00:00:00 --- log: started forth/07.01.10
00:13:52 --- join: ecraven (n=nex@eutyche.swe.uni-linz.ac.at) joined #forth
00:46:23 --- join: Cheery (n=Cheery@a81-197-54-146.elisa-laajakaista.fi) joined #forth
01:30:41 --- join: crest_ (n=crest@p5489517C.dip.t-dialin.net) joined #forth
01:42:32 --- quit: Crest (Read error: 110 (Connection timed out))
04:23:34 --- log: started forth/07.01.10
04:23:34 --- join: clog (n=nef@bespin.org) joined #forth
04:23:34 --- topic: 'Welcome to #forth. We discuss the Forth programming language and a variety of technical subjects. Introduction: http://tinyurl.com/kvawv | Starting Forth: http://tinyurl.com/rm7pq | Thinking Forth: http://tinyurl.com/nsy4j | Gforth compiler: http://tinyurl.com/s8uho | ANS/ISO Forth Standard doc: http://tinyurl.com/nx7dx | http://quartus.net/search | Paste: http://forth.pastebin.ca'
04:23:34 --- topic: set by Quartus on [Sat Nov 25 22:45:21 2006]
04:23:34 --- names: list (clog arke_ crest_ Cheery ecraven I440r__ grub_booter X-Scale timlarson Snoopy42 virsys Zarutian Shain Baughn @crc nighty Raystm2 absentia ayrnieu madwork @Quartus @Quartus_ @slava TreyB madgarden warpzero cmeme ccfg)
04:55:55 --- nick: arke_ -> arke
04:56:02 --- mode: ChanServ set +o arke
05:38:44 --- join: zpg (n=user@81-179-127-214.dsl.pipex.com) joined #forth
05:56:03 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #forth
06:05:27 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth
06:05:36 <zpg> Afternoon (morning), Ray
06:08:22 <Ray_work> Good morning. :)
06:08:53 <Ray_work> What's going on this fine day?
06:09:33 <zpg> Not too much, though you're spot on about the weather.
06:09:52 <Ray_work> Excellent. Love a pretty day.
06:10:14 <Ray_work> Nice here too. Starting out a bit nippy, gonna be great.
06:11:13 <zpg> :)
06:15:53 --- quit: X-Scale ()
06:22:01 --- nick: Raystm2 -> nanstm
06:22:22 <Ray_work> there's my sweetie. :)
06:23:34 <zpg> heh.
06:58:57 --- quit: ecraven ("bbl")
07:24:37 --- quit: I440r__ ()
07:29:51 --- join: JasonWoof (n=jason@c-71-192-26-248.hsd1.ma.comcast.net) joined #forth
07:29:51 --- mode: ChanServ set +o JasonWoof
07:31:58 <zpg> hi Jason.
07:46:29 <Ray_work> hi JasonWoof.
07:46:40 <Quartus> Hey.
07:46:57 <Ray_work> busy morning already here at wheels-r-us. 
07:47:01 <Ray_work> Hey Quartus :)
07:47:54 <zpg> Q
07:48:36 <Ray_work> Quartus: are you the Q from StarTrek NG?
07:48:50 <Quartus> I hope not.
07:49:03 <zpg> Who's that? Perhaps it's telling, but "Q" in that context is David Llewelyn for me.
07:49:17 <Ray_work> Wait, so it is. :)
07:49:28 <Ray_work> Now it's Cleese.
07:49:46 <zpg> *Desmond Llewelyn, even.
07:49:52 <Ray_work> sure. :)
07:49:57 <JasonWoof> hi zpg, Ray_work
07:51:01 <zpg> howdy.
07:51:13 <JasonWoof> breakfast accomplished!
07:51:48 <Ray_work> Well done sir!
07:51:57 <Quartus> zpg, John Delancie playing a semi-omnipotent fop.
07:52:53 <Ray_work> And that's not you? /me ducks for cover. :)
07:53:04 <Ray_work> Hope your in a teaseable mood.
07:53:09 <Ray_work> </tease>
07:53:09 <zpg> Quartus: ah okay, his face looks vaguely familiar.
07:53:22 <Quartus> I hope not. :)
07:53:52 <zpg> Quartus: thanks again for the late morning conversation, interesting stuff.
07:54:44 <Quartus> Sure, no problem.
07:59:37 <Ray_work> Big plans for the day, gentilemen?
07:59:45 <zpg> Calling me a gentile?
07:59:53 <zpg> My rabbi will be contacting you shortly.
08:00:29 <Ray_work> cool. I always wanted physical contact with a rabbi, these Catholic preists have no imagination.
08:00:58 <zpg> I don't know, Catholicism seems to be pretty trippy in a Lovecraftian sort of way.
08:01:15 <Ray_work> trippy is right.
08:01:39 <zpg> I take it you're from a Catholic background then?
08:01:49 --- quit: ayrnieu (Connection timed out)
08:01:53 * Ray_work 's last name is St. Marie.
08:02:08 <zpg> That's a fair point :)
08:03:24 <Ray_work> I found out this summer that Dad droped the e from Ste. 
08:04:03 <Ray_work> I don't know why he did, but I have to assume that the army fatigues stencil came back incorrect and he just kept it from there. 
08:04:44 <zpg> heh.
08:05:24 <timlarson_> on a similar note, Larson was Larsson before the boat trip to the US.
08:06:56 <Ray_work> timlarson when did your people come over?
08:07:04 <zpg> from TUNES -- 'We foresee a "grand unified theory of programming languages" which will show how they're all the same underneath... Or in other words, we foresee a time when there'll be nothing new to discover regarding programming languages, and all this nonsense will finally come to an end...' -- anyone follow that hype?
08:07:13 <timlarson_> my grandparents came over
08:07:40 <Quartus> zpg, they just didn't know werty yet
08:07:42 <timlarson_> some to the US others to Canada
08:08:30 <zpg> Quartus: :) I'm trying to ascertain whether there's anything actually going on with TUNES.
08:08:57 <Quartus> zpg, based on that sentence, no.
08:09:09 <Ray_work> Thanks timlarson, much like my own family.
08:09:38 <zpg> To me it looks like a specification for vapo(u)rware, but I haven't delved into the project.
08:09:55 <Ray_work> Quartus TUNES, that's crc's friend that started retroforth, isn't it.
08:09:58 * Ray_work checks.
08:10:04 <Quartus> is it? I don't know.
08:10:37 <zpg> Yes..
08:10:47 <zpg> Retro was started as a TUNES Forth project.
08:11:05 <Ray_work> I'm wrong.
08:11:17 <Ray_work> Oh it was. okay hmm. 
08:11:36 <zpg> 'Tom Novelli's original "Retro" was a Forth-based OS and a prototype low-level infrastructure for the TUNES project.' ==> http://tunes.org/wiki/Retro
08:11:52 <zpg> we've discussed his conclusions before.
08:12:16 <Quartus> how grand that description. I'm not knocking RetroForth, but it's just a straightforward Forth implementation. :)
08:13:08 <zpg> Quartus: sure, again there seems to be a bit of hot air enveloping TUNES. but to be fair i think the term "prototype" clarifies the nature of early retro.
08:13:28 <Quartus> To my way of thinking, a prototype is for something different and new.
08:13:51 <Ray_work> ya, thanks zpg, I recall now.
08:13:58 <zpg> I get the impression with hobbyist projects that "different and new" means "different and new for me"
08:14:00 <JasonWoof> TUNES has some cool ideas
08:14:05 <JasonWoof> but I doubt they will be implemented
08:14:24 <zpg> And "different and new for me" is 100% relevant for learning. Depends what you're trying to do of course.
08:14:45 <Quartus> zpg, relevance to the individual notwithstanding, that's a grand description.
08:14:48 <zpg> JasonWoof: I read through some of the bullet points, reads quite nicely. Much like early Smalltalk etc., everything-as-object, no such thing as an application, etc.
08:15:12 <Ray_work> the retroforth direction has changed a few times since Tom had it.
08:15:49 <Quartus> yes, it has converged on accepted practice.
08:16:12 * zpg nods
08:16:51 <zpg> It would be interesting to take a look at the retro words and see how everything's named.
08:17:04 <zpg> (for me, obviously)
08:17:12 <Quartus> You can get a good sense of what's renamed by reading my standard layer.
08:17:21 <zpg> Indeed, that's what I had in mind.
08:17:35 <zpg> I think I have retro sitting on a ssh linux box.
08:17:51 <zpg> Does it generate standalones by the way?
08:17:58 <Quartus> I'll msg you the short list.
08:18:04 <zpg> Neat.
08:18:18 <JasonWoof> zpg: I don't think so
08:18:34 <Quartus> zpg, I think under Linux you can clone off the whole system under another name.
08:18:34 <zpg> Ok.
08:18:49 <zpg> Ah, combine interpreter and currently compiled code into a new binary?
08:19:14 <Quartus> Just dump the entire process into an elf binary.  I haven't tried it.
08:19:23 <zpg> Okay, presumably with an entry point.
08:20:59 <Quartus> RetroForth words that have different names in Standard Forth:  | r ::    RetroForth words that use a Standard name for another purpose: # >number \ exit find forth i repeat ?
08:21:11 <Quartus> >number and find are just slightly different.
08:21:44 <Quartus> That's pretty much it.
08:22:08 <JasonWoof> what's R?
08:22:12 <Quartus> R is R@
08:22:18 <JasonWoof> ahh
08:22:20 <zpg> Why all this kaffufle over ANS versus non-ANS then?
08:22:30 <Quartus> zpg, I sincerely don't know.
08:22:39 <Quartus> Largely virl-generated.
08:22:43 <zpg> heh.
08:23:04 <Quartus> oh, add ;; to the first list.  That's what retroforth calls exit
08:23:15 <JasonWoof> at first glance it seems to matter, because ans-compatibility means you can run a bunch of existing code
08:23:35 <JasonWoof> but then you realize that it wouldn't be that hard to write an ans-compatibility layer
08:23:59 <Quartus> oh, if that's what you mean -- compliance means code-sharing.  So it's a good thing.  But virl's dreamy maunderings about how pretty rf is, and how ugly the standard by contrast, I fail to see.
08:24:34 <Quartus> JasonWoof, it's possible to write a forth dialect wherein a standard layer would be painfully difficult.
08:24:42 <Quartus> RetroForth isn't such a dialect.
08:25:04 <zpg> As Marcel demonstrated though, you can implement your own dialect on top of ANS rather than instead of it.
08:25:35 <zpg> So if you really wanted, you could write a set of macros etc. to make ANS look more like rf, then just take that set of definitions to any ANS you want.
08:25:36 <Quartus> yes, and though I absolutely hate the arbitrary renamings of words that he does, it's obviously the lesser of the two evils.
08:25:49 <zpg> Agreed.
08:26:09 <Quartus> zpg, there's a bit more to it.  I had to change a couple of other things about rf, like not making names visible until after ;
08:26:40 <zpg> Yeah, I can imagine it's not a straightforward and simple task. 
08:26:46 <zpg> My point is q. a general one though.
08:27:23 <Quartus> I guess so.  Though I wouldn't hunt around for justifications for arbitrary word renamings; it might encourage them.  They drive me nuts.
08:28:15 <zpg> Well, a shared language is obviously more conducive to collective understanding and development.
08:28:28 <Quartus> pacman
08:28:31 <zpg> If you're going for the original Moore-method of rewriting everything from scratch each time.
08:28:38 <zpg> ... then I guess it's arbitrary.
08:28:56 <zpg> tetris  ok
08:29:07 <Quartus> quick game
08:29:43 <Quartus> I have to be extremely motivated to try and read Marcel's code; the arbitrary renamings are a hard obstacle.
08:32:39 <JasonWoof> sometimes it's hard not to make standard word names do something different
08:33:16 <JasonWoof> oh, wait for/next aren't in the ans spec are they?
08:33:45 <Quartus> No. Those are common-usage words that implement (usually a countdown) loop in the most efficient fashion.
08:33:55 <JasonWoof> right
08:34:06 <JasonWoof> but there's no way in hell I'm going to loop the wrong number of times like gforth does
08:34:07 <Quartus> There are about 4 variants.
08:34:43 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth
08:34:43 --- mode: ChanServ set +o tathi
08:35:46 <JasonWoof> : foo 1 for ." this gets printed twice in gforth. " next ; foo
08:36:08 <Quartus> Right, it's zero-indexed.
08:37:11 <Ray_work> hi tathi.
08:37:18 <tathi> hi Ray
08:37:18 <JasonWoof> what?
08:37:37 <zpg> 1 - 0 = 1 loop
08:37:44 <Quartus> The loop.  It counts down to 0.
08:37:51 <Ray_work> 0 is the first number, 1 is the second. the first number is called the index.
08:37:56 <JasonWoof> right. and how could it not get to 0 after one loop?
08:38:05 <Quartus> It does. It terminates after the 0 .
08:38:07 <JasonWoof> is it checking for -1?
08:38:19 <Quartus> It's checking for crossing the boundary between 0 and -1.
08:38:33 <JasonWoof> so it's checking for -1
08:38:38 <JasonWoof> I think it should check for 0
08:38:44 <Quartus> Some implementations do.
08:39:05 <JasonWoof> I do this to fix gforth, and make "unloop" work:
08:39:07 <JasonWoof> : for 0 postpone literal postpone ?do ; immediate  ( x -- ) loop x times (handles case where x=0)
08:39:11 <JasonWoof> : next postpone loop ; immediate
08:39:16 <tathi> presumably it's just using +loop, which the standard defines to check that way
08:39:21 <tathi> but it does read like crap
08:39:23 <Quartus> Some count from limit-1 to 0.  Some count from limit to 1.  Some count from limit-1 to 1.  Some allow access to the index via I; some don't.  Some count upward instead of downward.
08:39:59 <Quartus> Though I know why for/next isn't standardized, this does underscore the value of standardization. :)
08:40:01 <JasonWoof> I understand the effeciency of counting down
08:40:26 <Quartus> and also, some permit unloop/leave, some don't.
08:40:41 <zpg> you could just add something like ": for postpone 1- postpone for ; immediate" in gforth
08:40:54 <tathi> Quartus: because there were already conflicting FOR/NEXTs in existence, and implementors wouldn't agree to change theirs?
08:41:08 <JasonWoof> zpg: yeah, I did it with do/loop so I could use "unloop exit"
08:41:24 <zpg> i see.
08:41:25 <Quartus> tathi, in this specific case, not so much that as that for/next is supposed to be the 'fast' loop, so it's implemented whatever way makes it go zoom on a given platform.
08:41:47 <tathi> ah
08:42:07 <JasonWoof> so what platform was gforth implemented for?
08:42:18 <Quartus> The C abstract machine.
08:42:55 <JasonWoof> I have a hard time imagining that it's easier to check for -1 than 0 in C
08:43:23 <JasonWoof> enough easier that it's worth breaking readability
08:43:25 <Quartus> I didn't write it, but I'd guess they cast about looking at other for/next implementations, and found most of them went to 0.  They do, for whatever reasons.
08:43:43 <JasonWoof> you mean -1
08:44:03 <Quartus> No, I mean they go to zero.  Zero is one of the values the loop includes.
08:44:11 <Quartus> Stops after that.
08:44:37 <JasonWoof> so the problem is it's decrementing after the check instead of before
08:44:47 <tathi> looks like it's because they just reused the condition checker from LOOP and +LOOP
08:45:10 <Quartus> Maybe.  Most of the historical implementations do go to 0.
08:45:18 <Quartus> So 5 for i . next  counts 5 4 3 2 1 0
08:45:25 <Quartus> (that's where you're even allowed to use i)
08:45:52 <Quartus> The practical upshot is that for/next is not a portable structure.  The portable loop structure is do/loop and friends.
08:46:33 <Quartus> JasonWoof, I'd guess that most of ye olde implementations checked for overflow in the loop register.
08:46:39 <Quartus> er, underflow, I mean.
08:46:51 <tathi> funny that there are so many differences.
08:47:08 <tathi> every processor I've used has a decrement-count-register-and-branch-if-nonzero instructon
08:48:41 <Quartus> http://groups.google.com/group/comp.lang.forth/msg/110193257c606ca1
08:49:02 <JasonWoof> and most (every?) language I've learned has a loop (usually called FOR) which loops x times, not x+1 times
08:49:57 <JasonWoof> in some languages you can loop x+1 just as easily
08:49:59 <Quartus> Standard Forth doesn't have for/next at all.  What languages are you thinking of?  The ones that come to mind for me have a start and a limit.
08:50:08 <Quartus> C, BASIC, Pascal.
08:50:21 <Quartus> Forth has do/loop, again start and limit.
08:50:52 <JasonWoof> C, perl, php, python
08:50:58 <JasonWoof> ppc asm
08:51:12 <Quartus> C?
08:51:16 <JasonWoof> I think lisp, though I don't know it well
08:51:25 <JasonWoof> while(x--)
08:51:32 <Quartus> Looping in Lisp is a whole 'nother subject.
08:51:42 <zpg> heh.
08:51:50 <Quartus> Oh. I don't think of while as a counting loop.
08:52:06 <Quartus> In that case you have the equivalent in Standard Forth -- begin/until  or begin/while/repeat
08:52:15 <JasonWoof> well, that's the easiest way if you've got the count in a variable
08:52:30 <Quartus> sure, different animal to my eyes though.
08:53:16 <JasonWoof> you can do the standard, verbose thing: for(x = 0; x < 3; x++)
08:53:34 <Quartus> Right.  Which terminates on whatever condition you please.
08:53:56 <JasonWoof> to me these are somewhat analagus to for/next vs do/loop in forth
08:54:12 <JasonWoof> while(x--) is like "x for ... next"
08:54:12 <Quartus> Somewhat.  So is begin/while/repeat.
08:54:34 <JasonWoof> while do/loop/begin/while/repeat get you the functionality you get with for() in C
08:54:55 <Quartus> You can replicate while(<cond>) with begin/while/repeat or begin/until
08:55:56 <Quartus> If you look at the thread from the link I posted, you can see our very own ASau being a considerable knob. :)
08:58:38 <Quartus> He refused to accept that for/next was implemented differently in different Forths.
08:58:52 <Quartus> In some kind of bizarre way that involved changing his argument with each reply.
09:02:40 <Quartus> Man, that's really annoying just to re-read.  I'd forgotten what a complete tool he is.
09:02:47 <Quartus> Coffee time. :)
09:04:25 <Quartus> oh yes, that's the other minor distinction -- some rely on a short-branch displacement, thus limiting the size of the code inside the loop.
09:16:32 <JasonWoof> woh
09:16:36 <JasonWoof> optomization is tricky
09:17:58 <Quartus> <PoppaVic> "optimizing code" is a never ending battle that is ultimately futile.
09:18:12 <Quartus> Soak in that PV wisdom. :)
09:18:21 <zpg> Having a pleasant conversation with PoppaVic is a never ending battle that is ultimately futile.
09:18:38 <Quartus> I think PoppaVic is never-ending and ultimately futile.
09:18:44 * zpg smiles
09:18:57 <zpg> On the other hand, I hear he has a lovely smile.
09:19:03 <Quartus> ew.
09:19:24 <Quartus> <PoppaVic> "optimizing" is almost 100% mechanical - or it's conceptual - WORK WITH IT.
09:19:42 <Quartus> JasonWoof, is this helping? :)
09:22:36 <Quartus> <PoppaVic> stop pretending to "optimize" - write the API and then write the apropos asm for the apropos platforms for the apropos linking.
09:22:44 <Quartus> I can keep going, these are all verbatim. :)
09:22:57 <zpg> Feel free to show mercy at any time.
09:23:02 <Quartus> hehe
09:23:45 <grub_booter> well, umm... at a certain level and approach, that's description of optimisation is more or less true...
09:24:51 <grub_booter> there's only so much you want to do at the code level itself - it's a trade off between readability/maintainability and performance...
09:25:14 <Quartus> grub_booter, PoppaVic is a font of... what's the word that means 'the opposite of wisdom'?
09:25:24 <grub_booter> heh
09:25:29 <grub_booter> i know him of old
09:27:32 <grub_booter> but wrt the comment above, i think he's not far off the money - at least, the bits of the code you would want to optimise should be 'trivial' and hence 'mechanical' because to do it any other way would be madness - you don't want to write complex code at that level
09:27:34 <Quartus> and I have absolutely no clue what that blither about 'mechanical or conceptual' is supposed to mean.  I like the WORK WITH IT though.  Makes me think of werty.
09:28:00 <Quartus> THINK !!!
09:28:47 <Quartus> well, as regards C (##c is where those brilliant quotes come from), some people try to write code that they think will translate into better machine instructions, which is both silly and difficult.
09:29:02 <grub_booter> yup
09:30:19 <grub_booter> which is why optimisation isn't something you focus on - you focus on breaking things down into discrete components, profile, find slow components, and refine - optimise small things first (and last)
09:30:29 <Quartus> Right. JasonWoof is writing an optimizer, though. 
09:30:41 <grub_booter> ah
09:30:45 <grub_booter> automated?
09:30:55 <Quartus> Anyway optimization can be taken up a level; a better algorithm is the best optimization.
09:31:10 <Quartus> JasonWoof is optimizing generated code in his forth-like dealie, I believe.
09:32:07 <grub_booter> well, if it's profiling and pinpointing small things which need attention and providing a replacement... that'd be cool if it would work...
09:32:46 <Quartus> Forth optimization is usually pinhole based, eliminating/replacing instruction sequences.
09:32:54 <grub_booter> yup
09:33:05 <Quartus> Some systems go further.
09:33:49 <grub_booter> well, automating it involves making the process mechanical...
09:34:14 <grub_booter> so i think perhaps pv was pointing out something jw had already seen?
09:36:59 <Quartus> I copied those quotes in from somewhere else.  Humour only.
09:37:07 <JasonWoof> I haven't decided how I want to do optomization in fronds
09:37:09 <grub_booter> dunno - mebbe jw wasn't seeing that and pv was telling him (would explain the 'work with it' comment?)
09:37:33 <grub_booter> yeah - np - just found them curious
09:37:44 <JasonWoof> I have a bunch of stuff I want to play with
09:38:23 <grub_booter> ah - k
09:38:26 <JasonWoof> including a sort of profiler that automatically marks things as being appropriate for inlining
09:38:47 <grub_booter> that'd be cool...
09:38:54 <grub_booter> have you used the intel profiler at all?
09:39:02 <grub_booter> it does an amazing job...
09:39:50 <grub_booter> doesn't provide replacements, but it really shows where most of the time is spent within an execution of the app
09:39:50 <Quartus> grub_booter, those were things poppavic said to other people at other times in another channel
09:39:57 <grub_booter> ah :-)
09:40:12 <grub_booter> i was wondering if i picked up the wrong end of the stick there :-)
09:40:36 <grub_booter> but like i said, regardless of who it was to, it wasn't all wrong..
09:40:51 <Quartus> If it seemed right at all, I suspect it was accidental. :)
09:41:00 <grub_booter> heh
09:42:33 <grub_booter> well, i've released my command line/scriptable forth inspired video editor today
09:44:06 <grub_booter> (just as an easter egg - i don't think it'll catch on ;-))
09:46:02 <grub_booter> (well, not as a command line thing...)
09:49:37 <zpg> hi grub_booter -- good stuff.
09:49:45 <grub_booter> hey zpg
09:50:22 <grub_booter> thanks (it's just on linux for now though - have asked another guy to look at os/x - don't care about windows)
09:50:31 <zpg> link?
09:51:10 <grub_booter> have a build and installer for ubuntu... that's all at the moment though
09:51:51 <grub_booter> the tar ball source code hasn't been made available, though the code is.. it's all in cvs
09:52:04 <grub_booter> www.sf.net/projects/openlibraries
09:52:53 <grub_booter> the ubuntu link is: http://repo.jahshaka.org/ubuntu/jah-dapper-install.sh
09:53:29 <grub_booter> it's a kinda 'unofficial'/beta release - final for the end of the of the week
09:53:40 <zpg> ok.
09:55:34 <grub_booter> most useful command is: jbatch rpn.oml ~/Desktop/*.mpg quad (plays out 4 vids at once)
09:56:35 <grub_booter> or: jbatch rpn.oml ~/Desktop/*.mpg quad a ! quad b ! quad c ! quad d ! a b c d quad (plays out 16 :-))
09:57:46 <zpg> good stuff.
10:13:28 <grub_booter> http://users.pandora.be/acp/jbatch/16.jpg
10:14:23 <grub_booter> pornxxx and pamela anderson were both fakes btw :-)
10:14:53 <grub_booter> (you'll find them all on google video - not may good ones though)
10:15:32 <zpg> looks neat!
10:16:06 <grub_booter> should see the function def at the top there...
10:17:48 <zpg> colon definitions working nicely then?
10:18:14 <grub_booter> yeah - that one worries me though - worked first time, so there's bound to be a bug in there...
10:18:30 <zpg> :)
10:19:46 <grub_booter> hmm - i wonder...
10:20:26 <grub_booter> damn - was going to try encoding a flash (append: 'op:file.flv store' to the command line...)...
10:20:40 <grub_booter> but no diskspace left on my web site
10:20:42 <grub_booter> sigh
10:21:30 <zpg> perhaps abandon the fake porn?
10:21:40 <zpg> free up some space.
10:21:46 <grub_booter> heh
10:22:06 <zpg> you sound curiously unconvinced :)
10:22:29 <grub_booter> i'd put it up on the company repo server, but that might get me into trouble :-)
10:23:01 <zpg> there's a chance.
10:23:27 <grub_booter> $ ls -l output.flv
10:23:27 <grub_booter> -rw-r--r-- 1 charlie charlie 1860180 2007-01-10 19:18 output.flv
10:24:10 <grub_booter> not a bad size - looks a bit grainy though [low quality by default]
10:24:31 * zpg nods
10:24:34 <grub_booter> lasts 34 seconds
10:25:17 <grub_booter> .. could reduce the frame rate i guess..
10:25:24 <grub_booter> and res
10:40:31 <grub_booter> http://users.pandora.be/acp/jbatch/output.flv - it's slightly distorted cos i didn't correct the aspect ratio (sar should be 1:1 for flash - this vid is 59:54)
10:40:48 <grub_booter> just curious if it works...
10:41:03 <zpg> hmm firefox doesn't like FLVs it seems
10:41:42 <grub_booter> hmm - checking
10:42:37 <JasonWoof> found a quote I like: "I think imagination is rare, despite its alleged ubiquity."  -- Maggie Balistreri
10:43:17 <grub_booter> zpg: yeah - getting the same - trying to download it...
10:45:34 <Ray_work> flv windows IE7 doesn't like it either...
10:45:59 <grub_booter> i wonder if i'm hitting the same issue i had with the ftp the other day...
10:46:15 <zpg> given that i can play videos in youtube, i gather there ought to be a plugin for flv's kicking around?
10:46:15 <grub_booter> iyr, i couldn't upload the doc...
10:46:19 <JasonWoof> ooh, this one's better: "We keep our insignificant blemishes so that we can blame them for our larger defects."  -- Stephen Fry
10:46:33 <zpg> heh, to wit.
10:47:45 <Ray_work> I always wondered why they let me come back to work here, and thanks to Fry, I know.
10:48:06 * zpg chuckles
10:48:18 <grub_booter> heh
10:53:38 <TreyB> For kicks:  http://www.cabel.name/images-post/2007/01/atari-forth.jpg
10:54:48 <Ray_work> TreyB: You draw that?
10:55:19 <TreyB> Nope.  A friend pointed me at it.  I asked if the Forth came with the girl ;-)
10:55:31 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth
10:56:45 <Quartus_> I have that book.
10:56:53 <Ray_work> Hehe, You don't wanna start a relation ship with a cartoon girl that is minus atleast on dimention. Please, take advice from someone who was hurt by WonderWoman.
10:58:54 <JasonWoof> Ray_work: huh?
10:59:10 <JasonWoof> 2d?
10:59:28 <Ray_work> Hehe, You don't wanna start a relationship with a cartoon girl that is minus at least one dimention. Please, take advice from someone who was hurt by WonderWoman.
10:59:33 <Ray_work> there I think that's better.
10:59:59 <JasonWoof> here you mean WonderWoman literally
11:00:02 <JasonWoof> I'm catching up
11:00:20 <Ray_work> Jason, I've known some one dimentional women. :)
11:00:28 <Ray_work> Maybe we all do.
11:02:02 <JasonWoof> I've never met anybody that shalow :)
11:02:20 <JasonWoof> everybody's got some depth
11:02:55 <Ray_work> Don't go into men's clubs much, do you. :)
11:03:06 * Ray_work neither do I honey
11:03:14 <Ray_work> Damn I know shes logging.
11:03:59 <JasonWoof> no
11:06:17 <Ray_work> I suppose the shallowness of a woman at work in a place like that is totally reciprical to the amount of money she can earn.
11:07:05 <JasonWoof> you think the shallow ones make more or less?
11:07:31 <Ray_work> Definately more!
11:07:40 <Quartus_> the poor drawing on that book cover can't even afford a whole shirt.
11:09:19 <Ray_work> Ya, and the girl can't either.
11:09:35 <Quartus_> tragic!
11:09:43 <JasonWoof> clearly fantacy. check out the shlong too
11:10:35 <JasonWoof> I don't think shallowness is really my thing, so I don't think I'd be drawn to it so much
11:10:44 <JasonWoof> but I'm probably not the typical clientelle at a strip club
11:11:00 <Ray_work> ya.
11:15:58 <TreyB> An old boss went to a strip club one evening and his wife (who didn't have a problem with it) asked if came back drunk and horny.  He replied "I am NOT drunk.".
11:19:28 <Ray_work> If Nan minds, she's never made a big deal about it.
11:20:21 <Quartus_> just so long as you keep paying her. :)
11:22:43 <Quartus_> Geez, tough room
11:23:28 <TreyB> :-)
11:24:54 <Ray_work> hahahah.
11:25:00 <Ray_work> expensive is not the word.
11:29:40 <Quartus> I should send you guys my wireless bill, I mean, this is comedy gold I'm laying down here. :)
11:29:50 <Quartus> This is the A material. :)
11:30:36 --- quit: Quartus_ ("used jmIrc")
11:32:30 <Quartus> From ##c:  <PoppaVic> velco: you know? in all these years? I've never really seen "erotic talk" online, and even less in a bedroom.. Weird.
11:32:32 <Quartus> Oh, what we're missing.
11:32:54 <Ray_work> man. hehe
11:33:35 <Ray_work> makes a big hairy man like you wanna /join ##c and whisper sweet nothings.
11:33:54 <Quartus> Whisper? SCREAM!
11:38:29 <JasonWoof> I wonder if you can get pv to shut up irl
11:38:52 <JasonWoof> that's generally a prerequisite to hearing any kind of talk
11:38:52 <Quartus> I'd settle for online.
11:39:21 <JasonWoof> on irc there's /ban, irl there's pointy sticks :)
11:39:30 <Quartus> heh
11:39:41 <zpg> irl?
11:39:45 <TreyB> In Real Life
11:42:33 <zpg> ah.
11:49:30 <grub_booter> zpg: http://users.pandora.be/acp/jbatch/output.swf - any better?
11:50:29 <grub_booter> (sorry about the sound - will fix up later..)
11:50:54 <grub_booter> (it will resample and mix, but you have to tell it to do so...)
11:51:18 <Quartus> dude, what are you, the Man Who Fell To Earth? :)
11:52:11 <grub_booter> heh
11:53:29 <zpg> grub_booter: yep, that works here .
11:53:45 <grub_booter> cool
11:53:46 <Quartus> is it in some way useful to play 16 video clips at the same time?
11:53:49 <grub_booter> thanks :-)
11:54:01 <grub_booter> Quartus: dunno - the beeb asked me for 4
11:54:07 <grub_booter> if i can do 4, i can do 16
11:54:11 <grub_booter> ;-)
11:54:22 <Quartus> I guess if it's one of those commercials where you briefly show a whack of screens all playing bits of different programs.
11:54:45 --- join: arke_ (n=chris@pD9E04C29.dip.t-dialin.net) joined #forth
11:54:47 <grub_booter> yeah - or one of those mosaic channels on sat - pulling in live feeds from lots of channels
11:55:06 <grub_booter> there's also elements of 24 in there... 
11:55:27 <grub_booter> (the show that is) - though obviously, they don't tend to hit 16...
11:55:45 <grub_booter> max out at about 3 or 4
11:56:11 <grub_booter> anyway - talking of which - i'm going back to the couch to watch some more of series 5 - ttyl
11:56:47 <grub_booter> and thanks for the tests :-)
12:11:51 --- quit: arke (Connection timed out)
12:13:23 --- quit: Shain (Remote closed the connection)
12:16:46 <Quartus> don't forget the Brady Bunch opening sequence
12:18:11 <slava> Quartus: FFT is 14 lines of factor code
12:18:45 <Quartus> what? No FFT primative in the factor kernel? :)
12:19:19 <Ray_work> any thing that can be written in 14 lines can be writen in 13. THINK!!!
12:19:33 <Quartus> YOU ARE ALL FOOLS !!!
12:19:44 <Ray_work> I AM ALL FOOLS!!!
12:22:49 <JasonWoof> what's fft again?
12:22:56 <JasonWoof> something like finite feline transform?
12:23:12 <slava> fast fourier transform
12:23:19 <JasonWoof> right!
12:23:21 <JasonWoof> I was close
12:23:23 <Quartus> http://en.wikipedia.org/wiki/Fast_Fourier_transform
12:23:36 * JasonWoof goggles
12:25:08 <Quartus> there are four methods shown here as part of the FSL: http://www.taygeta.com/fsl/library/dfourier.seq
12:25:11 <Quartus> I suspect they can be factored better.
12:26:21 <Quartus> What method are you using in Factor, slava?
12:26:29 <slava> it was contributed
12:26:32 <slava> i know nothing about fft
12:26:37 <slava> http://factorcode.org/repos/Factor/demos/fft.factor
12:26:59 <slava> notice how it has no loops, its all vector math, apl style :)
12:27:23 <Quartus> it's a natural for vectors
12:28:04 <Quartus> That 14-line block of code would need at least 14 comments for me to follow it.
12:28:28 <slava> or knowing what the words do
12:28:45 <Quartus> Even then. No information about the method used, how the output is ordered, any limitations, etc.
12:29:24 <zpg> what are [ and ] in factor, slava?
12:29:31 <slava> code blocks which are pushed on the stack
12:29:45 <slava> [ 2 2 + ] call  is like  2 2 +
12:29:53 <slava> and you can pass code blocks to other words which implement iteration, etc
12:30:03 <Quartus> ...an example of use, little things that I look for like that :)
12:30:06 <zpg> same syntax as smalltalk's blocks then?
12:30:12 <zpg> ifTrue: [ ... ] ifFalse: [ ... ]
12:33:00 <JasonWoof> except backwards
12:33:17 <zpg> oh sure, i'm just talking about the concept
12:33:55 <Quartus> just syntax markers denoting 'here a block begins' and 'there it ends'.  
12:34:09 <Quartus> Kind of embedded :noname definitions
12:34:31 <Quartus> retroforth has something akin to it, [[ and ]]
12:34:58 <Quartus> so        : foo 42 . [[ 17 3 * ]] execute . ;   foo -> 42 51
12:36:01 <JasonWoof> that's a nifty feature
12:36:18 <Quartus> In Forth, I think it runs counter to factoring, so I'm not sure I like it.  I don't know about Factor, perhaps it's necessary there.
12:36:53 <JasonWoof> yeah... I'm not sure if I'd want that feature, or if I'd rather just define the thing and pass the xt
12:37:13 <Quartus> It facilitates unnamed embedded factors.  I'm not sure that's good.
12:37:45 <JasonWoof> eg:  : inner-bit 17 3 * ;   : foo 42 . ['] inner-bit execute . ;   foo -> 42 51
12:38:20 <slava> they're not like xt's
12:38:46 <JasonWoof> slava: no?
12:38:49 <Quartus> Right.      rf has t/f also, so      : foo [[ 42 . ]] [[ 23 . ]] t/f ;  true foo -> 42  false foo -> 23
12:39:04 <slava> JasonWoof: they're actually just lists
12:39:11 <slava> you can take two code blocks, append them together, pass the result to a word
12:39:23 <slava> also, i don't think you'd want to factor out each branch of a conditional, like this
12:39:28 <slava> ['] foo ['] bar if
12:39:33 <slava> you'd write [ ... ] [ ... ] if
12:39:40 <JasonWoof> no
12:39:43 <slava> of cousre if each bbranch is long it can be factored but that should be up to the programmer
12:39:45 <Quartus> if that's the only way to express a conditional.  As I say, it may make more sense in Factor.
12:39:51 <JasonWoof> I'd write:  : foo .. if bar exit then baz ;
12:39:52 <slava> that's the only way, yes
12:39:56 <slava> sure
12:40:02 <slava> so in forth you might not need code blocks like this
12:40:16 <Quartus> that's akin to the t/f thing in rf, which I personally find hideous.
12:40:18 <slava> another thing is that you can pass a literal in a code block, like [ 123 ]
12:40:19 <JasonWoof> slava: exactly
12:40:23 <slava> you can't get an 'xt' of a literal
12:40:35 <Quartus> well, you can.     42 constant x  ' x 
12:40:44 <Quartus> or   :noname 23 ;
12:40:45 <JasonWoof> sure you can
12:41:02 <JasonWoof> oh, I guess you can't do   ' 23
12:41:06 <Quartus> no.
12:41:21 <JasonWoof> you can in my latest forth-likes though
12:41:39 <Quartus> I mean, you could extend ' so that if it can't find the name, it wraps it in a noname and sends that back.  Why you'd want to I don't know.
12:41:45 <JasonWoof> in fronds there's no difference between a constant and a literal
12:41:58 <slava> i sometimes have conditionals where one branch just pushes a literal
12:42:17 <JasonWoof> there might be a dictionary flag or something that says how it was created, I forget. If so I don't check it anywhere
12:42:43 <JasonWoof> I'm picky about syntax
12:42:51 <JasonWoof> I don't like:  [ ... ] [ ... ] if
12:43:10 <slava> 'if' is no more special than 'each', 'map', etc in factor
12:43:16 <slava> its just another word
12:43:23 <JasonWoof> I know
12:43:31 <slava> well, its a primitive, and the compiler turns it into an efficient branch, but that's detail
12:43:35 <JasonWoof> it's special in my forths :)
12:43:50 <slava> i don't use syntax for control structures because code blocks allow more flexibility
12:44:08 <slava> also in factor the parser doesn't compile code
12:44:11 <slava> it builds a tree
12:44:11 <JasonWoof> sure, but I still don't like them :)
12:44:25 <slava> [ ... ] [ ... ] if  parses as an array with 3 elements
12:44:57 <JasonWoof> yeah, I'm my "compiler/parser" thing goes from an array, to vm instructions
12:45:35 <slava> factor code can be manipulated and analyzed with array words
12:45:37 <JasonWoof> slava: right. you compile it later?
12:46:06 <slava> i code after parsing in most cases, however keeping the syntax tree around makes 'see' possible to implement
12:46:12 <slava> and things like 'usages'
12:46:29 <JasonWoof> you code after parsing? what's that mean?
12:46:36 <slava> compile after parsing, sorry
12:46:40 <JasonWoof> ahh
12:46:45 <slava> the compiler takes a syntax tree and produces machine code, basically
12:46:55 <JasonWoof> immediately after? or you wait until someone wants to execute it?
12:47:05 <JasonWoof> right
12:47:22 <slava> eg, after a source file is loaded, all new definitions are compiled
12:47:29 <JasonWoof> I guess mine does too
12:47:42 <JasonWoof> my fronds sources are essentially a tree
12:48:00 <slava> later i'll have a tree editor instead of a text editor
12:48:01 <JasonWoof> it's just that each node is a definition
12:48:04 <slava> using text for import/export
12:48:07 <JasonWoof> you cannot define more than one node on a line
12:48:17 <JasonWoof> that's where I'm headed
12:48:21 <JasonWoof> editor is unfinished though
12:48:36 <JasonWoof> you can browse and delete, but not insert
12:48:44 <slava> heh, same here
12:48:51 <JasonWoof> :)
12:48:54 <Quartus> ok, so in Gforth:  : '  >in @  ['] ' catch if  >in ! :noname  parse-word evaluate  postpone ;  else nip then  ;
12:49:01 <Quartus> So you can  ' 23 execute . -> 23
12:49:06 <Quartus> not sure what good it is :)
12:49:26 <slava> Quartus: the thing is that in factor [ ... ] is literal syntax for a data structure
12:49:33 <JasonWoof> Quartus: nice :)
12:49:35 <slava> [ 1 2 3 ] [ 2 2 2 ] v+ .
12:49:38 <slava>   [ 3 4 5 ]
12:49:41 <Quartus> slava, sure.
12:50:08 <slava> [ 2 ] [ 2 + ] append call .
12:50:08 <slava>   4
12:50:19 <Quartus> of course,    ' 23 . ' 23 . -> 418108224 418108248   different xt each time.  
12:50:22 <Quartus> Minor detail. :)
12:50:45 <Quartus> slava, I see.
12:50:45 <JasonWoof> Quartus: that's because you did :noname instead of :
12:51:07 <Quartus> JasonWoof, right.  
12:51:27 <Quartus> I have initial uneasy feelings about making named words for each integer.
12:51:52 <JasonWoof> :)
12:52:01 <JasonWoof> don't many forths do that for low numbers?
12:52:10 <JasonWoof> ie 0 constant 0 1 constant 1 2 constant 2 etc
12:52:15 <Quartus> 0 and 1, sometimes.  The code representations for them are smaller than the literal sequence.
12:52:23 <JasonWoof> oh
12:52:36 <JasonWoof> I thought it was so the parser didn't have to search whe entire dictionary for them
12:52:50 <Quartus> Well, I suppose in a really crude implementation that might be a factor too
12:53:10 <Quartus> cf, say.
12:53:12 <Quartus> :)
12:53:14 <JasonWoof> in fronds all numbers are in the dictionary
12:53:21 <JasonWoof> they don't have definitions though, just values
12:53:39 <JasonWoof> constants can have definitions, but obviously numbers like 23 don't need them
12:53:41 <Quartus> well, in Standard Forth all numbers are in the dictionary, too, in the broadest sense.
12:54:05 <JasonWoof> how's that?
12:54:16 <Quartus> Numbers are acceptable input.
12:54:32 <JasonWoof> I don't look at it that way
12:54:36 <Quartus> They are parsed and something happens when they're encountered.  So in that sense they're tacitly in there.
12:54:40 <JasonWoof> input can be a word or a number
12:54:53 <JasonWoof> that's how I describe the interpreter
12:55:00 <Quartus> Sure, that's also a valid description.
12:55:25 <Quartus> I drew up a BNF for Forth -- specifically for a Forth with no words in its dictionary.
12:55:29 <Quartus> It parses three kinds of numbers.
12:55:36 <Quartus> And identifies non-numeric sequences.
12:55:59 <JasonWoof> can it deal with creating words?
12:56:13 <Quartus> The BNF is for a Forth with no words at all in its dictionary.
12:56:24 <Quartus> No : no CREATE no CONSTANT no :nonane no nuttin'.
12:56:37 <Quartus> Adding words to Forth changes its parser.
12:57:55 <Quartus> Re your interpreter description, something like : 23 42 ; throws a spanner in the works.
12:58:13 <JasonWoof> heh
12:58:29 <JasonWoof> suppose that's ok though, since that would screw up humans too
12:58:32 <Quartus> Now you have to say input can be names, numbers, or numbers that are actually names.
12:58:47 <Quartus> Well, also       : 23 23 ;     2 base !  23 . -> 23
12:59:08 <JasonWoof> no...
12:59:22 <JasonWoof> prints 10111
12:59:44 <Quartus> right, sorry.  The value is accepted as input, however.  It wouldn't be if it weren't defined as a named word.
12:59:57 <JasonWoof> right
13:00:24 <Quartus> So it gets a little on the harder-to-completely-describe side.
13:01:43 <Quartus> for that matter, it's not just when the value is invalid in the given base.   decimal : 23 23 ;  hex 23 . -> 17
13:01:55 <tathi> er.  only if you insist on using "describe" in a BNF sort of way
13:02:08 <Quartus> No, even if you're just saying that input is 'names or numbers'.
13:02:18 <Quartus> Names can be numeric.
13:02:41 <Quartus> Numbers can be alphanumeric.  How you name words affects it.  What base you're in affects it.
13:02:48 <tathi> right
13:02:57 <tathi> I don't see that that invalidates the description though
13:03:14 <Quartus> It doesn't invalidate it, it's just that there are a series of sub-paragraphs if you want to fully explain the action of the thing.
13:03:50 <tathi> oh.  sure.
13:04:08 <Quartus> The BNF I wrote covers a dictionary with no words, because as soon as you add, say, a parsing word to the dictionary, the BNF description of the system changes.
13:06:10 <Quartus> I wrote it to ascertain exactly what behaviour the parsing of various numbers would have in different bases, and if there were any differences in such when relaxing the minimum Standard parsing requirements.
13:06:58 <Quartus> Rather than just messing with Gforth and assuming their implementation was canonical, which it slightly isn't.
13:08:05 <tathi> non-canonical as in having extra functionality?
13:08:42 --- join: ramserver (n=gschuett@4.38.41.141) joined #forth
13:08:42 --- mode: ChanServ set +b *!*@4.38.41.*
13:08:42 --- part: ramserver left #forth
13:08:49 <Quartus> IT WORKS! YAY!
13:08:51 <Quartus> That was gavino!
13:09:04 <tathi> hehe
13:09:04 <zpg> :)
13:09:29 <Quartus> where was I.  As in Gforth-specific relaxings of the rules. My notes on the differences are on another system slightly inaccessible at this moment, let me see if I can fiddle around and remember where the differences were.
13:09:52 <Quartus> boy, that ban dealie pleases me no end.
13:10:27 <tathi> just curious
13:10:46 <Quartus> Well, one minor point, Gforth accepts floats while in bases other than decimal, but only if they have an embedded sign in the exponent value.
13:10:54 <Quartus> There are other variances.
13:11:03 <tathi> ah
13:11:41 <Quartus> Nothing radical.  But little things that could change the strict interpretation of numbers in some conditions.  So I wanted a canonical model to compare against.
13:12:18 <Quartus> single-cell vs. double-cell vs. float in varying bases.
13:13:20 <JasonWoof> icky sticky
13:15:00 <Quartus> Stuff you'd probably never run into, and if you were worried you'd just follow the strict rules.  Good to know, though.
13:17:37 --- join: Jules_ (i=Jules@cp550544-a.landg1.lb.home.nl) joined #forth
13:20:06 <Quartus> Hi Jules_.
13:20:17 <Jules_> hi
13:21:16 <Quartus> New to #forth?
13:21:20 <Jules_> yes
13:21:29 <Quartus> Thought perhaps.
13:21:34 <zpg> welcome Jules_
13:21:46 <Jules_> thanks
13:30:08 <grub_booter> http://users.pandora.be/acp/jbatch/output.swf - final for now (family vids - no audio, but man, flash scales poorly... playing that with ffmpeg based stuff is much better)
13:30:31 <Quartus> it was in such lo-res last time I think it was family-safe then too :)
13:30:36 --- join: timlarson__ (n=timlarso@65.116.199.19) joined #forth
13:30:36 <zpg> heh.
13:30:39 <zpg> hi tim.
13:31:04 <grub_booter> heh - probably :-)
13:31:56 <Quartus> think you've got a ground-loop somewhere :)
13:32:17 <grub_booter> blame the monkey behind the cam :-)
13:32:31 <Quartus> unless it's part of the artistic effect to make me feel like I'm sitting too close to the projector
13:32:41 <grub_booter> heh
13:32:43 <Quartus> the audio, I mean
13:33:14 --- quit: timlarson__ (Client Quit)
13:33:15 <grub_booter> oh - umm - there's not supposed to be any audio (thought you were meaning the wobbly cam work..)
13:33:19 <slava> Quartus: you know? in all these years? I've never really seen "erotic talk" online, and even less in a bedroom.. Weird.
13:33:41 <Quartus> He went on to explain how familiar he was with penises, and how detailed pictures of hairy arms appeal to him.
13:33:42 <TreyB> Aieee!
13:33:48 <slava> lol
13:33:54 <Quartus> TreyB, that's PoppaVic from ##c, today.
13:34:02 <slava> is he married today?
13:34:06 <TreyB> Yeah, I know.  Then slava quoted it back.
13:34:06 <Quartus> I don't ask.
13:34:22 <zpg> and Jules_ looks perplexed.
13:34:34 <zpg> (one would assume)
13:34:39 <slava> just as well PoppaVic doesn't come here anymore.
13:34:53 <Quartus> Yes.
13:35:53 <Quartus> slava, I see you have your own personal antagonist over on reddit.
13:35:54 <Quartus> gwenhwyfaer 
13:36:07 <slava> no, you mean grauenwulf
13:36:21 <Quartus> oh. Something inanely Norse like that.
13:36:34 <slava> he's not my personal antagonist. he posts trash in every single thread
13:36:49 <Quartus> Well, they seldom stay focused on only one thread.
13:38:04 <slava> in a lisp thread, he lamented that lisp programmers don't innovate like microsoft does by adding direct literal XML syntax to the language.
13:38:09 <slava> or something like that.
13:38:33 <Quartus> At least he was talking about programming languages.  He's not in PoppaVic's league. :)
13:38:37 <slava> in vb you can write crap like  x = <foo><bar baz='1234'/></foo>
13:38:49 <slava> and x becomes a dom node
13:41:03 <JasonWoof> oh god
13:43:37 <Quartus> that's curious. What's the benefit?
13:43:54 <slava> easier to write code that spits out lots of xml, i guess
13:44:03 <slava> you can interpolate variable values in the xml literal
13:44:08 <slava> horrible thing to hard-code in your language, though
13:44:11 <Quartus> Yes.
13:44:22 <slava> sun is thinking of doing the same in java
13:44:40 <Quartus> Well, that's unfortunate.
13:46:57 <TreyB> Does anyone know how to temporarily disable one CPU on a dual-core WinXP machine?
13:47:09 <grub_booter> i've been thinking about xml a bit.. instead of having many nested levels of elements, just have a completely flat document which requires an rpn type of deserialisation process :-)
13:47:27 <slava> what's the point of that?
13:48:17 <grub_booter> point? none.. i just find it.. interesting :-)
13:48:25 <slava> its been dnoe
13:48:43 <grub_booter> ah - so not just me then :)
13:50:13 --- quit: timlarson_ (Connection timed out)
13:52:37 <grub_booter> i guess the main advantage is that the objects being reconstructed are forced to be simpler - they take 'inner' objects from the stack as the deserialisation occurs - don't think it restricts anything (... at least, not in my domain..)
13:54:18 <slava> recursive descent parsers are not that hard
13:54:26 <Quartus> That's true.
13:55:14 <grub_booter> true - but it's not just deserialise - during a serialise, each object just push their inner objects on to the serialising stack and then dump their own state...
13:55:33 <slava> recursive descent printers are not hard either :)
13:55:37 <grub_booter> heh
13:55:46 <grub_booter> but neither is this :-)
13:55:57 <slava> but its not human readaable
13:56:16 <grub_booter> heh - why not?
14:02:05 <grub_booter> i don't see any reason why it wouldn't be considered human readable - at least, no more so than an n-th degree nested xml doc.. the only advantage with the nesting is the collapsible nature of the deep nested element (in a gui editor type of way)
14:04:17 <Quartus> what are you talking about?  instead of <foo>...</foo> you'd do...?
14:04:28 <slava> ... 3 foo
14:04:34 <slava> where 3 is the number of items in ...
14:04:50 <slava> i'd simply prefer (foo ... )
14:05:08 <slava> a lisp parser is only a dozen or so lines of code in factor
14:05:20 <Quartus> Which is syntactically similar to <foo> ... </>
14:05:21 <slava> or you can use factor's own parser, but that's somewhat unsafe because your data could execute arbitrary code at parse time
14:05:32 --- quit: zpg ("back in a bit")
14:05:34 --- quit: Jules_ ("Ik ga weg")
14:05:42 <grub_booter> well, depends on the objects being serialised - i was thinking about video filter graphs.. so each element would be either a video or a filter
14:06:48 <grub_booter> then you would have: <video blah=/><filter name=something args=... /><filter blah=/> 
14:07:12 <grub_booter> and from a human point of view, inserting or removing filters would be trivial
14:07:30 <grub_booter> admittedly, i wasn't thinking about use in the grander scheme of things :-D
14:07:56 <Quartus> THINK !!!
14:08:04 <grub_booter> heh - FOCUS
14:08:06 <grub_booter> ;-)
14:09:13 <grub_booter> the typical way of doing that involves an increasingly complex nesting
14:09:51 <Quartus> so you're just talking about tagging.
14:11:18 <grub_booter> hmm? no - storing/sending filter graphs as xml between processes and to/from files for recovery
14:12:55 <Quartus> yeaah, I mean in your markup above. Looks like you're associating a name with a video and a set of filters.
14:14:48 <grub_booter> more like <video uri="video.avi"> <filter type="black_and_white"> etc
14:15:12 <Quartus> so coupled by proximity, or what?
14:15:21 <grub_booter> order, yup
14:16:14 <Quartus> so you could as easily do <video uri="video.avi" filter="black_and_white"> etc.
14:16:17 <grub_booter> when it parses, it pushes video.avi on the stack, then the filter applies by grabbing the object on the stack and attaching itself to it and pushing itself back
14:16:55 <grub_booter> ah - sorry - me being terse - yeah, you could - but some of these filter have many arguments...
14:18:22 <grub_booter> so like: <video uri="video1.avi"> <video uri="video2.avi"> <filter type=composite x=0.5 y=0.5 w=0.5 h=0.5> [for example]
14:18:41 <Quartus> at which point the < and > are simply there to make people think you've got some lovely structured format. :)
14:18:58 <grub_booter> and it is :-)
14:19:36 <Quartus> so go full-hog.    "video1.avi" video   "video2.avi" video  composite 0.5 x 0.5 y 0.5 w 0.5 h filter
14:19:44 <grub_booter> yup
14:20:08 <Quartus> Quit pretending you're ??ML of any kind. :)
14:20:16 <slava> what about Ocaml?
14:21:01 <Quartus> No, nor worml or its cousin squirml.
14:21:18 <grub_booter> heh - like i said, found it interesting - the 'xml' structure is identical to the way the stack expects data...
14:22:08 <grub_booter> i dunno - does xml mandate nesting? don't think so ;-)
14:25:57 <grub_booter> (well, ok - one level would be needed like <graph><video><blah></graph>)
14:26:10 <slava> xml requires exactly one top-level element
14:26:19 <grub_booter> yup - ack'd that
14:27:55 --- join: Quartus_ (n=Quartus_@209.167.5.2) joined #forth
14:27:56 --- mode: ChanServ set +o Quartus_
14:28:17 <Quartus> gavino msg'd me. Evidently he is confused as to why he's banned.
14:28:30 --- mode: Quartus set -b *!*@4.38.41.*
14:32:43 <grub_booter> well, i like it cos serialisation and deserialisation is all uniform with the construction phase - there can be complex nested things interleaved but that's down to what a filter or video element wants to put there (and since 3rd party libs are typically used, that'd probably happen as a matter of course...)
14:33:31 <grub_booter> dunno - haven't been convinced enough to implement it, but i think it'd work
14:33:38 <Quartus> if it's meant to be parsed and acted upon, why not write it in Forth?  Why write it in some kind of format that requires parsing?
14:34:16 <Quartus> I can understand if you need to talk to other software somewhere that expects a certain format for certain data.
14:34:19 <Quartus> But if it's local to you...
14:35:15 <grub_booter> hmm - well, i'll certainly be using my forth 'dialect' for comms
14:35:43 <grub_booter> (have already been playing with a playout server arrangement that receives the 'commands' as text)
14:36:20 <Quartus> parsing something fancy seems like unnecessary overhead, unless you're complying with some arbitrary format.
14:36:48 <grub_booter> well, with what i have, there's no compilation involved there, so it's just a stream of data and is interpreted as such on the other side
14:37:09 <Ray_work> don't think i didn't see that cf dig,  Quartus. FOOLS!!!
14:37:31 <Quartus_> heh
14:37:40 <Ray_work> For that gavino gets your home address, hehe :)
14:37:49 <Quartus_> grub -- compilation?
14:37:51 <Ray_work> What is your home addres. :)
14:38:08 <Quartus_> pretty easy to find, ray
14:38:26 <Ray_work> ya I know, funnier the way I stated it. Come on! FOOLS!!!
14:38:30 <slava> somewhere in Thailand?
14:38:51 <Ray_work> MeLoveYouLongTime Lane.
14:39:12 <Quartus> Ten dalla.
14:39:37 <Ray_work> sucky-fucky-ten-bucky
14:39:53 <grub_booter> meaning that you can inject words on the fly (just like you would in forth, but the domain is restricted - unlike forth where injection of new words could change the running state in a big way...)
14:39:57 <Ray_work> yikes, kids the children THINK!!!
14:40:33 <Quartus> grub_booter, you can do that with Forth, too.  Isolate your allowed words in their own wordlist.
14:40:53 <grub_booter> ah - cool
14:40:53 <Quartus> Set the search-order appropriately.
14:41:10 <Ray_work> I suppose that's what you do when you don't want to give access to the entire compiler, no?
14:41:33 <Quartus> Or you just want to be sure that what you're executing is from a given set of words.
14:41:37 <Ray_work> Be carefull about parsing words? what else?
14:42:32 <Ray_work> I'm thinking in terms of ChuckBot-Palm and I'll need to expose some words for the programming interface but really not experianced enough to do such a thing. This will be my first turnkey.
14:42:40 <Ray_work> Turkey? no Turnkey.
14:42:49 <Ray_work> okay turkey.
14:42:52 <Quartus> The dictionary isn't present in a standalone app generated by Quartus Forth.
14:43:04 <Ray_work> cool.
14:43:10 <Quartus> So you can't expose some words.
14:43:28 <Ray_work> FOOLS!!!
14:43:30 <Ray_work> really?
14:43:37 <Quartus> Probably better to have labeled buttons, and take action on them.
14:43:45 <slava> i'd like to get factor running on apple's iphone
14:43:49 <Quartus> Really. There's no dictionary headers.
14:44:04 <Ray_work> okay, I get it.
14:44:12 <Ray_work> know what to do, thank. :)
14:44:13 <Quartus> No FIND. No EVALUATE.
14:44:17 <Ray_work> +s
14:44:46 <grub_booter> so, would it be common for a running instance of forth to allow word creation from the outside (via a socket for example)?
14:44:48 <Quartus> If you want it to run inside Quartus Forth, you've got all that stuff.
14:45:18 <Quartus> grub_booter, just some random implementation of Forth waiting for code to come in?  Or have you got some purpose in mind?
14:45:33 <grub_booter> specific purpose
14:45:56 <Quartus> well, I'm not sure how to answer the 'would it be common'.  What are you thinking of doing?
14:47:21 <grub_booter> thinking playout servers - like the basic one would provide simple video playout, but broadcasters would be able to customise things to add their own rules - apps connecting wouldn't need to be changed to conform with the rules
14:47:23 --- join: mjanssen (n=mjanssen@ip54518373.direct-adsl.nl) joined #forth
14:48:12 <Quartus> I don't know enough about the problem domain to advise.  Allowing clients the ability to arbitrarily reconfigure a server such that it affects all other clients sounds really bad to me.
14:48:57 <grub_booter> well, it kinda depends - like, most broadcasters have a watermark which is applied to all video that's played out - some don't
14:49:16 <grub_booter> some change the watermarks for certain programs, some don't
14:49:50 <grub_booter> others have rules about adverts (volume handling for example)
14:50:04 <Quartus> ok. So the server needs to be aware of the client's preferences.
14:50:50 <grub_booter> yeah - but being flexible enough to handle some variance in a single execution (and typically, we're talking preferably *long* uptimes)
14:51:19 <Quartus> I can't see anything but bad in allowing or expecting your clients to sit down and code up little Forth routines to modify the server's behaviour.
14:51:38 <Quartus> Passing preference info, sure. Something that can be screened and validated for error.
14:51:42 <Quartus> Different thing.
14:51:53 <grub_booter> yeah - well, it would be the client doing the writing - just triggering custom ones
14:52:09 <grub_booter> (or if they want to, they can.. their risk)
14:52:10 <Quartus> If they're only triggering custom pre-written routines, then sure. 
14:52:20 <Quartus> well, not their risk -- the server's risk.
14:53:02 <Quartus> Anyway you can seal-off a set of words, such that those are the only ones available.  This is sometimes done, to provide an interactive but restricted interface to a forth program.
14:53:29 <grub_booter> yeah - that sounds like a sane approach to me
14:53:42 <Quartus> In a system where a turnkey app doesn't allow dictionary access, you'd need to associate the words with their actions in your app, which isn't a big deal either.
14:53:53 <grub_booter> indeed
14:54:37 <grub_booter> most of the time, these things would all be custom developed and the app would just provide a means to switch on/off pre-defined stuff
14:55:19 <Quartus> I wouldn't be comfortable providing any interface for a client to execute arbitrary code on a server.
14:56:33 <grub_booter> yeah - which is why i like the subset approach (which is essentially what i have - it's just that in the subset, i allow quite a lot of 'mucking about')
14:58:08 <grub_booter> to the point where you could (theoretically) tell all clips to play backwards or have a picture of a naked woman embossed in the top right hand corner, or even have parallel playback with a live feed...
14:58:38 <grub_booter> (like cnn style outside broadcast stuff)
14:59:20 <Quartus> If you left Gforth wide-open, a client could execute shell commands, or what-have-you.
15:00:07 <grub_booter> yup - not in the shell i provide of course, but that's what i was thinking about wrt forth as a general recipient of commands over a socket...
15:01:06 --- join: ttuttle (n=tom@gentoo/contributor/ttuttle) joined #forth
15:01:17 <ttuttle> Quartus: How much junk snail mail do you get from your domain registration?
15:01:20 <Quartus> It's ttuttle! Here to tell us that he's made 17 cents today. :)
15:01:28 <ttuttle> Quartus: No.
15:01:36 <ttuttle> Quartus: I haven't been on mturk in a while.
15:01:48 <grub_booter> actually, being able to attach filters which provide tickers would also be part of this (news, stock prices, sms messages...)
15:02:13 <Quartus> ttuttle - snail mail? Very little.
15:03:02 <ttuttle> Quartus: Oh.
15:03:14 <ttuttle> Quartus: Do you have a WHOIS protection thingy?
15:03:23 <Quartus> one or two a year.  There were a couple of attempts to convince me I was being sent a registration renewal, which was bogus.
15:03:32 <Quartus> ttuttle, I don't think so. Anybody can look it up.
15:03:44 <Quartus> let me see
15:05:19 * ttuttle is transferring his domain.
15:05:22 * ttuttle crosses his fingers.
15:05:50 <Quartus> yeah, it merrily reports my mailing address.
15:06:12 <Quartus> So send that spare cash. :)
15:08:30 <Ray_work> Than ends another broadcast day. Oh say can.....
15:08:32 --- quit: Ray_work (Read error: 131 (Connection reset by peer))
15:08:48 <grub_booter> heh
15:08:53 <Quartus> ttuttle, I'm not paranoid about mail, snail or e.
15:09:51 <Quartus> I filter email. I'm on the national don't-send-me-crap junkmail and telephone list.  Otherwise, I deal with problems as they arise.  None have.
15:12:04 <Quartus> aside from the bogus renew-o dealies
15:12:06 <grub_booter> well, way i see it - each broadcaster has their own requirements - and trying to grow to accomodate every possibility in a fixed way just seems to lead to a bloated and largely redundant codebase - by allowing flexible/on the fly customisation we possibly open up the system for abuse, but then we aren't talking internet connected boxes here and access is restricted by default - you wouldn't want a junior introducing porn vids at 2pm for ex
15:12:06 * ttuttle has transferred his domain from Namecheap.com to his registrar.
15:12:11 <ttuttle> Quartus: The renew thingies are funny.
15:12:18 <slava> grub_booter: main problem is that forth is not type safe
15:12:23 --- join: zpg (n=user@81-178-126-78.dsl.pipex.com) joined #forth
15:12:33 <grub_booter> yeah - that's always a risk
15:12:36 --- nick: arke_ -> arke
15:12:43 --- mode: ChanServ set +o arke
15:12:47 <Quartus> it just means your user-visible words have to validate their inputs.
15:12:54 <grub_booter> indeed
15:13:34 <ttuttle> Quartus: /me now happily has his domain name, DNS service, hosting, and email forwarding from one company.
15:13:45 <Quartus> ttuttle, all eggs in one basket, eh? :)
15:13:47 <grub_booter> that's a given really - everything going to air needs to be validated - all stack entries need to match the expectations of the words used
15:14:03 <ttuttle> Quartus: Yeah, I figure if one goes out the rest are pretty useless anyway.
15:14:09 <ttuttle> Quartus: Except maybe the email, but I don't use it much.
15:14:13 --- quit: jackokring (Read error: 148 (No route to host))
15:14:50 <grub_booter> but anyway - was an interesting chat (for me anyway :-)) - turning in now
15:14:54 <Quartus> k
15:17:30 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/")
15:20:59 <ttuttle> Quartus: /me is wondering why the NFSN page says "Action Status: Transfer waiting for Losing Registrar Approval" and "Transfer Status: InActive".
15:21:40 <Quartus> I don't know.
15:22:26 <zpg> Hi.
15:22:37 <Quartus> Hey.
15:24:03 <Quartus> ttuttle, at a guess, the registration is inactive until the transfer is approved.
15:24:56 <ttuttle> Quartus: okay
15:26:17 <ttuttle> Quartus: I have a feeling that "losing" registrars aren't very speedy about approving domains for soon-to-be-ex-customers.
15:26:48 <Quartus> There's some code of conduct they're supposed to comply with; if they behave badly, they might be behaved badly towards.
15:27:20 <Quartus> They make money from inbound transfers as well.
15:27:47 <ttuttle> Quartus: Any experience with NameCheap?  (They're my "losing registrar".)
15:27:54 <Quartus> Never used'em.
15:28:15 <Quartus> I'm not Joe Website, though.  I have only a small number of domains.
15:29:08 <ttuttle> Quartus: I have one.
15:39:45 <JasonWoof> registrars always seem to be slow about opproving domain transfers
15:39:55 <JasonWoof> I assume they are a pain in the ass on purpose
15:44:27 --- nick: nanstm -> Raystm2
15:45:08 --- quit: Quartus_ (Read error: 104 (Connection reset by peer))
15:46:54 --- join: Quartus_ (n=Quartus_@209.167.5.1) joined #forth
15:46:54 --- mode: ChanServ set +o Quartus_
15:50:38 <Raystm2> This begins another broadcast day.
15:50:53 <zpg> Broadcast away.
15:51:19 <Raystm2> Damn! I keep missing Ray_work. It's as if he knew I was about to log in.
15:51:29 <Raystm2> I'm getting a complex.
15:52:03 <Raystm2> Man, we were busy, I say. Busy!
15:52:52 <Raystm2> Turns out there are wheels and casters on _everything_ .
15:53:08 <Quartus> Even on the wheels and casters.
15:53:19 <JasonWoof> not on me
15:53:21 <Raystm2> Bearings count?
15:53:25 <zpg> I'm lost.
15:53:30 <Quartus> Bearings do not count.
15:53:31 <Raystm2> JasonWoof: you don't own inline skates?
15:53:42 <JasonWoof> zpg: turn right at the dunkin' dognuts
15:53:47 <Raystm2> A wheel is by definition a bearing
15:53:49 <Raystm2> .
15:53:54 <Quartus> dognuts?
15:53:59 <Raystm2> hehe
15:54:05 <JasonWoof> Raystm2: don't think so
15:54:09 <Quartus> no no no. Wheels have axles.
15:54:12 <Raystm2> that one caught me, good one JasonWoof.
15:54:42 <JasonWoof> I had inline skates at some point, but I'm assuming they've moved on to greener pastures
15:54:44 <Raystm2> A wheel, by definition is actually a lever.
15:54:49 <Quartus> is it at all alarming that my cutlery is non-magnetic?
15:55:05 <Raystm2> hmm. How do you check for buckshot?
15:55:24 <tathi> why would you want magnetic cutlery?
15:55:33 <Raystm2> To check for buckshot.
15:55:37 <Quartus> I just figured it to be stainless steel.  Apparently it ain't.
15:56:22 * JasonWoof goes downstairs to magnetize his knives ;)
15:56:25 <tathi> oops.  getting magnetic and magnetized mixed up
15:57:02 <tathi> that is a little odd.  wonder what it's made of?
15:57:07 <Quartus> I have absolutely no idea.
15:57:26 <Quartus> No response even to some of those really strong rare-earth magnets.
15:57:34 --- quit: ttuttle ("Lost terminal")
15:59:31 <Quartus> Chromed aluminum, maybe?
16:00:26 <Raystm2> nonmagnetic steel Definition:  Austenitic steels such as the 14% manganese steels and the 303 type 18/8% chromium-nickel stainless steels.
16:00:42 <Quartus> Maybe the latter.
16:00:48 <Raystm2> Whilling to bet the latter, 
16:00:49 <Raystm2> ya.
16:01:14 <Quartus> I remember the cutlery we had when I was a kid, it was hellaciously magnetic, despite being stainless.
16:01:21 <Quartus> Things used to stick to each other.
16:01:45 <Raystm2> Ah sure. Now that makes sense for having non magnetic.
16:02:16 <Raystm2> Who wants to draw the entire drawer every pull.
16:03:39 <Raystm2> new Burke this week. Two infact,
16:03:48 <Quartus> neat. Still watching the other ones.
16:04:10 <Quartus> zpg, from yesterday -- add >IN to that list.
16:04:14 <Raystm2> I lie, it's 3
16:04:34 <Raystm2> Seems to be adding one a day now.
16:04:35 <Quartus> >IN is neither a converter nor a relocator
16:04:38 <tathi> dunno, sounds like austenitic steels would still be paramagnetic...but maybe not
16:05:01 <Quartus> Paramagnetic I can't easily test for, I'd need to attach a magnet to my drillpress or something
16:05:25 <tathi> oh.  I thought it meant it would still stick to a magnet, just not become a magnet itself.
16:05:36 <tathi> never mind then.
16:05:55 <Quartus> Ah. Hmm, I thought paramagnetic referred to things like aluminum, where you can generate heat in'em with sufficient magnetic provocation.
16:06:21 <Quartus> And they'll respond, albeit oddly, to really intense magnetic fields.
16:06:48 <Raystm2> Like in a MRI? or more powerful?
16:06:56 <tathi> yeah, looks like you're right
16:07:04 <Quartus> Aluminum is affected by an MRI field, but in curious ways.
16:07:37 <Quartus> You get resistance while moving the aluminum in the field, if I remember correctly.
16:07:41 <zpg> Quartus: yeah, I omitted >IN as I was talking about the convention of >xyz meaning "convert to xyz"
16:07:43 <Raystm2> The metal shards from the nail I stepped on in may really reacted badly during the MRI. I sware my foot was on fire.
16:07:50 <zpg> whereas >IN doesn't mean "convert to in"
16:07:57 <Quartus> no, nor 'relocate to IN'
16:08:01 <zpg> indeed.
16:08:04 <Quartus> another of ye olde holdeovers
16:08:07 <zpg> just to befuddle everyone.
16:08:39 <Raystm2> This is just as we discussed that day Quartus asked me for opinions to consider for 'the book'. 
16:09:04 <Raystm2> I'm thinking a history of words in context helps develpers understand the 'why'.
16:09:15 <Raystm2> developer too.
16:09:27 <Quartus> yes.
16:09:44 <zpg> a sound idea ray.
16:09:48 <tathi> yeah, and a lot of it isn't easily findable on the net
16:10:01 <tathi> appendix A or whatever in the standard has a bunch of stuff, but by no means all
16:10:17 <Raystm2> and Chucks site does as well, but thatt'
16:10:23 <Raystm2> 's the real early stuff.
16:12:43 <Raystm2> Current words in the standard is sufficient tho. Only others when the words in the standard need to be clarified, /me thinking <build does> ( did I do that right?)
16:14:42 <tathi> I haven't actually read Thinking Forth or whatever closely enough to know exactly how <build does> works
16:14:47 <tathi> And I don't really care.
16:14:51 <tathi> :)
16:15:14 <Raystm2> Reason I mention is because we still have DOES> but bot <BUILD
16:15:20 <Raystm2> not 
16:15:54 <Raystm2> instead <BUILD has been deprecated for CREATE and frankly, I wish I knew why.
16:16:15 <tathi> so what does <BUILD do?
16:16:38 <Raystm2> IT was part of the DOES> thing back when and I assume it did CREATE.
16:17:05 <Quartus> <BUILDS was a half-assed version of CREATE
16:17:06 <Raystm2>  that's why DOES> has a > cause <BUILD had a < and they were taken together.
16:17:10 <Raystm2> I see.
16:17:38 <zpg> yeah it still reads well
16:17:51 <zpg> i mean, "create" could be "create:" i guess, but even so
16:18:35 <zpg> and <BUILD would be confusing if it, like CREATE, forward-parsed.
16:18:57 <tathi> found it in f83.txt, where it seems indistinguishable from CREATE except that it's not really fully described.
16:19:26 <zpg> so   : constant <build , does> @ ;
16:19:31 <zpg> ?
16:19:57 * Raystm2 picks up 12string, rips "From the begining" ELP.
16:21:02 <Quartus> Have a look at Starting Forth's description of it.
16:21:42 * Raystm2 desides it's bare with out the twenty minute keyboard solo.
16:21:44 <tathi> ah.  I was looking in Thinking, which doesn't talk about it.
16:22:07 <Raystm2> I think the MVPforth tutor has something about it...
16:22:09 <zpg> Quartus: the SF I had in hard-copy was already using CREATE I think.
16:22:38 <Quartus> <BUILDS isn't fully defined, and I'm not at all sure you can use it in a word separate from DOES>.  I'd have to go dig back into it to be sure.
16:24:47 <Raystm2> no i'm wrong, the MPV tutor has create does>
16:27:24 <Quartus> Ok. So it looks like the FIG-era <BUILDS worked like CREATE, but you had to have a DOES> or there'd be no default action defined.  So : CREATE <BUILDS DOES> ; would give you a CREATE-equivalent.
16:28:10 <tathi> ah
16:29:09 <Raystm2> http://www.colorforth.com/HOPL.html has an early history, if you didn't already know and too bad if you did cuz there it is.
16:30:37 <Raystm2> 'early history' of forth words not <build does> sorry if confusion.
16:31:18 <zpg> well there's The Evolution of Forth too.
16:36:19 --- join: nizchka (n=dLdGJozH@e33014.upc-e.chello.nl) joined #forth
16:36:42 <Raystm2> God : create <build does> ; Forth. It didn't evolve. You watch it or you will be excommunicated, which is much easier now due to the communication medium. 
16:37:43 <zpg> Huh? Are you creating God in RPN?
16:38:31 <Raystm2> no that would be doG
16:39:30 <zpg> is that colorForth?
16:39:46 <zpg> etaerc doG
16:39:52 <zpg> the new paradigm.
16:40:02 <Quartus> not with an uppercase letter in it
16:40:24 <zpg> good point.
16:40:25 <Raystm2> Ya, not at the end. only at the begining. 
16:40:34 <Raystm2> ALSO:
16:40:36 <zpg> so, what colour is "dog"?
16:40:53 <Raystm2> Snoopy is black and white.
16:40:59 <Raystm2> ALSO:
16:41:22 <Raystm2> if you try to do an all caps word, the caps break at the EXTENtion.
16:41:31 <Raystm2> BUG?
16:41:43 <zpg> i'm waiting for the "colorForth uses colours to distinguish what a token does. to further simplify the system, Chuck has decided to strip out the colours."
16:41:58 <Quartus> oh no, doubtless it's there by design, to prevent you from making a name that's got too many CAPS in it!
16:42:17 <zpg> Raystm2: are you serious?
16:42:50 <Raystm2> Its because the all caps tag appears in the first 'word' of the name and the second one has tag 0 or text.
16:43:03 <Raystm2> f+or
16:43:18 <Quartus> Yes, it sucks *that much*.
16:43:26 <zpg> so RAy is two tokens?
16:43:42 <Raystm2> not in 3 letters but your point is valid.
16:44:11 <zpg> to quote Dante, "fucking hell"
16:44:11 <tathi> that has got to be a bug
16:44:20 <Quartus> design feature.
16:44:27 <Raystm2> 4-5 and sometimes a couple more in really well packed and letter-well-chosen words.
16:45:27 <zpg> Raystm2: run while you still can.
16:45:47 <Quartus> If somebody came on here and said he was going to implement Forth, and this was his idea about how to do it, he'd be strongly advised to take up other work.
16:45:55 <zpg> Ha!
16:46:22 <Raystm2> to make it work, a tag 0 extention would have to parse every previous tag to see if it is following an all caps tag. or the all caps tag would have to modify the extention in such a way that it still runs the -trainling-like routine in the editor that makes it display like one word.
16:46:27 <zpg> I can't believe even the most fundamental idea in Forth -- the token -- is fucked.
16:47:03 <Quartus> Ray, to make it work, TO MAKE IT WORK, you store names as *names*, and not in some half-assed imbecilic 32-bit huffman-compressed dingbat way!
16:47:15 <Raystm2> Granted.
16:47:33 --- quit: nizchka ("Azureus 2.5.0.0")
16:47:41 <Quartus> This is supposed to be *simpler*?
16:48:50 <Raystm2> Gentilemen, someone has to know where the keep out signs are kept. But somebody really has to know why to keep out and if you don't have someone who does, the mistakes get repeated. In here, I'm that for colorFroth. <intended>
16:49:14 <Quartus> Dude.
16:49:51 <Raystm2> You could have found out yourself, Quartus, but you've asked me what I know... Saved your life.
16:50:04 <Raystm2> Here's your sign. :)
16:50:14 <Quartus> We don't need a prancing village idiot to remind us not to eat dirt by example.
16:50:29 --- join: jackokring (n=jackokri@static-195-248-105-144.adsl.hotchilli.net) joined #forth
16:50:31 * Raystm2 dances into the kitchen for a bite. brb. :)
16:51:05 <Quartus> cf is a monstrosity.  It is either a long and not particularly funny joke, or some kind of aberration designed specifically to be nothing like what works.
16:52:32 <Quartus> Doubtless my words will summon an apologist who will tell me that no, all these stupidities work together to guide me toward being a better coder.
16:53:32 <Raystm2> I think I did that already, but then you know my level of experiance allows for such a comment.
16:53:57 <Quartus> I expect a future revision to re-invent 3+length, and then slowly move toward FIG.
16:53:58 <zpg> Apologies for uncontrollable cursing btw.
16:55:09 <Quartus> heh
16:55:13 <Raystm2> zpg not a problem here. :)
16:55:22 * Raystm2 an old sailor from way back.
16:55:46 <zpg> :)
16:56:01 <slava> Quartus: no, a future revision will feature no ascii input, no stack, will run on arm cpus only, touch screen input, mid-levels, YOU ARE ALL FOOLS!!!
16:56:09 <Quartus> heh
16:56:21 <Quartus> I AM THE WORLD'S GREATEST SYSTEMS PROGRAMMER"
16:56:43 <Quartus> Unless Chuck went insane and moved to Thailand, I think we can count on that being fiction. :)
16:56:44 <zpg> Anyone else noticing the shift from humour to despair here?
16:57:27 <arke> ColorForth is something that I think anybody should take a look at, even if briefly, just to get an idea that "hey, look, its actually possible to do things completely different". 
16:57:32 <Raystm2> Not I. Comedy of tradegies.
16:57:45 <arke> Not something to use regularly or for commercial projects or so though.
16:57:59 <Quartus> arke, there's always stupid ways to do things.  cf has a remarkable number all packed into one place.
16:58:01 <Quartus> Huffman packed.
16:58:19 <arke> CF also has some cool stuff. :)
16:58:23 <Quartus> such as?
16:58:23 <arke> But yeah.
16:58:44 * zpg weeps
16:59:09 <Raystm2> Introducing BalognaForth. Each class of word a different sausage.
16:59:30 <arke> Quartus: the colors. 
16:59:34 <arke> Quartus: ;)
16:59:40 <Raystm2> Editable hell it's edible.
16:59:54 <Quartus> The colors are stupid.  There, it has been said.  They are intensely stupid.  
17:00:27 <JasonWoof> man, some people are annoying
17:00:27 <arke> I prefer normal code too but its something to be investigated.
17:00:41 <JasonWoof> I live with some cool people, but they have some really stupid friends
17:00:49 <arke> JasonWoof: welcome to my life :)
17:00:58 <zpg> Welcome to everyone's.
17:01:04 <JasonWoof> hey arke :)
17:01:06 <Quartus> They present an obstacle to sharing code in anything other than HTML.  They are inaccessible to anyone with even moderate colour differentiation problems.
17:01:17 <JasonWoof> how's life (asside from annoying)
17:02:11 <JasonWoof> man, did somebody bring up cf again?
17:02:16 <arke> Quartus: so for that fact alone you don't think its worth looking at from a research point of view? I think it is. I think anything should be looked at, just to see how different things do things, how it can be done, how it works well, how it doesn't, and as food for new ideas.
17:02:40 <Quartus> arke, I have looked at it far more than I would have liked, thanks to its being brought up in a couple of contexts.  I didn't stop at the colour.  I should have.
17:02:55 <Raystm2> Truth be told that you don't gain much from the current scheme that is not already covered with word position and coding practice. They could still have color and a colorblind person wouldn't have a problem understanding the code.
17:02:58 <JasonWoof> cf is badly written
17:03:04 <Quartus> cf is badly designed.
17:03:54 <zpg> arke: check the logs, far more than colour has been discussed by Quartus and others.
17:03:56 <Raystm2> JasonWoof: People like to tell me i'm an idiot. I'm used to it. Some sort of magnetism. 
17:03:56 * Raystm2 invests in a non-magnetic set of cutlery, maybe that will help. :)
17:04:15 <JasonWoof> lol
17:04:43 <JasonWoof> I certainly wouldn't want my knives to stick to my stupidity
17:04:46 <JasonWoof> or something
17:05:22 <Raystm2> Well when you stick a knife in your own back right out of the programming box, the rest is academic physics.
17:05:41 <zpg> What does that mean?
17:06:00 <Quartus> I was decidedly put off by the externally visible stupidities. But I've had occasion to delve deeper, to find that there are far worse things the further in you go.
17:06:07 <Raystm2> I learned colorForth totally before I learned any other language totally.
17:06:36 <JasonWoof> here's my attitude about colorforth: sure, there's things in it we could probably all learn from, probably not that much for people who have been programming a lot for 10+ years
17:06:41 <zpg> Raystm2: but colorForth doesn't have IMPORT
17:06:52 <JasonWoof> but there's plenty of easier and faster ways to learn about programming than to dig through that crap.
17:07:02 <arke> Fact is, I'm really fond of forth but these days I still write mostly in non-forth languages, because thats what gets the job, thats what I'm intimately familiar with, and I get a high throughput with it. There is a large number of available libraries, etc. etc. etc., but I still look at forth alot, I idle here, I look at whats going on, I educate myself, I ask questions, I incorporate Forth ideas into my projects ... Its a learning process and no
17:07:03 <arke> thing should be let out becuase in the end, its all just food for new ideas, and one new idea may eventually turn out to be a really big new idea. I know colorforth is flawed in many ways, I'm not defending that. I'm defending its place as something to be looked at ... something to study and to learn from. If you learn that its mostly methods of not doing it, then thats equally as valid as learning that there are good things about it, and for that
17:07:03 <arke>  reason alone I think its something everybody should at least briefly investigate. Holy hell, I just typed alot, probably inconsistent and repeating bullshit :)
17:08:07 <JasonWoof> the people who seem to be pushing colorforth don't seem to be good programmers
17:08:15 <Raystm2> zpg: the system is considered native even tho there are hosted ports to learn and practice from. there would be no need for import in such a system. a new block of code need only be appended to the current running image and loaded to run new code.
17:08:35 <zpg> arke: Sure, don't be ignorant seems to be what you're saying. That's not the argument though. 
17:08:43 <JasonWoof> I find it extremely strange that colorforth is designed by and for someone with 30+ years of experience, and it attracts people with little to none
17:08:53 <JasonWoof> and they can't do much with it, but they're all excited about what little they can
17:08:56 <zpg> Raystm2: It was a Doty-joke.
17:09:08 <Quartus> JasonWoof, good summary.  I find it strange too.
17:10:20 <JasonWoof> arke: we have checked it out.
17:10:26 <JasonWoof> even I ran the damn thing and I've got a mac
17:10:39 * Raystm2 doesn't know Doty, but thinks he understands this to be a gavino nay a werty a-like.
17:10:44 <arke> Thats good. :)
17:10:50 <Quartus> Yes, I'm not spouting off without having seen what I'm talking about.
17:11:34 <zpg> I took a look but am largely spouting off with ignorant petulance.
17:11:35 <Quartus> The Emperor has no clothes.
17:12:09 <zpg> Quartus: so, where's your _ then?
17:12:19 <Quartus_> here
17:12:21 <JasonWoof> I did take some ideas from colorforth
17:12:23 <zpg> magic.
17:12:44 <JasonWoof> like the colored display and the A register
17:14:05 <Quartus_> You're using syntax coloring -- different animal.
17:14:40 <JasonWoof> no, I'm storing color bits with each word
17:14:42 <Quartus_> unless I am not recalling correctly?
17:14:55 <JasonWoof> though I only use 3 colors
17:14:56 <zpg> this is Herk?
17:15:03 <JasonWoof> other things I use a prefix characters for
17:15:04 <JasonWoof> yes
17:15:13 <JasonWoof> I went by Herkamire for a long time
17:15:18 <Raystm2> I liked it as a newb because in any other language an newb stares at the screen for 3 weeks looking for something he's capable of codeing and understanding himself. A newb understands nothing of the internals in a cf. But instantly you have a display and a keyboard hacked and off you go, learning you should have never used a system that easy to operate, no matter how badly implemented.
17:15:36 <JasonWoof> my forth was herkforth, but I'm rewriting it. new project is called fronds
17:15:39 <zpg> Raystm2: I don't follow.
17:16:54 <zpg> Raystm2: Most scripting languages -- say, Python -- offer a pretty clean interface with a tonne of documentation and data types at your disposal.
17:17:14 <zpg> Raystm2: Not to mention a display and keyboard.
17:17:42 <Raystm2> zpg, the underlying system, no matter how fraught with disaster, leaves an interface and display that , with a few minutes practice and much like a computer game, you can start making what you think of work right away. 
17:18:18 <Raystm2> zpg Python is nice and WAY to big and complicated for a newb. I loved Python, did all of the tuts. Read all of the documentation. 
17:18:29 <JasonWoof> the only significant and working app for cf that I've heard of is okad
17:18:37 <Raystm2> Went to code something on my own and never figured out where to start.
17:18:46 <zpg> Raystm2: sure, but you don't have to comprehend the entire system to write your applications.
17:19:02 <Quartus_> jason, has said app ever been seen?
17:19:04 <zpg> Raystm2: i understand the desire for knowing exactly what you're working with, but even so.
17:19:14 <Raystm2> No. You THINK!!! you do tho.
17:19:16 <JasonWoof> Quartus: I've seen it in a video :)
17:19:52 <JasonWoof> Raystm2: what is it about cf that makes it easy for you to get started?
17:20:26 <Raystm2> All colorforth programs are essentially a version of LOGO.
17:20:33 <zpg> Bah?
17:20:42 <Raystm2> the colorforth word logo, not the system LOGO.
17:20:47 <zpg> Ah.
17:20:48 <JasonWoof> I'm guessing the problem you had with python is that you didn't know how to do computer programming.
17:21:03 <JasonWoof> most computer language tutorials are to teach you the language, not the skill of programming
17:21:20 <zpg> Agreed. You need to build up a conceptual understanding before programming into a language.
17:21:58 <JasonWoof> Raystm2: what's that?
17:21:59 * Raystm2 has been programming longer then many of the people in this room have been alive. /me looking in JasonWoof's direction.
17:22:20 * JasonWoof blinks
17:22:34 <JasonWoof> Raystm2: what kind of stuff have you written? and in what language(s)?
17:22:44 <zpg> I think I fall into Jason's age-bracket too, for the record (before this gets all agist)
17:24:41 <Raystm2> I made my first "SNOOPY" spreadsheet in BASIC on a mainframe ( name escapes me) in '74.
17:26:48 <Quartus_> is that the whole history? :)
17:27:04 <Raystm2> Oh heck no...
17:27:50 <Quartus_> I don't know what a "SNOOPY" spreadsheet is.
17:28:25 <Raystm2> it's a giant Snoopy on a real spreadsheet. :)
17:28:33 <zpg> heh
17:28:50 <Raystm2> tractor driven paper. some may not remember...
17:29:03 <zpg> what's paper?
17:29:07 <Quartus_> ah.
17:30:09 <Raystm2> there was programming the next 7 years mostly in fortran and cobol, but in the last 4 we built our own little 'computers' out of intel 4004's and such.
17:31:19 * Raystm2 went to a tech highschool and studied electronics in a lab every other week. We'd make up for the lost two each each month by doubleing up the school work on the 'other' weeks.
17:32:47 <Raystm2> But sure. I wasn't a total geek, most of those hours spent in the radio station we built as the dj and general manager granted. But i've always had a hand in programming something some how. CNC machines for a while. 
17:33:06 <Quartus_> cnc is pretty neat.
17:33:09 <Raystm2> ya.
17:33:49 <Raystm2> I guess I didn't start picking up c and Python till '96 tho, internet and all. 
17:33:59 <Raystm2> Python was 1.5.2
17:34:36 <Quartus_> ok, so you officially have no excuse for thinking cf is worth the floppy it boots from. :). 
17:35:29 <zpg> what's a floppy?
17:35:37 <Raystm2> Oh but it did. I didn't need windows or linux or anybody but my own imagination. Sure I could have done that other ways. I found cf first. and there ya go.
17:36:19 <Raystm2> Interest in cf forced me to learn how such a system works and is put together.
17:36:29 --- quit: mjanssen (Read error: 110 (Connection timed out))
17:36:40 <Quartus_> ok. My fault, I said that doG couldn't be a cf name. Which is true. So let's leave cf for #c4th and all the brilliant minds there.
17:36:40 <Raystm2> there was no documentation. it was very painfull
17:37:07 <Raystm2> I'm okay with that. 
17:37:36 <zpg> Another day, another trip to the coda.
17:38:03 <Quartus_> good luck with that 'hello world' project, let us know if it ever bears fruit. :)
17:38:07 <Raystm2> funny how it can't be mentioned ( much like that Shakespear play) with out an hour of wind down.
17:38:11 <JasonWoof> LOL
17:38:31 <zpg> Raystm2: heh, very true.
17:38:42 <Raystm2> DS al CODA, just saw it three lines above this one.
17:39:00 <Raystm2> oops start with 0 :)
17:45:53 <Raystm2> Since '96 theres been pascal, html, several 'newer' basics, Lisp, Scheme, a touch of perl, what else...
17:46:30 <Raystm2> What I wish i had found early on was IRC. 
17:46:41 <Raystm2> then I would have had some better direction.
17:47:53 <Raystm2> A lawer who defends himself has an idiot for a client. /lawyer/self-taught-programmer...
17:48:25 <Quartus_> I'm self-taught. Key is to actually study, though.
17:48:31 <zpg> I don't think the joke translates.
17:49:03 <Raystm2> maybe not in every case.
17:49:05 <zpg> Quartus_: not just that. there's a sprawl of books and languages, often poorly written.
17:49:22 <zpg> especially all those damned C books that are 4 inches thick.
17:49:35 <Quartus_> zpg, yet you'd have no trouble naming the shining stars among them.
17:49:59 <zpg> Quartus_: of course. but you have to either (i) know what you're looking for or (ii) plod for a long time before recognising what you should look for next.
17:50:46 <Quartus_> or ask.
17:50:57 <Raystm2> took a long time for me to ask. 
17:51:05 <zpg> my favourite C book -- the one i finally got somewhere with -- was Alcock's.
17:51:17 <zpg> It isn't stocked in any of the major chains here. A shame.
17:51:21 * Raystm2 makes note.
17:51:46 <Raystm2> I've never even considered programming as a living so there was no real reason to bother anybody.
17:52:36 <zpg> Hobbyists enjoy helping hobbyists too. Perhaps moreso.
17:53:04 <Raystm2> I would hang some in the PythonTutor chat. Very helpful bunch.
17:53:27 <Raystm2> But I was scared of IRC cuz of all the bad press.
17:53:30 <zpg> hang as hang out, or as in lynch?
17:53:40 <Quartus_> I had K&R 1st ed., then 2nd. Good book.  Plauger, Ritchie, Knuth, Hoare, Dijkstra.  These weren't hard to find, or hard to find out about.
17:53:56 <Raystm2> zpg: I would imagine some of the prior and much of the later.
17:54:27 <zpg> Yeah, the thickness ratio of Wrox/SAMS books : Ritchie et. al is absurd
17:54:39 <zpg> Raystm2: heh.
17:54:47 <Raystm2> Quartus indeed. those names came up alot. My hobby only afforded me what I could find online, the computer was more then expensive for us then. :)
17:55:10 <zpg> Yeah, these books aren't too economical.
17:55:22 <Quartus_> I wasn't wealthy. I read a lot in the local + university library.
17:55:44 <zpg> Did I mention that I had to get SF and TF dug out from the basement of the Cambridge UL?
17:55:53 <Raystm2> Those books that I purchased for you, besides internet for dummies and a python i found at the library, oh and the first C book at the library, was all I had.
17:55:57 <zpg> Quite funny. Somone had to go and burrow around especially for some Brodie.
17:56:01 <Raystm2> for=from sorry.
17:56:12 <Quartus_> ye olde sf, or the slightly newer folio?
17:56:28 <zpg> I can't recall the date. But it was 2nd edn or newer.
17:56:39 <zpg> I could check the library database actually. Hang on.
17:57:09 <zpg> Raystm2: did you buy books from Quartus_?
17:57:10 <Raystm2> Hard to find programming stuff in this area. You have to be at UTA or live in Richardson to talk that kind of tech with the populace.
17:57:33 <Quartus_> inter-library loan, ray.
17:57:35 <zpg> The British Library are /finally/ letting people register without 20 questions and proof of research.
17:57:38 <zpg> Useful.
17:57:40 <Raystm2> zpg, ya, he had some doubles. My favorite investment of all last year.
17:57:52 <zpg> Raystm2: what did you get?
17:57:59 <Raystm2> Quartus: wasn't interested enough for something like that.
17:58:34 <zpg> Quartus_: http://ul-newton.lib.cam.ac.uk/cgi-bin/Pwebrecon.cgi?v1=1&ti=1,1&Search%5FArg=starting%20%20forth%20brodie&SL=None&Search%5FCode=FT%2A&CNT=25&PID=KZnom3L0tXT000Ae87IEW4ewJc&SEQ=20070111015418&SID=1
17:58:39 <zpg> oh hang on, you're mobile.
17:58:41 <Quartus_> He got the 3 hardcover Knuth, 1st ed., and a pile of Forth stuff.
17:58:46 <Raystm2> Starting Forth for one... 
17:58:48 <zpg> SF: "Englewood Cliffs ; London : Prentice-Hall, c1981."
17:58:49 <Raystm2> them ya...
17:59:02 <Quartus_> that's the first ed I think
17:59:12 <zpg> Hey, sounds like a good batch of books.
17:59:26 <Raystm2> ya that's the one I got. 81
17:59:29 <zpg> I read a bizarre bit of Knuth in Borders recently. Something about compsci and God.
17:59:43 <zpg> TF was "Englewood Cliffs ; London : Prentice-Hall, c1984."
17:59:55 <Raystm2> I only have the online version of that;
18:00:19 <zpg> I think I got most of what I could from SF
18:00:50 <Quartus_> zpg, I don't know if he's religious. He plays church organ.
18:01:27 <zpg> It seemed pretty relaxed, you know, elegance of maths etc.
18:01:32 <zpg> Only thumbed through it.
18:01:38 <zpg> http://ul-newton.lib.cam.ac.uk/cgi-bin/Pwebrecon.cgi?SC=Subject&SA=FORTH%20%28Computer%20program%20language%29&PID=k4A1kypX3A715xCG5pUeDJTre5&BROWSE=1&HC=30&SID=2
18:01:46 <zpg> a few offerings in the UL.
18:01:51 <Quartus_> he's a mathematician with a keen understanding of language. Good combo.
18:01:58 <zpg> Agreed.
18:02:12 <zpg> Quartus_: do you have a mathematical background btw?
18:02:33 <Quartus_> after a fashion. I'm not a math geek.
18:02:47 <zpg> :)
18:03:07 <zpg> Let me find the blurb for the Knuth, I've intrigued myself.
18:03:28 <zpg> "In the fall of 1999, computer scientist Donald E. Knuth was invited to give six public lectures at MIT on the general subject of relations between faith and science. The lectures were broadcast live on the Internet and watched regularly by tens of thousands of people around the world, and they have remained popular many months after the event. This book contains transcripts of those lectures, edited and annotated by the author."
18:03:36 <Quartus_> I have an interest, but I lean more toward language than math.
18:03:42 <zpg> title = "Things a Computer Scientist Rarely Talks About"
18:03:59 <Quartus_> zpg, interesting. I'll look for it.
18:03:59 <zpg> Quartus_: which aspects of language?
18:04:37 <zpg> Hey, neat -- "You can still watch the webcasts of the original lectures; indeed, to the author's surprise, those downloads are still ranked #1 by Dr. Dobb's after being in their archives for more than 2000 days!"
18:04:56 <Raystm2> Threaded Interpretive Languages ( liked that one ) , Dr.Dobbs Tool Book of forth second edition, Forth Fundementals 2 language glossary, FOrth applications ( really liked that one too ) ,,, think that's it.
18:05:05 <Quartus_> zpg, most of 'em. Communication, I suppose, is the right word. That covers my interest in art (film + music) too.
18:05:14 <zpg> + information theory?
18:05:37 <Quartus_> yes, all part and parcel. 
18:05:46 <zpg> yeah, very interesting field.
18:05:50 <zpg> or fields.
18:06:10 <zpg> you can't really slice up the boundaries between disciplines when you get past a certain stage.
18:06:17 <Quartus_> I feel it's all one field, but we don't have a collective term.
18:06:29 <zpg> "life theory" heh
18:06:48 --- quit: Snoopy42 ()
18:06:49 <Raystm2> good as any.
18:08:02 * Raystm2 see's my study as communications as well, starting with electronics, modulated to music and 'other' forms of comm in the Navy ( read flags, signals, semiphore and flashinlight ) and then the recording studio in Tahoe and then computing.
18:08:38 <Raystm2> not to mention all of those garage bands. :)_
18:08:54 <zpg> Well, this is the beginning of the digital decade after all.
18:09:40 <Raystm2> Funny, Bill G was just saying that...
18:09:49 <Raystm2> Catchy I guess.
18:10:18 <zpg> Yeah. He must be some sort of sage.
18:10:23 <Raystm2> How about Apple's next new thing?
18:10:43 <zpg> iPhone?
18:11:06 <zpg> Apparently it's revolutionary. I think that's apple-speak for "shiny".
18:12:42 <Raystm2> see, I've been telling them all, "until it is your remote control AND your keyless entry into everything, never mind all the stuff it already does..." they ain't got dookie.
18:13:41 <Raystm2> It should plug into your dash. It should take voice commands and hear you yell cuz the shower is too hot and it should ajust appropriately.
18:14:14 <Raystm2> FOOLS!!! I AM THE GREATEST iphone PROGRAMMER!!!
18:15:13 <slava> the iphone looks pretty neat
18:15:36 <Raystm2> The new E-motion phone should cry with you, where ever you go. 
18:15:38 <slava> i might get one to port factor to
18:15:50 <Raystm2> That would be cool.
18:21:47 <Quartus> will it be programmable?
18:29:02 <Quartus> no clue? :)
18:34:18 <Quartus> what, is everybody off watching Seinfeld?
18:34:28 <zpg> Ahh, genius Knuth comment. "They call it open source whicih means you have to keep open (sic) 10 books..."
18:34:41 <Raystm2> Sorry, helping a BURKE newb. :)
18:34:53 <Quartus> what's with the (sic) and why is that insightful? Am I missing context?
18:34:55 <Raystm2> i'm Burkeing just now...
18:35:00 <zpg> Quartus: Watching a google video Knuth Q&A session. Not the same as the one paste earlier (real audio didn't work)
18:35:26 <zpg> Quartus: yeah, you're missing the humourous context I'm afraid. The chap asking the question was around 75. 
18:35:43 <Quartus> so the (sic) indicates...
18:35:51 <zpg> my idiocy -- ignore it.
18:35:56 <Quartus> ok :)
18:36:07 <zpg> mind was still on the video.
18:36:15 <Raystm2> video?
18:36:21 <JasonWoof> I just read apple's intro to the iphone
18:36:26 <JasonWoof> just the flashy pages
18:36:39 <zpg> the question, to complete the context, was whether programming had changed in 30 years, and whether the chap who programmed 30 years ago would be surprised by reading modern source listings.
18:36:40 <JasonWoof> I read the term "multi-touch" about 4 times before I realized what they mean
18:37:09 <Quartus> So 3rd party apps?
18:37:11 <JasonWoof> if that phone is anything like that multi-touch demo movie I saw on google video.... this could be amazing
18:37:19 <zpg> in retrospect, you can't divorce knuth's comments from the video and maintain the humour.
18:37:23 <JasonWoof> no, a touch screen that can sense multiple touches at once
18:37:38 <Raystm2> video?
18:37:40 <zpg> Raystm2: http://video.google.co.uk/videoplay?docid=3147833455273735180&q=knuth+god
18:37:46 <Raystm2> ah :) thanks. 
18:38:00 <zpg> Raystm2: you'll note i was searching for the stuff i mentioned earlier
18:38:03 <Quartus> shame that chap doesn't have any way of seeing any modern code.  Somebody should invent the internet.
18:38:29 <zpg> the what now?
18:38:45 <Quartus> gah that video starts with a goddamn test pattern, thank you kindlyh
18:39:11 <JasonWoof> http://video.google.com/videoplay?docid=6379146923853181774&q=multi-touch+display
18:39:12 <zpg> oh sorry. if it helps, i was deafened by that too.
18:39:17 <Raystm2> does it start with an iritating COLORBARS screen and a high pitched frequency?
18:39:23 <zpg> heh, yup.
18:39:44 <zpg> the first question is complete turkey too.
18:39:49 <Raystm2> indeed it does. :)
18:40:10 <zpg> prefaced by 5-10 minutes of fluff.
18:40:11 <JasonWoof> Quartus: check the multi-touch video if you haven't already. most of the begining is kinda silly screen-saver stuff, but once they get into editors and games and stuff later it's pretty impressive
18:40:22 <zpg> JasonWoof: this is the iPhone?
18:40:44 <Quartus> Frick.  Google video crashed Opera again.
18:41:21 <Raystm2> ah yes i've seen that movie JasonWoof.
18:42:33 <JasonWoof> zpg: the video is some big display in apples research labs
18:42:44 <Raystm2> very neat.
18:42:56 <zpg> JasonWoof: ah okay. will take a peek later. two videos at once leaves bruising.
18:43:10 <JasonWoof> the tiny bit of info on apple's site about iPhone says it has a "multi-touch" screen and says something about being able to zoom in and out on web pages
18:43:10 <Quartus> the new phone will have that display?
18:43:47 <Quartus> everything will come down to how well it works, and battery life.
18:43:47 <JasonWoof> Quartus: I'm sure it won't be that cool, but they did say it would be multi-touch and mention zooming
18:44:18 <Quartus> bear in mind it's a tiny screen, so single-touch is sometimes a challenge. :)
18:45:36 <Quartus> gah, the music is killing me
18:46:50 * Raystm2 needs a choppingBoardComputer.
18:46:54 <Quartus> I wonder what the resolution of the multi-touch is, nothing there suggests it's more than about an inch across on that big screen
18:48:09 <zpg> Raystm2: heh.
18:48:12 <zpg> Raystm2: hokey, huh?
18:49:00 <Raystm2> Hokey you say? Most functional use of a computer i've seen in a long time. hehe ;)
18:49:21 <Raystm2> Heck, people are beating on them all the time already.
18:50:46 <Quartus> I can't play that video.  I have to forward past the fucking test pattern, and every time I do that with a google video, it seizes up about 2 minutes later.
18:50:57 <Quartus> What fantastic tool recorded the test pattern?
18:51:44 <zpg> God knows. Can't you just mute and wait?
18:51:51 <zpg> Come but here for 2 mins.
18:51:56 <zpg> *back here
18:52:11 <Quartus> Maybe later.
18:52:21 * zpg nods
18:52:28 <zpg> I wonder if there's more Knuth kicking around.
18:52:34 <zpg> (video lectures)
18:52:58 * Raystm2 reminds himself of the SICP lectures still to watch.
18:53:00 <Quartus> yes, tons of classroom stuff at stanford
18:53:08 <zpg> neat.
18:53:16 <Quartus> bit tiring, actually.  He stammers.
18:53:22 <zpg> i think i'll abort this at minute 35.
18:54:28 <Quartus> hmm, Knuth is slightly younger than my father. :)
18:54:44 <Raystm2> year please?
18:54:47 <zpg> i watched a 4 part hyperscope presentation one sunday morning c. 5am to pass the time. needed to be up early, opted for an all-nighter. my advice is don't go near the damned thing.
18:54:49 <Quartus> '38
18:56:22 * Raystm2 pauses at 13:04 the first question for later consumption.
18:56:36 <Raystm2> My dad is two years older.
18:57:19 * Raystm2 googles hyperscope
18:58:06 * zpg growls
18:58:14 <zpg> none of these ddj knuth videos will play.
19:02:00 <JasonWoof> the iPhone looks amazing: http://video.google.com/videoplay?docid=6634817287115134475&q=iphone
19:02:58 * Raystm2 hates ie7. 
19:03:13 <Raystm2> it should play quicktime right out of the box darn it!
19:03:46 <JasonWoof> it should play theora out of the box
19:03:51 <JasonWoof> screw quicktime
19:04:10 <JasonWoof> I have no use for proprietary file formats
19:05:09 --- join: juri_ (i=[+RlP1mj@volumehost.com) joined #forth
19:05:29 <zpg> iphone looks pretty neat.
19:06:40 <Quartus> it's juri_, she's slumming :)
19:07:32 <juri_> slumming? i'm presently coding in perl and tintin++.. i'm touring the ritzy languages. ;P
19:07:32 <zpg> Quartus: http://www-cs-faculty.stanford.edu/~knuth/316.html
19:07:39 <Quartus> hehe. I mean coming back here. :)
19:08:24 <juri_> FWIW, i'm in #machinevision (which has NO traffic) and #openmosix, on this network. :)
19:08:40 <Quartus> sure, leave us poor Forthers all by ourselves unsupervised.
19:09:21 * zpg redefines 0 by mistake & sniffles
19:09:22 <juri_> well, the christian "stuff" really gets on my nerves sometimes. :P
19:09:25 <Quartus> zpg, Asimov wrote extensively on the bible himself, but wasn't religious.
19:09:39 <Quartus> juri_, me too.
19:09:53 <zpg> Quartus: i'm not implying Knuth is religious, or not for that matter. the book looks interesting.
19:10:07 <zpg> Quartus: and thought you might be intrigued by the link given the above discussion.
19:10:21 <Quartus> Checking it out, since that video bombed out again even while on pause.
19:12:09 <juri_> so, i'm a full-time tintin++ coder again. :)
19:12:23 <Quartus> The Rin version?  
19:12:36 <juri_> Rin? nani?
19:12:39 <Quartus> Rin tin tin.
19:12:45 <juri_> oh ha ha. :P
19:12:45 <zpg> what on earth are you two talking about?
19:12:48 <Quartus> I've never heard of tintin++.
19:12:59 <Quartus> nor even tintin.  Nor tin.
19:13:03 <juri_> its a mud^Wautomation client. ;)
19:13:18 <Quartus> parsing.... automation, ok.  Client, ok.  Together, null result.
19:13:32 <juri_> i use it to automate rollouts of symbol^Wmotorolla wireless routers.
19:13:44 <zpg> what's with this ^W business?
19:13:53 <Quartus> it's emacs humour
19:13:55 <juri_> emacs sends its love. :)
19:14:30 <zpg> i'm using emacs, but evidently am not well versed on why kill-region is appearing this frequently.
19:14:40 <juri_> ^W == "backspce over previous word".
19:14:40 <zpg> oh i mean
19:14:41 <zpg> har har
19:14:53 <Quartus> it's olde telecom humour, suggesting that the previous word was the truth, but has been deleted in favour of the subsequent one.
19:14:54 <zpg> C-w is kill-region here
19:14:58 <zpg> which is standard.
19:15:25 <zpg> Quartus: yeah I gathered it might be something of that nature.
19:15:30 <Quartus> the stupider^H^H^H^H^H^H^H^Hmore primtive version is as shown.
19:15:51 <zpg> Quartus: I had more difficulty following why symbol and mud were being deleted.
19:16:01 <Quartus> I'm not sure either, but here we are. :)
19:16:03 <juri_> symbol just changed names to motorolla today.
19:16:12 <Quartus> oh did they? Curious.
19:16:24 <juri_> tintin is a mud client. but i dont say that too loudly arround here, as i use it for automation. :)
19:16:33 <Quartus> tintinabulation.
19:16:36 <zpg> you mean you don't use emacs for mudding?
19:16:37 <zpg> gosh.
19:17:31 <juri_> zpg: i think the language would be too sequential.
19:17:38 <zpg> isn't ^W vi anyway?
19:17:42 <zpg> in insert-mode.
19:17:51 <Quartus> jeepers. It's a nerd joke, just roll with it. :)
19:17:56 <juri_> i use tintin for its "virtually stateless" scripting. ;)
19:18:03 <zpg> christ, i'm confused.
19:18:19 <zpg> let it be noted that i'm totally lost. 
19:18:26 <juri_> you have to write the state into the variables in the language, and declare them all ahead of time. there is no "present point" in the program.
19:18:49 <Quartus> But if you find the +100 pike, you can vanquish all the Kobolds in the dungeon.
19:18:52 <juri_> i like to say tintin executes programs left-to-right, instead of top-to-bottom ;)
19:20:56 <Quartus> how oblique.
19:23:54 <Raystm2> Tintin where do I turn? Tintin is this the right direction?
19:24:04 <slava> juri_: is it purely functional?
19:24:51 <juri_> slava: define the term?
19:24:56 <slava> never mind
19:25:03 <Raystm2> Use tintin to automate rollouts, but TomTom is the map of the future. 
19:25:27 <Quartus> sounds like some kind of state-table dealie
19:25:37 <Raystm2> Ad brought to you by those who use UpChuckForth, a colorful derivite.
19:27:20 <Quartus> ray's into the schnaaps again? :)
19:27:36 <Raystm2> Drambuie but I've not touched it I sware.
19:27:52 <Raystm2> Diabeties you know.
19:28:04 <Raystm2> Alcohol gives me false low readings.
19:28:19 <juri_> alcohol gets your feet cut off, dude.
19:28:29 <juri_> <-- diabeties you know.
19:28:34 <Raystm2> Dangerous when you dip your oreos in Drambuie.
19:29:04 <zpg> i think my irc client must have a virus.
19:29:07 <zpg> tea time.
19:29:36 <Raystm2> Did not know you had it, don't know you, did know that feet are in jeopardy if not taken seriously.
19:30:35 <Raystm2> Eyes too. I could bolt down to a seguay if I absolutely had to but like to see what i'm crashing into.
19:33:05 <Quartus> Just looking at the FFL from http://ffl.dvoudheusden.net/
19:34:18 <zpg> oh wow, not seen this before
19:35:15 <Quartus> Badly in need of example code, it took me several minutes to sort out how to use the random number generator.
19:35:27 <slava> factor has all of that except the AVL tree, and terminal escape sequences.
19:35:38 <Raystm2> ya, thanks Quartus. nice to know about this.
19:35:46 <Quartus> well, get coding, slava :)
19:36:07 <Raystm2> slacker! :)
19:36:29 <slava> avl trees are planned
19:38:35 <zpg> Quartus: just took at look at gforth's engine.c -- not exactly pretty reading :)
19:39:21 <Quartus> Not terribly complex, really.  It defines macros for the engine's operation, depending on the type of engine selected.
19:40:31 <zpg> Yeah, I gather more than a cursory glance is necessary.
19:40:59 <Quartus> It's quite sophisticated, so if you're just looking for a simple implementation, you might do better to look at a simpler implementation. :)
19:41:13 <zpg> This is true, I might start small.
19:41:18 <zpg> Perhaps hForth.
19:41:34 <JasonWoof> what's the terminology for the optomization of ; (or exit) where it checks if it's immediately after a call, and if so it changes that call to a branch (so it no longer pushes the rstack)
19:41:41 <Quartus> Gforth is built over gcc, and uses some pretty detailed expert knowledge of gcc to squeeze performance out of the resulting build.
19:41:43 <JasonWoof> is that "return-call optimization"?
19:41:53 <Quartus> tail-call optimization
19:41:58 <JasonWoof> thanks
19:42:47 <zpg> Quartus: yeah, that i gathered from Ertl's discussion of 'labels as values'
19:43:03 <Quartus> so it's bound to be more obscure than a naive implementation would be
19:43:10 <slava> gforth is incredibly obscure
19:43:12 <Quartus> MAF is a good study.
19:47:56 <Raystm2> juri_, if your not to terribly put off by my lack of tact and sensitivity, I'm newly diagnosed. Do you recommend a good chat?
20:00:25 <juri_> ray: no idea.
20:00:48 <juri_> my rules of thumb are avoid sugars, eat very regularly, small amounts, and pray. :)
20:01:40 <Quartus> so eat a page of the bible every 30 minutes.
20:01:50 <slava> works wonders
20:01:55 <zpg> literally.
20:01:59 <juri_> makes great rolling papers. ;P
20:02:34 <Quartus> Leviticus is especially high in fiber.
20:02:57 <juri_> yea, but job really knew how to party. :)
20:03:13 <zpg> seems like a bit of a miserable bugger.
20:07:38 <Quartus> Poor Job.
20:08:09 <zpg> quite.
20:12:10 * Raystm2 starts with 3:16
20:12:33 <zpg> ahh, may the knuth be with you.
20:13:02 <Raystm2> And also with you.
20:13:51 * zpg bows reverently
20:14:53 <Raystm2> Bless me father. May it do you well, 9 9 9 lit if ." Number" . lit if ." Number" . lit if ." Number" . then then then ;
20:16:14 <JasonWoof> oh, that's rich
20:16:25 <Raystm2> very nearly Beatles in colorForth. the WhiteText album. 
20:16:37 <JasonWoof> I made a new ; that does tail-cal optomization
20:16:45 <Raystm2> neat. :)
20:16:47 <JasonWoof> but it crashed
20:16:47 <zpg> in herk/fronds?
20:16:54 <JasonWoof> yeah, for fronds
20:17:16 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-134-234.pools.arcor-ip.net) joined #forth
20:17:18 <JasonWoof> up until recently none of my loops overflowed the return stack, so I ignored it for a while
20:17:30 <zpg> Raystm2: you might want to scrap your "lit"s
20:17:35 <JasonWoof> my bug was that I forgot to make it an immediate!
20:18:10 <slava> java factor's tail call optimization only worked on and off, and i never noticed because it just wasn't fast enough to execute a loop which could fill up the return stack in any reasonable length of time
20:18:22 <slava> and java has a growable stack with a max size of 10mb or something
20:18:30 <Raystm2> zpg and what do you know of cf? hmm? oh yeah it's ?lit at that doh! 
20:18:43 <zpg> how can that be cf?
20:18:46 <zpg> you're using ." ..."
20:19:12 <JasonWoof> heh, my return stack holds 1024 items or so
20:19:41 <JasonWoof> I like my forth
20:19:48 <JasonWoof> I can write working code even when I'm tired
20:19:50 <slava> factor's default call stack sizes are 64kb
20:19:54 <slava> s/call//
20:20:02 <JasonWoof> ok, now what did I want tail-call optomization for again?
20:20:08 <slava> no reasonable code overflows that
20:20:12 <Raystm2> I defined ." ( sticks out tongue ) where ya been? huh? It's actually s:  but you all know it as ." and it was defined by Roman ( searching for last name starts with P ) author of the windows versionl.
20:20:58 <JasonWoof> I heard someone say that they profiled a large body of their code and found that it was rare that either stack had more than 16 items
20:21:35 <JasonWoof> so I figured 4K for each stack aughta be perfectly reasonable
20:21:38 <JasonWoof> or maybe tathi decided
20:21:42 <Quartus> Probably fine.
20:21:45 <zpg> Raystm2: :) good stuff
20:21:49 <JasonWoof> it's the smallest amount of memory that you can mprotect
20:24:51 <Raystm2> Pavlyuk. Roman Pavlyuk.
20:25:46 <zpg> does this code actually do what Drake claimed his did?
20:25:51 <zpg> i.e., work properly?
20:25:59 <Quartus> pacman
20:26:13 <zpg> tetris  ok
20:31:22 <Quartus> Raystm2 is away; he's printing strings on the screen for the FIRST TIME EVAR.
20:32:33 <zpg> who needs to prints strings when you can write cursive?
20:34:34 <JasonWoof> dejavu
20:34:43 --- quit: jackokring ("What?! Open source isn't good enough for you? Bersirc 2.2 [ http://www.bersirc.org/ - Open Source IRC ]")
20:35:23 <JasonWoof> nope, I want Free Software
20:35:39 <JasonWoof> irssi all the way
20:36:02 <slava> i don't care about using all free sfotware anymore
20:36:18 <JasonWoof> I'm fine with using some that's just open source
20:36:32 <slava> i don't draw a distinction
20:36:33 <Quartus> I have never found software religion.
20:36:36 <JasonWoof> but it's still not good enough for me
20:36:41 <JasonWoof> :)
20:37:05 <slava> time is worth money too, and if a free alternative is inferior, then i'm wasting money
20:37:13 <JasonWoof> I dig the philosophy stuff on gnu.org pretty strongly
20:37:19 <Quartus> Apparently tremendous excitement can be derived from GPLv3, or discussions of temples vs. bazaars.  But not for me.
20:37:54 <JasonWoof> GPLv3 is important I think, but not exciting
20:38:07 <slava> i oppose drm, but not closed source
20:38:17 <JasonWoof> I'm in favore of bizare development style
20:38:37 <Quartus> well, I don't know.  I only know I don't have whatever virus makes you all passionate about insisting that everything be free.  Stallman retrovirus A.
20:38:50 <JasonWoof> I've got my fuzzy purple hat on and my Goofy sweatshirt
20:38:59 <slava> i think linux is heading in the wrong direction anyway
20:39:13 <zpg> slava: howso?
20:39:46 <JasonWoof> I don't know where linux is headed
20:39:48 <slava> well, i was helping a friend install fedora the other day, and i was amazed and how much stuff just plain doesn't work, or needs excessive configuration to get into a usable state, etc
20:39:58 <JasonWoof> but I'm frustrated that Xfree86 doesn't seem to be getting replaced
20:40:12 <slava> unix, x11 and c need to die
20:40:20 <JasonWoof> slava: yeah, a lot of distros are a pita
20:40:25 <slava> all distros are a pita
20:40:27 <JasonWoof> some are easy though
20:40:33 <slava> if you find ap roblem in one distro, a linux advocate just tells you to try another distro
20:40:35 <zpg> slava: yes, that's redhat though. the ubuntu team aim to hide all that stuff behind the WM of choice.
20:40:38 <JasonWoof> I wouldn't say all
20:40:43 <JasonWoof> ubuntu is easy
20:40:48 <slava> then you find a problem in that distro, and the same advocate tells you to try a different one
20:40:52 <JasonWoof> knoppix is easy as hell. I just burned the CD and it worked
20:40:55 <slava> with this same friend, before fedora we tried ubuntu on his laptop
20:40:56 <slava> buggy as hell
20:41:04 <slava> these guys do no testing whatsoever, they're a bunch of teenagers
20:41:12 <zpg> heh.
20:41:17 <Quartus> Knoppix does tick along pretty nicely.
20:41:29 <slava> zpg illustrates that nicely, he told me to try ubuntu after i mentioned trying fedora
20:41:33 <JasonWoof> slava: easy there killer
20:41:34 * TreyB swears by Gentoo, but has done UN*X for so long that he *likes* compiling.
20:41:40 <slava> but if i try ubuntu again and it has a problem, zpg will tell me to try mandriva
20:41:47 <slava> life's too short to be messing around with an os
20:41:54 <JasonWoof> slava: so did I. that's because ubuntu is trying to do what you want
20:41:55 <slava> an os is there to run applications, not to amuse a bunch of teenagers
20:42:14 <slava> who think c and manual memory management and endless configuration and buggy drivers are good
20:42:26 <JasonWoof> what's this crap about teenagers?
20:42:37 <zpg> slava: no i won't, i don't use linux. i'm merely suggesting that redhat always seems to have been ill thought out and configured. and that ubuntu claim to address that. the distribution thing -- and configuring every component 900 different ways -- drove me nuts. hence no linux here.
20:42:37 <Quartus> yeah, most of them are old guys like me at this point. :)
20:43:02 <slava> i run ubuntu on my amd64 box and it works reasonably well
20:43:05 <slava> except the gui updater tool doesn't work
20:43:08 <slava> and except a few other issues
20:43:12 <slava> but on my friend's laptop it was totally busted
20:43:18 <JasonWoof> slava: there are distros that you don't have to touch configurations. like knoppix and ubuntu and now gNewSense
20:43:29 <slava> see? the 'try another distro' thing again
20:43:43 <JasonWoof> sure, not everything works.
20:43:47 <slava> in 2 months gNewSense will die and you'll have to reinstall it again because the teenagers will ahve moved on and won't make security fixes for it
20:43:53 <JasonWoof> on the computer downstairs, we don't get 3d accelleration with gNewSense
20:43:56 <Quartus> I ran SLS many years ago, installed off floppy, and had surprisingly little trouble with it.  Lucky, I guess.
20:43:58 <JasonWoof> but we do with knoppix
20:44:15 <slava> debian is the only distro i've had good experience with
20:44:16 <Quartus> wait, SLS?  Yes, I think so.  Slackware was in there too at some point.
20:44:22 <slava> but debian has its own problems
20:44:25 <JasonWoof> slava: what are you compairing these linux dostros to?
20:44:26 <slava> installing all the plugins/codecs for multimedia takes ages
20:44:29 <slava> JasonWoof: mac os x
20:44:32 <zpg> everyone has their own problems.
20:45:07 <JasonWoof> slava: ok. see, most people use windows, which seems to me to be less functional than stock linux installs on random hardware
20:45:24 <slava> i don't care about having 32482 window managers, i only want one that lets me write code and stays out of the way
20:45:30 <Quartus> I had a customer once who could barely get a Palm to work, and failed to ever keep one working.  I don't know what the blazes he did to them, but he had an endless story of failure to tell.
20:45:44 <slava> i know how to use linux
20:45:51 <slava> i have servers that stay up for hundreds of days serving data
20:45:57 <Quartus> I'm not suggesting you don't, just that luck seems to bunch up for some people, good and bad.
20:46:02 <slava> but for a development workstation os x is superior
20:46:15 <JasonWoof> perhaps
20:46:20 <JasonWoof> apple makes good stuff
20:46:28 <JasonWoof> it just doesn't suit me
20:46:52 <Quartus> Maybe it's that I don't give a hoot about graphics accelerators.
20:46:55 <JasonWoof> I want to get hardware that works well with freeBSD and run that
20:47:12 <JasonWoof> I don't like apples gui
20:47:19 <JasonWoof> it's better than most, but it's not what I want
20:47:38 <Quartus> I'm happiest at a command line. :)
20:47:44 <JasonWoof> yeah
20:47:46 <zpg> obviously you get than in osx
20:47:48 <JasonWoof> I've tried lots of file managers
20:47:53 <JasonWoof> and I still don't get the point
20:48:02 <slava> i use the finder occasionally
20:48:06 <JasonWoof> I get that they're useful for people who don't know how to use the commandline
20:48:08 <slava> mostly i do file maangement in the command line
20:48:11 <JasonWoof> or can't type decently or something
20:48:19 --- quit: tathi ("leaving")
20:48:21 <JasonWoof> but for me, they aren't any use
20:48:22 <Quartus> I can certainly get from 0 to a command line with every distro of linux I've ever used.
20:48:23 <TreyB> It will become more and more difficult to find machines without fancy graphics accelerators.  Eventually we'll drop access to the frame buffer altogether.
20:48:44 <slava> i don't care about framebuffer access
20:48:50 <slava> that's for colorforth to worry about
20:49:25 <Quartus> I mean I'm happy when there's a display.  I don't care if the framerate isn't ideal, or what shade of beige the clock is.
20:50:11 <slava> heh
20:50:17 <Quartus> So I may completely miss subtle nuances that would lead somebody else to proclaim the entire distro broken.
20:50:43 <slava> well, something like the dvd burner not working at all is not a subtle nuance
20:50:45 <TreyB> Apple and MS both have resolution-independent UIs on the horizon, which bodes well for the aging population of developers.
20:50:53 <Quartus> If the mouse didn't go, or I couldn't start X (and I needed X, which I dont always), or if the network failed, or I couldn't read/write the drives, then sure.
20:51:13 <slava> in fedora on my friend's laptop, we installed the ntfs driver (which was not bundled)
20:51:17 <slava> then we plugged in his usb hard disk
20:51:26 <slava> the os auto-detected it and tried to mount it, which amazed me actually
20:51:32 <slava> then it proclaimed that the ntfs file system was not supported
20:51:37 <slava> in the command line, manually mounting it works
20:51:38 <Quartus> Last I touched down, the ntfs stuff was read-only and still somewhat broken.
20:51:45 <zpg> ditto
20:51:46 <slava> but the hald or whatever fedora uses to detect hotplugging doesn't mount it
20:51:52 <slava> Quartus: its read/write now
20:51:55 <Quartus> probably the driver returns warning messages.
20:52:11 <Quartus> Is it in release? I thought it was still wobbly, at any rate.
20:52:13 <slava> probably, but this type of thing should work
20:52:25 <slava> usb has been around for what, 10 years now?
20:52:27 <slava> ntfs is not rocket science
20:52:41 <Quartus> well, failure to automount a pre-release driver wouldn't freak me out.  ntfs is apparently pretty hard to do, they've been at it for aeons.
20:52:42 <JasonWoof> yeah, linux dosen't do everything
20:52:45 <slava> its not pre-release
20:52:46 <Quartus> Read isn't tough. Write is.
20:53:01 <slava> there's a done driver, its just not bundled with fedora
20:53:12 <Quartus> ok. Maybe they have some reason why.
20:53:19 <JasonWoof> :)
20:53:42 <Quartus> I wouldn't format an external USB drive with NTFS, myself.
20:54:14 <JasonWoof> I wouldn't format any drive NTFS
20:56:01 <Quartus> anyway if full ntfs support in Linux is available now, it's pretty recent.  Let me see.
20:56:02 <slava> the fedora extras repository has an rpm for the nvidia vidoe card driver
20:56:02 <slava> with a hardcoded dependency on the i586 kernel package
20:56:02 <slava> installed the i686 kernel package? you have to install the driver yourself
20:56:02 <slava> the very SAME driver
20:56:02 <slava> except some kid fucked up the rpm
20:56:02 <slava> things like this is why i talk about teenagers, because its exactly how teenagers write code
20:56:35 <Quartus> oh, trust me, 40-year-olds fuck up procedures too :)
20:57:16 <slava> well my friend is a former windows user, are you going to tell him to reformat everything?
20:57:16 <slava> what if he needs to transfer files from his windows machine?
20:57:16 <zpg> night all.
20:57:16 --- quit: zpg ("bon nuit")
20:57:21 --- quit: crc (zelazny.freenode.net irc.freenode.net)
20:57:21 --- quit: slava (zelazny.freenode.net irc.freenode.net)
20:57:48 <Quartus_> suffering a netsplit over there
20:57:50 --- join: crc (n=crc@pdpc/supporter/active/crc) joined #forth
20:57:50 --- join: slava (n=slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
20:57:50 --- mode: irc.freenode.net set +oo crc slava
20:57:58 <Quartus> netsplit.
20:58:31 <Quartus> slava, if the ntfs support isn't ready for prime-time, then yes, he's stuck.
20:58:49 <slava> the ntfs support works. its the usb hotplug that's broken when it comes to loading the correct file system module
20:58:58 <JasonWoof> why would he need read/write to transfer his files?
20:59:01 <Quartus> he should rescue his data from ntfs using Windows, and move the data onto something that both can read reliably. 
20:59:25 <slava> even if you mount it read only the same problem will occur with the hotplug daemon
20:59:40 <slava> its not the read/write that is the problem
20:59:45 <Quartus> so what is the problem?
20:59:45 <JasonWoof> I don't know anything about the hotplug stuff
20:59:48 <JasonWoof> except that it's improving
20:59:56 <slava> Quartus: hotplug doesn't load ntfs automatically
21:00:02 <Quartus> right. Why?
21:00:03 <slava> JasonWoof: its always improving, but never really works
21:00:08 <slava> i don't know, because its buggy c shit
21:00:17 <Quartus> ok.
21:00:20 <JasonWoof> as apposed to windoze?
21:00:27 <JasonWoof> which is buggy c shit
21:00:32 <slava> windows has similar issues
21:00:38 <slava> so does os x, except it has less in my experience
21:01:17 <JasonWoof> mac os x is superior in many ways
21:01:26 <JasonWoof> but I still won't install it
21:01:31 <JasonWoof> I hate the gui
21:01:40 <slava> i'm not trying to convince you to install it
21:01:42 <JasonWoof> and I want to be able to configure everything
21:01:49 <slava> i don't
21:01:54 <JasonWoof> don't then
21:01:57 <slava> in fact, too much configurability is detrimental to stability and performance
21:02:07 <slava> you can run x11 on mac os, but that defeats the purpose
21:02:12 <JasonWoof> dunno about stability
21:02:23 <Quartus> is it ntfs-3g under Fedora 6?
21:02:49 <JasonWoof> my computer crashes every couple months or so
21:02:54 <JasonWoof> and it's over 6 years old
21:02:58 <JasonWoof> I think I'm doing fine with stability
21:07:40 --- quit: slava (zelazny.freenode.net irc.freenode.net)
21:07:40 --- quit: crc (zelazny.freenode.net irc.freenode.net)
21:08:02 --- join: crc (n=crc@pdpc/supporter/active/crc) joined #forth
21:08:02 --- join: slava (n=slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
21:08:02 --- mode: irc.freenode.net set +oo crc slava
21:18:37 <JasonWoof> I like working with my own code!
21:19:01 <JasonWoof> I accidentally spit binary data to my vm terminal
21:19:12 <JasonWoof> and what do you know? it doesn't totally (*&@#$( up my terminal
21:19:26 <JasonWoof> it just displays some symbols
21:19:27 <Quartus> But it should.  Fix that.
21:19:34 <slava> presuambly your terminal doesn't have state like a vt100 does
21:23:42 <JasonWoof> sure as hell doesn't
21:23:45 <JasonWoof> hate that crap
21:23:56 <JasonWoof> I don't think programs should be able to break your terminal
21:24:00 <JasonWoof> but maybe that's just me
21:24:05 <slava> state should be minimized
21:24:07 <Quartus> where's the spirit of adventure?
21:24:25 <JasonWoof> I also don't think that apps should be able to take sole posession of your keyboard
21:25:33 <slava> in what instance does this occur?
21:26:26 <JasonWoof> when you run fullscreen games
21:26:46 <JasonWoof> probably fullscreen anything
21:26:59 <JasonWoof> but it's the crappy fullscreen games that crash and leave me without a functioning monitor, mouse, or keyboard
21:27:28 <TreyB> Full-screen anything presents a problem when running un-trusted code on a machine.
21:27:43 <JasonWoof> only because the system is badly designed
21:28:02 <slava> all code is untrusted
21:28:08 <TreyB> OS types that deal with this sort of thing usually reserve a Secure Attention Key (or combination) that the app can't catch.
21:28:23 <JasonWoof> eg I don't think it's a problem on mac os
21:28:31 <JasonWoof> and in windoze you can cope
21:28:51 <JasonWoof> in linux you have to go to another computer, and log in remotely to get your keyboard back on
21:29:03 <TreyB> Generally, such OSes don't permit full-screen display because it allows the app to trick the user into compromising passwords and such.
21:29:07 <JasonWoof> of course
21:29:22 <JasonWoof> I want my window manager to get first crack at key presses no matter what
21:29:34 <JasonWoof> I don't give a crap if something's running fullscreen
21:29:40 <JasonWoof> when I press F2 I want a terminal
21:29:45 <JasonWoof> no questions asked
21:29:51 <Quartus> But, F2 is hyperspace.
21:30:33 <JasonWoof> when I press ctrl-alt-f2 or whatever it is I want to get out of x and go to the terminal that the kernel supplies
21:30:45 <JasonWoof> but, the way X is designed, these only work sometimes
21:31:25 <JasonWoof> fronds now makes a new word and inserts it into the source when you press a key
21:31:28 <JasonWoof> fancy pantsy
21:31:49 <JasonWoof> now I just need to make it so you can make words that are more than one letter long
21:32:02 <JasonWoof> heh heh heh
21:33:15 <JasonWoof> well... not "just" that
21:33:45 <JasonWoof> but that would be nice
21:33:56 <JasonWoof> and dictionary lookup to, so you can use existing words
21:34:04 <JasonWoof> s/to/too
21:34:13 <Quartus> don't forget to huffman encode all your names to 28 bits.
21:34:22 <JasonWoof> doh!
21:34:42 <TreyB> feh
21:34:46 <JasonWoof> how could I have forgotten!
21:35:13 <JasonWoof> and there I was thinking I wouldn't have to bother, since word names are not used at compile time
21:35:52 <JasonWoof> see, what I've done here is optomized my datastructures for performance!
21:36:01 <Quartus> again with the performance!
21:36:34 <JasonWoof> I store my strings in raw, uncompressed form for the fastest possible accessing and display
21:37:02 <JasonWoof> thus saving the editor precious nanoseconds every time I look at the source code!
21:37:09 <Quartus> fool! do you think memory is free?
21:37:26 <Quartus> what, you need longer than 5 character names?  Kids today!
21:37:44 <slava> i remember fox saying something about below average IQ!
21:38:10 <JasonWoof> ooh! I can delete this crap I inserted :)
21:38:15 <JasonWoof> man, this editor is going to take a long time
21:38:20 <JasonWoof> I don't think I like writing interfaces
21:38:31 <Quartus> slava, yes, apparently I'm an idiot compared with Jeff Fox.  
21:38:45 <JasonWoof> LUDDITE!
21:39:02 <Quartus> Hey, if you make all your word names from the letters r t o e a n and i you can squeeze seven whole characters in.
21:39:22 <JasonWoof> neat
21:39:32 <JasonWoof> rotten
21:39:53 <JasonWoof> arrrrr
21:40:02 <JasonWoof> yeah, you can be quite expressive ;)
21:40:57 <Quartus> so trainee, iterate, intertia, and terrain. :)
21:41:03 <Quartus> oops. inertia.
21:42:08 <JasonWoof> wow, now doesn't that just eat your breakfast?
21:48:06 <Quartus> narrate, aerator, tearier, neonate, entreat, retreat, rainier, retrain, nattier, and nitrate.
21:48:36 * JasonWoof boggles
21:49:46 <Quartus> That may not be a complete list, but it's pretty close.  
21:51:46 <JasonWoof> I just  tried an internet anagrammer thing and it couldn't find anything will all the letters once
21:51:50 <JasonWoof> guess that's a lot of vowels
21:52:12 <JasonWoof> anyway, I'm off to bed
21:52:16 <Quartus> cya!
21:52:18 <JasonWoof> thanks for hangin'
21:52:23 <Quartus> anytime.
21:52:23 --- quit: JasonWoof ("off to bed")
23:59:59 --- log: ended forth/07.01.10