00:00:00 --- log: started retro/10.09.07
07:53:48 --- join: docl (~luke@97-120-241-117.ptld.qwest.net) joined #retro
12:44:08 <crcx> docl: any instructions on the trigger code you posted yesterday?
13:20:36 <docl> no, I was just kind of randomly typing stuff at work and wanted to make sure I had access to it later
13:20:57 <docl> the idea though is to be able to make spaces optional
13:25:57 <docl> the character dict code I pasted previously might be a better starting point. you could insert it into the accept loop and make each trigger have an on/off boolean. I should check through my old code, I think I had something like that almost written.
13:46:04 <docl> as of the latest rx-core, !@foo does not seem to work.
13:48:34 <docl> were nested prefixes removed?
13:49:24 <crcx> yes
13:49:42 <crcx> they were being compiled incorrectly in some cases
13:51:21 <docl> ah, ok
13:52:10 <docl> I liked having !@ for pointers, but it's not a huge deal.
14:00:23 <crcx> I hope to add them back soon
14:03:11 <crcx> or we could do a new prefix for the !@ action...
14:21:11 <docl> yeah. but we will eventually run out of combinations, won't we?
15:15:21 <docl> : __, .data ` tuck ` @ ` ! ` ++ ; immediate
15:16:53 <docl> this lets you comma something to any location where you have a pointer variable. (e.g. ,heap is the same as , )
16:21:54 <docl> I don't know if comma is the best character to use. Maybe > would be better?
16:27:16 <docl> : __> .data ` tuck ` @ ` ! ` ++ ; immediate
16:27:16 <docl> : __< .data ` dup ` -- ` @ ` @ ; immediate
16:27:16 <docl> 10 20 >heap >heap
16:27:17 <docl> <heap <heap .s
16:29:26 <docl> now, if we wanted to get really fancy we could use duck typing to make it cover linked lists as well
16:33:19 <docl> but to do that you'd need to store the flag somewhere in the data structure. not as simple.
16:58:18 <crc> if I ever get around to adding unicode support to the vm, we'd have plenty of prefixes...
17:03:14 <crc> with the new vocabularies, we can repurpose prefixes a bit by altering search order
17:04:04 <crc> though if this is done a lot, it could get confusing
17:05:08 <docl> http://gist.github.com/569363 duck typed lists
17:07:06 <crc> doesn't work in definitions
17:11:58 <crc> scratch that; I typed it in wrong :(
17:34:13 <crc> btw, I expanded your parable-in-retro code a bit (added dip, loop, and loopi, and did a couple of reorganizations): http://gist.github.com/569391
17:53:31 <docl> nice :)
17:53:48 <crc> is this ok to include in the 10.7 release?
17:53:54 <docl> sure
17:53:59 <crc> thanks :)
17:57:45 --- join: roarde (~sixforty@pdpc/supporter/active/sixforty) joined #retro
18:17:15 <crc> hi roarde 
18:17:35 <roarde> good evening
20:50:48 <roarde> 10.09.04 13:30:49 <crc> foucist: 10.7 breaks compatibility with the 10.6 release; it's not a bugfix release :)
20:50:59 <roarde> how do I catch a break here?
21:01:31 <foucist> crc: : __< .data ` dup ` -- ` @ ` @ ; immediate
21:01:52 <foucist> i've completely forgotten forth code so i'm gonna go ahead and say it looks retarded
21:02:00 <foucist> how about
21:02:15 <foucist> __< do ... end
21:02:31 <foucist> ` and @ and ! and ; all look retarded :P
21:02:43 <foucist> also you have 'immediate'
21:03:10 <foucist> if you were gonna be like chuck moore you'd call it 'imm' or some crazy symbol
21:04:16 <foucist> actually, thinking about it, . would be a good symbol for dup
21:05:02 <foucist> rather than display or w/e it is
21:06:27 <foucist> realistically the forth language vocabulary needs to be completely evaluated and changed to something much more readable
21:06:52 <foucist> no matter how "readable" it may be to you, it can always be MORE readable
21:06:55 <foucist> wayyyy more
21:06:57 <roarde> logo?
21:07:11 <foucist> roarde: ruby
21:07:38 <foucist> roarde: ruby is the MOST readable programming language that is also terse and high-level
21:07:56 <roarde> so, why use a forth rather than ruby?
21:08:12 --- quit: crc (Ping timeout: 240 seconds)
21:08:15 <docl> forth is nice and low level :)
21:08:27 <foucist> forth is also high level
21:08:45 <foucist> docl: stop wallowing in the low level and build forth to a high level :P
21:08:49 <roarde> foucist: whose forth is high-level?
21:09:04 <docl> forth is flexible
21:09:05 <foucist> roarde: forth itself is both low level and high level.  forth is like lisp
21:09:28 <docl> foucist: have you looked at parable?
21:09:30 <foucist> roarde: it's fully extensible
21:09:36 <foucist> docl: nope
21:09:55 <roarde> cool, foucist. so extend it for readablility, already
21:10:05 <docl> crc is renaming and rewriting a lot of the words in it
21:10:18 --- join: crc (~charlesch@184.77.185.20) joined #retro
21:10:24 <docl> wb crc
21:10:26 <foucist> roarde: i'm talking about renaming the actaul core vocabulary, dropping all the lame names etc
21:10:55 <docl> maybe we should prototype a no-lameness retro and see how it looks :)
21:11:00 <foucist> roarde: also fat chance, i'm not a coder
21:11:23 <docl> http://gist.github.com/569391 is my reinterpretation of parable, as fixed up by crc
21:11:39 <roarde> I'm gonna drop all the names and just use symbols, like the artist formerly known as the artist formerly known as Prince used to do.
21:12:07 <docl> but I've been working on better words for custom stacks. which is what you mocked above
21:12:44 <foucist> docl: what's it? a DSL for parable?
21:12:49 <foucist> google isn't showing any parable programming language
21:13:14 <docl> parable is a language crc has been writing, based on toka
21:13:15 <docl> http://rx-core.org/dev/parable.fossil/index
21:13:23 <docl> which is also something crc wrote
21:13:37 <docl> the difference being parable is written in retro
21:13:42 <docl> toka was written in c
21:13:44 <foucist> docl: i wasn't mocking the __< part, i was mocking forth itself, ` @ ! dup and : ; too
21:14:06 <docl> even the colon, eh?
21:14:13 <foucist> yep
21:14:18 <docl> well in parable colon is only a prefix for making a symbol
21:14:27 <docl> so you'd do :foo [ ... ] def
21:14:56 <foucist> UGH, camel case?!
21:14:57 <docl> the brackets make a quote or code block, and def links them together.
21:15:13 <foucist> to_number guys..  or even to_n
21:15:22 <foucist> ruby has "foo".to_i for integer
21:15:24 <foucist> etc
21:15:33 <foucist> to_f, to_a (array), to_h (hash)
21:15:57 <foucist> sorry for bitching ;)
21:16:49 <foucist> i see a heavy heavy influence of forth on parable
21:19:36 <foucist> i suppose i could attempt to design something like parable with a strong ruby influence
21:19:46 <roarde> what's comparable to retro, but sits still long enough to grok?
21:21:33 <foucist> roarde: based on my knowledge of the forth world from 2006-2007, i would say that retro is the most forth-like of the "modern" PC forths that are cross platform, etc
21:23:33 <roarde> foucist: k, but if it were a child, I'd bypass ritalin and go straght for a heavy mallet and quick-set cement
21:24:24 <roarde> maybe an old machine forth
21:26:22 <docl> ouch
21:28:21 <foucist> roarde: what are you looking from forth?
21:28:26 <foucist> what are your goals
21:28:50 <docl> do ... end :foo <- is this good?
21:29:53 <roarde> adding to my repertoire which is rusty and includes only "copy con > myprog.com"
21:30:07 <foucist> docl: well ideally  how about def foo  ... end like ruby
21:30:37 <foucist> roarde: and what do you want to do with your repertoire?
21:31:42 <foucist> roarde: what about using linux?  now there's a repertoire..  cat something | sort | uniq | awk '{print $1}' | sed 's/foo/blah/g' > newfile
21:31:53 <roarde> foucist: the decision to use forth next is already made. Just have to find one that'll stand still for a month
21:32:16 <foucist> roarde: retro will stand still for a month :P
21:32:26 <foucist> roarde: you don't always have to use the absolute latest version
21:33:05 <foucist> just use 10.6 and then bug crc for all support and then refuse to upgrade when he suggests that
21:33:22 <roarde> foucist: If I'm sharing, viable options are: 1. use the latest  2. fork
21:33:31 <foucist> sharing what?
21:34:07 <roarde> is there anyone around with more answers than questions?
21:34:11 <roarde> don't look at me
21:35:34 <foucist> docl: what's the state of colorforth these days?
21:35:48 <foucist> docl: do ppl run it in windows/linux/osx?
21:35:53 <foucist> or just in vm?
21:36:18 --- quit: crc (Remote host closed the connection)
21:36:20 <roarde> foucist: see arrayforth. windows or a few native machines only.
21:36:33 --- join: crc (~charlesch@184.77.185.20) joined #retro
21:36:40 <roarde> vm's don't even work reliably
21:43:01 <foucist> ah
21:44:16 <docl> : alias create @last !d->xt &.word reclass ;
21:44:16 <docl> &: alias def
21:44:16 <docl> &; alias end &.macro reclass
21:44:16 <docl> def foo .s end
21:45:30 <docl> foucist: it's pretty trivial to redefine words ;)
21:47:35 <docl> ampersand gets the xt of a word. kind of like ' used to. ' is currently used for getting a character. yeah I guess it does look a little ADHD
21:52:11 <roarde> I need serious professional help. I keep falling for this crap
21:52:42 <roarde> yet, I'm off to try installing plan 9 from today's "latest iso"
21:52:57 --- quit: roarde (Quit: and don't dare mention p9p!)
22:04:36 --- join: roarde (~sixforty@pdpc/supporter/active/sixforty) joined #retro
22:06:22 <docl> foucist: we can rename the words easily enough... the question is whether we really gain anything by that
22:08:01 <docl> I suspect we would...
23:39:26 --- quit: docl (Ping timeout: 264 seconds)
23:59:59 --- log: ended retro/10.09.07