00:00:00 --- log: started forth/06.10.15
00:20:28 --- join: crest_ (n=crest@p54894B2F.dip.t-dialin.net) joined #forth
00:30:36 --- quit: Crest (Read error: 110 (Connection timed out))
00:36:41 --- quit: JasonWoof ("off to bed")
01:29:21 --- nick: crest_ -> Crest
01:45:35 --- quit: virl (Remote closed the connection)
01:58:17 --- quit: nighty_ (Read error: 104 (Connection reset by peer))
02:33:35 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #forth
02:48:25 --- join: Cheery (n=Cheery@a81-197-19-23.elisa-laajakaista.fi) joined #forth
03:55:15 --- join: vatic (n=chatzill@pool-162-84-156-148.ny5030.east.verizon.net) joined #forth
04:00:04 --- quit: Teratogen ("Leaving")
04:00:52 --- join: Teratogen (i=leontopo@slashsnot.org) joined #forth
04:17:05 --- join: zpg (n=user@user-54417e84.l6.c5.dsl.pol.co.uk) joined #forth
04:28:24 <zpg> afternoon
04:40:52 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth
04:41:14 <zpg> hi virl 
05:14:21 --- quit: virl (Read error: 110 (Connection timed out))
05:29:35 --- join: erider (n=erider@unaffiliated/erider) joined #forth
05:29:38 <erider> hi all 
05:34:15 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth
05:35:17 --- quit: k4jcw (Remote closed the connection)
05:36:10 <zpg> hey erider 
05:36:22 <erider> hi zpg
05:36:31 <zpg> how's it going?
05:36:47 --- join: k4jcw (n=jcw@adsl-065-006-151-062.sip.asm.bellsouth.net) joined #forth
05:46:06 <erider> good 
05:46:24 <erider> doing some exploration in lush 
05:46:40 <zpg> neat.
05:47:16 <erider> I'm looking at it for the lisp side of fisp 
05:48:12 <zpg> fisp?
05:49:58 <erider> new language that we are working on
05:50:36 <erider> forth + lisp and some features from other languages 
05:53:26 <virl> ehm? reminds me on factor
05:58:59 <zpg> got a website?
05:59:02 <zpg> and who's "we"?
06:00:10 <erider> yes
06:00:42 <erider> http://thin.bespin.org/keowiki//
06:02:32 <zpg> neat
06:02:44 <zpg> so you're doing postfix heads of lists in parentheses?
06:03:10 <zpg> or just every phrase has its own parenthesis?
06:03:16 <zpg> ==>  ( _ _ _ ?exit _ _ _ ?loop )
06:04:04 <erider> do you know C zpg
06:04:26 <zpg> yes
06:04:28 <erider> I can email you the source code 
06:04:35 <zpg> oh it's working?
06:05:08 <erider> yes some of it we still have to work on some things 
06:05:55 <zpg> neat stuff.
06:06:05 <zpg> just compiling something else at the moment, which has proven a nightmare so far.
06:06:12 <zpg> getting somewhere, finally...
06:06:16 <erider> I'm working on the compiler but the interactive mode and the interpreter works 
06:06:50 <zpg> neat stuff.
06:07:06 <erider> its going to be compiler interpreter and interactive 
06:08:08 <zpg> i don't like   "[]___] Step aside" sort of syntax. "()...)" ?
06:08:23 <zpg> it does, certainlty to a lisper, but i'm sure to everyone else too, like it's mismatched.
06:09:35 <zpg> erm, "it looks mismatched" is what that was supposed to say.
06:12:52 <erider> you can use [ ( { they are interchangeably 
06:16:19 <zpg> so "(]___)" ?
06:20:48 <erider> yes that could be one notation
06:21:45 <zpg> hmm ok
06:48:52 <virl> eh what?
07:08:18 <zpg> 'eh' to who?
07:08:34 <zpg> this is erider fisp, i was commenting that the unmatched paren syntax was confusing.
07:08:38 <zpg> *erider's
07:18:58 --- quit: erider (Read error: 110 (Connection timed out))
07:19:21 --- join: erider (n=erider@unaffiliated/erider) joined #forth
07:23:43 --- quit: k4jcw (Remote closed the connection)
07:24:27 --- join: k4jcw (n=jcw@adsl-065-006-151-062.sip.asm.bellsouth.net) joined #forth
07:24:57 <k4jcw> I wish like hell I knew why X or KDE forgets I have a keyboard.  The USB mouse is plugged into the keyboard, and it keeps working, so it's not a USB driver/hardware issue.
07:25:07 <k4jcw> I suspect modular X is still a little stupid, at times.
07:25:19 <Quartus> Never had it lose my keyboard. Suspect you have something else going on.
07:28:38 <k4jcw> Can't imagine what.  It's done it across a number of kernels.  I've replaced the keyboard.  And I don't remember this ever happening until after I switched to modular X.
07:29:10 <virl> X is always stupid, it's a kind of black magic
07:29:15 <k4jcw> 1;
07:29:25 <I440r> maybe the keyboard itself is flaky ?
07:30:02 <k4jcw> As I just said, I've replaced the keyboard.
07:30:45 <virl> config on one version of X enables usb and on another newer one version it doesn't detect it. 
07:30:45 <k4jcw> Oh, and as soon as I exit KDE, the keyboard works at the shell.  So it's not a kernel USB driver losing it.
07:31:57 <k4jcw> I didn't know that X was actually USB-aware.  
07:33:34 <I440r> could be kde fscking up hehe 
07:33:41 <k4jcw> Yes, it could well be.
07:34:00 <I440r> try using some other wm for a while to see if it still happens
07:34:10 <k4jcw> I'd trying running straight X, but it happens too infrequently.
07:39:30 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
07:40:15 <snowrichard> good morning
07:40:22 <Quartus> hey
07:40:47 <snowrichard> added a switch statement to my interpreter
07:41:14 <zpg> hi Quartus
07:41:23 <snowrichard> hi zpg
07:41:34 <zpg> hey snowrichard
07:42:04 <zpg> gtk+ nearly installed, but now i need cups
07:42:11 <zpg> (a newer version, presumably)
07:42:46 <zpg> then hopefully i can try gtk+ and gforth. the idea as it stands is this: try the demo suggested. if that works, try glade,  see how it produces code for gtk+. once that's done, perhaps write some wrappers for gtk+ widgets. then try writing gui apps in gforth.
07:45:33 <zpg> (and perhaps check out the retroforth code for gtk bindings, though i imagine this isn't directly compatible)
08:04:15 --- quit: snowrichard ("Leaving")
08:14:24 <vatic> hi Quartus: can you offer any general advice on structuring the termination of a WHILE loop on 2 conditional tests?
08:14:50 <vatic> Say GOAL is a cell address that would be found in a car of a list..
08:15:05 <vatic> and CLAUSES is the list being searched...
08:15:06 <Quartus> as in ....     begin  test1 while do-something  test2 while do-something-else  repeat  ?
08:15:37 <vatic> Oh, you can "cascade" WHILEs?
08:15:54 <Quartus> not quite like that.   begin   test1 while   test2 while   repeat then       is the actual form.
08:15:57 <vatic> like that?
08:16:18 <Quartus> LIke that.
08:16:52 <vatic> Is this mentioned somewhere, or something that an experienced programmer has picked up?
08:17:30 <Quartus> It's a natural feature of the way control structures are implemented.  They all work with dests and origs, so they can be assembled in various ways.
08:18:57 <Quartus> Here's a practical example:  : scan ( c-addr1 u1 char -- c-addr2 u2 )  >r  begin  dup while  over c@ r@ <> while  1/string  repeat then ;
08:19:00 <vatic> I think it'll do it. I'd want to terminate on CLAUSES cdr nil =   and   GOAL CLAUSES car =
08:19:05 <Quartus> "1 /string" of course.
08:19:11 <zpg> (howdy vatic)
08:19:20 <vatic> hi zpg!
08:19:30 <zpg> how's it going?
08:19:40 <vatic> ok
08:20:08 <zpg> good good.
08:20:13 <vatic> Quartus: thanks! I think I can take it from there, and will be on the lookout for these kinds of combinations...
08:20:18 <Quartus> Ok.
08:21:03 <vatic> Quartus: I got my search to work with a single while, but it would take the car of nil which wouldn't be too cool!
08:21:21 <Quartus> My car and cdr now return nil for nil.
08:21:38 <Quartus> : car dup if >car @ then ;
08:22:29 <vatic> Quartus: so I won't be blowing up my TX with 0 car? ;-)
08:22:38 <k4jcw> In C, we have structures.  What's the most idiomatic way of implementing structures in Forth?  Assume something like a structure that contains an int for an ID, an int for birth year, and 2 20 character fields for names.  Ideally, should a 5th field need to be added, it should be painless (in C, we'd simply add the field, and for any memsets or memcpy()'s, sizeof (foo_t) takes care of the change in size).  
08:22:39 <Quartus> No resets required.
08:22:44 <zpg> vatic: are you using Q's list code?
08:22:50 <zpg> vatic: or one of the other implementations?
08:22:53 <k4jcw> I can think of several ways to do this, of course, but I don't know which would be consider mo-better.
08:23:12 <k4jcw> Left out a word.  An array of structures.
08:23:19 <Quartus> k4jcw, there are some very simple approaches.  Commonest is a       struct{  1 cells field: foo  }struct my-struct    kind of thing.
08:23:20 <vatic> zpg: Quartus' code of course! It runs better on Quartus Forth! ;-)
08:23:30 <zpg> heh heh
08:23:48 <zpg> vatic: so you're hacking on a PDA?
08:24:01 <Quartus> k4jcw, my-struct winds up being a word that returns the size of the entire struct.
08:24:24 <k4jcw> And what do accesses to it look like?  Say foo [3].ID ?
08:24:47 <k4jcw> foo is my structure, not your field.
08:25:31 <Quartus> How could they look like that? This is Forth. Field names, like my field foo, are offset-adders. So    create instance my-struct allot ...  5 instance foo !
08:25:34 <vatic> zpg: got a used Palm keyboard and Quartus Forth is the bomb! Keyboard is a tremendous amplifier for a PDA...
08:25:46 <Quartus> The palm keyboard is a handy gadget.
08:25:59 <k4jcw> "Say" as in "in C, we'd do it like this"
08:26:08 <Quartus> ah ok.
08:26:11 <k4jcw> Obviously that's not a Forth syntax.
08:27:40 <k4jcw> Oh, apparently in that Vx+GPS auction I won, which includes the folding keyboard, he's also throwing in a "thumb keyboard", which I think is the little guy that plugs into the bottom of the Vx, and gives a small PDA-like keyboard.
08:27:48 --- quit: erider (Read error: 110 (Connection timed out))
08:27:57 <Quartus> Lots of toys.
08:28:12 --- join: erider (n=erider@unaffiliated/erider) joined #forth
08:28:50 <k4jcw> Is this struct{ thing defined in DPANS somewhere?  And/Or can it be implemented in Forth fairly easily?
08:28:57 <k4jcw> I'm thinking I'd like to add that to mine.
08:29:22 <Quartus> It's not a standard word.  It's very easy to implement.
08:30:14 <Quartus> : struct{  0 ;
08:30:34 <Quartus> : }struct constant ;
08:30:57 <Quartus> : field:  create  over , +  does>  @ + ;
08:31:50 <k4jcw> Hmm.  Looks like namespace collisions can occur between different structures field name, then.
08:32:12 <Quartus> Two ways to go; modules, or make the name fields unique.
08:33:25 <Quartus> When porting C structures across from Windows, I set up each .h equivalent in its own module, and expose them as needed.
08:34:00 <Quartus> C .h files tend not to re-use field names inside a particular .h.  Any conflicts require a slightly modified name, or a prefix.
08:34:20 <k4jcw> I like the modules approach better.  Having to uniquely name fields is kludgy.
08:34:26 <zpg> vatic: oh, neat. my keyboard should be arriving early this week. which model PDA are you using?
08:34:50 <zpg> (and i agreet re: QF (thanks Neal) -- the RsrcEdit stuff is pretty neat too, amazingly concise composition of GUI programmes)
08:34:54 <zpg> *agree
08:34:55 <vatic> zpg: I have a TX...
08:34:55 <Quartus> What I'm saying is that even in the modules approach, you *could* have a conflict.  Certainly I wouldn't make a module for each struct.
08:35:07 <Quartus> zpg, I'm glad it's going well.
08:35:12 <zpg> Quartus: yeah, enjoyable stuff.
08:35:59 <zpg> okay, the gtk+ moment of truth...
08:37:03 <zpg> woohoo!
08:39:17 <zpg> ==> ezil.sourceforge.net/screenshots/gtk-osx.jpg
08:39:55 <Quartus> what are we supposed to notice from that?
08:40:08 <zpg> sorry, that's my relief at having spent hours getting GTK+ to work on OSX.
08:40:16 <Quartus> ah
08:40:29 <zpg> so now, time to recompile gforth with ffcall support, then hopefully get a window on the screen....
08:40:55 <Quartus> k4jcw, you'd use align as appropriate to get aligned fields.
08:43:27 <vatic> Quartus: so the THEN as in REPEAT THEN provides resolution to the second WHILE's forward reference?
08:43:40 <Quartus> To the first one, in a two-while loop.
08:44:12 <Quartus> It's a stack, so REPEAT resolves the most recent WHILE, leaving the other one to be resolved by THEN, or ELSE THEN.
08:44:29 <vatic> Quartus: proceeding from inner to outer as in LIFO?
08:44:43 <vatic> or a stack...
08:45:05 <Quartus> Control-flow items go on a stack; it's almost always the data stack, in fact.
08:45:33 <vatic> during compilation, you're talking about?
08:45:37 <Quartus> Right.
08:45:44 <vatic> Ok, thanks!
08:47:46 <zpg> typical -- gforth won't compile out-of-the-box with the ffcall library installed.
08:48:35 <vatic> zpg: you'll need the CVS version for that. Had luck compiling for Debian. Didn't finish my struggle under OSX...
08:48:45 <zpg> hmm okay.
08:48:53 <zpg> apparently there's a long long problem on PPC OSX
08:49:06 <zpg> CVS is supposed to solve this?
08:49:44 <vatic> zpg: I don't have any probs with the release version on OSX, but the issues with compiling the CVS are magnified on OSX.
08:50:10 <vatic> check the gforth mailing list on chaossolutions and the Forth-Gesellschaft sites...
08:50:31 <zpg> okay. i'm trying 0.6.2 here incidentally. i haven't come across any information for checking out a cvs branch.
08:50:39 <zpg> it compiles without ffcall, doesn't with it.
08:50:46 <zpg> i found this thread -- http://www.mail-archive.com/gforth@chaossolutions.org/msg00346.html
08:50:59 <zpg> but there's no explicit workaround that i can see.
08:52:07 <vatic> zpg: dunno...
08:54:59 * zpg sighs
08:56:22 <vatic> zpg: did you check the release info at the tuwien CVS?
08:56:37 <zpg> haven't come across it... have you the link?
08:56:40 <vatic> that message was a while ago...
08:56:56 <zpg> yes indeed
08:57:10 <zpg> as i said, i may be missing an obvious link but haven't seen a cvs repository link for gforth
08:57:24 <zpg> http://www.jwdt.com/~paysan/gforth.html <-- just release info
08:57:34 <zpg> ah, http://www.mail-archive.com/gforth@chaossolutions.org/msg00437.html
08:58:02 <vatic> zpg: http://www.complang.tuwien.ac.at/forth/gforth/cvs-public/
08:58:16 <zpg> yeah thanks, the chaos... page linked there too :)
08:58:19 <vatic> zpg: googling "anton ertl gforth" got  me there... ;-)
08:58:31 <zpg> ah nice
08:58:33 <zpg> "gforth cvs"
08:58:35 <zpg> for me
08:59:58 <zpg> vatic: the TX looks really nice btw, would you recommend it?
09:00:04 <vatic> zpg: the install is cranky as I said above...
09:00:40 * zpg nods
09:00:45 <zpg> i'll try it and see.
09:01:00 --- quit: erider (Read error: 110 (Connection timed out))
09:01:06 <zpg> i can always post a message to the list to see what the fix is for getting gforth to compile on osx with ffcall
09:01:08 <vatic> zpg: as opposed to what?
09:01:19 <zpg> vatic: what opposed to what?
09:01:23 --- join: erider (n=erider@unaffiliated/erider) joined #forth
09:01:59 <vatic> what PDA are you using? (although I'm hardly an expert on PDAs...)
09:02:03 <zpg> m130
09:02:20 <zpg> aha!
09:02:33 <zpg> ./configure ac_cv_sizeof_long_long=0 <-- "If C's "long long" do not work properly on your machine (i.e., if the tests involving double-cell numbers fail), you can build Gforth such that it does not use "long long":"
09:03:37 <vatic> zpg: you might want to see if the release version has that config option before proceeding down the rabbit hole of CVS compilation for OSX...
09:03:55 <zpg> i just tried on the release version (0.6.2)
09:04:01 <zpg> but alas, still doesn't work.
09:17:48 <zpg> let's hope this works --> http://www.complang.tuwien.ac.at/forth/gforth/Known-problems.html, there's a patch for OSX
09:34:37 <zpg> ahaha!!
09:34:39 <zpg> http://ezil.sourceforge.net/screenshots/gtk-forth.jpg
09:35:26 <erider> looks like someone has a MAC
09:35:36 <k4jcw> So what's the Forth code look like?
09:35:37 <zpg> yes it does.
09:35:48 <zpg> it's taken from http://www.shockfamily.net/cedric/fflib.html
09:36:30 <zpg> here: http://ezil.sourceforge.net/devel/gtk-forth.html
09:36:56 <k4jcw> Cool.  Now you can write the bindings to get Sleepycat's Berkeley DB working under gforth :)
09:37:03 <zpg> hehe
09:37:13 <zpg> the code is pretty hard to follow as it stands, and i can't remember anything about gtk+
09:37:30 <zpg> but as the screenshot indicates, it's operational, and there isn't too much code to get this up and running.
09:37:43 <zpg> for starters, we wrap the libgtk calls into a header.
09:44:49 --- join: JasonWoof (n=jason@c-71-192-33-206.hsd1.ma.comcast.net) joined #forth
09:44:49 --- mode: ChanServ set +o JasonWoof
09:45:28 <zpg> hi JasonWoof 
09:48:21 --- join: Raystm2- (n=NanRay@adsl-68-93-41-121.dsl.rcsntx.swbell.net) joined #forth
09:49:26 <vatic> Hi Quartus, code is much easier to read also:
09:49:28 <vatic>    : find-clause ( goal @clauses1 -- goal @clauses2 )
09:49:29 <vatic>       begin 2 dup
09:49:31 <vatic>          dup nil <> while
09:49:32 <vatic>          car <> while
09:49:34 <vatic>       cdr repeat then 2drop ;
09:53:44 --- quit: Raystm2 (Read error: 104 (Connection reset by peer))
09:55:07 --- join: nighty__ (n=nighty@sushi.rural-networks.com) joined #forth
09:59:37 <Quartus_> good deal
10:02:58 <vatic> Quartus: does look OK for beginner code? :-)
10:03:02 <vatic> that
10:05:48 --- quit: nighty_ (Read error: 110 (Connection timed out))
10:05:56 <Quartus_> not too bad!
10:11:48 <vatic> Quartus_: wouldn't expect perfection at this stage! ;-) thnx...
10:13:09 <vatic> Quartus_: right now, I'm sort of working "middle->out" on this backtrack search, so I'm taking Pete Townsend's word for what params should be on the stack.
10:13:22 <JasonWoof> hi zpg 
10:13:23 <vatic> Might find out later that some other organization is better...
10:18:00 <JasonWoof> vatic: I don't understand the control constructs there
10:19:20 <vatic> JasonWoof: how so? it's supposed to test for cdr being not nil and a match between car and what's on the stack..
10:19:50 <JasonWoof> I don't know what happens when you put two "while"s in
10:19:54 <JasonWoof> or what the "then" does
10:20:24 <vatic> JasonWoof: talk to Quartus about that ;-) His suggestion!
10:20:29 <JasonWoof> hehe
10:20:43 <vatic> JasonWoof: the then resolves the forward ref of the first while...
10:20:55 <JasonWoof> oop, gotta go.
10:20:56 <JasonWoof> later
10:21:01 <vatic> bye!
11:17:31 <Quartus_> yes, the then resolves the first while.
11:19:19 <Quartus_> with an else then, you can have different post-loop code for each while.
11:25:53 --- quit: timlarson ("Leaving")
11:26:35 --- join: timlarson (n=timlarso@user-12l325b.cable.mindspring.com) joined #forth
11:35:08 --- nick: Raystm2- -> Raystm2
11:35:15 <zpg> hey ray
11:37:36 <vatic> Quartus: I think I need to know about ELSE THEN because my 2drop should happen when cdr dup nil <> proves false...
11:39:08 <Raystm2> Hello, men!
11:39:20 <vatic> begin code1 test1 while code2 test2 while code3 repeat post1 else post2 then
11:40:07 <vatic> Quartus_: so when test1 is false and the while terminates, it runs post2?
11:40:13 <zpg> Raystm2: howdy, man.
11:40:18 <Raystm2> :)
11:40:35 <lukeparrish> hiya guys
11:40:49 <vatic> Raystm2: hi!
11:44:07 <Raystm2> Hi vatic. hi lukeparrish.
11:48:59 <vatic> Quartus_: answered my own ?, thanks!
11:50:39 <vatic> still a little gun-shy from crashes... ;-)
11:51:54 <Quartus_> heh. Sorry, mobile. Not always looking.
11:52:35 <vatic> Quartus_: it was simple enough to just try, but I'm glad you mentioned it in the thread. I had overlooked a termination condition.
11:52:55 <vatic> wasn't sure where to go... ;-)
11:54:48 <vatic> Quartus_: is there a way to dump q4th history into a memo?
11:57:02 <zpg> Quartus_: I don't have the ability to 'see' definitions on the palm, do i?
11:57:52 <vatic> zpg: I don't think that's provided. RTFM! :-)
11:58:28 <zpg> heh
12:10:02 <Quartus_> zpg, yes -- include disasm
12:10:28 <Quartus_> vatic, there's no history dump; after a bit you should be able to code one up, though!
12:12:16 <vatic> Quartus_: true!
12:12:53 <vatic> Quartus_: the setup with peditPro is pretty nice...
12:14:06 <vatic> though.
12:30:45 <zpg> Quartus_: thanks.
12:30:51 <zpg> just wrote a bit of code on the Palm, very neat.
12:32:23 <vatic> Raystm2: do you run Windows?
12:40:02 <Raystm2> yes.
12:45:43 <vatic> Raystm2: OK, I have Windows CDs for your Vx...
12:46:47 <Raystm2> You do? vatic, I couldn't be more excited. It's really kind of you to do this for me. I've been dieing to try the Quartus Forth.
12:48:34 <vatic> Raystm2: strange thing is, I have no idea where it came from. With an extra cradle and USB connect kit...
12:51:28 <Raystm2> really, you mean, you didn't bye it?
12:51:37 <Raystm2> buy even.
12:52:24 <vatic> Raystm2: must've snuck in the front door when I wasn't looking, around 2000 ;-)
12:53:20 <Raystm2> hehehe. 
12:53:33 <Raystm2> What do you use now?
12:54:26 <k4jcw> Geoworks
12:56:05 <Raystm2> geoworks website is unavailable. :(
12:56:30 <vatic> k4jcw: NAPLPS...
12:56:58 <k4jcw> *snort*
12:57:20 <vatic> k4jcw: comes with my subscription to Times Mirror Videotext...
12:57:29 <vatic> Videotex...
12:58:33 <k4jcw> Why?
12:58:37 <vatic> k4jcw: did that Pocket Mirror work for you?
12:59:15 <k4jcw> Yes!  Quite well.  Thank you.
12:59:38 <vatic> k4jcw: cool! I'll scrap it off my disk now...
12:59:41 <vatic> scrape
12:59:55 <k4jcw> Seriously, why do you have a Times feed?
13:02:24 <vatic> k4jcw: Sorry, guess I went too far back in the wayback machine. mid-to-late 1970s Times Mirror had a Videotex service trial in Florida, along with Dean Witter and a bunch of others.
13:03:09 <k4jcw> I believe the Times is still running theirs, is it not?
13:03:16 <vatic> They used NAPLPS for the user interface. Can't imagine the data rate was better than 1200bps over the phone lines...
13:03:58 <vatic> k4jcw: They put it out of their misery in the very early 1980s. If they have something now, there isn't any continuity...
13:04:09 <k4jcw> I always wanted to play with that stuff, back then.  Wanted to build my own decoder. 
13:04:25 <vatic> k4jcw: there was software.
13:05:15 <k4jcw> Yea, but all the services were pay services, IIRC.  And I couldn't afford/couldn't justify it.
13:05:22 <vatic> I wrote a disassembler for it, never got the graphics routines all done...
13:06:09 <vatic> k4jcw: In the early 1980s there were a few BBS's running NAPLPS. Chariot in Colorado was one. Quite famous at the time.
13:06:35 <vatic> I hosted a lot of NAPLPS art on Compuserv's ArtSig also...
13:07:14 <I440r> a disassembler for what ?
13:07:21 <k4jcw> I didn't know about the BBSs.  I used to run one, back in the 80's.  Apple II, single 140K disk, all integer basic and assembly.  But never anything with NAPLPS.
13:07:32 <vatic> Toronto was a big hot spot for that kind of work too.
13:08:49 --- join: iano (n=iosgood@sub26-46.member.dsl-only.net) joined #forth
13:08:57 <vatic> k4jcw: Microstar software of Nepean, wrote a NAPLPS terminal program for IBM's online offering (I forget the name)
13:09:21 <vatic> so it was available free for you to hack, and you could buy it from Microstar for about $30...
13:09:29 <k4jcw> http://bbslist.textfiles.com/404/   Scroll down to 404-256-1549
13:10:58 <vatic> Hahaha! Very cool, JC! Brings back memories...
13:11:39 <k4jcw> Did you ever used to read Readers Digest?
13:11:57 <vatic> k4jcw: right Prodigy's give-away terminal program displayed NAPLPS and you could hack it to dial something other than Prodigy...
13:12:24 <vatic> k4jcw: no, not unless I was stuck at some Summer cottage...
13:13:57 <k4jcw> I don't know if you remember the "I am Joe's <organ>" stuff, but one of their more memorable one was a guy named Joe Trott.  Somehow he sustained some brain damage (birth or accident, I dont' remember).  Anyway, he was a friend of mine, and the guy I sold my Apple II to.  Unfortunately, I failed to keep a copy of my BBS software.  I have a hard copy listing of the BASIC portion from an older version, but none of the 3000 lines of un
13:13:58 <k4jcw> commented assembly.
13:15:21 <I440r> haha
13:15:28 <I440r> nice 6502 uncommented asm :)
13:15:44 <vatic> I am Joe's Man Gland!
13:29:41 <k4jcw> Jeez, you really know how to shut down a channel :)
13:30:40 <vatic> k4jcw: That's all I remember...
13:33:30 <zpg> hmm, given that <# takes an unsigned double, how would one go about prettily printing a triple (or, for that matter, printing one at all)?
13:34:06 <Quartus> zpg, you'd need to successively divide the triple by the base, and display the value.  <# and friends work on doubles, so you'd need to put together a new set.
13:34:42 <zpg> ah, i see
13:35:15 <zpg> so literally print one number at a time?
13:35:23 <zpg> as we work through the columns?
13:36:11 <Quartus> that's how <# #s #> works.
13:36:24 <zpg> neat, i'm not going to do this now, but i've just been playing with doubles on the palm.
13:36:26 <Quartus> Except the process returns your digits in reverse, so you have to manage that.
13:36:33 <zpg> 2. 32 exp 1 d- ud.
13:37:01 <zpg> yes, i'm aware of the <# reversal ... i wouldn't want to write a triple word that behaves any differently
13:37:03 <Quartus> what's '32 exp 1' ?
13:37:04 <zpg> would be confusing.
13:37:10 <zpg> sorry -->
13:37:17 <zpg> 2. 32 exp  1 d-  ud.
13:37:30 <zpg> (2^32) - 1, print it.
13:37:39 <zpg> since 2^32 requires a triple it seems.
13:37:42 <Quartus> 1 d- will do something odd, because 1 is not a double.
13:38:02 <zpg> oh sorry, my transcribing error, yes it should read 1.
13:38:07 <zpg> 2. 32 exp  1. d-  ud.
13:38:17 <zpg> (that caught me out when i wrote it too heh)
13:38:49 <Quartus> 2. 32 exp  would overflow 32 bits, it'd be 0.
13:39:05 <Quartus> Assuming exp does what it probably does.
13:40:18 <zpg> yes, the stack reads 0 0 afterwards.
13:40:56 <zpg> hence the one subtraction -- i was just checking the layout of a double in memory, and tinkering about.
13:42:39 <zpg> oddly,  1. 1. d-  ud.  still produces 0
13:42:48 <Quartus> Why is that odd?
13:43:00 <zpg> both "2. 32 exp" and "1. 1. d- ud." both leaving "0 0" on the stack
13:43:22 <Quartus> Sure.  2. 32 exp results in 2^32, the lower 32 bits of which are 0.  1. 1. d- ud.  subtracts 1. from 1., and leaves 0.
13:43:46 <zpg> oh i see what i'm doing wrong.
13:43:47 <zpg> duh
13:44:09 <zpg> 0. 1. d- ud.
13:44:16 <zpg> better :)
13:44:20 <Quartus> or more explicitly,  -1. ud.
13:44:27 <zpg> ah ye
13:44:31 <zpg> +s
13:44:53 <zpg> sorry, i've just been spending the evening with QF, doubles and the stack
13:45:11 <Quartus> there's lots of ways to generate 0. :)
13:45:35 <zpg> 0 dup 1 over d-
13:45:40 <zpg> :P
13:46:26 <I440r> that way probably wouldnt conform to the standard :)
13:46:28 <I440r> just a guess
13:46:36 <zpg> why?
13:46:59 <zpg> my understanding is that the least significant double cell sits below the most
13:47:05 <I440r> because it assumes the higher order word of the double is at top of stack. isnt that tied to the processors word order ?
13:47:29 <zpg> i thought the stack order of doubles was irrespective of endianness
13:47:33 <Quartus> It's not.  The high-cell of a double is always on top.
13:47:38 <zpg> cool
13:47:41 <Quartus> It has nothing to do with endianness.
13:47:47 <I440r> does the standard specify that ?
13:47:56 <zpg> so my silly example is still ANS?
13:48:00 <Quartus> Yes, as did FIG, Forth 79, and Forth 83.
13:49:36 <zpg> that was my understanding, hence the tinkering.
13:49:50 <zpg> btw -- i don't think i've asked but why are people actively anti-ANS?
13:49:57 --- join: Anbidian (i=never@S0106000fb09cff56.ed.shawcable.net) joined #forth
13:50:04 <zpg> i read a chap's site today saying in pretty puritanical terms that this was the case for him.
13:50:28 <Quartus> zpg, I really don't know.  I haven't met anyone who needs portability who speaks out against it; all the negativity comes from people who work in some variant to which they've become emotionally attached.
13:50:39 <zpg> i see.
13:51:05 <zpg> i thought your ANS layer for retro was pretty neat ... prevent too much fragmentation.
13:51:23 <I440r> mostly because i HATE most of their word names.  but i also dislike the thinking behind certain words
13:51:23 <Quartus> Really fun to write.
13:51:35 <I440r> i think "portabilty" is a complete non issue
13:51:38 <I440r> and an impossability
13:51:52 <Quartus> but you fit the profile, I440r; you don't write for more than one Forth; you write just for your own personal variant, to which you have become emotionally attached.
13:52:20 <zpg> I440r: hate .... what?
13:52:53 <I440r> show me portable ans sources and ill show you the bullshit visuall clutter DUCT TAPE  code  at the top of the source file
13:53:02 <I440r> not always. but its there most times
13:53:15 <I440r> #ifdefined this compiler #ifnotdefined this word blah blah blah
13:53:16 <Quartus> I am in the process of writing a substantial body of portable code, so I'll have the opportunity to show you that.
13:53:44 <I440r> quartus you will have to cater to the differences between all the ans compliant compilers
13:53:50 <I440r> is this word defined, if not define it
13:54:25 <Quartus> I'm managing that process all in one common include, but it's not to smooth differences -- it's to implement non-standard common-usage extensions.
13:54:54 <I440r> i still consider it duct tape code
13:54:57 <Quartus> And conditionals are only to avoid redefining those words if they're already there.  Not necessary, but more elegant.
13:55:21 <zpg> http://forthlinks.com/blog/ <-- new, usenet post
13:55:22 <I440r> i wont even support conditional compilation.  i would have but i saw the SHIT people do with it
13:55:59 <Quartus> I've hit my thumb with a hammer a few times, but I keep one around anyway.
13:56:04 <I440r> most sources i look at look no better than an UNMADE BED - i take great pains to keep my sources formatted neatly and CONSISTENTLY!!!!
13:56:18 <Quartus> yes, I think good formatting is key for any programming.
13:56:37 <I440r> most people dont format nicely
13:56:53 <I440r> or consistently either
13:56:55 <Quartus> Often they've never been shown good examples.
13:56:57 --- join: erider_ (n=erider@unaffiliated/erider) joined #forth
13:56:58 <I440r> mov <tab> eax, #1
13:57:07 <I440r>    push <tab> <tab>< eax
13:57:25 <I440r>  jmp <space>  foo
13:57:36 <I440r> no an excuse
13:57:37 <zpg> Quartus: do you have any special settings in memo pad. i mean, it's pretty usable, but perhaps fixed width font or turning off capitalisation post words ending in full-stop?
13:57:39 <I440r> lol
13:58:31 <Quartus> zpg, under OS4 and below I use ShiftHack for that (my own thing, it's free at my site).  I stay with the font that's there.
13:58:39 <zpg> k
13:58:52 <I440r> shifthack ?
13:59:17 <Quartus> Palm OS thing, stops the auto-capitalization the Palm has.  It's handy when you're writing normal text, but not so handy for Forth.
13:59:36 <zpg> Quartus being the font of all things good for palm, heh :)
13:59:40 <I440r> does os5 do that too ?
13:59:44 <Quartus> heh. I won't claim that. :)
13:59:56 <I440r> i mean. thats moronic and hacks arent possible now
14:00:00 <Quartus> OS5 has autocapitalization too.  There's a system extension to turn it off too, I don't recall the name right now.
14:00:47 --- quit: erider (Read error: 110 (Connection timed out))
14:03:57 <zpg> Quartus: Shifthack works well, thanks.
14:04:09 <Quartus> :) Glad it's still getting some use.
14:05:21 <zpg> heh
14:05:35 <zpg> the m130 is suddenly surprisingly useful.
14:06:15 <Quartus> Nearly everything I've written was in answer to some itch of my own that needed scratching.
14:08:13 --- join: erider__ (n=erider@nc-71-50-113-74.dyn.embarqhsd.net) joined #forth
14:08:37 --- nick: erider__ -> erider
14:15:06 --- quit: erider_ (Read error: 110 (Connection timed out))
14:16:57 <juri_> i like the sony clie peg20. ;)
14:18:07 <Quartus> The Sony palm devices were unfortunately slightly incompatible in certain ways with the mainstream Palm OS.
14:44:44 --- join: arke (n=Chris@pD9E049C4.dip.t-dialin.net) joined #forth
14:44:44 --- mode: ChanServ set +o arke
14:57:29 <I440r> why is that ?
14:57:36 <I440r> the sony devices LOOK better lol
14:58:28 <Quartus_> sony made some mods for their own reasons that weren't reflected on any other device.
15:00:35 <I440r> guess they had to make the product "theirs" lol
15:02:13 <Quartus_> they had collapsible input areas, rotating screens, different resolutions, but they hacked in add-ons for them instead of working toward a shared portable solution.
15:09:11 <Quartus_> in consequence it became extremely difficult to write a palm app that could access such facilities reliably on all pal devices.  A real-world example of the need for portability!
15:11:12 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/")
15:15:59 --- quit: arke (Read error: 104 (Connection reset by peer))
15:16:00 --- join: arke_ (n=Chris@pD9E049C4.dip.t-dialin.net) joined #forth
15:19:37 --- nick: arke_ -> arke
15:19:41 --- mode: ChanServ set +o arke
15:35:54 <Anbidian> Quartus: one problem with that example, Palm wanted portability and variance.  Now, if someone created a PDA with the philosophy of pursuing portability is confused, you might get a very different result.
15:36:25 <Quartus_> I'm afraid I don't follow.
15:36:57 <zpg> incidentally, i've been tinkering with [defined] and looking into find; now, "bl word" seems to push an address, whereas parse-word, word and of course s" myword" will return an address length pair. does find check for both, will it only work on an address...?
15:37:01 <Anbidian> You point to Palm as an example of the need for portability, but Palm itself endorses the philospohy of Portability.  
15:37:28 <Quartus_> right. The sony divergence from that stands out by contrast.
15:38:12 <Anbidian> Now, if Palm had not accepted portability as an approach, then there may have been very different results that *may* have been better (depending on your criteria).  So infact that may be an example of why portability is not needed.
15:39:05 <Quartus_> well, on the contrary. As a developer, it would have been marvellous to portably be able to handle collapsible input areas. Because of sony, we couldn't.
15:39:38 <Quartus_> there were some workarounds, but they were all clumsy.
15:41:11 <Anbidian> I get what you're saying.  I see how portability would have served Palm if all the manufactures were forced to work within certain confines or interfaces.
15:41:36 <Anbidian> I'm just asking what would have been possible on the other side too, if portability was rejected as a goal.  No PDA company has done this to my knowledge, so I can't be sure.
15:41:46 <Quartus_> sure. A common interface means your apps run everywhere the interface is valid.
15:42:48 <Quartus_> I'm not sure what rejecting portability means. If a developer has to re-write his app for every pda, well, how is that good for anybody?
15:43:03 <arke> It isn't, pure and simple
15:43:20 <arke> :)
15:43:24 * arke quiets down again
15:45:19 <Anbidian> Quartus: That wouldn't be the point.  If one rejected portability from the start, then you would save all the hassles of trying to be portable, and get easier, smaller, better apps (I am assuming).  How you managed variance would then become a different matter.  Either you had variance only on certain dimensions but kept all other parts identical, or you had different units with different OS/software.
15:45:55 <Quartus_> I'm sorry, Anbidian, perhaps it's me, but I'm not following you.
15:46:50 <I440r> write code for a specific system and your code will be alot tighter
15:46:51 <k4jcw> He's saying that apps would be smaller if you didn't have to worry about portability.  That there is overhead in allowing for portability.
15:47:05 <I440r> there is in almost every case
15:47:12 <Anbidian> Well, if one took the approach of rejecting portability then they would have to be careful how they approached variance: so either they would have little or no variance, or they would allow variance but have no portability.  Then you would create variance to a larger degree because it wouldn't make sense to rewrite apps for something roughly similar.
15:47:30 <Quartus_> there isn't. Palm apps are very small. Coding to a different but equivalent api would not make them smaller, just incompatible.
15:48:08 <Anbidian> Quartus: you're telling me there are zero sacrafices made on any level to keep different palm devices portable?
15:48:16 <Quartus_> perhaps you're using the words 'portable' and 'variance' in ways I don't understand.
15:49:36 <Quartus_> anbidian, with the exception of situations like the sony one, there are no compromises that come to mind at all.  Hmm.  When you build a bitmap that you want to display in b/w and colour, you have to provide both.
15:50:10 <Quartus_> well, 'have to' -- the b/w will display the color one, albeit awkwardly.
15:50:35 <Anbidian> different screen resolutions, different placement and quantity of buttons, different processing power, different interfaces, different fonts, etc. isn't there all of that and more?
15:51:26 <k4jcw> Are you assuming portability across all PDA OSs?  WinCE, PalmOS, etc?
15:51:46 <Quartus_> the screen size (modulo sony) is abstracted, as is the harware button interface. There are extra fonts beyond the base on some systems.
15:51:50 <k4jcw> Because portability is not something some of these people desire.  They want to lock you into their device because a particular app is available on it.
15:52:59 <Quartus_> I'm not talking about one app that runs on wince, symbian, psion, and palm. I'm just talking about Palm. Amazing portability is possible within that platform (again, excepting sony).
15:54:13 <Quartus_> you can write apps that run, binary-compatible, on the first USR Pilot and the latest OS5 Treo.  That's ten years of hardware.
15:54:15 <k4jcw> You're not.  I get the impression that Anbidian is.
15:56:55 <Anbidian> k4jcw: portability within a single platform, like PalmOS
15:57:41 <Anbidian> They now have different processors, from those motorala ones to the new ARM ones, so I don't get how you can say they haven't made any compromises for portability, what can be a larger compromise than running software on an emulator on a different processor?
15:57:58 <Quartus> The compatibility is impressive.  There's a common subset of the current API that runs back all the way.  Sony's divergence is a noteworthy exception to that.  Now, it may not be all Sony's fault, but wherever the blame may lay, it stands out sharply.
16:04:07 <Quartus> It is, of course, always possible to write an app that depends on a specific feature that is not present on all devices.
16:04:21 --- quit: virl (Remote closed the connection)
16:04:40 <arke> OT - anybody know of a Forth system for TI-89? In particular, ones that I can create a forth program in, have it compile, and run it from TI89-BASIC?
16:04:56 <arke> (well, the second feature isnt required but would be nice)
16:08:17 <Quartus> arke, google can't find one.  Just went through the possible hits.
16:09:27 <arke> thats why I asked ... googling for it is hard, since forth is a common english word, and programming is a common word on ti89 sites
16:09:46 <Quartus> "in forth" helps, or excluding -"so forth" -"back and forth" and a bunch of others.
16:11:51 <Quartus> This is the closest hit, but it's not a Forth: http://www.paxm.org/symbulator/download/rpn.html
16:20:05 --- quit: vatic (Read error: 60 (Operation timed out))
16:22:02 <arke> Thanks
16:22:13 <arke> an interesting self-fulfilling article I just found: http://www.embedded.com/story/OEG20010731S0028
16:22:32 <arke> (and no, I don't agree with the author, or at least not completely)
16:22:52 <Quartus> He makes quite a few flat-out errors in his article.  The responses below it are very good.
16:24:22 <arke> Yeah
16:24:24 <arke> like the comment thing
16:24:47 <I440r> did they EVER post an opposing view of forth to counter that article ?
16:25:00 <I440r> ive seen that before too
16:25:01 <Quartus> Right. This is not a guy who has ever used Forth.  He may have seen it once, or had it described to him by somebody else who was badly misinformed. 
16:25:05 <I440r> its totally bogus
16:25:25 <Quartus> I suppose it's faintly possible he used some amazingly fucked-up weekend-coder's attempt at a Forth.
16:25:31 <I440r> exactly. so someone should write a counter article for them and insist it be published
16:26:09 <Quartus> http://www.embedded.com/showArticle.jhtml?articleID=9900308
16:27:50 <arke> there it is, hehe
16:28:20 <Quartus> Notice it got no dissenting comments.
16:28:25 <I440r> :)
16:28:39 <I440r> i did see that. but then.. THAT article was written by a cluebie :)
16:29:45 <arke> HAH
16:29:46 <arke> thats a new word
16:29:50 <arke> "cluebie" :D
16:30:08 <Quartus> http://ganssle.com/
16:30:36 <Anbidian> arke, where did you find that article?
16:30:53 <Anbidian> I just found it today too
16:30:56 <arke> dont remember, somewhere searching for forths for calculators
16:31:05 <arke> i think it was on some forth links ring
16:31:12 <I440r> "Forth sucks?" Is that what passes for thoughtful analysis these days? (sigh)
16:31:14 <I440r> lol
16:31:20 <Quartus> He leaves that article out of his web-site. I wonder why.
16:31:37 <I440r> lol
16:31:44 <Anbidian> it's the interweb, anyone can publish anything ;)
16:32:27 <Quartus> The negative press he got from that article must have had some impact on even him.  I'm sure he'd like embedded.com to remove it.
16:32:54 <Anbidian> I put up a Forth blog today, though I may and try and get other people to contribute to it too because if it's just me it'll be too infrequently updated and have a limited area of focus
16:33:31 <I440r>  Please let me know when you'll be in Southern California, and I'll be happy to show you what modern Forths look like.
16:33:31 <I440r> Elizabeth D. Rather
16:33:31 <I440r> President
16:33:31 <I440r> FORTH, Inc. 
16:33:32 <I440r> lol
16:33:36 <I440r> Thwap!
16:34:43 <Quartus> Yes. I doubt the fellow ever bothered.  Why be informed?
16:37:20 <I440r> heh 
16:37:34 <I440r> im more qualified to hate forth than him :)
16:37:39 <I440r> lol
16:38:08 <I440r> eep! 
16:38:15 <I440r> ive never been "qualified" before :)
16:39:41 <arke> Since when do you hate Forth?
16:40:06 <arke> I thought you just hate C and are bringing the things you do like about C into isForth :)
16:40:30 <I440r> i dont have forth - i do dislike certain aspects of the ans standard tho
16:40:54 <I440r> theres not very much To like about c
16:41:55 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-151-054.pools.arcor-ip.net) joined #forth
16:43:39 <arke> I think C does have likeable aspects
16:43:43 <arke> and some very hateable aspects too
16:43:47 <arke> but you can say that about any language
16:43:58 <I440r> actually its not C i hate really
16:44:00 <I440r> its how its used
16:44:03 <arke> Which is why there is no one language to rule them all
16:44:08 <Quartus> If you're Jack Ganssle, you can say it without even knowing the language.
16:44:15 <I440r> hehe
16:44:24 <I440r> because he is smarter than us mere forth coders
16:44:26 <arke> you have to choose your tool based on availability, performance requirements, convenienve, and last but not least, preference.#
16:44:32 <arke> Hehe :)
16:44:44 <I440r> which is why i code in C when im on a contract
16:44:46 <I440r> like now :P
16:44:50 <Quartus> I think he's 'choosing his tool' a little too often.
16:45:32 --- quit: iano ()
16:45:40 <arke> I think the correct terminology here is "being a tool" :)
16:45:56 <I440r> i was just about to say taht lol
16:46:05 <Raystm2> that's funny.
16:46:07 --- join: slava (n=slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
16:46:08 --- mode: ChanServ set +o slava
16:46:17 <arke> hi ray
16:46:24 <I440r> hi slava
16:46:30 <slava> hi
16:46:32 <arke> VOODOO VOODOO RAY da daaaa du daaaa da daa
16:46:40 <Raystm2> hey Chris, vie ghets? ( is that even close? )
16:46:56 <arke> if you replace the v with a w and switch the h and e, its perfect :)
16:47:07 <Raystm2> ah okay wie gehts.
16:47:23 <arke> Gut, und dir?
16:47:30 <Raystm2> ~
16:47:49 <arke> :D
16:48:08 <Raystm2> hi slava.
16:48:25 --- join: zpg` (n=user@user-54478c30.wfd83b.dsl.pol.co.uk) joined #forth
16:49:06 <arke> sleep time for arke
16:49:07 <arke> good night
16:49:24 <Raystm2> g'night arke.
16:50:37 --- quit: Snoopy42 (Read error: 145 (Connection timed out))
16:51:02 --- nick: snoopy_1711 -> Snoopy42
16:51:41 <I440r> wish we had a disk defragmenter in linux that could defragment a fat32 file system
16:51:58 <Quartus> If only there was a programmer around.
16:52:11 <I440r> something like nortons dos defrag
16:52:33 <slava> no programmers, only luddites.
16:52:40 <Quartus> THINK!!!
16:52:43 <I440r> ?
16:52:53 <slava> I am the worlds best systems programmer!
16:53:03 <Quartus> You are all fools!!!
16:53:11 <I440r> yea and im the worlds best guitarist :)
16:53:34 <slava> I440r: we're making fun of werty from comp.lang.forth
16:53:44 <I440r> aha. 
16:53:53 <I440r> invite him in here :)
16:53:59 <Quartus> by simply quoting him, which is impressive.
16:55:10 <I440r> lol
17:00:18 <Quartus_> so write a fat32 defrag in forth, i440r.
17:01:13 --- quit: zpg (Read error: 110 (Connection timed out))
17:01:45 --- quit: timlarson (orwell.freenode.net irc.freenode.net)
17:02:00 --- join: timlarson (n=timlarso@user-12l325b.cable.mindspring.com) joined #forth
17:05:30 <erider> Who in here is a vim user?
17:07:26 <k4jcw> me
17:08:15 --- join: vatic (n=chatzill@pool-162-84-156-148.ny5030.east.verizon.net) joined #forth
17:11:10 <I440r> quartus i dont trust myself to not blow away my fat32 file system lol
17:11:35 <Quartus_> I suggest testing on a dummy one.
17:11:40 <I440r> heh
17:11:56 <I440r> actualy right now i dont have one because the drive decided to go belly up instead
17:11:59 <Anbidian> Welcome vatic.
17:12:04 <I440r> im rmaing it and will have to reinstall :/
17:13:12 <k4jcw> Must be a Seagate.
17:13:16 <I440r> no
17:13:20 <I440r> or... yes
17:13:22 <I440r> hehe 
17:13:34 <I440r> its a maxtor and seagate acquired them i belive
17:13:40 <I440r> maxtors SUCK
17:13:46 <k4jcw> Bah.  You are WRONG.
17:13:57 <I440r> ?
17:13:58 <k4jcw> Maxtors are good.  Seagate non-SCSI drives suck ass.
17:14:14 <I440r> ive had more maxtor failures than any other make of drive
17:14:20 <I440r> western digital are way better
17:14:48 <I440r> THESE two drives both died at exactly the same moment. i shut down, removed them. inserted them in a new box and...
17:14:53 <I440r> neither drive will even spin up now
17:14:58 <k4jcw> We seriously do not agree here.  I'd consider WD a step below Seagate, and Seagate a step below an IOmega zip drive.
17:15:11 <k4jcw> Sure you plugged in the power cables?
17:15:14 <I440r> ive had that happen many times before with maxtor drives
17:15:22 <I440r> posative
17:15:28 <I440r> and i didnt static zap them either
17:15:36 <k4jcw> Actually...
17:15:43 <I440r> nd zip drives are horrible. i think they are on the bottom of the list lol
17:15:43 <k4jcw> Are you sure the +12V is good on your power supply?
17:15:57 <I440r> none of the other drives are complaining
17:16:04 <k4jcw> ah.
17:16:14 <k4jcw> With both drives down, it kinda sounds like +12V is missing.
17:16:29 <I440r> its a 700 W drive
17:16:44 <k4jcw> ?
17:16:57 <I440r> erm its a 700w power supply i mean lol
17:16:59 <I440r> sorry heh
17:17:19 <k4jcw> Yea, but you *could* have a bad cable, or even a bad 12V section in the PS, was more my thinking.
17:17:20 <I440r> and ive swapped power cables between drives. i dont think its the power
17:17:28 <k4jcw> Yea, probably not then.
17:17:52 <k4jcw> Will they spin up without the data cable connected?
17:18:01 <I440r> no
17:20:28 <I440r> these are not my only bad experience with maxtor, almost every maxtor drive iev ever had has gone the way of the dodo while still in warrantee
17:20:44 <I440r> ive had some WD's do that too but far fewer
17:21:25 <k4jcw> I've had one Maxtor go bad in 6 years.  And I've got 23 drives spinning between various systems.
17:23:13 <I440r> this sytem actually just had a WD go down, a 200 gig sata that was a replacement for one i RMA'd a few weeks ago but this is VERY rare for wd's for me
17:23:29 <I440r> the replacement says "refurbished" im going to BITCH like a mofo tomorrow
17:23:40 <I440r> when i rma a brand new drive i want a freeking brand new replacement
17:23:46 <I440r> not someone elses hacked on rma
17:23:50 <k4jcw> It should be illegal to replace items covered by warranty with refurbished junk.
17:24:03 <I440r> ya
17:24:16 <I440r> luckilly that drive was my windows install and i have a backup of it on dvd
17:24:24 <I440r> somewhere
17:24:25 <I440r> lol
17:25:10 <k4jcw> I was tearing down some old boxens at work, and found one of those Seagate Bigfoot drives it.  Single platter 5.25" HD.  Don't remember the capacity.  I took the cordless drill and drilled a hole in it to make SURE no one could use it ever again.
17:25:24 <k4jcw> I swear those things had a 110% failure rate.
17:25:37 <slava> drilling that hole probably made it work better
17:26:56 <k4jcw> It's a bitch to drill through a platter, I'll tell you that.  Bits don't seem to want to bite into them very easily.
17:27:17 <I440r> its made of titanium glass i belive 
17:27:51 <k4jcw> That's mostly the 3.5" drives, IIRC.  The 5.25"s and old 3.5"s are hardened aluminum.
17:27:51 <I440r> they make nice mobiles.  hang a bunch of them from fishing wire outside your door
17:28:05 <I440r> they have a very pleasant sound when they collide :)
17:28:20 <I440r> aha
17:28:24 <I440r> junk :)
17:29:34 <k4jcw> I've started saving the magnets from the drives I destroy (older smaller drives.  Anything 40GB and under).  Those magnets are scary powerful.  You can really pinch the heck out of your finger.
17:44:56 <I440r> yea i collect those too heh
18:07:32 --- quit: Quartus_ (Read error: 60 (Operation timed out))
18:24:35 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #forth
18:33:39 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #forth
18:33:39 --- mode: ChanServ set +o Quartus_
18:38:20 --- quit: nighty__ (Read error: 110 (Connection timed out))
18:43:50 <Zarutian> say, is there an implmentation howto/guide for implementing Forth on new archs?
18:45:57 <Anbidian> Zarutian: one sec
18:46:13 <Anbidian> eForth was built around an easy to implement forth for new archs, IIRC
18:46:39 <Anbidian> " I wrote eForth so that it would be easier for me to develop Forth systems for many processors. Processors included 56000, 6502, 68332, 68HC11, 8051, 8080, 80C166, 80C196, 80x86, RTX2000, RTX2001, RTX2010, SC32, StrongARM, TMS320, Z80, etc"
18:46:51 <Quartus> well, sort of. It's a teaching system, not a practical one.  Brad Rodriguez' "Moving Forth" series, though a bit dated, would be more useful.
18:47:16 <Anbidian> see: http://www.baymoon.com/~bimu/forth/ and http://www.offete.com/eforth1.html
18:47:56 <Anbidian> moving forth: http://www.zetetics.com/bj/papers/moving1.htm
18:48:38 <Quartus> eForth may be simple to port (I've never tried it), but when you're done, you'd only have eForth, which is perhaps non-ideal for your needs.
18:49:05 <Anbidian> what do you feel eForth would be lacking?
18:49:18 <Quartus> efficiency, for one. It's a slow implementation.
18:49:54 <Anbidian> how slow is it?
18:50:50 <Quartus> Most of it is implemented in Forth source, and it's unoptimized, so it's slower than most.
18:51:17 <k4jcw> It also has no block or file storage facilities, at least last time I looked at it.
18:51:30 <Anbidian> so how does it store a program?
18:51:35 <k4jcw> You don't.
18:51:38 <k4jcw> You type it in.
18:51:45 <k4jcw> Or stream it in with a terminal program.
18:51:49 <Anbidian> there is an eForth2, as well, but Im not sure what the diff is
18:52:28 <Anbidian> "eForth model was designed for portability, and only 31 primitive FORTH
18:52:28 <Anbidian> words are machine dependent and must be coded for specific underlying
18:52:28 <Anbidian> CPU.  It is slow except for the very fast CPU's like 80x86.
18:52:46 <k4jcw> "very fast".  Heh.  Tells you how dated *that* is.
18:52:55 <Quartus> If it's a model of implementation you want, I would recommend MAF first and foremost.  If it's a guide, "Moving Forth". 
18:53:44 <Zarutian> Quartus: a url to MAF would be nice
18:54:01 <Anbidian> Zuartian, scroll up
18:54:09 <k4jcw> MILF
18:54:15 <k4jcw> Machines I'd Like to Forth :)
18:54:24 <Anbidian> ahahhahah, didn't see that coming
18:54:34 <Quartus> ftp://ftp.taygeta.com/pub/Forth/Applications/ANS/maf1v02.zip
18:55:18 <Zarutian> Quartus: danke
18:55:20 <Anbidian> oops, MAF not = Moving Forth
18:57:04 <Anbidian> Night all.
18:57:06 --- quit: Anbidian ()
19:01:20 --- quit: erider ("I don't sleep because sleep is the cousin of death!")
19:04:41 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
19:13:31 --- quit: vatic ("*poof*")
19:17:43 --- quit: snowrichard ("Leaving")
19:43:22 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
19:43:49 <snowrichard> hi
19:44:00 --- join: Bushmills (n=l@wpc3131.amenworld.com) joined #forth
19:44:04 <Quartus> hi Snoopy42
19:44:05 <Quartus> oops
19:44:06 <Quartus> hi snowrichard
19:44:13 <Quartus> who is Snoopy42 anyway? Always here but always silent.
19:44:18 <zpg`> evening guys.
19:44:27 <snowrichard> updated my interpreter release on freshmeat
19:44:29 <Quartus> hi zpg. Isn't it 4AM for you?
19:44:44 <snowrichard> v0.3 now
19:44:45 <zpg`> 3:45, yes
19:45:09 <zpg`> i've begun to notice that CMoore is pretty reticent when it comes to bringing 'foils' or laptops to lectures he's given.
19:46:40 <Quartus> I have one of his fairly recent chats, I think he had some visual aids.
19:48:31 <Quartus> snowrichard, what's new in it?
19:49:13 <zpg`> Quartus: oh, neat. the google video ones seem sparing.
19:49:26 <zpg`> this is intriguing -- not sure what they're talking about though: http://video.google.com/videoplay?docid=8026011931655804826&q=colorforth 
19:49:27 <Quartus> well, it's not a televisual feast or anything.
19:49:31 <zpg`> seems to be a forth environment with nice visuals
19:49:55 <zpg`> :r4color or something
19:50:44 <Quartus> http://www.geocities.com/redaforth/
19:51:45 <zpg`> ah, that must be it.
19:51:52 <Quartus> It is.
19:51:58 <zpg`> looks kinda smalltalky/logo-esque in those photos.
19:53:48 <Quartus> It's a crude Forth-ish thing.  The sources that are in the .zip have questionable Forth style.  Variables.  PICK.  Etc.
19:54:12 <zpg`> hmm ok
19:54:29 <snowrichard> a bug fix, and a switch statement
19:54:38 <snowrichard> some input ops
19:54:53 <Quartus> I see.
19:58:06 <snowrichard> the bug was get_opcode returning 0 in some cases
19:58:24 <snowrichard> when it should have defined a new dictionary entry
20:01:17 <Quartus_> bugs are eternal.
20:03:07 --- quit: snowrichard ("Leaving")
20:03:47 <zpg`> does 'vocabulary' see much use btw?
20:04:06 <I440r> i prefer that name over "word list" bleh
20:04:16 <zpg`> har har
20:05:00 <zpg`> no, i was watching Ribles Sudoku talk, which was actually pretty neat. he redefined + and - within a vocabulary so that he could use them to parse the sudoku table using the standard parser.
20:05:25 <Quartus_> Common-use vocabulary replaces the first wordlist in the search order. I've used it on occasion.
20:05:28 <zpg`> i.e., he just let normal parsing take place, and blanked out + and - where boxes met / intersected.
20:05:49 <zpg`> Quartus_: but i gather it's not standard stuff? i couldn't find it in QF for example.
20:06:07 <Quartus_> oh, ok. That's just using wordlists. The particular mechanism is not critical.
20:06:28 <Quartus_> there's a Vocabulary def in the appendix.
20:06:51 <zpg`> appendix to ...?
20:07:14 <Quartus_> standard.
20:07:19 <zpg`> ah ok.
20:07:53 <zpg`> also, i was looking into [defined] which is sitting in one of the QF memos. it uses word to scan from the input buffer, so that find can search for the given word.
20:07:58 <Quartus_> it's trivially defined on top of the standard wordlist stuff.
20:08:02 * zpg` nods
20:08:27 <zpg`> i was expecting s" dup" find to work too, then i noticed that find was expecting an address, not address length pair.
20:08:37 --- quit: Raystm2 ("Should have paid the bill.")
20:08:42 <zpg`> however, i also though WORD was being phased out in favour of PARSE/-WORD
20:08:43 <Quartus_> I find I use [undefined] is even more useful.
20:08:55 * zpg` nods
20:09:04 <zpg`> but my confusion came over the format returned by WORD
20:09:10 <zpg`> and the incompatibility with s" dup" find
20:09:17 <Quartus_> sure. Word returns a counted string that find needs, so they're good together.
20:09:48 <I440r> why do s" dup" find
20:09:51 <I440r> instead of ' dup
20:09:52 <I440r> ?
20:10:27 <zpg`> because if i do
20:10:29 <zpg`> ' qweiocj
20:10:32 <zpg`> it throws an error
20:10:45 <zpg`> i want to check whether qweiocj is defined.
20:11:11 <I440r> then use defined
20:11:18 <I440r> : defined bl word find ;
20:11:18 --- join: Raystm2 (n=NanRay@adsl-68-93-41-121.dsl.rcsntx.swbell.net) joined #forth
20:11:22 <I440r> no ?
20:11:25 <zpg`> yes
20:11:29 <zpg`> i looked at the definition
20:11:35 <zpg`> i was wondering why WORD was being employed.
20:11:41 <zpg`> not what to use.
20:11:51 <Quartus_> because of find's requirements.
20:12:08 <zpg`> okay.
20:12:19 --- part: Bushmills left #forth
20:12:26 <zpg`> i didn't know counted strings existed until this evening. i thought strings were starting address and length
20:12:32 <I440r> word USUALLY parses the next token delimited string out of the input stream and puts it at HERE
20:12:42 <I440r> find wants the string to be searched for at HERE
20:12:48 <zpg`> ahh
20:12:59 <Quartus_> modern strings are c-addr u, but counted strings crop up now and then.
20:13:25 <I440r> wtf is c-addr u ?
20:13:27 <zpg`> so how to counted strings work internally? where is the count held.
20:13:27 <I440r> oh
20:13:30 <I440r> address and length
20:13:32 <I440r> on the stack
20:13:41 <I440r> count converts a counted string into that format
20:14:02 <Quartus_> first char of the string, zpg.
20:14:25 <I440r> which limits your strings to 255 bytes :)
20:14:46 <zpg`> ahh, this is getting very interesting. thanks for clearing that up.
20:15:10 <zpg`> time to test this.
20:15:17 <Quartus_> counted strings are also known as pascal strings.
20:15:21 * zpg` nods
20:17:17 <I440r> how do you resize a partition in linux ?
20:17:27 <zpg`> cfdisk?
20:17:31 <zpg`> or fdisk
20:17:42 <slava> I440r: carefully
20:17:55 <I440r> thers no data on it
20:18:08 <I440r> i just want to shrink the first prtition and grow the second one down
20:18:13 <I440r> without having to delete either
20:18:22 <zpg`> okay, so count does pascal->addr u
20:18:33 <zpg`> is there a simple builtin for addr u -> pascal ?
20:18:42 <zpg`> or does this have to be written manually?
20:18:42 <I440r> drop 1-
20:19:01 * slava is writing a pastebin
20:19:03 <I440r> which will only work if the address actually points 1 char beyond the count byte of a counted string
20:19:08 <I440r> in forth ?
20:19:10 <I440r> hehe
20:19:10 <slava> and a web framework to go with it!
20:19:21 <zpg`> slava: forth or factor?
20:19:26 <Quartus_> c"foo" makes a counted string.
20:19:33 <zpg`> ahh ok
20:19:39 <slava> factor is a forth :)
20:20:04 <zpg`> the drop 1- example --> if addr u has been put on the stack ": histring s" hi" ;", we can't guarantee that 1- the address is going to contain the count can we?
20:20:23 <Quartus_> well... I'd say Forth is untyped.
20:20:59 <Quartus_> It isn't, in that example.
20:21:28 <slava> forth has types, but no type checks
20:21:37 <Quartus_> right.
20:21:58 <Quartus_> zpg, a s" string has no count byte.
20:22:08 <zpg`> indeed. so this drop 1- seems incorrect for conversion.
20:22:19 <Quartus_> right.
20:22:47 <Quartus_> PLACE is a common-isage word for making a counted string in memory.
20:22:49 <I440r> linus t doesnt like forth for that very reason... maybe he would like factor :)
20:24:46 <zpg`> Quartus_: thanks.
20:25:02 <zpg`> i read more into pfa, create ... does> today ... forth gets more and more intriguing :)
20:25:09 <zpg`> been writing a lot more in QF too.
20:25:18 <Quartus_> it does it by copying the string to a specified target.
20:26:37 <zpg`> yeah, i can see how it would be easy to allocate space for ((length) characters + 1), then copy the string into the 1+ offset and write the length to the first byte.
20:27:14 <Quartus_> that's the way.
20:27:21 <zpg`> :) 
20:30:16 <slava> i need to debug my xml parser
20:38:26 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
21:06:54 <zpg`> night all
21:06:57 --- quit: zpg` ("ERC Version 5.1.3 (IRC client for Emacs)")
21:30:28 --- quit: segher (Nick collision from services.)
21:30:35 --- join: segher_ (n=segher@dslb-084-056-136-068.pools.arcor-ip.net) joined #forth
21:45:52 --- quit: snowrichard ("Leaving")
21:46:42 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
21:48:51 --- quit: snowrichard (Client Quit)
22:05:01 --- quit: nighty_ ("Disappears in a puff of smoke")
22:46:21 --- quit: arke (Read error: 104 (Connection reset by peer))
22:48:15 --- quit: slava ()
22:49:01 --- join: arke (n=Chris@pD9E049C4.dip.t-dialin.net) joined #forth
22:49:01 --- mode: ChanServ set +o arke
23:18:50 --- join: snowrichard (n=richard@12.18.108.162) joined #forth
23:19:11 <snowrichard> hey
23:28:10 --- quit: arke (Read error: 110 (Connection timed out))
23:41:23 --- quit: snowrichard ("Leaving")
23:42:45 --- join: arke (i=Chris@x458.vpn.hrz.tu-darmstadt.de) joined #forth
23:42:45 --- mode: ChanServ set +o arke
23:49:50 <arke> good morning
23:59:59 --- log: ended forth/06.10.15