00:00:00 --- log: started retro/06.10.15 01:45:35 --- quit: virl (Remote closed the connection) 01:58:17 --- quit: nighty_ (Read error: 104 (Connection reset by peer)) 02:33:33 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #retro 02:48:29 --- join: Cheery (n=Cheery@a81-197-19-23.elisa-laajakaista.fi) joined #retro 04:09:40 now I've my virtual machine. 04:09:43 Damn it's primitive. :) 04:40:52 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #retro 05:14:21 --- quit: virl (Read error: 110 (Connection timed out)) 05:29:30 --- join: erider (n=erider@unaffiliated/erider) joined #retro 05:34:15 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #retro 07:18:58 --- quit: erider (Read error: 110 (Connection timed out)) 07:19:23 --- join: erider (n=erider@unaffiliated/erider) joined #retro 08:27:48 --- quit: erider (Read error: 110 (Connection timed out)) 08:28:14 --- join: erider (n=erider@unaffiliated/erider) joined #retro 09:01:00 --- quit: erider (Read error: 110 (Connection timed out)) 09:01:25 --- join: erider (n=erider@unaffiliated/erider) joined #retro 09:25:17 good afternoon 09:34:31 hi 09:35:39 hi 09:36:18 crc how is your gtk+ coming 09:36:26 the basic bindings are done 09:36:51 I have done a couple of simple programs with it; but still have to work on bindings for more widgets 09:37:00 crc: have you seen the commenting system lukeparrish or somebody introduced? 09:37:07 I was just looking at a jpg of what zpg had done 09:37:18 Cheery: the prefix based one? 09:37:34 yes, one which used backslashes. 09:37:38 yeah 09:37:43 \comment\ 09:37:47 \comment 09:38:00 What about adding it to rx? 09:38:01 :) 09:38:05 And removing the | 09:38:23 I will consider it 09:38:58 I think it is enjoyable. 09:39:52 Have you seen this? http://retroforth.net/paste/?id=223 09:40:49 I had not seen that 09:42:07 Cheery: I am working with an older version of ipkg as a basis for the library update/installation tool 09:43:11 cool 09:43:33 I should have something ready to show off later today 09:47:42 did you see http://ezil.sourceforge.net/screenshots/gtk-forth.jpg 09:47:50 crc 09:47:59 yes 09:48:07 hello world is a trivial app in gtk :) 09:48:21 --- join: Raystm2- (n=NanRay@adsl-68-93-41-121.dsl.rcsntx.swbell.net) joined #retro 09:48:38 http://ezil.sourceforge.net/devel/gtk-forth.html 09:49:46 http://retroforth.net/paste/?id=201 09:50:26 using my wrappers over the raw GTK+ api 09:52:10 cool 09:52:11 that yields http://retroforth.org/gtkhello.png when run 09:53:44 --- quit: Raystm2 (Read error: 104 (Connection reset by peer)) 09:54:42 cool! 09:55:05 --- join: nighty__ (n=nighty@sushi.rural-networks.com) joined #retro 09:55:16 I wish I could read forth better 09:58:50 heh crc 09:59:34 virl: hi 10:01:00 reading forth is a problem? 10:02:23 I can read assembly easier then forth 10:05:03 for me it's the opposit 10:05:21 virl: you find forth easy to read? 10:05:48 --- quit: nighty_ (Read error: 110 (Connection timed out)) 10:05:49 when it's well written, yes. 10:06:03 crc: what window manager do you use? 10:07:15 hmm.. something with Desktop, which ones have a desktop directory? KDE for sure, Gnome? and which others? 10:07:22 at the moment I am using compiz+gnome 10:07:33 ok, yes.. 10:07:48 I also use ratpoison, blackbox, and sometimes WindowMaker 11:18:29 ok, the basic install/remove functionality is working 11:25:53 --- quit: timlarson ("Leaving") 11:26:35 --- join: timlarson (n=timlarso@user-12l325b.cable.mindspring.com) joined #retro 11:34:26 * Raystm2- likes | pipe comments in Retro. This tells me from the first comment, without any other context, that the forth source i'm looking at is a retroforth source. 11:35:08 --- nick: Raystm2- -> Raystm2 11:35:15 --- mode: ChanServ set +o Raystm2 11:37:29 now I can install, remove, list all installed libaries, and query individual libraries for some information about them 11:38:43 crc: query? like looking in a 'not-all-ready-installed' library, looking for the source listing of the word you need? 11:39:57 no, each library provides a brief description with things like the "needs" line to use 11:40:08 okay. 11:40:19 rxforth info would give this 11:40:53 I still have to code the "needs" word 11:41:17 for now, I am putting the libraries in /usr/share/rxforth/library 11:41:17 Needs is an import word that only imports once? 11:41:24 eventually, yes 11:41:32 it also has to know where the library modules are 11:55:03 other than not importing more than once, I have a functional "needs" now 12:35:41 now it's loading from ~/.rxforth/library/ 12:41:21 crc am I doing something wrong here... 12:41:42 ? 12:41:47 I'm creating a page in projects/glypher called notes 12:41:54 ok 12:42:11 i've hi-jacked and made changes to book calling it notesbook. 12:42:20 crc: have you add it to the rxforth yet? 12:42:34 Cheery: it'll be up in about an hour 12:42:38 ok. :) 12:43:00 notesbook is supposed to do what book does for handbook only for notes, generate the html, ps, txt, from the rdml. 12:43:12 when I call notesbook it never returns. 12:46:18 Raystm2: it should work now 12:46:32 rdml files need to end with \end 12:47:09 :) 12:47:16 yikes okay thanks. 12:47:54 I should port rdml over to rxforth... would make some things easier and less troublesome :) 12:48:33 yes, indeed. you could have a fully inclusive documenting facility, right on board. 12:49:12 infact, with some hacking, the whole docstring thingy might could use the power of the rdml. 13:09:41 rxforth 1c is up on the google group 13:10:48 cool. 13:11:23 neat 13:13:02 to run the gtk+ examples, you'll have to use rxpkg to get the current library module :) 13:13:05 Raystm2: would pipe stack diagrams work? I wonder... : foo | a b -- c \ bar baz ; 13:13:33 stack diagrams are in parens, tipically... 13:14:04 ooh, would be backwards-compatible with most existing pipe comments since parse goes to end of line anyway 13:14:27 yeah. I was thinking a key easier to hit would be better 13:14:45 I strongly disagree with that idea. 13:14:53 oh okay. I get the problem now. I was wondering why all the fuss. 13:15:47 : foo |a b -- c\ bar baz ; 13:17:27 parens are counter-intuitive. in math, we are used to them meaning either "do this first" or "feed this into a function". having them mean "do nothing" is kind of strange. 13:17:47 They're being used parenthetically, as in written prose. 13:19:14 Or, perhaps more illustratively: they're being used parenthetically (as in written prose), not mathematically. 13:19:33 "say this in a lower tone, it is important information that would otherwise impede the flow of dialogue" 13:19:49 Besides, all forthers grow up with the stack diagram in parens. I've always like the pipe commentuer because you immediately recognise that you are in a Rx based forth and that things will be retroForthish from that point on. 13:20:02 Correct. So they're not counter-intuitive, unless you're confusing Forth with mathematics. 13:20:12 Confuse it instead with prose, and the parens make perfect sense. 13:20:19 forth *is* math. 13:20:31 I disagree. 13:20:31 The universe is math. :) 13:20:37 functions, numbers. it's all math. 13:20:48 Language, and abstraction; it's language. 13:21:00 well it's really more like a machine, but it's a math machine. 13:21:10 fancy calculator! 13:21:20 math is language... /me persists for no-good reason. 13:21:24 :) 13:21:34 A narrow view, and one at odds with the aspects of good Forth. Naming is key; naming is language. 13:22:04 naming of math symbols. 13:22:42 besides, RPN makes the need for mathematical parens obsolete, doesn't it? 13:22:44 * lukeparrish persists for no good reason as well :) 13:22:49 Math is a subset of language; Forth is a subset of language. That doesn't mean math=Forth, as they're different subsets, and use different alphabets, different semantics, different concepts. 13:23:59 here is a Venn diagram intersecting those languages, as you can see, there are intersections and there are outersections... 13:24:00 hmm. forth is a subset of language, which includes math as it's (major?) component. 13:24:53 I guess programming (of any kind) includes both linguistic and math elements. 13:26:11 (some programming might exclude language elements... but that's debatable, since any semantic could be termed "language" by definition.) 13:26:12 At any rate, the use of ( ) for comments in Forth is not at all at odds with the idea of Forth as a human language, like English. 13:27:25 well, it seems to mess with the flow of it for me. maybe I'll get used to it eventually, but that doesn't mean there's no better way to approach it. 13:28:55 It's one of a few conventions that goes back to the earliest days of Forth; it's well-accepted. To vary it is to ghettoize your own code. That may not concern you, of course. 13:29:36 I'm bold that way. When it suits me. :) 13:30:00 :) 13:30:34 It seems to me that | is no easier to type than (. 13:30:39 "Forth, things written in stone are softer then they appear." 13:30:47 hey I could do something really strange like use colors instead of punctuation :) 13:31:13 NO! Stop! Go back! It's a trap! 13:31:48 it's easier. less stretching. also, the closing mark in that case could be \, which has no shifting. 13:32:30 (could vary in easiness depending on the keyboard. mine is right by my pinky.) 13:32:54 * Raystm2 is so used to hitting backspace that the pipe combo is actually a smidge easier to hit. 13:32:56 Stretching? ( is left-little-finger on shift, right-second-finger on 9. | is left-little-finger on shift, right-little-finger on \. Much further away. At any rate the effort is pretty similar. 13:33:59 also ( and ) are on different keys. | and \ are on the same key. less mental switching. 13:34:27 They are not, however, paired delimiters in any accepted way. 13:35:01 why would a forther care about that? : and ; aren't, outside of the forth community. 13:36:02 They're not used as paired delimiters in Forth, though. : is a defining word that takes an argument. ; is a sentinel of a sort to end the definition. Like the words BEGIN and END in some languages; those aren't paired delimiters either. ( ) are. [ ] are. { } are. 13:39:19 not sure I see the distinction. 13:39:51 every : ends with a ;, so they are paired. 13:40:09 since they start and end something, I'd call them delimeters. 13:40:58 s/delimeters/delimiters/ 13:41:57 I wouldn't. But supposing you do call them delimiters; | and \ have no history behind them at all as a pair of delimiters. The semantic meanings of \ in Forth are already well known, and | has other computing connotations. They don't pair visually or conceptually. 13:43:25 besides, what do you do when you want to parse a directory and you need \ ? 13:43:46 use / 13:43:55 :) okay. thanks :) 13:44:34 it is recognized by both windows and unix systems, so there's no reason to force \ as the directory separator inside a program 13:44:50 I knew that. 13:46:07 the questions that count are the visual one and the convenience one. that is probably how Chuck Moore would look at it. 13:46:54 He did, and I believe he eliminated the need. 13:46:58 does it work visually? does it impede one's ability to type code smoothly? 13:47:15 I sincerely doubt he would agree that visual and convenience issues are paramount, but I'm not talking about the opinions of a reclusive eccentric genius, but rather accepted semantics and general usefulness. 13:47:22 : foo | a b -- c \ bar baz ; 13:48:08 lukeparrish: your argument is compelling to me, that does look okay. 13:48:11 I think I could get used to it visually 13:48:50 it certainly seems ergonomically preferable to me, personally. 13:49:25 the next question is how many other people would find it to be such (on either or both counts) 13:49:53 but something just as or more powerfull already exists. Your membership to the society as a whole is in jeopardy and your fee will be returned to you, should the counsil deam you are to be excomunicated. 13:50:10 OH NO! :P 13:50:14 hehe. :) 13:52:06 --- join: Anbidian (i=never@S0106000fb09cff56.ed.shawcable.net) joined #retro 13:52:18 welcome, Anbidian 13:52:34 Anbidian ; 13:53:55 hello Anbidian 13:54:43 Greetings all. 13:55:27 * lukeparrish has been experimenting with a new commenting syntax 13:55:40 : foo | a b -- c \ bar baz ; 13:56:20 That is new? 13:56:35 Anbidian, he's using | and \ to replace ( and ) 13:56:38 oh wait the ; is at the end 13:56:44 this is as a stack diagram 13:56:47 --- join: erider_ (n=erider@unaffiliated/erider) joined #retro 13:57:07 ah, I see Quartus 13:57:58 Quartus: what do you use Retro for? 13:58:06 the two questions are: 1) is it really easier to type? 2) is it as easy or easier to read? 13:58:47 Anbidian, I use RetroForth as an ANS Standard Forth with my layer on top of it. 13:59:12 3) Is it necessary (and I'm not saying it's not, I don't know, I just remember reading once Mr. Moore saying how he initially started using stack diagrams but abandoned them in the end because he was doing things too complicated if he found a need for them) 13:59:23 you have to compare them together: 13:59:24 : foo | a b -- c \ bar baz ; 13:59:24 : foo ( a b -- c ) bar baz ; 13:59:54 Quartus: Do you use it on Windows or Linux? 14:00:17 Windows and FreeBSD, mostly Windows. 14:00:28 hi erider_ 14:00:32 even more: 14:00:32 : foo | a b -- c \ bar baz ; | this 14:00:32 : foo ( a b -- c ) bar baz ; | this 14:00:47 --- quit: erider (Read error: 110 (Connection timed out)) 14:01:01 Hmm. 14:02:42 Apparently, in example one here, pipe does double duty, commenting out till end of line unless first terminated by backslash. 14:02:57 right 14:03:33 I guess it would have to be smart about that in multi-line functions with comments. 14:03:38 doesn't hurt much to add to pipe, I suppose. 14:04:12 in the current implementation of parse, when in the interpreter loop it terminates at the end of line anyway. 14:04:39 right so this is just parsing for the \ and exiting early. 14:04:48 so \ would simply be an early escape from | 14:04:53 ya. 14:06:00 or on compile \ backtracks to | and compiles ( and then terminates itself with \ as ) in it's place. 14:07:25 this eliminates the need for branching the | everytime it is encountered with something you can do at compile time. 14:07:53 Raystm2, maybe it's just me, but I can't make head nor tails of what you just said. :) 14:07:55 no, on the compile it just skips the intervening part using a macro/self 14:08:16 --- join: erider__ (n=erider@nc-71-50-113-74.dyn.embarqhsd.net) joined #retro 14:08:26 Quartus: sad thing is, neither can I. 14:08:37 --- nick: erider__ -> erider 14:09:18 * Raystm2 lost his perspective on that one. 14:09:32 crc: do you have some packages to test? 14:10:56 for rxpkg? 14:11:05 the gtk and x11 packages are online 14:11:15 those are all I've setup for the library so far 14:11:20 thing is, \ is already a commentuer in languages, so where is the comment, | here \ or here ? 14:12:08 http://retroforth.net/paste/?id=224 14:12:28 the first here :) 14:12:37 |comment\ 14:13:12 right but if i'm used to seeing... 14:13:15 \ comment 14:13:18 it might be confusing if you're really used to \ comments 14:13:26 in my language, i begin to get confused. 14:13:26 and never heard of | comments 14:13:34 ya. 14:13:49 * Raystm2 actually makes a point and marks day on calendar. 14:14:00 :) 14:14:12 I wonder how long that confusion would last. 14:14:34 not long, about as long as learning that | was the commenter. 14:14:42 and how it worked. 14:15:06 --- quit: erider_ (Read error: 110 (Connection timed out)) 14:15:17 its backwards compatable, and you haven't quit the original operation of ( and ) . 14:15:29 its mostly harmless. 14:15:47 mostly ;) 14:17:01 mostly because it adds complexity no matter how small, to the system as a whole and to the knowledge one must aquire to adjust to the system. 14:17:29 But it's danceable, I give it an 92. :) 14:19:32 another thing is making it a prefix, lets you leave spaces out from between the pipe and comment. 14:19:42 |comment\ instead of | comment \ 14:20:12 |comment \ <-- not this? 14:20:20 that would work. 14:20:35 but it's completely optional 14:21:14 : foo|blah \bar ; would not work because there's no space once the parsed area is snipped out. 14:21:43 but : foo |blah\bar ; or : foo|blah\ bar ; would. 14:22:22 so now : has to parse for | while createing the name of the new word? 14:22:36 no wait, : foo|blah\ bar; would not work. 14:22:57 | must be preceded by a space. 14:23:02 So the process of being different is to take something simple and well-established, and complicate it? Forth words are space-delimited throughout, unless they're in this special list of exceptions that you have to remember. Maybe a post-it note on the monitor. 14:23:52 parsing prefixes effectively make the following text part of the same word. 14:24:03 not in what you just poste.d 14:25:36 | is now a prefix (or optionally a word that does exactly the same thing, if you care to add a space.) \ is a delimiter, so it doesn't need a preceding space. 14:27:54 : test |abc--abc\.s ; works fine because the parsing word stops before .s 14:28:31 but I expect most people will find : test |abc--abc\ .s ; more aesthetically appealing, so they'll go with it. 14:28:35 the old stupid sayings..' ANS makes it right' 14:29:11 virl: you mean "trusted and well-established makes it right" ;) 14:29:54 yes, that's what I meant, and I hope it's clear that I find it stupid. 14:30:28 arbitrarily renaming existing facilities is not progress. 14:30:43 ANS Forth isn't Forth but a language with the same name. I love this quote of chuck. 14:31:15 The only thing Chuck thinks is Forth is whatever variant he's currently using. That means, to follow Chuck, you'd be using colorForth now and nothing but. 14:32:08 Quartus: this is a real mechanical change, not a simple renaming. removing an unnecessary space requirement, and using a more accessible and readable key. 14:32:37 whatever, I think it's worth experiementing and those who like to do crazy things should be in #retro the others who love ans and co should go to #forth 14:32:41 To provide a facility, under a new name, that exists already in well-accepted form. 14:33:33 when it would be well accepted then there wouldn't be some people who do it on other ways 14:33:46 virl: I don't like kicking out those with opposing views. 14:34:19 very bad, since they are a sort of error-checking mechanism (if you don't let it get under your skin :/ ) 14:34:47 lukeparrish, me to, but I don't like it when they reduce the fun in the channel. 14:35:16 (even if kicking out is metaphorical, i.e. telling them they aren't welcome or such.) 14:35:18 I don't oppose progress and experimentation; I just don't think arbitrary renaming, even with questionable minor changes, is progress. 14:35:55 This is a discussion channel, virl; we're discussing. You're the one trying to end the discussion; perhaps you're the one who is in the wrong place. 14:36:16 well, in my mind the naming was/is arbitrary to start with. it's sufficiently obvious that whatever comes after the name of a function is supposed to be a stack diagram. 14:36:28 I never was at the wrong place 14:36:38 In fact, the pipe is unnecessary. you could make colon parse for \ (or whatever) anyway 14:37:53 : : wsparse entry '\ parse 2drop ] ; 14:38:51 virl is welcome here, in my book. so is Quartus. I would appreciate very much if they do not quarrel over personal dislikes. 14:39:01 crc: :) 14:43:29 hmm. I wonder if that would seriously be a good idea. 14:44:05 automatic stack diagrams 14:50:19 it would make stack diagrams automatically more popular. people not used to using them would be able to skip with little effort. 14:50:56 : foo a b -- c\ bar baz ; 14:51:13 doesn't have the visual appeal of seperating the comment from the name though 14:51:38 : foo \ bar baz ; 14:51:49 seperates the code from the name pretty good though 14:52:16 : foo word1 word2 ; a b -- c 14:52:45 : test code with a descriptive name | a b -- c \ bar baz ; 15:02:58 Anbidian: I think placing the stack diagram next to the name is probably better. definitions have widely varying lengths, so finding what you want to know wouldn't be as easy. 15:11:12 --- quit: Cheery ("Download Gaim: http://gaim.sourceforge.net/") 16:04:21 --- quit: virl (Remote closed the connection) 16:41:55 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-151-054.pools.arcor-ip.net) joined #retro 16:50:37 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 16:51:02 --- nick: snoopy_1711 -> Snoopy42 17:01:45 --- quit: timlarson (orwell.freenode.net irc.freenode.net) 17:02:00 --- join: timlarson (n=timlarso@user-12l325b.cable.mindspring.com) joined #retro 17:23:06 I updated the RxForth 1c prerelease snapshot with a better copy of rxpkg 18:07:32 --- quit: Quartus_ (Read error: 60 (Operation timed out)) 18:24:33 --- join: nighty_ (n=nighty@sushi.rural-networks.com) joined #retro 18:38:20 --- quit: nighty__ (Read error: 110 (Connection timed out)) 18:57:06 --- quit: Anbidian () 19:01:20 --- quit: erider ("I don't sleep because sleep is the cousin of death!") 20:08:37 --- quit: Raystm2 ("Should have paid the bill.") 20:11:18 --- join: Raystm2 (n=NanRay@adsl-68-93-41-121.dsl.rcsntx.swbell.net) joined #retro 20:12:24 --- mode: ChanServ set +o Raystm2 22:05:01 --- quit: nighty_ ("Disappears in a puff of smoke") 23:59:59 --- log: ended retro/06.10.15