tja> So any variable references need relocation.
22:08:15 <Herkamire> you want to be able to make libraries?
22:08:24 <kc5tja> Or, more bluntly, any use of an absolute address needs relocation.
22:08:41 <Herkamire> my code is not relocatable
22:08:43 <kc5tja> Herkamire: Not right away.
22:08:56 <kc5tja> It'd be a nice feature to have though.
22:09:16 <kc5tja> Especially when writing software using unit tests.
22:09:30 <Herkamire> if you want to be able to compile a binary then you would have to have a symbol table. but if not, and all your linking is dynamic, I don't see why you would need a symbol table.
22:09:50 <kc5tja> OK, I'm really not sure how many times I need to explain this.
22:09:58 <kc5tja> Nor am I sure of how to make it any clearer at all.
22:10:08 <Herkamire> oooh :) I like that idea. then your unit tester could just load your forth and call a function :)
22:10:14 <kc5tja> I need a symbol table because FS/Forth needs to invoke dlopen, dlclose, and dlsym.
22:10:21 <kc5tja> These functions are not Linux API calls.
22:10:29 <Herkamire> oh.
22:10:30 <kc5tja> They exist solely in libdl.so.
22:10:35 <kc5tja> E.g., itself, a shared library.
22:10:39 <Herkamire> so you are statically linking to libdl.so
22:10:41 <kc5tja> Hence, I must link against that shared library.
22:11:04 <Herkamire> is statically linking the propper terminology?
22:11:08 <kc5tja> No
22:11:36 <kc5tja> If you have libfoo.a (e.g., an archive of .o files), that's statically linking, because a physical, self-contained, 100% complete copy of libfoo is embedded smack in your binary's file.
22:11:45 --- join: Serg_Penguin (~z@212.34.52.140) joined #forth
22:11:57 <Herkamire> right
22:12:08 <kc5tja> If you have libfoo.so (e.g., a shared object file), that's dynamic linking, because you include only references to the library, which is loaded at load-time, and linked just before run-time.
22:12:44 <Herkamire> for what you're doing, you simply say "fsforth links to libdl" right? 
22:12:58 <kc5tja> What I'm trying to do is something equivalent to this:
22:13:06 <kc5tja> ld -o fsforth fsforth.o -ldl
22:13:13 <Herkamire> right
22:13:24 <Herkamire> so then you need the libdl symbols in your ELF's simbol table
22:13:31 <kc5tja> Yes
22:14:07 <Herkamire> would you consider having just those symbols? (downside being that you could not load your forth _as_ a library?
22:14:55 <kc5tja> Well, I need to do a complete feasibility study.  But I'd rather not impose artificial limitations if I can help it.
22:15:15 <Herkamire> right
22:15:20 --- join: matt_ (~matt@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
22:15:33 <kc5tja> I guess what I'm saying is that ELF is not hard to implement.
22:15:40 <Herkamire> that really sucks that the dlopen and dlsym are not syscalls
22:15:41 <kc5tja> But ELF may be hard to maintain in a live Forth system.
22:15:54 <kc5tja> Well, not even hard.
22:15:58 <kc5tja> Just cumbersome. 
22:16:07 <Herkamire> I think it would be cumbersome to compile relocatable code
22:16:13 <kc5tja> And I am seriously questioning whether the payoff is worth the implementation complexity.
22:16:19 <kc5tja> It's not.
22:16:27 <kc5tja> mov eax,address_of_foo
22:16:28 <Herkamire> actually, you're right
22:16:38 <Herkamire> 99% of my code is relocatable
22:16:49 <kc5tja> is perfectly relocatable, as long as you flag that use of address_of_foo as needing a fixup in a relocation table.
22:17:02 <arke> bye
22:17:09 <kc5tja> laters clog
22:17:12 <kc5tja> FOASIJDF:
22:17:20 <kc5tja> I hate this program sometimes.
22:17:21 <kc5tja> later arke
22:17:39 <kc5tja> Tab completion should never change to tab enumeration.
22:17:55 <Herkamire> it's pretty easy most of the time to be relocatable without the need for fixup on PPC at least. just use relative branches, and use branch tables instead of tables of function addresses.
22:18:24 <Herkamire> kc5tja: you mean it should work like it does in vim not bash?
22:18:25 <kc5tja> With x86, you only have relative branch addresses.
22:19:01 <kc5tja> Herkamire: When I type foo<tab> in Bash, it'll complete, but it will NOT auto-enumerate (e.g., each <tab> does not cycle through a list of matches prefixed with foo).
22:19:09 <kc5tja> Bash exhibits perfectly correct behavior.
22:19:58 <Herkamire> what exactly do you mean by enumerate? you mean display all possible completions? or try the first possible enumeration?
22:20:06 <kc5tja> Type "arke" in x-chat, then hit tab.
22:20:09 <kc5tja> Watch what happens.  :)
22:20:19 <Herkamire> don't have xchat
22:20:30 <Herkamire> irssi completes the first match of multiple matches
22:20:31 <kc5tja> It'll respond with clog, because it's the next item in the user list.
22:20:48 <Herkamire> and doesn't cycle through them if you hit it again
22:20:52 <kc5tja> Herkamire: So does x-chat.  But then, after that, it'll enumerate.
22:21:21 <Herkamire> enumerate throught things that done't match?
22:22:00 <kc5tja> Yep.
22:22:02 <Herkamire> irssi completes with the matching nick that most recently said something in the channel. I think that's the most useful
22:22:18 * kc5tja prefers the Bash solution -- hit tab, and it beeps.  Hit it again, and it shows a list of choices.  Pick one.  :)  (I admit vim's is nicer, but not essential.)
22:22:23 <Herkamire> althought it does give the wrong nick sometimes
22:22:50 <Herkamire> I think it depends on context
22:23:01 <Herkamire> when I don't know what's there, I very much prefer bash.
22:23:17 <Herkamire> but when I'm opening files I open a lot (like programming on my own projects) I prefer vim
22:23:43 <Herkamire> I actually wish bash completed the same way as vim (with another key to show you all matches)
22:23:50 <kc5tja> I've used vim for years, and never have I ever opened more than one file in vim.  Nor have I ever used the :r command.  :)
22:24:17 <Herkamire> :) actually vim does have another key to show all matches (ctrl-D)
22:24:33 <Herkamire> wow
22:24:43 <Herkamire> I use multiple files a lot
22:25:00 <Herkamire> do you use split?
22:25:05 <kc5tja> Split?
22:25:10 <Herkamire> :split filename
22:25:13 <kc5tja> No
22:25:19 <kc5tja> One window, one console, one editor.
22:25:36 <Herkamire> interesting.
22:25:47 <Herkamire> so you just quit and open another file?
22:26:00 <kc5tja> Yep.  Or I open another shell window.
22:26:25 <kc5tja> My feeble mind isn't sophisticated enough to memorize all the useful commands of vim and it's multi-file support commands too.
22:26:48 <Herkamire> it did take a while before I knew enough commands to enjoy multiple files
22:27:17 <Herkamire> the one that makes it all worth it is ctrl-shift-6 (ctrl-^) which switches to the last file you were looking at
22:27:19 <kc5tja> I need to pick up a good book on vim.
22:27:49 * kc5tja has the O'reilly book Learning VI, which is a *very* good book.
22:27:51 <Herkamire> much quicker that :q<enter><up arrow><enter>
22:28:15 <kc5tja> But it doesn't go into specifics with respect to vim.
22:29:05 <Herkamire> I've hardly used straight vi, so I'm not terribly aware of what features are only in vim
22:29:15 <kc5tja> LOTS.  :)
22:29:19 <Herkamire> :)
22:29:46 <kc5tja> Having used raw vi, I'd never choose vi over vim.  It's just too useful.
22:30:14 <Herkamire> I agree
22:30:42 <Herkamire> I'd choose vim simply for the syntax coloring.
22:30:51 <kc5tja> Another feature I don't use.  :)
22:31:20 --- quit: Sonarman ("leaving")
22:31:45 <Herkamire> maybe it's my dyslexia or something, but I find it way easier to see what's going on with color
22:32:30 <Herkamire> BTW vim doesn't really edit multiple files at once (unless you split the screen and show them both at once.)
22:32:37 <kc5tja> Yes, I know.
22:32:42 <Herkamire> it more keeps track of what files you've had open, and allows you to switch
22:32:57 <kc5tja> But unless I have a good user interface for supporting multiple files, which neither Emacs nor vim support to my liking, I just won't use it.
22:33:12 <Herkamire> the only way I use that history is ctrl-^ to switch to the last file
22:33:18 <kc5tja> I find I can keep things in my head better if I can actually see the files that are open, instead of remembering which window has what file tucked away underneath something.
22:33:42 <Herkamire> that's the way vim works
22:33:52 <Herkamire> in vim the only open files are the one(s) you can see
22:33:55 <kc5tja> The only thing I miss doing it this way is cut-n-paste, beyond that which fits on a single window at least.
22:34:19 --- quit: arke (Connection timed out)
22:34:33 <kc5tja> I'm not too sure about that.
22:34:35 <Herkamire> vim saves it's clipboard across sessions
22:34:48 <kc5tja> I distinctly remember having a session that had three files open at once, but only one was on screen at a time.
22:34:58 <kc5tja> Herkamire: Across sessions.  Not across windows.
22:35:02 <Herkamire> when you do :e it closes the file you're looking at (you must first save or discard changes) before it opens the one specified
22:35:52 <Herkamire> it then saves the filename of the one you were just editing so you can go back and edit it if you like
22:36:22 <Herkamire> kc5tja: right. not accross windows. that's one reason I often do multiple files in one vim session
22:37:03 <kc5tja> I don't use it often enough to be a major concern.
22:37:09 <kc5tja> But :e seems like it'd be useful to me.
22:37:15 <kc5tja> I wasn't aware of its existance.
22:38:11 * kc5tja just learned of the :e! command -- basically "revert to last saved."  :)  That can be of immense value.
22:38:23 <Herkamire> nice :)
22:38:36 <Herkamire> makes perfect sense :)
22:38:55 <Herkamire> the default file is the one you're looking at. ! generally means without saving
22:39:11 * Serg_Penguin screwed my mind w/ VI
22:39:27 <Serg_Penguin> i can never imagine more weird editor ;((
22:39:45 <Herkamire> Serg_Penguin: ever tried ed?
22:40:15 <Serg_Penguin> in a fresh install of FReeBSD, i had to edit /etc/rc.d to raise network
22:40:27 <Serg_Penguin> it was a big deal of pain !
22:40:32 <Serg_Penguin> ed ? no, never
22:42:34 <Herkamire> I like :e! now I don't have to do :q! and open the file again
22:43:37 <Herkamire> I still think ctrl-d for listing possible completeions is my best new vim descovery. awesome with the :help command
22:43:50 <Herkamire> especially since I can't seem to find the command to go to the next matching help page
22:43:53 <kc5tja> Herkamire: Yes.
22:44:00 * kc5tja is experimenting with split windows now.
22:44:13 <kc5tja> vim can handle vertical panes as well as horizontal panes.
22:44:26 <Herkamire> :vsplit
22:45:09 <Herkamire> ^W^W switch panes
22:45:14 <matt_> hmm, trying to use quartus forth... really need one of those keyboard's for the palm
22:45:29 <Herkamire> [number]^W+ make pane bigger
22:45:34 --- nick: matt_ -> fridge
22:46:49 <Herkamire> I love'd my targus keyboard
22:47:03 <Herkamire> too bad they didn't have dvorak support until after my visor died
22:49:10 <fridge> at the moment, I write code on the pc, export it to the palm, then "include file"... even typing that takes a bit of effort, I'm not a fan of pen based input methods
22:49:34 * kc5tja must be the only person who really doesn't have that much of a problem with it.  :)
22:49:38 <Herkamire> fridge: have you seen dasher?
22:49:41 <kc5tja> But I agree that a keyboard is nice to have.
22:49:59 <kc5tja> Oh crap!  I forgot to do laundry tonight!  :( :(
22:50:18 <fridge> Herkamire, I think I have
22:50:29 <fridge> I thought it was only for PocketPC
22:50:41 <Herkamire> I have wet laundry in the washer
22:50:53 <Herkamire> kc5tja: than's for reminding me :)
22:50:55 <fridge> if is a stream of letters zooming in?
22:50:58 <Herkamire> brb laundry
22:51:24 * kc5tja is going to put his laundry in to let it run over night.
22:51:28 <kc5tja> Then dry it in the morning.
22:51:39 <Herkamire> dasher is the zooming letters. it's mainly a linux project I think
22:52:30 <Herkamire> I think I heard that palm is not fast enough to run dasher
22:54:29 <kc5tja> Certainly the ARM-based units are...
22:59:32 <kc5tja> Hmm...Dasher looks very interesting.  :)
23:01:55 <Herkamire> :)
23:02:53 <Herkamire> there is an mpeg of someone using dasher on an ipaq
23:04:59 <Herkamire> my uncle (who is a chiropractor and also works with people with severe disabilities) was very interested in the possibility of some people who can't type (or maybe even people who can't move their arms at all) being able to type with dasher
23:05:02 <kc5tja> I looked at the eye-movement video.  That's pretty neat.
23:05:17 <Herkamire> yeah :) I want to try :)
23:05:40 * kc5tja wonders how hard it'd be to write software that works like Dasher.
23:06:01 <Herkamire> donno
23:06:10 <Herkamire> I think they did a damn good job though
23:06:53 <jdrake> can anyone translate this letter (or jist of it) into a more english rather than the format it is in now? http://darwinports.gene-hacker.net:8080/430
23:10:14 --- quit: Serg_Penguin ()
23:11:11 <Herkamire> jdrake: yes :) it's quite amusing
23:12:02 <jdrake> i can't quite understand it too much
23:14:16 <kc5tja> Dear Editor
23:15:13 <Herkamire> I am afraid that the letter in the last issue about Forth Inc's using only three letter name name fields has had the opposite effect from what the writer wanted. His letter (like this one) showed that saving only three letters and count is just about optimal in terms of a trade off between saving memory and keeping legible. We still don't se the need for 31 character names in the general case.
23:15:19 <Herkamire> Yours Truly
23:15:21 <Herkamire> Chuck Moore
23:15:24 <Herkamire> Forth Inc.
23:15:57 <I440r> i disagree with that big time
23:16:17 <I440r> full names COST NOTHING if you keep headers seperate and turnkey
23:16:25 <kc5tja> I440r: I agree with it, considering the usual case of Forth programming.
23:16:37 <kc5tja> I440r: You're changing the argument.
23:16:49 <I440r> no - its saying 3 letters and a count is enough
23:16:51 <I440r> i disagree
23:16:55 <kc5tja> That's not what it's saying.
23:16:57 <jdrake> how about this idea - back then machines were slower
23:17:00 <kc5tja> That's not AT ALL what it's saying.
23:17:16 <I440r>  We still don't se the need for 31 character names in the general case
23:17:38 <kc5tja> It's saying that three letters and a count are enough in terms of a trade off between saving memory and keeping legible.  That is a signficiant difference.
23:17:46 <I440r>  three letters and count is just about optimal in terms of a trade off between saving memory and keeping legible
23:17:53 <I440r> 3 letters is illegible
23:17:58 <I440r> so the trade off isnt there at all
23:18:02 <kc5tja> ummm
23:18:04 <kc5tja> ummm
23:18:09 <Herkamire> I440r: don't know about you, but I could read the letter just fine
23:18:14 <kc5tja> you do realize that the source still has more than three letters in it, right?
23:18:15 <Herkamire> it is legeble
23:18:21 <I440r> i didnt look at it lol
23:18:28 <jdrake> how about C64 - it only kept the single character i think (basic)
23:18:40 <kc5tja> jdrake: But it didn't let you have long variable names at all.
23:18:41 <I440r> so is the letter where they swap arround all the middle letters of every word
23:18:44 <Herkamire> I440r: read:  http://darwinports.gene-hacker.net:8080/430
23:18:57 <kc5tja> What it's saying si that you can have any word you want, but only the first three letters are significant with respect to scanning the dictionary.
23:19:21 <kc5tja> It does ******NOT****** say that you're restricted to just three letters.
23:19:28 <I440r> having to read that crap in source files would be a nightmare
23:19:37 <kc5tja> Oh for fucking god damned christ's sake.
23:19:39 <jdrake> thank you for the translation and invoking conversation :-)
23:19:49 <I440r> you have to think at each word
23:19:56 <Herkamire> I440r: I don't
23:19:56 <I440r> if the words were all there in full you would not have to
23:20:01 <I440r> also the letter is contrived
23:20:06 <kc5tja> I440r: Please stop.
23:20:15 <jdrake> yes please
23:20:21 <kc5tja> You're making accusations that are wholesale wrong and uninformed,  You have demonstrated a total ignorance of what Chuck is driving at.
23:20:21 <I440r> im sure i could come up with a letter just like it that was almost impossible to read
23:20:22 <Herkamire> please
23:20:41 * kc5tja sighs
23:20:44 <Herkamire> I440r: good for you. that's not the point.
23:20:47 * jdrake notices kc5tja's Godliness at the moment
23:20:51 <I440r> no. i see exactly what he is driving at and i disagtree
23:20:57 <kc5tja> o, you clearly don't.
23:21:05 <kc5tja> You insist that three letter words are illegible.
23:21:08 <kc5tja> Which may be true.
23:21:17 <Herkamire> I440r: you didn't even read the thing before you decided it was wrong. now drop it
23:21:21 <kc5tja> But what is being discussed is not source code consisting exclusively of three-letter words!!!
23:21:36 <kc5tja> What is being discussed is DICTIONARY SEARCH KEYS consisting of three-letters and a count.
23:22:07 <kc5tja> Anyway, I'm going to bed.
23:22:15 <kc5tja> i can't take much more of this, and I have to work tomorrow.
23:22:15 <Herkamire> goodnight kc5tja :)
23:22:25 <Herkamire> don't forget your laundry :)
23:22:26 <I440r> bah. try using a decompiler or debuggger on those words
23:22:29 <I440r> nightmare
23:22:39 <kc5tja> I440r: You don't need a decompiler or debugger -- you have the block source.
23:22:45 <jdrake> is there any place where I can get a quick overview of forth syntax and common practice?
23:22:53 <kc5tja> jdrake: No
23:23:05 <kc5tja> There is no such thing as common practice in Forth.
23:23:24 <jdrake> how about the syntax part
23:23:37 <I440r> thers almost no such thing as syntax too heh
23:23:43 <jdrake> how about this - is there a nice standard that can be read
23:23:49 <kc5tja> As far as Forth syntax is concerned, probably the best in-print reference is the "Forth Handbook," by Elizabeth Rather.
23:23:50 <I440r> no
23:23:57 <I440r> thers an evil standard that can be read thoi
23:23:59 <I440r> tho
23:24:04 <jdrake> forth-83?
23:24:14 <jdrake> kc5tja, any online references?
23:24:21 <I440r> no i prefer that to ans for sure
23:24:32 * kc5tja overwhelmingly prefers ANSI to F83.
23:24:52 <I440r> the ans std is designed to let people NOT have to know forth
23:24:56 <kc5tja> ANSI basically *is* F83, but it's been modernized and rationalized.  And is platform independent.
23:25:01 <kc5tja> I440r: Another baseless accusation.
23:25:05 <I440r> no
23:25:09 <kc5tja> Prove it to me.
23:25:18 <I440r> postpone
23:25:25 <I440r> nuff sed
23:25:28 <kc5tja> Wrong.
23:25:32 <kc5tja> To use it, you must know Forth.
23:25:38 <I440r> no
23:25:46 <Herkamire> one point for kc5tja
23:25:47 <I440r> postpone is horrible
23:25:51 <kc5tja> In fact, that is THE quintessential word you must know Forth about to use it, lest you crash the environment.
23:25:59 <I440r> it will do [compile] AND compile both
23:26:12 <kc5tja> I440r: PRECISELY WHY YOU MUST KNOW FORTH TO KNOW WHEN TO FUCKING USE IT!
23:26:14 <I440r> thus releaving you of the responsability of know how to use those words
23:26:25 <I440r> nope
23:26:35 <I440r> you need to know forth to know compile and [compile]
23:26:45 * kc5tja sighs
23:26:49 <I440r> postpone baby's you so you dont need to know if the word your compiling is immediate or not
23:26:50 <kc5tja> Whatever.
23:26:59 <kc5tja> That is just plain not true.
23:27:45 <jdrake> hmm, kc5tja reminds me sort of like smerdy and I440r reminds me for arke  both from #hprog
23:28:00 <jdrake> only kc5tja is consistantly correct and rational :)
23:28:19 <kc5tja> And in a pissy mood because of all the horseshit misinformation flinging around here.
23:28:21 <I440r> no he just argues better
23:28:41 <kc5tja> Anyway,
23:28:48 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
23:29:19 <I440r> the ans standard does NOTHING helpfull, it RESTRRICTS whaqt you can do
23:29:22 <I440r> forth doesnt
23:29:24 <I440r> ans is not foreth
23:29:25 <I440r> forth
23:29:27 <I440r> period
23:29:29 <Herkamire> I440r: it's not just that he argues better. it's that he has experience and logic behind his opinions
23:29:54 <I440r> Herkamire, ive got over 20 years of coding forth
23:30:07 <I440r> erm over 15
23:30:07 <Herkamire> I440r: where as you're making shit up half the time.
23:30:08 <I440r> i lied
23:30:09 <I440r> lol
23:30:13 <Herkamire> I440r: you don't get to decide what forth is
23:30:34 <I440r> forth is what you make it. but ans is just horrible
23:30:56 <Herkamire> I440r: you regularly spout out things that are blatently wrong. like "the ans standard does NOTHING helpfull"
23:31:08 <I440r> im using ans forth at work and i spend 80% of my time fighting with it trying to do shit as simple as fetch from address a and store to addres b
23:31:15 <I440r> "invalid numerical inlut"
23:31:22 <I440r> address a is not valid in this blah blah blah
23:31:23 <Herkamire> I don't like the ans standard iether, but you can't go around saying it doesn't do anything useful.
23:31:31 <I440r> it doesnt
23:31:34 <I440r> not a fucking thing
23:31:49 <I440r> its a means of hurding you in like sheep. bahhhh!
23:32:03 <Herkamire> you can't prove that
23:32:04 <I440r> its like typing in c. 
23:32:09 <Herkamire> however, I can disprove it
23:32:20 <I440r> you cant store that char at that address because that address is reserved for BYTE data
23:32:41 <I440r> so you have to keep adding bullshit visual clutter red tape to your sources
23:32:51 <jdrake> how about colorForth
23:32:56 <I440r> just to get it to do simple fetches and stores
23:33:20 <Herkamire> ans is useful for sharing code across implementations and platforms. which is useful for learning if nothing else.
23:33:21 <I440r> i think isforth speaks for itself. or will when its documented
23:33:25 <I440r> no
23:33:28 <I440r> BULLSHIT
23:33:31 <I440r> absolute bullshit
23:33:42 <I440r> theres no such thing as portable codxe
23:33:57 <I440r> marcel hendrix has a library of benchmarks
23:33:59 <I440r> so do forth inc
23:34:04 <Herkamire> I440r: what chat client are you using?
23:34:04 <I440r> you can download both of them
23:34:16 <jdrake> I440r, go stick your head in the cold shower, you are a little hot
23:34:20 <I440r> every single file in there has "#if defined this system" conditional compiles
23:34:24 <I440r> you dont have ONE source fits all
23:34:33 <I440r> you have 389465289374658972435 sources all interleaved into the same files
23:34:35 <jdrake> CTCP VERSION reply from I440r: xchat 2.0.5 Linux 2.4.20-gentoo-r9 [i686/1789MHz]
23:34:36 <I440r> just like in c
23:34:42 <jdrake> that is what he runs
23:35:05 <Herkamire> xchat is portable code.
23:35:09 <I440r> no
23:35:11 <I440r> its not
23:35:25 <I440r> i can guarantee that there is conditional compile code scattered throughout the soruces
23:35:30 <jdrake> how can I compile xchat on any system that is unix?
23:35:34 <jdrake> I440r, making it portable
23:35:39 <I440r> next your going to tell me that the linux kernel sources are portable
23:35:45 <I440r> NO thats NOT portable code
23:35:56 <I440r> thats multiple versions of the same code for different platforms
23:35:57 <Herkamire> I440r: the linux kernel has been ported to 100s of platforms
23:36:00 <I440r> ALL platform specific
23:36:04 <Herkamire> you can't go around saying that it's not portable
23:36:10 <jdrake> I440r, that is simply incorrect
23:36:20 <I440r> no it fucking isnt
23:36:22 <Herkamire> I440r: YOU CAN'T MAKE UP YOUR OWN DEFINITIONS FOR WORDS
23:36:38 <I440r> how the fuck ncan you call soruces with conditional compil directives portable
23:36:42 <jdrake> there is a lot of code reuse between the ports, such as interrupt control or pci drivers, etc
23:36:43 <I440r> thats NOT one source fits all
23:36:50 <I440r> tahts one file with lots of specific sources
23:36:53 <I440r> and some common code
23:36:55 <jdrake> hmm, does anyone have /kick ability?
23:37:14 <I440r> jdrake, FUCK YOUJ i fucking own this channel
23:37:27 <Herkamire> From The Free On-line Dictionary of Computing (27 SEP 03) [foldoc]:
23:37:27 <Herkamire>   portability
23:37:27 <Herkamire>        
23:37:27 <Herkamire>           <operating system, programming> The ease with which a piece of
23:37:45 <Herkamire>           software (or {file format}) can be "ported", i.e. made to run
23:37:45 <Herkamire>           on a new {platform} and/or compile with a new {compiler}.
23:38:18 <I440r> ok. so i write an assembler function for an x86 and then write the same function for the ppc. i put them both in the same source file with conditionls
23:38:27 <I440r> is the x86 function portable ?
23:38:28 <I440r> NOI
23:38:29 <I440r> NO
23:38:48 <Herkamire> was it easy to make it run on PPC?
23:38:50 <I440r> there is NO such thing as portable code.
23:39:20 <Herkamire> I440r: ever heard logic vs sex?
23:39:21 <I440r> when you wirte an x86 emulator for the ppc them my x86 code is portable - maybe
23:39:34 <Herkamire> I have an x86 emulator on my pc
23:39:38 <Herkamire> my ppc
23:39:58 <jdrake> why is ans forth not 'ansi forth'
23:40:30 <I440r> herc freebsd has linux emulation. isforth runs under it. but isforth is NOT portable - take my sources and compile them under fbsd and they STILL wont run without linux emulation
23:40:39 <I440r> portability is a complete and utter MYTH
23:40:49 <I440r> "nearly portable" i will believe in
23:40:59 <Herkamire> I440r: you don't know what portable means
23:41:01 <I440r> but you have to massage the sources to port them
23:41:17 <Herkamire> that doesn't mean they are not portable
23:41:22 <I440r> : foo 10 0 do i . loop ;  <-- thats portable
23:41:41 <I440r> non trivial code is always NOT portable without some tweaking here and there
23:41:42 <Herkamire> I440r: ok, therefor NOT ALL code isn't portable
23:41:57 <Herkamire> I440r: you cannot prove that
23:41:58 <I440r> sure : noop ;
23:42:06 <I440r> thats fucking portable too but it doesnt do much usefull
23:42:10 <I440r> its TRIVIAL
23:42:12 <Herkamire> so
23:42:18 <Herkamire> it still makes you obviously wrong
23:42:30 <I440r> no. it makes you splitting hairs
23:42:38 <I440r> ill refine my above statements jsut for you
23:42:46 <Herkamire> if all it takes is a trivial and stupid example to prove that you are full of shit.....
23:42:50 <I440r> any NON trivial application is ALWAYS NON PORTABLE
23:43:00 <Herkamire> ok, still wrong
23:43:33 <Herkamire> programming is the art of piecing together trivial things into more complex and usefull programs
23:43:41 <I440r> ok. show me a non trivial application that is portable accross multiple operating systems and processors WITHOUT ANY conditional compilation crap
23:43:58 * I440r waits
23:43:59 <jdrake> I440r, explain how a console python application can run on a darwin system, or a linux system with no modification?
23:44:40 <I440r> can it ?
23:44:42 <I440r> always ?
23:44:46 <I440r> every time ?
23:45:07 <jdrake> actually yes it can
23:45:07 <Herkamire> http://www.campchaos.com/cartoons/napsterbad/napsterbad_56k.html
23:45:15 <jdrake> http://www.pygame.org/gamelets/  <-- pick one
23:45:32 <I440r> is  every single possible applicatition ALWAYS 100% portable without any massaging and conditioanls in python ?
23:46:05 <I440r> i seriously doubt it unless you are guanranteed to have exactly the same bugs in the interpreters on all platforms
23:46:10 <jdrake> not every possible, nearly every program
23:46:28 <I440r> oooh NEARLY!!! - ok ill concede that SOME languages are more portable than others
23:46:33 <I440r> but NONE are 100%
23:46:37 <I440r> forth definatly isnt
23:46:43 <I440r> but its more portable than c is
23:46:46 <jdrake> if you use the stuff available in the standard library then it always is
23:47:10 <Herkamire> I440r: you keep changing the subject
23:47:11 <I440r> even accross different forth standards its easier to port than it is to port c accross different compilers on the SAME platform
23:47:15 <I440r> no i dont
23:47:18 <Herkamire> we aren't talking about weather languages are portable
23:47:24 <I440r> they arent
23:47:29 <I440r> im still on that subject
23:47:37 <I440r> show me where i changed the subject
23:47:51 <Herkamire> you started talking about weather languages were portable just now
23:48:04 <I440r> YOU keep misdirecting the argument into other arguements not i
23:48:09 <Herkamire> before you were saying that code isn't portable
23:48:15 <I440r> i stand by my original statement
23:48:27 <I440r> and anything else ive said is to expand on that arguement
23:48:30 <I440r> NOT to change the subject
23:48:36 <I440r> which is what you are doing
23:48:37 <Herkamire> quote: 02:43 < I440r> any NON trivial application is ALWAYS NON PORTABLE
23:48:48 <I440r> true
23:49:28 <I440r> make that 99.9999999999999999999999999999999999999999999999999999999999% if all non trivial applications
23:49:33 <Herkamire> 02:46 < I440r> oooh NEARLY!!! - ok ill concede that SOME languages are more portable than others
23:49:36 <Herkamire> 02:47 < I440r> but NONE are 100%
23:49:39 <I440r> close enough to ALL that i could care less
23:50:01 <I440r> your the one switching and baiting not me.
23:50:01 <Herkamire> also you haven't really addressed the issue that you are making your own definition of the word "portable"
23:50:04 <I440r> your splitting hairs
23:50:22 <I440r> err portable is "compiles and runs identically"
23:50:30 <Herkamire> where does it say that.
23:50:32 <I440r> WITHOUT modification
23:50:48 <Herkamire> what basis do you have to assert that definition
23:50:49 <I440r> if you have a different definition - or if someone else has one then its wrong
23:50:57 <Herkamire> fuck you
23:51:02 <I440r> err. the english language ?
23:51:04 <Herkamire> speak english like rest of us
23:51:28 <Herkamire> remember when I printed out the definition of "portability" in the dictionary of online computing?
23:51:47 <jdrake> i might be back tomorrow when it is more civil, ngith
23:52:00 <Herkamire> goodnight jdrake
23:52:14 <jdrake> merci
23:52:45 <Herkamire> jdrake: sorry for the arguing.
23:53:16 <Herkamire> jdrake: I'm getting pretty sick of I440r's pigheaded stupidity. but perhaps I'm the same for trying to reason with him.
23:53:16 <fridge> haha, I love it
23:56:49 <fridge> python code is portable, python itself isnt'
23:57:08 <I440r> by your definition herk ALL code is portable because for ANY given source file you can write the same thing on a difrent platform
23:57:20 <I440r> so by "porting" it the original was portable
23:57:37 <I440r> you did NOT compile the original on the target, you evvectivly rewrote it
23:57:41 <I440r> thats NOT portable
23:58:24 <I440r> i can port almost any assembler application to almost any other processor
23:58:32 <I440r> MOST people would call assembler NON portable
23:58:44 <I440r> so whose definition of portable is wrong 
23:58:54 <jdrake> I440r, can't you understand the argument is finished and pointless (READ: DEAD)
23:59:11 <I440r> i say YOURS is. but i dont fucking call you a pinhead because i disagreee with you
23:59:46 <jdrake> you continually argue about something that nobody cares about anymore if they did to begin with
23:59:57 --- mode: ChanServ set +o I440r
23:59:59 --- log: ended forth/03.12.07