00:00:00 --- log: started forth/06.09.20 00:34:29 --- join: koaftder (n=koft@cpe-071-077-035-155.nc.res.rr.com) joined #forth 00:34:36 --- part: koaftder left #forth 01:57:02 --- join: futhin (n=wunderwa@bespin.org) joined #forth 01:57:02 --- mode: ChanServ set +o futhin 02:52:18 --- quit: TreyB (Read error: 110 (Connection timed out)) 03:11:28 --- quit: JasonWoof ("off to bed") 04:26:13 --- join: PoppaVic (n=pete@0-1pool67-185.nas22.chicago4.il.us.da.qwest.net) joined #forth 04:38:51 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 04:43:57 --- quit: uiuiuiu_ (Remote closed the connection) 04:47:21 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth 04:50:58 --- quit: Quartus__ (Read error: 60 (Operation timed out)) 04:59:02 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 05:04:07 --- quit: uiuiuiu_ (Remote closed the connection) 05:05:52 --- join: TreyB (n=trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 05:34:22 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 05:39:27 --- quit: uiuiuiu_ (Remote closed the connection) 05:49:03 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/") 06:16:20 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth 06:18:19 Good morning. 06:18:39 howdy 06:20:00 What do ya know, PoppaVic? 06:20:20 I know it's still early.. And, hell - it's chilly outside. 06:21:41 Ray_work: rereading Thinking Forth (the color pdf).. It's sorta' useful. 06:22:39 Yes, indeed. You know, I've never finished that book. Thanks for reminder. 06:23:12 I fear digging out my hardcopy, and the pdf is sorta' convenient in all ways except bathroom-runs 06:24:46 I found a few interesting sections for rereading, and I've been trying to follow 2 of them, while the third is one of those "things that make you go hmmm" 06:26:58 Ya, I always find something I miss. The book is 'denser' ( for want of a better word) then it appears on a first scan thru. 06:28:00 it's more universal than it appears, or it's predecessor 06:28:39 I've been trying to work up a "functional spec", using it. Before I approach a "design spec" 06:29:02 But, it also stirs up some ideas. 06:29:06 Sounds like a good idea. 06:29:08 ya. 06:31:53 I was getting peeved with one page - 100 of the pdf - before I saw the recant-footnote, and then sorta' been glaring at it's ideas since 06:31:59 110, I mean 06:33:43 hehe /me makes note to look that up when I get home. 06:33:59 What's the upshot? 06:34:32 He was declaiming against Objects - and ranted awhile before the footnot that admits his exposure was some one guys Forth-package. 06:34:48 ..and then he suddenly realized that he had not seen alternatives 06:35:00 Oh yeah, sure. He ( Brodie) said something about that in the forward. 06:35:15 right, it links to that 110 page 06:35:19 "Please disregard my rant on page..." 06:35:24 righto 06:35:24 I see. 06:36:10 I guess, since writing that, he had seen a good implementation of ooForth. 06:36:22 Personally, I'm thinking of these "Objects" and I can envision a foothold in wordlists, Vocs, and such... Which makes me wonder if the vocstack is... sensible. 06:36:46 brb - lemme' recycle the ISP 06:36:48 --- quit: PoppaVic ("Pulls the pin...") 06:38:45 --- join: PoppaVic (n=pete@0-1pool73-219.nas24.chicago4.il.us.da.qwest.net) joined #forth 06:39:03 Where was I? 06:49:02 You were ... "Objects" and I can envision a foothold in wordlists.... 06:49:18 wonder if the vocstack is ... sensible. 06:50:08 ahh 06:50:28 Yeah, acting like "namespace", but not exactly the same 06:52:22 Retroforth has added some voc stuff, along with Quartus' ANS section: -- /section , and the local namespaces loc: ;loc. 06:53:12 Yeah, I'm not sure what they meant, end up with, or why... I'm just thrashing the concept. 07:05:29 I've been playing with those. They do seem to make things a bit more convenient. Allowing me to hide some words from the user and such. 07:06:02 And we should.. a flat namespace is a god-AWFUL thing. 07:09:55 Ray_work: I'm even toying with offloading some 'words'/headers into databases. Internalized-access might qualify as an example. 07:10:21 I really, really despise :noname 07:14:56 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 07:20:01 --- quit: uiuiuiu_ (Remote closed the connection) 07:36:29 * Ray_work sorry about delay, getting busy here... 07:36:48 ya. I've never used :noname, but then i'm still sorta noobish. 07:37:41 Ray_work: it's just an oddity 07:38:18 Years and years ago, there was a HUGE debate on "state-sensitive words"... I've never seen much of a solution. 07:38:44 ..they'd also debated flow-words and ds/rs to death 07:49:00 Does anyone here do SuDoKu? 07:49:17 * PoppaVic sighs 07:49:18 No 07:49:50 Does programming or masturbation or sex count? 07:50:37 I thought about "doing" the Kama-Sutra, but I ain't too limber and pages are not forgiving. 07:50:55 No :-) My question relates specifically to soluiton strategies, in particular, if some puzzles actually require back-tracking. 07:51:04 ewwww 07:51:22 Can't you just jerkoff or have a gf? 07:51:37 Not back-dooring. Get a grip ;-) 07:51:46 yuck 07:52:22 I got that grip, a wife, AND the back-door - I just don't "do foreign & puzzles" 07:52:52 It strikes me as another silly Fad 07:54:22 It seems like a relatively straight forward logic puzzle to me. I've done a couple of "hard" puzzles over the last couple of days, and none of them required back-tracking although one did have multiple solutions. 07:54:22 My Ma' just sits there reading Romances and doing Xword puzzles.. I'd like to avoid that mindset 07:55:26 TreyB: sounds like you might just enjoy a Boolean Algebra class 07:55:46 I've taken one already. 07:55:55 Or, maybe Matrix Algebra (eww) 07:56:30 For the record, a "hobby" is not Doing Puzzles. 07:57:01 If it were, then "taking a shit" would be akin to "taking a break" ;-) 07:57:43 I don't do them for a hobby. In fact, I never bothered until yesterday. 07:57:50 hmm, otoh... employers often see those as... NM 07:58:30 TreyB: yeah, I pass... It's too similar to vegetating with a puzzle-book or a romance. 07:59:55 Maybe reading a SF/mystery or watching a same or "action film" is the same - but, in those cases I am TRYING to "shut off forebrain" 08:00:38 it's weird, what folks think is "entertainment" or "hobby" 08:01:23 I find calculating/guesstimating, reloading and shooting "a relief". 08:15:18 --- quit: PoppaVic ("Pulls the pin...") 09:05:52 Why would anyone hate :noname? 09:07:39 TreyB, it's a good thing PoppaVic set you straight about those puzzles. : 09:07:39 :) 09:13:07 i dont like noname 09:13:12 i prefer to behead later 09:14:36 That's fine while coding, and if a given Forth supports beheading, but you may well want to generate a new word at runtime. 09:20:28 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 09:25:33 --- quit: uiuiuiu_ (Remote closed the connection) 09:29:33 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 09:32:02 hi 09:32:10 re 09:50:40 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 09:55:45 --- quit: uiuiuiu_ (Remote closed the connection) 10:25:55 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 10:31:00 --- quit: uiuiuiu_ (Remote closed the connection) 10:31:59 --- join: rabbitwhite (n=roger@136.160.196.114) joined #forth 10:33:51 --- join: Quartus__ (n=Quartus_@209.167.5.1) joined #forth 10:35:36 I ran into implementation-specific issues with :noname under pForth just yesterday whilst throwing intentionally mal-formed code at the compiler. 10:36:34 such as? 10:36:41 --- quit: rabbitwhite () 10:36:43 I wanted to recover the dictionary space used by the partially-compiled definition. 10:36:57 backup here's value 10:37:09 a forget type thingy 10:37:42 that word shouldnt be recognised until after the word's header is "fixed" tho 10:37:43 right? 10:38:10 which is what 'recurse' does 10:38:16 :nonanme words don't have headers in pForth. 10:38:17 to "complete" a word's header 10:38:19 ok 10:38:34 but isnt the concept the same as with headless words? 10:39:18 i mean, as i understan, the XT to a headless word points to which; the cfa or the pfa? 10:39:41 and that would be the first cell after the last or latest word well-defined 10:39:44 wouldnt it? 10:40:39 In pForth XTs point into the code/data space. It has routines to walk through the dictionary to match an XT to a name if required. 10:40:58 ok, so then you have two spaces to cleanup 10:40:59 They do this so they can drop the headers. 10:41:28 yea, i understan that 10:41:31 Two for failed normal secondaries, and one for failed :noname secondaries. 10:41:51 but what are the pointers to code/data that mark where latest is? 10:42:10 those should be what you tweak 10:42:22 Latest points to the last named word. For :noname I just abandon the space. 10:43:06 but what about the here var that points to the next free cell in code space; wouldnt you have to tweak that? 10:43:19 iow, what is adjustable to "abandon the space"? 10:43:24 I'd have to re-write :noname to save the XT/here at the start of :nonanme. 10:43:25 what word 10:43:39 ok, so that makes that new field the cfa 10:43:55 It places it on the stack, but I can't depend on it when compilation fails. 10:43:59 and it points to the actual code for the headless 10:44:13 but the space that was going to be used is "static" ina sense 10:44:29 so the stacked xt is still or would be valid still 10:44:41 Right, but the compiler might have already ,-d things in already. 10:44:56 so an extra dup is needed before , 10:45:24 the assumption is false that the xt is not needed after; the word doing the , shouldnt be consuming it 10:45:28 But the stack doesn't survive the ABORT that happens when the compile fails (say, for a missing word). 10:45:35 ok 10:45:46 then the abort is preamture 10:46:32 it's premature bc an object exists in data-/code-space that compromises the task 10:46:44 and that's a no-no. 10:46:56 i call that a bug in logic 10:47:01 not well thought out 10:47:07 I never claimed pForth didn't have warts. 10:47:10 i know :) 10:47:28 i dont konw pforth spezivigally; i'm just sayin... :) 10:47:50 is it yours? 10:48:05 We use it strictly as an embedded scripting language, and no human directly interacts with the interpreter. 10:48:10 ok 10:48:41 I actually feed it from binary-encoded XML body tags. 10:48:42 ok, so what word calls abort? 10:48:52 tentative-ew :) 10:48:55 heh 10:49:54 WBXML looks a lot like the old AIFF file format, IMO: a stream tag/length/data tripples. 10:50:04 kk 10:50:24 no need to be xml detailed; i barely know xml either 10:50:50 The compiler calls abort if it couldn't find the word or convert it to a number. 10:51:12 ok, so that's a standard quit, right? 10:51:23 Esentially, yes. 10:51:28 maybe quit at that point is 10:51:29 umm 10:51:49 In forth you don't get many other choices :-) 10:51:49 another word is needed to check for "dangling" nonames 10:51:57 only bc of lack of foresight :) 10:52:22 so a proper clean-up can occur before quit is called 10:52:26 oops, abort 10:52:40 Right. If I cache the XT/here at the start of :noname (in a global, say), then I can do it. 10:52:52 yea but, in a global? ew 10:52:54 heh 10:53:19 How many compilers can you run at once in a single VM? 10:53:23 what about using a here or last/latest to mark the true top of next free "space" 10:53:39 globals interfere with recusion 10:53:43 limitations 10:53:47 recursion 10:54:03 ANS prohibits nested :noname . 10:54:21 Hell, it prohibits nested definitions period. 10:54:31 sure but that doesnt mean you cant layer higherlevel constructs/concepts to make life easier 10:54:49 whats teh point in doing nested definitions anyway 10:54:55 thats just sphagetti code 10:54:59 well ans is a committee rhinocerpig :) 10:55:11 ans is a cammel 10:55:18 re I 10:55:21 a horse designed by a committi 10:55:30 nested defs in other languages have a porpose 10:55:40 obfuscate the issue 10:55:41 i call em rhinocerpigs :) 10:55:46 the do that nicely anyway :) 10:55:50 heh 10:55:58 so anyway 10:56:01 lol 10:56:05 Anyhow, I don't really need any more. When I catch this problem, I log an error and it shows up in the report, and gracefully continue. 10:56:06 :) 10:56:13 ok 10:56:20 bbl 10:56:22 chores to do 10:56:27 start new contract in arizona monday 10:56:37 Cool. 10:56:47 coo 11:04:23 --- join: rabbitwhite (n=roger@136.160.196.114) joined #forth 11:08:37 --- join: JasonWoof (n=jason@c-71-192-33-206.hsd1.ma.comcast.net) joined #forth 11:08:37 --- mode: ChanServ set +o JasonWoof 11:08:53 --- quit: snowrichard ("Leaving") 11:09:47 --- quit: rabbitwhite () 11:21:32 --- join: Cheery (n=Cheery@a81-197-19-23.elisa-laajakaista.fi) joined #forth 11:30:50 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 11:44:44 hi 11:49:15 --- quit: snowrichard ("Leaving") 12:12:46 --- quit: Anbidian (Read error: 60 (Operation timed out)) 12:18:51 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 12:21:58 GRRRRRRRRR 12:22:13 how do you get CVS to update properly 12:22:30 I type "cvs up" and it won't download a new sub-directory that has been added to the repository 12:22:31 Switch to SVN? 12:22:39 it just says: ? dpip 12:22:44 I have no file/folder named dpip 12:22:53 but there's one in the repository which I'd like to have 12:24:11 cvs up -dP 12:25:41 ahh, thank you 12:25:55 I tried "cvs up -d" and "cvs up -P"... 12:26:36 Individually they should have done it, I think. 12:29:48 maybe it did and I didn't notice 12:29:55 anyway, hopefully I'll never have to use CVS again 12:30:12 svn doesn't seem to have so many weird problems 12:31:11 No, it doesn't. I've used Perforce for 10+ years now, so I've grown accustomed to it, but I set up an SVN respository for my personal work at home. 12:32:10 "svn up" actually updates your local copy to match the repository head (plus any local modifications) 12:32:45 that directory showed up, but I'm still getting file-not-found errors trying to build 12:32:53 guess I'll just check out again 12:33:25 What do you want to build? 12:36:26 dillo 12:36:59 it became unhappy about the crypto lib it was linked to going missing (I probably upgraded it) 12:37:12 so I figured I'd update dillo since I was rebuilding it 12:37:21 turning out to be more trouble than it's worth 12:37:51 So it seems. 12:39:11 maybe if I go away and eat something it'll all start to work :) 12:39:46 Your lunch emits that many nutrinos? 12:40:11 s/neutrinos/ 12:40:31 Although, from a nutrition standpoint... 12:40:51 now autoconf fails 12:40:55 joy! ;) 12:41:32 *whew*, no PappaVic to get riled up. 12:42:34 * TreyB runs some errands. 12:42:45 oops 12:42:54 * JasonWoof follows the installation instructions 12:58:48 hi 12:58:58 Hi slava. 13:00:14 follow the yellow bits on your screen 13:00:41 and find the quiznos sub! mmmm! 13:01:45 heh 13:19:04 I don't know what bugs pForth may have, but you can use RECURSE in :NONAME words. Rollback of an incomplete word would have to be implementation-specific. To do it in a strictly Standard program, you'd set a MARKER and call it to roll back. 13:34:53 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/") 14:37:52 --- quit: virl (Remote closed the connection) 14:40:15 --- nick: Quiznos -> GoogleMaster 14:40:31 --- nick: GoogleMaster -> Quiznos 14:50:40 Quartus: I understand. I just needed to extend it enough so that I could drop a bad block of code, report the error, and keep going. 14:51:57 I don't have any illusions of, or even requirements for, standards conformance. 14:51:59 Definitely an implementation-specific thing. 14:52:11 Yep. 14:52:32 I just found it odd that a number of us had dealings with :noname on the same day :-) 14:53:03 I don't know, it was just Doofus^h^h^h^h^h^h^hPoppaVic saying he 'hated' :noname, which is easily the stupidest thing I've heard this week. 14:53:34 Some days... 14:57:25 I have an allergic reaction to anything that inane. I hope the silent onlookers in the channel realize that he spouts nothing but utter nonsense on any subject. 15:03:40 plus his overreaction to treyb's sudoku interest 15:03:59 i mean fine, he doesn't like it, but he doesn't have to waste time denigrating it heh 15:04:06 Yeah, what were you thinking enjoying yourself with a puzzle, TreyB? Man, I hate that almost as much as I hate :noname. :) 15:04:11 he probably should be banned, but i don't really hang on this channel 15:04:27 :) 15:04:44 futhin: did you need me a couple days ago? 15:04:46 I don't (and won't) sweat it unless he gets in the way of a serious discussion. 15:04:49 futhin, he's no more than really stupid, and really irritating. I don't know if that's sufficient grounds to boot him off. 15:04:49 Ray_work: nope? 15:05:19 Ok. thougth I saw you call out my name, but I saw it after I left. 15:05:29 Quartus: heh, if thats not sufficient grounds, what is? :P he's been in this channel for more than a year and hasn't shown any signs of growth has he? 15:05:31 I mean after you left 15:05:48 i only call out your name in my sleep 15:06:05 Your sweet. :) 15:06:05 heh 15:06:13 futhin, no he hasn't. I don't know, I think the threshold for banning is higher. If he starts spouting racist dogma again, I'd have no hesitation about booting him off. 15:06:22 TreyB: what do you mean by the backtracking? 15:06:33 * Ray_work will catch up from home. 15:06:35 --- quit: Ray_work ("User pushed the X - because it's Xtra, baby") 15:07:34 Quartus: sometimes if i think a person should be banned but he hasn't made that clear cut enough, i'll subtly troll him until he crosses the line 15:07:46 futhin: I saw some examples of sudoku solvers that work by "guessing" and then "backtracking" to point if it can't find a solution. 15:07:50 altho that method hasn't really worked in #forth before 15:08:34 futhin, my sincere concern about him is that he misinforms and drives newcomers away. 15:08:47 I wondered if anyone had seen a puzzle that *required* that technique, or they just didn't put in the effort to implement all the logical reductions. 15:09:29 TreyB, I suppose a lot of games use that kind of approach. Chess, for instance, though it isn't what I'd call backtracking, but I suppose it technically is -- following lines of play and discarding them if they score too low. 15:09:43 TreyB: hmm, are you sure you don't look at a square and say "what if this is 3?" and then look along the lines to see if anything conflicts/supports that? 15:10:34 its hard to solve sudoku entirely by logical reductions because sometimes you just don't see the next step.. 15:10:59 Of the "easy", "hard", and "very hard" puzzles I've solved over the last two days, none of them required guesses. One of them had multiple solutions, though. 15:11:17 * slava has been working on a logical problem for 6 months 15:11:29 still can't see the next step 15:11:30 hmm, i don't think there's supposed to be multiple solutions.. 15:11:40 slava: heh thats nuts 15:12:30 TreyB: somehow i don't believe that none of them required doing "what if", unless you're some sort of superhuman :P 15:12:47 But he is. He used to work for Be and Palm. :) 15:13:42 futhin: You can construct a puzzle with two pairs of swapped numbers in two ways. 15:14:07 Quartus: that just proves that I know how to ride a company into the ground. 15:14:21 TreyB, that's true. But that's a superpower if I ever saw one. 15:14:31 heh 15:15:59 lol 15:16:04 radioactive man! 15:16:15 he has a half-life 15:16:37 futhin: like PoppaVic, you're rarely on topic 15:16:37 Half a life is better than no bread. 15:16:57 PoppaVic is often on-topic, unfortunately. He's just completely wrong. 15:17:12 --- join: uiuiuiu_ (i=ian@dslb-084-056-244-048.pools.arcor-ip.net) joined #forth 15:17:25 I don't put a whole lot of stock in his input. 15:17:36 If he just waxed on about his siblings, and guns, and his mother, he wouldn't confuse anybody about Forth. 15:17:38 slava: settle down, go smoke some more weed instead of trying to flame me :P 15:18:16 do you smoke weed with paul graham? 15:18:43 that doesn't make any sense 15:18:45 Coincidentally I was reading Paul Graham by accident a couple of days ago. It's like having your feet stuck in wet mud. 15:20:51 is it cool to hate paul graham now? i better get on that bandwagon 15:21:08 i don't hate paul graham, i just asked if you had smoked weed with him 15:21:26 i don't smoke weed 15:22:17 --- quit: uiuiuiu_ (Remote closed the connection) 15:22:20 he tells me otherwise 15:22:38 lol 15:24:29 --- part: futhin left #forth 15:46:06 --- quit: neceve ("Leaving") 15:52:14 --- nick: nanstm -> Raystm2 16:16:36 I don't have any idea if it's cool to hate Paul Graham. I don't hate him, and I'm unconcerned with whether I'm considered cool, but I do think he's a blowhard windbag, and that he's getting worse over time. 16:17:16 okay /me needs to google this Paul Graham to get context. 16:17:58 http://www.paulgraham.com/articles.html 16:19:32 You are kind. 16:20:10 I've been here before. I think about a programming contest over several languages. He had a bent about forth if I recall. 16:20:42 Like many others, he thinks the programming world starts and stops with what he knows about it. In his case, that's Lisp. 16:21:16 I see. 16:21:36 Increasingly he's moved away from talking about Lisp into more general topics about which he knows considerably less. 16:23:46 hehe. 16:24:06 He's drinking his own Kool-aid, as they say, in larger and larger quantities. 16:24:07 thanks for saving me the time i would have taken to get to that same opinion. 16:24:40 Heh. I may have saved you the same sensation of being trapped in wet mud. 16:24:41 :) 16:25:30 been there. 16:27:31 It's like reading a Herbert Schildt book on programming C. He says many things with great confidence and assurance that are simply flat-out wrong. 16:28:03 --- join: Anbidian (i=anbidian@S0106000fb09cff56.ed.shawcable.net) joined #forth 16:28:11 You saved me again. :) 16:28:22 heh 16:39:14 Quartus: have you ever heard a concern about forth that couldn't be fixed in forth? 16:40:46 From people brand-new to the language, I often hear them bemoan the lack of string-handling. They want to see the C library functions, or equivalent, and feel lost without them. 16:41:01 Of course that can be addressed in the form of the C library functions, if you like, in Forth, though there are better and more Forth-like ways. 16:41:45 I see. Sure string stuff. hmm. 16:41:47 I read somebody complaining of the lack of an associative-array type, but that's trivially overcome as well. 16:41:54 right. 16:42:04 Let's see. What else... RPN. FFI. 16:42:07 ffi ya. 16:42:30 but no _real_ issues tho, am I assessing this correctly? 16:43:09 A given individual might strongly feel that any or all of those are 'real' issues, especially people who are still trying to write C in Forth. 16:43:19 Oh the only other complaint i see... "that forth doesn't have what this forth has"... 16:43:28 Stack-juggling, too, is a common complaint. 16:43:40 practice fixes that last one. 16:43:53 Sure, and working with a few items at a time, which is a byproduct of learning to factor. 16:43:54 c in forth, ack. 16:44:02 ya. 16:44:16 Okay, cool. It's as I thougth. 16:44:35 Forth isn't a simple Algol-type derivative, and the differences confuse people who've only ever used those, and thus these complaints arise. 16:44:49 * Raystm2 nods* 16:44:52 Speaking of type, the lack of types is another complaint. 16:45:00 Or rather, the lack of type-checking. 16:45:11 Ya, but if you really need that... 16:45:24 you roll yer own, no? 16:45:49 Sure. I don't find it any kind of limiting factor, and that's not apologia, but the truth. But it's a complaint I've heard from others. 16:46:00 right 16:46:10 Okay. cool. 16:46:27 What else... blocks were widely hated. Lack of floating-point was, though that's history in most Forths now. 16:46:42 I just wanted to be right when I formulate my next complaint buster. 16:47:05 3-chars+length dictionary naming was hated even among Forth people. Likewise the lack of namespaces, but again, both are now history. 16:47:15 hehe :) 16:47:23 I used that to make a point the other day. 16:47:34 The 3-letter names thing that is. 16:47:48 Clever letter, Chuck wrote. 16:48:10 Carefully-worded letter, and I don't think it changed anybody's mind. Chuck is a hard-core minimalist. 16:49:00 ya. 16:49:02 :) 16:49:22 Someone has to be the anchor on that end of the chain of programmers. 16:49:34 Might as well be the author. 16:49:55 --- quit: segher (Nick collision from services.) 16:50:06 I don't know, I didn't feel adrift before he went mad on minimalism. :) Anyway that's all that comes immediately to mind when I think of things people complain about in Forth. 16:50:09 --- join: segher (n=segher@dslb-084-056-176-157.pools.arcor-ip.net) joined #forth 16:50:15 I'm intreged(sp?) by the minimalist approach. 16:50:48 ... as you prob'ly gleened. 16:50:50 Keyboards with fewer keys, screens with fewer characters, these things do not interest me. 16:51:37 I know. I still like you. lol :) 16:51:42 heh 16:52:21 Anyway these days -- I can't think of a full-featured Forth that doesn't have FFI. 16:52:57 Right. 16:53:11 Quartus, thanks for your opinion on this matter. :) 16:53:23 Or wordlist support, or floating-point, etc. 16:53:24 Sure. 16:53:50 I'd love to re-write my bosses custom business software in an ANS forth. 16:53:58 It's in BASIC now. 16:55:39 Out of curiousity, why an ANS forth? 16:56:06 so that another programmer comming behind me ( when I die or something ) can just pick up and go with it. 16:56:22 the idea is that the software would become a profit center for his business. 16:56:34 Portability is important for that. 16:56:46 Absolutely. 16:56:55 I see. 16:58:07 Even in .bas it's good back-office software. Even has POS. 16:58:34 If you have any thought for the future, it makes no sense not to write to the standard as far as is possible, while clearly identifying and modularizing anything that's implementation- or system-specific. 16:58:44 But it's old, slow( just fast enough now but slowing) and could be extended to do much much more. 16:58:58 Agreed. 17:00:26 I have no business interests or concerns of what others would be able to do with my code so ANS doesn't appeal to me. 17:01:29 Though I am still unsure if there really is any relationship between standardization and ease of maintainence. 17:01:53 The devil is in the details, of course. 17:02:53 I assume that doing the code in ANS would mean that the new programmer would beable to get help on anything they didn't already understand, even if they never ANS'ed before. 17:03:13 With out me haveing to provide help on every single word. 17:05:18 Which I'd prob'ly provide anyway... 17:05:19 That would be one approach. The other approach would be one in which more learning would be required and the learner would be more self sufficient and better able to understand code of any type without relying on a premade standard. 17:05:52 One benefit of standard programs is that moving to another platform is made simpler. 17:05:55 There is that. Hey, I know where you're comming from. I colorforth on purpose. :0 17:06:19 Ray, another complaint -- a boss might not want an app made in Forth because he believes (rightly or wrongly) that he won't be able to find another Forth programmer to maintain it. 17:06:21 And this system would need to span many machines. 17:06:26 Quartus: *nods* that has its advantages and disadvantages. 17:06:27 In other words, he wants whatever is being taught in colleges now. 17:06:39 Anbidian, I can't see a disadvantage in being able to port. 17:07:46 Quartus: Have you ever looked at Chuck Moore's discussions on why portability is inefficient and unnecessary? That is where I am coming from. 17:08:59 Quartus: acknowledged. 17:09:54 Anbidian, Chuck is brutally minimalist, and has no interest in portability. He has a programming audience of one. 17:10:04 Anbidian: inefficient, yes. Current practice, yes. And for very good reason. The most popular OS requires a degree to program it properly. 17:11:01 You can't just ask a business owner to roll his own accounting software and make it work with all of his machines. 17:11:19 Quartus: I have no disagreement. That approach is my preference at this time. 17:11:53 Anbidian, if you are writing code that you believe will never have any use outside of the platform it's currently on, then portability isn't something you need to consider. 17:11:56 Raystm2: Of course. If you want to deal with anything in the mainstream you have to accept their ways. 17:12:06 Quartus 17:12:33 : Right. 17:13:49 The other side of the coin is that while you may believe that your code will never need to run anywhere else, you may be wrong. 17:14:11 Another benefit of writing with an eye to portability is the ability to test your code on multiple compilers and platforms. 17:14:39 You also get to share your code with a larger community of developers, and make direct use of their code. 17:14:40 Yes. But I think what is being proposed is that having code needed to be run elsewhere is almost always going to inherently an inelegant. 17:14:47 *-an 17:15:02 ya. this is true. :) 17:15:15 * Raystm2 is a devout minimalist. 17:15:42 There is nothing inherent in writing portable code that requires it to be inelegant. 17:16:12 I'm questioning that. 17:16:26 I think Anbidian is speaking to the fact that ANS has a rep for being inelegant already. 17:16:43 Raystm2, do you have a colorForth page? I'm not sure, but I think remember your name from adding links to my site 17:17:00 Ya, I have a badly neglected colorforth.info 17:17:27 I've gotten into other projects and my new disease takes some of my old programming time. speaking of which... 17:18:06 Raystm2: :| that wasn't in my list. And I have a comprehensive colorforth section. Don't know how I missed it. Should that be linked to as colorforth.info or http://ray.rx-core.org/? 17:18:10 Anbidian, perhaps you can provide an example of what you consider inelegant code, or what it is about standard code that you find inelegant, or what it is about portable code that you find inelegant; or the opposite, what you find elegant about code which is non-portable and non-standard. 17:18:48 um I think colorforth.info as it is more to the point specific. 17:19:20 Anbidian: i didn't mention because i'm ashamed of the shape that it is in. 17:19:31 But now this might kick me into gear. 17:19:37 Quartus: I was speaking more basically in the sense of if you have two different contexts, two different platforms or processors or however you want to think about it, there is no way one soultion could be maximally optimal on both contexts if there is infact true difference. I have no specific examples to offer you, but I recall coming across many listed at ultratechnology.com 17:20:01 Raystm2: *nods* 17:20:13 Raystm2, there are so few colorForth resources that anything is better than nothing 17:20:28 I think that crc has proven that you can modularize the problem to a tiny file for each port. 17:20:28 For any given application and platform, there may be bottlenecks that require non-standard optimization. 17:20:38 Anbidian: agreed. :) 17:21:55 Anbidian: what is that page that has your forth links, please? 17:22:45 Quartus: I have so little experience with programming I probably ought not to be speaking. But what I have ascertained is that there appears to be a minimalist approach that produces the best results. However, there is a cost to taking the minimalist route. If that cost is prohibitive or not practical for a given need, then there are other ways to go about it. 17:22:55 Raystm2: http://forthlinks.com 17:23:55 Anbidian, I probably have more experience; I can assure you that it is entirely possible to write elegant and efficient code without twisting the dial all the way over to non-portable minimalism. 17:25:18 Quartus: Yes, I have heard that proposal before, and I don't know what is actually the case. My investigations have inclined me more towards the brutally minimalist approach. 17:26:31