00:00:00 --- log: started forth/06.09.26 00:01:15 does HYPE pass the objects on the stack, not "selectors" ? 00:03:22 I don't know. 00:03:41 I don't use OO, I don't believe in OOP. 00:04:38 I'm follower of FP. 00:05:31 I don't understand why I need these "objects". 00:06:10 i was just wondering the same about FP :P 00:06:17 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 00:06:17 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-111-243.pools.arcor-ip.net) joined #forth 00:07:01 I can explain why FP. 00:07:21 We have several centuries of science history. 00:07:53 We used functions for modelling the real word with great success. 00:09:12 There're no models similar to "objects" in OOP sense. 00:10:39 OOP has one major disadvantage: you can't find a way to 00:10:40 compose objects within OO paradigm. 00:11:05 --- quit: snoopy_1711 (Read error: 145 (Connection timed out)) 00:11:05 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-111-243.pools.arcor-ip.net) joined #forth 00:11:07 --- quit: snoopy_1711 (niven.freenode.net irc.freenode.net) 00:11:07 --- quit: Snoopy42 (niven.freenode.net irc.freenode.net) 00:11:07 --- quit: EdLin (niven.freenode.net irc.freenode.net) 00:11:07 --- quit: Quiznos (niven.freenode.net irc.freenode.net) 00:11:08 You have to invent "god" object to do this. 00:11:56 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-111-243.pools.arcor-ip.net) joined #forth 00:11:56 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-111-243.pools.arcor-ip.net) joined #forth 00:11:56 --- join: EdLin (n=chacha@as5300-6.216-194-22-43.nyc.ny.metconnect.net) joined #forth 00:11:56 --- join: Quiznos (i=1000@unaffiliated/quiznos) joined #forth 00:13:47 ASau: you can't compose objects but you can compose the functions that make up their implementation... 00:14:39 it's all the same in the end, the object metaphore scales better for real-world instances (in large numbers) wheras math has only really modeled abstract ideas... :P 00:14:49 If you have big diversity in object types and relations 00:14:51 betweem objects, you end with a mess, when you want to nail 00:14:51 "methods" to "objects". 00:15:24 So, you start nail "methods" to newly invented "god" object. 00:15:39 whereas in FP, everytihng is a global? 00:16:42 No. You may use local models, modules were not invented in 00:16:45 OOP, they're from procedure era. 00:17:51 Functions scale to any extent, in contrast to objects, 00:17:53 which can't work in anything more complex than hierarchial trees. 00:17:59 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 00:18:16 --- nick: snoopy_1711 -> Snoopy42 00:18:44 We use functions for much more complex things, that come from real world. 00:19:58 Real world is not hierarchial, we've come far away from 00:20:00 Ancient Greece and Dark Ages. 00:20:11 is it possible to plug in functional programs to a database elegantly? 00:20:28 Of course! 00:20:44 so there you go, both concepts are perfectly useful together 00:20:55 SQL is almost functional language. 00:21:24 i see what you're implying. you may well have the same data underneath the functional lanugage but it's exposed in a more elegant way. 00:22:36 You may use any sublanguage you want. 00:22:55 what's you FP language of choice? 00:22:57 +r 00:23:01 If you work with databases, you don't use more complex 00:23:06 combinators than "map". 00:23:22 Personally, I prefer Scheme. 00:24:00 interesting, closures in scheme are essentially equivalent to objects. you just get the power to define your own accessor logic 00:24:51 I think we can't exorsize procedures completely, so Haskell 00:24:53 people go wrong way. 00:25:51 that's what i was curious about 00:26:00 i'm not sure you can just "abstract away" side effects 00:26:10 Of course, you can modell "C++" objects (which are not 00:26:11 objects of OOP) with closures. 00:27:17 Sure you can. 00:27:31 Side effects model time run. 00:29:57 so you go back in time by backtracking... 00:30:22 interesting, i've not done that in the context of scheme 00:30:35 If you have several objects, that map their internal state 00:30:37 S_i -> S_i', then the overall state S_1 x S_2 x ... S_n -> 00:30:37 S_1' x ... x S'_n' = s', where s = S_1 x ... x S_n is compound state, 00:30:37 state of the "Universe'. 00:31:01 Backtracking is reversal function. 00:31:12 It is not a function. 00:31:32 In generic case. 00:32:28 isn't it super easy in Haskel? 00:32:55 I don't know Haskell to that extent. 00:33:54 Haskell is not easy, even his addicts are not aware of 00:33:56 consequences imposed by laziness of their language. 00:34:49 heh 00:37:19 What's the weather in Wien? 00:37:45 It seems we will have rain in 2-3 days, but I'm not sure. 00:39:12 it's very nice here 00:39:17 http://en.wikipedia.org/wiki/Joy_programming_language 00:39:24 where are you based? 00:39:35 Moskau. 00:42:33 Hmm. 00:42:34 not too cold yet? 00:43:39 Cheery, I should better ask you, what's the weather in Finland? 00:44:32 AJC: No, it's better to ask those who are located in Finland or Sweden. 00:46:19 AJC, do you use Forth for practical tasks? 00:46:34 i'm doing research at the moment 00:46:51 not in forth directly, but a similar interpreter with ideas from joy 00:47:26 CS student? 00:48:22 Why do you need such weird hybrids as Joy et al. 00:48:25 ? 00:48:59 no, i'm a game developer, but my publisher ran out of money, closed the studio, and paid us off... so i have time to play around 00:49:10 i'm using joy because i want to be able to undo everything :-) 00:49:17 Slava is/was CS student, it's his area. 00:49:19 backtracking in program search 00:49:35 Have you read articles of Gassanenko? 00:49:42 no, do you have a url? 00:50:00 He did researches in backtracking field. 00:50:28 http://forth.org.ru -> members -> Gassanenko 00:50:46 IIRC, he is called "mlg". 00:52:00 interesting 00:52:11 lol, "Forth Dimensions" 00:52:25 I don't think there's any need to be able to undo everything. 00:52:37 It can't be done. 00:52:42 The Second Law. 00:53:58 ASau: i need to be able to undo most things so that my search algorithm doesn't run out of memory 00:54:15 it works in all cases so far 00:55:59 --- join: Amanita_Virosa (n=jenni@ppp-70-243-36-192.dsl.hstntx.swbell.net) joined #forth 00:56:17 Guten Tag, Amanita_Virosa! 00:56:47 hiya 00:57:13 AJC, you pay for this in complexity of interpreter. 00:57:26 no, the interpreter is very simple 00:57:42 ASau: it's the implementation of each operation that's responsible for undoing 00:57:56 It may be easily be done recording original state and 00:57:58 resetting to it, when you want this. 00:58:47 Amanita_Virosa, I don't remember you, what do you do with Forth? 00:59:10 ASau: possibly, but this has been much easier so far though. 00:59:19 well, i wrote a forth OS 00:59:30 Why you need it?? 00:59:40 and an assembler that uses a FORTH script system to handle the internal conversion 00:59:45 and another handful of forths 01:00:01 Do you use Forth for practical things? 01:00:25 why am i suddenly getting the forth degree here? 01:00:29 err, the fifth degree :P 01:01:19 Get the seventh degree, and you are free. :) 01:02:07 heh 01:02:10 May I ask you for more elaborate description of your OS? 01:02:28 cyvos.com 01:02:36 the embedded one 01:03:48 AJC, I don't think that supporting reversal function for 01:03:50 everything is easier. 01:04:03 Ah! Now I recall you! 01:04:25 ASau: i have to support reversal for each atomic operation, then the composites can reverse automatically 01:04:47 Jenni, are there great reasons to renaming? 01:05:10 renaming? 01:05:29 i usually just go by "Jen", btw. 01:06:09 You called yourself other name 2 years ago. 01:06:36 eh? oh, you mean my nick 01:06:39 it evolves over time 01:07:08 Well, you should not expect me to be fluent in English. 01:07:58 oh 01:10:42 well okay 01:10:45 AJC, you can remember state, which is equivalent to 01:10:46 continuation, and return to it atomically. 01:11:04 the truth is, i really didn't change my nick 01:11:31 Continuations are even simpler then reversals. 01:11:48 Amanita_Virosa, really? 01:11:54 i took an alternate form of my original, "solar_angel"... specifically, "destroying_angel", and then converted it into the scientific name of a species with the same vernacular name. 01:12:15 and this is what resulted. 01:12:22 so yes, not really a big change. 01:13:32 ASau: that's kind-of what i do now. i have a continuation, which during search, yields the result state, then continues and undoes the change. 01:13:43 Two not as easy evident transformations are not really a 01:13:45 big change. 01:14:00 ASau: that's not the problem anyway, but thanks for the suggestions! 02:32:00 --- join: PoppaVic (n=pete@0-1pool65-75.nas22.chicago4.il.us.da.qwest.net) joined #forth 02:34:43 --- quit: Amanita_Virosa ("Wewps.") 02:34:47 ASau: here's calm and bright. 02:35:12 Actually, functions are a quite unsuccessful in mapping the reality. 02:35:35 because there is much more than they can handle. 02:36:25 when you map a reality into function, you assume reality works in 'bags' or isolated sections. 02:36:56 when reality really works in it's wholeness; functions ignore the synergy. 02:48:06 --- quit: JasonWoof ("off to bed") 02:54:14 --- quit: segher (Nick collision from services.) 02:54:24 --- join: segher (n=segher@dslb-084-056-151-056.pools.arcor-ip.net) joined #forth 02:56:36 Cheery: It is completely wrong. 02:57:17 We have more than two centuries behind functional approach. 02:58:41 The whole science, be it physics, chemistry or any other 02:58:42 technical field, is based on functions and relations of similar kind. 02:59:30 In contrast, OO paradigm started in late 70-s. 03:00:02 That means, we have have ~30 years for OO paradigm. 03:00:23 Even less for "event-driven" 03:01:12 That's why "OOP" is really hype. 03:01:37 I always find "oop" a lot less fun than "oo" 03:02:02 OO implies "contexts", at least. 03:04:00 The limitations of OOP are well known: the only example of 03:04:00 composition breaks OOP to pieces. 03:04:26 comp? 03:04:45 you mean like only/also ? 03:04:56 Event-based approach seems to be more functional. 03:05:08 PoppaVic: Sorry? 03:05:21 ASau: the words only/also/definitions, etc 03:05:42 ASau: not sure what you mean by compos. above 03:05:55 Ah! 03:05:55 (it's very early here) 03:06:18 I mean this: 03:06:39 in real world models (sciences) you have many relations, 03:07:08 LOL 03:07:26 ASau: when I said that functions are quite unsuccessful in mapping the reality, I didn't say anything about OOP. 03:07:35 :) 03:07:51 Cheery: illuminate. 03:08:14 when you compose an object of some kind from two objects of 03:08:17 other two types with the same rank, you don't have class, where 03:08:17 you composing function belongs to. 03:08:42 oh, NOW yer talking about that insane multi-inherit/polymorphism 03:09:00 Cheery, you said that "functions are quite unsuccessful in 03:09:01 mapping the reality". This is obvious wrong. 03:09:20 No, consider what is the reality. :) 03:09:27 And how does it work, you'll understand. 03:09:29 We have 300 years of functional approach, which is the 03:09:31 _only_ approach in sciences. 03:09:55 Reality is a lovely Concept. Metaphysics are slippery. 03:10:00 Cheery, if you refer to Popper et al, they don't know 03:10:00 what's reality. 03:10:30 Hmm. If I've understood that object orientation (class->object) is over 2000 years old. 03:11:09 I do not agree it is completely correct, but neither functional perspective is. :) 03:11:14 oh, now yer comparing to class/phylum, etc 03:11:18 I think they both sucks. 03:11:21 Ancestry relation is the simplest form of relation. 03:11:37 subclassing 03:11:52 We have moved far away from Ancient Greece for these years. 03:12:19 ASau: well, actually.. The system still works in the _gross_ 03:12:20 Now we have 300 years of successful functional approach. 03:12:41 nah, not really, we have just developed that information further. 03:12:47 It's just no good if they throw methods into the mix. 03:12:50 And that's correct approach. 03:13:33 You neither should say OOP is inferior to FP, you should think what's good in OOP and what's good in FP, then maybe understand something neat. 03:13:41 And, once again: I can see value to OO, single-inheritance or superclassing. I see far less of value in OOP 03:13:55 We know, that real world is more diverse to be able 03:13:57 described within "class-object" framework. 03:18:37 I don't know how you can describe several independent 03:18:38 classifications within OO framework. Do you have to decide, 03:18:38 which classification is of first rank? I think you have to. 03:18:48 notice the object-delegation-object is already much more capable, and that computer and structural order is really it's own world, very little of what we see in real works in such environment. 03:19:03 *realm 03:20:29 ASau: yes, if you were to use OO "objects" w/ single-inherit/superclass, you do indeed take responsibility for designing the proper structs/objects to get from A to Z 03:24:51 There's one thing in OOP, which particularly I hate: OOP 03:24:53 addicts speak their unique language, they don't speak the 03:24:53 language, known to scientists or community, they speak of 03:24:53 "objects" which are not objects, "methods" which are not 03:24:55 methods, "delegates" which have nothing common within regular 03:24:57 meaning of this word, instead of well studied and well known 03:25:00 terms of relation, (de)composition and structure. 03:25:28 the same is true of the socket volken - nothing like term twisting/abuse 03:25:52 Thought, I do not see point in such arguing, because I do not see much generic differences in OO and F, OO is just much more bullshit-verbose and just much more full of filter -stuff. 03:26:33 while the both structural patterns does the exactly same thing. 03:27:56 The main difference between FP and OOP is the orientation: 03:28:08 sophisticated handling of functions as data is much more powerful than OOP, thought. 03:28:15 and much more intuitive. 03:28:16 Personally, I like structure - "Chaos Programming" is for kids ;-) 03:28:20 FP focuses on relations, OOP focuses on "objects". 03:29:24 I'd love to talk about models instead of functions, and apply those models together different ways to make a program. :) While the program itself would propably yet work these models. 03:29:27 OOP is very contradictory: they say "objects", but objects 03:29:29 are of no value as such. 03:29:40 but it's not OOP 03:30:06 And, what do YOU see as "models"? 03:30:09 it's kind of, using functions as patterns and employing their strengths with new ways. 03:30:35 function sum is a model for me, a quite lot. :) 03:30:38 too vague - keep going 03:31:03 well, remove the syntax from: (sum 1 2) 03:31:40 slipping into lisp is not helping - keep elaborating on "models" 03:31:53 sum is a kind of model which does something for slow models it gets, and makes an another slow models which works as a mathematical sum of the earlier slow models. 03:32:18 PATTERNS! 03:32:19 :> 03:32:26 clear? 03:32:29 * PoppaVic sighs 03:32:40 FP is more elaborate: if you have goal (that's why you say 03:32:42 "programming"), you have means to achieve the goal (computational 03:32:42 functions), you have original data (which are real objects), seq. 03:32:42 your task is to compose function to achieve your goal. 03:32:43 Not really... Looks like twisted forth 03:33:48 forth is the best human readable model you can get from flow of program, but not good because you don't only have a program's flow when you have a program. 03:33:53 ASau: yes.. I've been cogitating for quite a while the idea of "types" and "objects" and such.. They look damned similar. 03:34:05 Cheery, you can't get far, if you keep at such primitive 03:34:06 models as "sum". 03:34:19 Cheery: that didn't even make sense 03:34:50 PoppaVic, smalltalkers have more real names for things: 03:34:51 instances and classes. 03:35:05 Unless... you might be trying to talk about self-modifying/adaptive programs? 03:35:25 lets say, program is much more than it's execution paths. but neither object oriented or functional programming paradigm considers it. 03:35:59 ASau: yeah, and I consider an "instance" to be akin to "variable/const" while "objects" and "types" are pretty "inert" 03:35:59 Cheery, functional program has no execution path. 03:36:19 Cheery: yer a strange duck. 03:36:39 it has. when you describe a function, you describe what a thing does to other things. 03:37:47 PoppaVic: self-modifying&adaptive programs are highly connected to this discussion, but not in the spotlight now. 03:38:04 sounds like a fine bit of insanity, to me 03:38:09 thought, maybe they tell one intresting notification. 03:38:24 If you write functional program you end describing domain 03:38:26 We aren't able to do them properly because we think that execution paths are static. :) 03:38:28 of operation in scientific style: relations, which are 03:38:28 predicates, compositions, or causal sequences. 03:38:57 and we model those execution paths by code. 03:39:21 It doesn't really change whether you was using C or java or anything else instead of forth. 03:39:46 language has no influence to this stuff. 03:39:54 all you are doing is twisting FP into a new shape with the same issues. 03:40:24 Cheery, language is of great importance. Sapir-Worf thesis. 03:40:27 cut "with the same issues" away. ;) 03:40:39 Asau, not in this kind. 03:41:03 even language is of great importance yes. :) 03:41:12 because we need to communicate. 03:41:58 but see, whether you say "flowers seem beautiful" in swedish or japanese, the content doesn't change. 03:42:23 If you want to get the result fast, you choose one 03:42:26 shortest way to go. You don't need "dynamic" execution path. 03:42:50 yep. 03:43:11 but deterministic programs would be nice to create. 03:43:13 anyway. 03:43:32 brb, going to get my little brother from school. 03:43:50 "One way to go" is the deterministic way. 03:44:27 argh, I meant to say determining programs, or whatever you call the programs which decides by complex logic what do they do. 03:44:58 kind of, advanced AI. 03:45:40 vtables, varrays Everything is decisions. Even Random#'s are used for decisions. 03:47:31 You don't have random numbers, where you think to have 03:47:32 random numbers, in real you need sequences or sets of numbers 03:47:32 with some specific properties. 03:48:07 --- join: Bourbaki (n=Miranda@p50916C39.dip.t-dialin.net) joined #forth 03:48:10 hidiho 03:48:17 ASau are you online? 03:48:36 see /whois 03:48:50 I mean, "Use /whois." 03:49:01 i take this as a yes :) 03:49:14 AJC told me that you might like to see an idea of mine 03:49:35 i said you guys are on the same functional wavelength :P 03:49:39 just searching the site sec 03:49:52 all the same :) 03:51:11 hm i cant access haskell.org atm ... 03:55:05 hm ok the server seems to be down ... care to listen to the general idea anyway? 03:58:31 --- join: uiuiuiu_ (i=ian@dslb-084-056-220-226.pools.arcor-ip.net) joined #forth 03:59:52 I care to listen almost any idea. 04:00:07 cool :) 04:00:24 I may seem to be off line: I'm merging sources. 04:00:46 the idea is to build a framework functional languages 04:01:08 that is a visual language that is based on graphs on top of them for the semantics 04:01:23 these graphs are topologies 04:01:35 and category theory diagrams 04:01:44 and of course graphs 04:01:55 the nodes of these graphs are typed containers 04:02:02 the edges are functions 04:02:21 the network is if processed updated like a petri net 04:03:11 now you can use topological operations on these graphs and also you can can "fold" the graph to a single edge as long as its non cyclic for now 04:03:33 youd need some more analysis for this because not all cyclic structures are alike 04:03:36 --- quit: uiuiuiu_ (Remote closed the connection) 04:04:54 the folding uses the composition operator of the category theory diagram that is in case of a haskell program it would use arrows which allow you more constructions than just composition but also baking two functions to a new one that executes both parallely 04:09:25 Sorry, I'm not mathematician, I don't know Petri nets etc. 04:11:14 I see one major problem with graphs, the same as with OOP: 04:11:15 mapping Decartes multiplication to new set is not expressable. 04:11:35 Try working with glade and such, where you can "draw connections" 04:12:57 ASau why? 04:13:36 I can introduce a special node, which should express 04:13:38 Decartes multiplication of income nodes. But I don't see it 04:13:38 expressive enough. Too many nodes. 04:14:11 but you can build the product of any two nodes in this graph 04:14:19 because you can fold the edges 04:16:28 Maybe I haven't understood your idea yet, I need time to 04:16:30 ASau: i think the idea is to model with a graph exactly what happens with a function, so it's equally expressive -- if i understand correctly 04:16:31 think. I'm not mathematician, I may have to translate your terms 04:16:31 into plain ones. 04:17:23 I've understood that, I think the goal is to model the 04:17:24 whole knowledge (sub)domain. 04:17:57 The problem is: I've done something similar. 04:18:39 I succeded clearing problem, I was solving. 04:18:54 But I don't think it is useful in general. 04:19:56 I may be wrong here: it is field for deeper research. 04:20:27 but it is because you can connect any two graphs using their topology and its a concise mathematical framework 04:20:37 ASau: I'd prefer folks test, instead if sitting in their "towers" ;-) 04:20:37 It involves psychology, and I'm not psychologist. 04:21:51 PoppaVic, I understand, I even could be volunteer. 04:22:01 hmm. 04:22:27 ASau: Too many math-heads just write docs... Too much "theory". 04:22:34 The problem is: if I volunteer to be tested, I can't work 04:22:37 on tests and procedures themselves. 04:22:54 Now I understand one thing about past. 04:23:19 People who I've been gathered my information from has learned much more than me from the learning event. 04:23:57 I think that's just the Heisenberg think. 04:24:17 It's incredible to notice. :) 04:24:29 Epiphanies are painful 04:24:42 Bourbaki: I think I understand the idea. :) 04:25:18 you don't happen to think allowing any kinds of directions and effects by doing petri nets, do you? 04:25:57 Cheery if you have not folded the graph it would more or less act like a coloured petri net with a different notion i think 04:26:17 but the cool thing is that an extended version would include different interpreters for the graphs 04:26:31 so you also can have an interpreter for cellular automata and so on 04:27:04 What I would like to see is well-factored diff(1) code in 04:27:06 and while the interpreter is a functor the composition of interpreters would be a natural transformation 04:27:06 Forth or Scheme. 04:27:22 Of course, I can port it myself. 04:27:40 But it takes _too_much_. 04:27:57 BTW. 04:27:59 Check http://asau.hotbox.ru/expand.tar.gz 04:36:27 --- quit: AJC ("Leaving") 04:37:32 All this stuff with graphs is neat, but I look for simple 04:37:33 application to compare texts, I see nothing. 04:38:04 I like simple.. and simple bolted-together, etc 04:52:29 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #forth 04:52:54 Guten Abend! 04:53:02 laters 04:53:19 hmm... recycle... brb 04:53:21 --- quit: PoppaVic ("Pulls the pin...") 04:56:18 --- join: PoppaVic (n=pete@0-1pool73-122.nas24.chicago4.il.us.da.qwest.net) joined #forth 04:58:50 --- join: uiuiuiu_ (i=ian@dslb-084-056-220-226.pools.arcor-ip.net) joined #forth 05:03:55 --- quit: uiuiuiu_ (Remote closed the connection) 05:11:38 --- join: PoppaVic1 (n=pete@0-1pool75-206.nas24.chicago4.il.us.da.qwest.net) joined #forth 05:11:54 --- quit: PoppaVic (Nick collision from services.) 05:11:56 --- nick: PoppaVic1 -> PoppaVic 05:28:31 I've fixed bug 05:28:45 Check http://asau.hotbox.ru/expand.tar.gz 05:31:20 ASau: for? doing? why? 05:31:49 I want to use native tools. 05:32:17 The main reason is: I don't have tools I need on PalmOS platform. 05:32:31 ahh 05:32:46 You don't have a shell under Palm OS, so what does getopt work on? 05:33:05 I sloooooowly move things to Forth. 05:33:41 I want test all these things on *IX. 05:34:09 The file paradigm is completely different under Palm OS, as well. 05:34:22 Also, it is training. 05:34:52 Maybe someone find native forth tools useful for their *ix box. 05:36:47 I don't know how to process PalmOS TEXt/REAd documents 05:36:48 other way. 05:37:15 They seem to be essentially sequential. 05:37:32 It's not a native Forth tool -- it calls a function out of libc, or am I misreading it? 05:37:43 BTW, is there API function to compress and decompress them? 05:38:20 I work on moving necessary libc parts. 05:38:44 great gnu 05:38:50 But there's still no shell, and thus no command-line, and so get-opt has noplace from which to get opts. 05:39:29 It does not matter at this moment. 05:40:10 I repeat missed question: 05:40:40 Is there PalmOS function to compress and decompress TEXt/REAd blocks? 05:41:07 TEXt/REAd files are called 'Doc' files. They're not flat files, but segmented PDB databases, with indexed records. 05:41:39 There are no built-in API functions to manage the compression, it's a 3rd-party format. 05:41:39 I'm full aware of it. The question remains as it is. 05:42:20 Does it come with ROM library? 05:42:27 ROM library? 05:42:41 Library, that is in ROM. 05:43:16 I.e. it is accessible as library. 05:43:40 Perhaps someone has written a shared library; I'm not aware of one, however. 05:44:18 Maybe, there's description of RCH1/DTGP format? 05:44:40 What is RCH1/DTGP? 05:44:56 Documents2Go texts. 05:45:19 Perhaps. I've never needed to encode or decode them. 05:45:57 These things are very rarely referred to by their type/creator-ID, by the way, so when you're searching for info you'll do better to use the colloquial names. 05:49:59 --- part: Bourbaki left #forth 05:55:24 Does forth really do a linear search on the dictionary to find words? Or am I mistaken? 05:55:39 Depends on the imp 05:55:43 It's up to implementor. 05:55:46 LoganCapaldo, it is effectively a last-defined to first-defined search, but it doesn't need to be linear. 05:55:52 some hash, some use trees, etc 05:55:57 k 05:56:07 MOSTLY it's "time-based" 05:56:35 You can think of it as a linear search conceptually, but it is commonly optimized for speed. 05:57:00 Anyone have links to an explanation of the "threaded" thing? I have difficulty searching for it ;) 05:57:08 well, you've got "newer vs older" and then cases, lengths, chars, etc: 05:57:11 Thank you 05:57:24 threaded thing? 05:57:25 LoganCapaldo: "threaded" how? 05:57:39 direct indirected, with the implementation 05:57:42 words? pcode/secondaries? 05:57:44 Hmm. 05:57:46 whatever that is exactly? 05:58:00 Quartus: I've got an idea. 05:58:04 Not the coop thread words 05:58:10 ahh - see the FIG site and everything up to F-PC and including Gforth. 05:58:30 Quartus: Do you know of any open-source reimplementation of MemoPad? 05:58:36 yer talking about storing "binary-code" addresses 05:58:39 No threaded code for dummie ssite eh? 05:58:54 ASau, there are several of them, as Palm put out the memopad source years ago. 05:59:12 buy loeliger and Ting books 05:59:40 Quartus, really? I should get it then. 05:59:53 Quartus, maybe, you know names/sites/etc? 06:00:05 Brodie, much as I like 'im, worked for "Forth, Inc." more often than not. As did Chuck. 06:00:13 I don't know if it's still available, in fact. http://palmos.com -> Developers, register, have a look around. 06:00:53 LoganCapaldo: it's not a difficult concept.. Same as using func-ptrs or (eww) basic GOTO 06:01:03 that's all it is? 06:01:09 LoganCapaldo, I remember a fairly concise description in comp.lang.forth, I'm having a look. 06:01:14 for some reason I thought it had to be more complicated than that 06:01:19 hence my confusion 06:01:27 It is more complicated than that. 06:01:36 direct-threaded is code-ptrs... INDIRECT-threaded is something like a ptr to a code-ptr 06:02:21 then you get flavors - like near, far, indexed, etc. 06:02:26 You may look at eforth, if you're asking for cooperative m/t. 06:02:53 ASau: nah, he's worried about the core direct/indirect 06:03:12 I hope "worried" equals "curious" in #forth ;) 06:03:12 FIG-UK site? 06:03:19 "Moving Forth"? 06:03:29 there is a FIG site that is international 06:03:38 taya or some damned thing 06:04:15 Remember: FIG is international.. All others are related or stand-alone. 06:04:52 LoganCapaldo: It's worthy of a few looks, for sure.. But, it takes a lot of head-wrapping 06:04:57 LoganCapaldo, before you get completely drowned in irrelevancies, here's a useful link: http://www.zetetics.com/bj/papers/moving1.htm 06:05:11 Thank you 06:05:57 Quartus: not a bad page - good call 06:06:29 LoganCapaldo, many modern systems do not use a threading method at all, but rather generate optimized native code. 06:06:40 Ah, fascinating 06:06:54 Yeah.. diff between interactive and compiler. 06:07:08 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 06:07:13 No, the native code systems are also interactive. 06:08:58 I misspoke: interactive and generator 06:09:14 Quartus: Does QuartusForth use the token threaded technique? 06:09:28 the generator can do a lot of things behind the scenes, the interactive-program can't/ 06:09:46 LoganCapaldo, no! It's an optimizing native-code compiler. 06:09:53 aha 06:09:54 BUT, meanwhile: that's a good url 06:09:57 cool 06:10:55 --- join: AJC (n=AJC@chello213047059068.25.11.vie.surfer.at) joined #forth 06:11:14 There is no externally visible difference between a threaded and a native-code compiling Forth; both take the same source and turn it into something the computer can run. I have no idea what 'generator' means to you, or how 'interactive' comes into play when talking about compilation methods. 06:11:26 hmm? a famewar again? 06:11:44 Quartus: no biggie 06:11:54 well, thanks a lot guys, you've been most informative 06:12:14 Now I have to go to class and learn about digital circuits 06:12:23 Enjoy! 06:12:32 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #forth 06:27:59 --- quit: ASau (Read error: 104 (Connection reset by peer)) 06:42:44 --- join: ASau (n=user@195.98.180.3) joined #forth 07:06:57 --- join: Raystm2- (n=NanRay@adsl-69-149-58-83.dsl.rcsntx.swbell.net) joined #forth 07:09:27 --- join: Raystm2_ (n=NanRay@adsl-69-149-58-83.dsl.rcsntx.swbell.net) joined #forth 07:10:17 --- quit: Raystm2_ (Client Quit) 07:17:49 --- join: TreyB_ (n=trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 07:18:12 --- quit: TreyB (Read error: 104 (Connection reset by peer)) 07:19:38 --- nick: TreyB_ -> TreyB 07:34:06 --- quit: EdLin (Read error: 145 (Connection timed out)) 07:45:41 --- join: jcwren (n=jcw@adsl-065-006-151-062.sip.asm.bellsouth.net) joined #forth 07:48:02 --- quit: jcw (Read error: 110 (Connection timed out)) 08:00:12 --- quit: PoppaVic ("Pulls the pin...") 08:18:58 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 08:34:04 --- quit: ASau (Remote closed the connection) 08:49:24 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #forth 10:14:49 --- quit: AJC ("Leaving") 10:21:47 --- quit: snowrichard ("Leaving") 11:12:03 --- join: JasonWoof (n=jason@unaffiliated/herkamire) joined #forth 11:12:03 --- mode: ChanServ set +o JasonWoof 11:12:35 --- join: EdLin (n=chacha@as5300-6.216-194-21-245.nyc.ny.metconnect.net) joined #forth 11:15:47 --- join: uiuiuiu_ (i=ian@dslb-084-056-220-226.pools.arcor-ip.net) joined #forth 11:20:52 --- quit: uiuiuiu_ (Remote closed the connection) 12:03:31 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 12:03:33 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-118-243.pools.arcor-ip.net) joined #forth 12:05:29 --- quit: Snoopy42 (Nick collision from services.) 12:05:38 --- quit: snowrichard (Client Quit) 12:06:07 --- nick: snoopy_1711 -> Snoopy42 12:11:04 --- join: uiuiuiu (i=ian@dslb-084-056-220-226.pools.arcor-ip.net) joined #forth 12:12:53 --- part: EdLin left #forth 12:16:14 --- quit: uiuiuiu (Remote closed the connection) 12:16:41 --- join: jwest- (n=jwest@unaffiliated/jwest/x-422957) joined #forth 12:16:45 --- part: jwest- left #forth 12:20:02 --- join: Astrobe (n=fred@c-real.rouen-wireless.net) joined #forth 12:27:09 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 12:46:24 --- join: uiuiuiu_ (i=ian@dslb-084-056-220-226.pools.arcor-ip.net) joined #forth 12:51:29 --- quit: uiuiuiu_ (Remote closed the connection) 13:01:39 --- join: uiuiuiu (i=ian@dslb-084-056-216-124.pools.arcor-ip.net) joined #forth 13:20:35 --- join: Quartus__ (n=Quartus_@209.167.5.1) joined #forth 13:35:53 --- part: Cheery left #forth 13:41:19 --- quit: Astrobe ("leaving") 13:50:06 hello 13:53:30 --- join: N-Dex (n=kevin@c-67-182-180-252.hsd1.ca.comcast.net) joined #forth 13:54:12 Does anyone know if there ever was a native, i386 protected mode polyforth? 14:35:42 --- quit: snowrichard ("Leaving") 14:45:13 don't all chime in at once! :-) 14:46:33 I have no idea. Why do you need to know? 14:46:48 Want to build a server. 14:47:28 and you've decided it has to be polyforth? 14:48:10 Pretty much. Haven't seen a forth that squeezes as much out of the machine as poly. 14:48:47 forth.com is your best point of contact, then. 14:49:11 Pretty much. Haven't seen a forth that squeezes as much out of the machine as poly 14:49:25 oops, darn'd chat client. 14:50:42 Seems I asked them about it awhile back and they don't seem to be very interested in marketing PF anymore. 14:51:03 I remember polyforth as a threaded implementation. I thought it had been supplanted by swiftforth, which is native-code. 14:53:01 looks like it's using "embedded DOS." blech! 14:53:20 Hmm? 14:54:14 oh, that was swiftX. Swiftforth 3.0 looks like it's windows captured 14:55:10 I know there's a windows swiftforth, I don't know enough about the company's products to say what other platform soutions they support under that or any other mark. 14:57:27 'Squeezing x out of a machine' depends entirely on what you mean by x, of course. An optimizing native-code forth will outperform a threaded one on both speed and code size, though, at least in the x86 world. 14:57:54 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 14:58:47 Yeah, guess I'm going to have to roll my own. 14:58:56 I'm sure if you write forth.com and give them your requirements -- non-hosted 386, or whatever you're after, they'll tell you if they have a product in that space. 15:02:16 --- quit: segher (Nick collision from services.) 15:02:26 --- join: segher (n=segher@dslb-084-056-163-153.pools.arcor-ip.net) joined #forth 15:08:53 --- quit: Ray_work ("User pushed the X - because it's Xtra, baby") 15:11:35 --- quit: snowrichard ("Leaving") 15:12:29 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 15:20:50 From what I recall, forth inc has, aside from their cross-compiler embedded solutions, moved toward hosted forths. 15:24:36 --- quit: snowrichard ("Leaving") 15:33:05 --- join: I440r (n=mark4@67.135.84.40) joined #forth 15:48:57 --- join: Jim7J1AJH (n=jim@221x115x224x2.ap221.ftth.ucom.ne.jp) joined #forth 15:56:34 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 16:00:50 --- quit: Raystm2- ("Should have paid the bill.") 16:04:01 --- quit: virl (Remote closed the connection) 16:46:11 --- quit: lukeparrish ("leaving") 16:58:50 --- quit: neceve ("Leaving") 17:00:39 --- quit: I440r (Excess Flood) 17:01:38 --- join: I440r (n=mark4@67.135.84.40) joined #forth 17:05:10 --- quit: I440r (Client Quit) 17:05:18 hi 17:05:23 re 17:10:33 how are you? 17:10:54 warming up my coffee brb 17:13:29 --- join: rabbitwhite (n=trip_n_s@c-68-34-1-138.hsd1.md.comcast.net) joined #forth 17:13:52 back 17:42:41 --- quit: rabbitwhite () 17:49:07 --- join: Bensel (n=Bensel@d88.as7.gdrp.mi.core.com) joined #forth 17:52:59 --- quit: snowrichard ("Leaving") 18:18:16 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 18:19:03 hi 18:22:24 --- quit: snowrichard (Client Quit) 18:53:47 --- join: nighty_ (n=nighty@H123.C72.B0.tor.eicat.ca) joined #forth 19:10:41 --- join: snowrichard (n=richard@12.18.108.162) joined #forth 19:10:53 hi 19:11:03 lo 19:11:51 up down 19:16:22 ( well, I meant "lo" as in "hello", but that interpretation works too :)) 19:17:36 --- quit: snowrichard ("Leaving") 20:24:05 --- quit: Bensel ("bed") 20:26:17 --- join: EdLin (n=chacha@as5300-6.216-194-21-245.nyc.ny.metconnect.net) joined #forth 20:39:21 --- part: EdLin left #forth 20:51:32 --- quit: segher (Nick collision from services.) 20:51:42 --- join: segher_ (n=segher@dslb-084-056-129-163.pools.arcor-ip.net) joined #forth 20:58:07 --- quit: nighty_ (Remote closed the connection) 22:13:00 --- join: I440r (n=mark4@12.197.217.242) joined #forth 22:27:51 --- nick: I440r -> i_refuse_to_regi 22:27:58 --- nick: i_refuse_to_regi -> I440r 22:28:01 bastards 22:28:16 Just register, you lame bugger. :) 22:28:25 fuck no 22:28:30 i am registerd. 22:28:36 i fucking chose if i log in or not 22:28:51 i think its fucking LAME to enforce me to 22:29:05 Not everything needs to be a battle. 22:29:13 it doesnt do a damned thing to stop spam because in all the years ive been here ive NEVER been spammed 22:29:23 it pisses me off 22:29:30 i cant /join #gentoo 22:29:37 because of this stupid fucking polacy 22:29:50 So login. 22:29:55 why 22:30:01 Why not? 22:30:03 why do i have to type that SHIT in every time i come here 22:30:09 because its a pain in the fucking ass 22:30:19 Automate it, then. 22:30:39 thats even more of a pain in the ass 22:31:37 fuck off 22:31:39 --- part: I440r left #forth 22:31:52 --- join: I440r (n=mark4@12.197.217.242) joined #forth 22:31:58 not you lol 23:04:51 --- quit: I440r (Excess Flood) 23:06:08 --- join: I440r (n=mark4@12.197.217.242) joined #forth 23:07:04 --- join: ASau (n=user@195.98.180.3) joined #forth 23:07:25 Good morning! 23:37:59 I440r: once again, you are proving that profanity is the language programmers know best! :-) 23:38:18 heh 23:38:22 hi guys and gals 23:38:38 hi 23:43:16 Hi fission 23:44:47 Quartus ! :D 23:48:19 how's the retroforth stuff? 23:48:39 Working on a few bugs with crc. 23:54:03 I was trying hard, but I don't think I can resist having a Pimm's No 1 cup 23:59:59 --- log: ended forth/06.09.26