00:00:00 --- log: started forth/07.04.17 00:05:31 --- quit: ASau ("ERC Version 5.0.1 $Revision: 1.726.2.3 $ (IRC client for Emacs)") 00:13:09 --- quit: skas_wk ("Leaving") 00:17:41 --- join: ecraven (i=nex@eutyche.swe.uni-linz.ac.at) joined #forth 00:33:54 anybody around? 00:34:56 hey 00:35:12 feelin' chatty 00:35:23 I got going about database design 00:35:53 --- join: nighty-- (n=nighty-@66-163-28-100.ip.tor.radiant.net) joined #forth 00:35:58 Once I got past the screwy naming conventions and way the author was thinking about it 00:36:06 it turns out that the database structure was the way I'd do it 00:36:46 the table with all the important data (the only thing I need to fetch and display on my front-end) was in a table that the database author has labeled as a many-to-many table 00:37:29 I think I should be a consultant some day 00:38:00 i'm wondering why my GC fails if the heap is larger than 1gb 00:38:08 because I get really engaged talking about what the code really does, and trying to find the best way to do it and to describe it 00:38:24 slava: that's an odd barrier 00:38:43 slava: 2GiB or 4GiB would be less suprising 00:38:56 it only happens if you run code which repeatedly allocates larger and larger structures, doubling the heap twice from 256mb to 1gb 00:39:05 starting with a 1gb heap doesn't trigger it 00:39:32 huh 00:39:44 hope your computer is newer than mine 00:39:49 the test is easy, allocate 16 million identical arrays in a loop 00:39:59 I've only got 256MB 00:40:27 i have 2gb, and i've found various scalability problems by maxing out 00:40:37 :) 00:40:46 gc takes 2.4 seconds with a 1gb heap, i need to work on that too 00:40:59 hehe 00:41:10 I can't imagine needing that much ram for anything 00:41:17 scientific calculations, mainly 00:41:19 data mining... 00:41:23 but I suppose it'll happen some day 00:41:28 no reasonable end-user application will need that much data 00:41:41 video editing perhaps 00:41:51 I can understand a cache being able to use as much ram as you have 00:42:17 my vision for my little forth is to have it run in a 1MiB block 00:42:40 i'm thinking if i can optimize for huge data sets on my powerpc g5, then i'll get good performance running small programs on handhelds :) 00:42:44 and any extra memory can serve the external cache thing 00:42:56 cool. 1mb is more than enough for a lot of functionality 00:42:57 heh 00:43:17 Terminal.app is using something like 40mb right now :/ 00:43:30 the only thing that's tempting me to increase that number is that my frame buffer is bigger than that :) 00:43:54 it's awefully convenient to be able to plot pixels all over an offscreen buffer 00:44:11 i hear modern video cards don't even like giving you direct frame buffer access anymore 00:44:26 bastards :) 00:44:47 --- quit: nighty- (Read error: 110 (Connection timed out)) 00:44:57 i want to write a compiler for a forth-like language which generates Cg code -- the OpenGL vertex shader language 00:45:02 you can do parallel calcualtions on the GPU 00:45:16 woh! you can write code for the GPU?! 00:45:20 I've always wanted to do that 00:45:26 with newer video cards, yes 00:45:32 they support procedural textures and shading 00:45:33 sweet 00:46:06 there's a language with C-like syntax but not really like C, and it's part of the OpenGL (2.0 ?) spec 00:46:20 cool 00:46:38 do you pass that text to the graphics card? or compile it first? 00:46:58 you pass it to your OpenGL implementation, it might execute it in software it in the card, or a bit of both 00:47:35 right 00:47:43 can you ask if it'll do stuff on the card? 00:47:47 i'm not sure how general purpose a GPU is, it might need the CPU to send it an instruction stream in real time 00:47:51 or do you just have to try it and see how long it takes? 00:48:14 probably best to time it anyway 00:48:17 i don't know, in general the opengl api doesn't provide a way of knowing if you're doing software or hardware rendering 00:48:24 since what you really want to know is "is it fast?" 00:48:33 and with some cards enabling certain features incurs drastic slowdowns 00:48:40 like the fog effect on older cards 00:48:42 right 00:48:59 so it doesn't matter if it's done on the card or not. 00:49:05 only whether it's fast or not 00:49:10 it does, because it might be unusably slow if it's in software 00:49:19 i don't think there are any fast software opengl implementations 00:49:33 you can write a program that does simple 2d graphics with x11 and opengl, it will be much slower unless you have a 3d card 00:49:43 well, if I write a simple 3d game that doesn't do anything too fancy 00:49:57 I'm sure 5 years in the future or so, there'll be a software renderer that could handle it fine 00:50:01 opengl is fast for 3d, but for 2d graphics it's slower than libraries optimized for 2d 00:50:13 because all rendering goes through a complex geometry/rendering pipeline 00:50:15 and, as you pointed out, just because it's done on the card, doesn't mean it's done fast 00:50:33 i still like opengl though, it's pretty much the only way to do portable graphics, 2d or 3d 00:50:49 yeah 00:50:59 it seems like a silly way to do 2d, but... 00:51:03 it's what works 00:51:16 os x uses the 3d card for 2d 00:51:21 every window is an opengl textured quad 00:51:22 yeah 00:51:31 yep :) 00:51:34 in the future i think all graphics apis will be 3d 00:51:59 at some level 00:52:02 microsoft has some new stuff in vista 00:52:17 there'll still be the basic moveto(), lineto(), setcolor() api available as a layer on top though 00:52:48 guess you mean the API at the OS level though 00:52:56 well os x has various high level 2d apis 00:53:06 like what you described except you can set the pen width, stroke shape etc 00:53:07 as it should 00:53:18 but there's nothing like that which is cross-platform 00:53:22 cairo is still immature 00:53:41 gtk isn't well supported 00:53:50 gtk requires x11 on mac os 00:53:52 SDL works pretty well, but doesn't have the feature set you want 00:54:01 sdl doesn't have geometric primitives 00:54:13 they're easy 00:54:18 I'm sure you could find some 00:54:24 i used to use SDL_gfx 00:54:28 have you checked for an SDL extension that gives you them? 00:54:43 my main problem with SDL is that it is limited to a single window, and doesn't abstract the clipboard 00:54:47 I was guessing SDL_gfx might have the line-drawing functions 00:54:53 right 00:54:55 I know 00:55:24 the one-window thing makes it so you can't use it for lots of things 00:55:36 although it's a limitation I wish all applications had 00:55:41 i ended up essentially rewriting sdl in factor 00:55:50 i have my own abstraction layer 00:55:52 I don't much like popups, or dialog boxes, or floating toolbars 00:56:04 nor do i, but i like having distinct windows for different applications 00:56:27 I wish programs would find a place to put that crap that's not just layered in front of themselves or other programs in a random-ish place chosen by your window manager 00:57:00 yeah, like a single floating window which contained all pending progress status bars, alerts and query dialogs 00:57:13 no, don't float it 00:57:19 that doesn't solve the problem 00:57:19 tile it, whatever 00:57:25 it just makes it your window managers problem 00:57:55 i find floating windows make the most sense on a large screen, but i still like it when applications reuse one window because it is just more keyboard friendly 00:58:17 I like it when they just use a strip of the main window 00:58:20 like gqview 00:58:40 it uses the left two inches or so for the file selection and thumbnails 00:58:55 i have a 19" crt for e-mail and irc, and a 30" lcd with three editor windows next to each other 00:58:57 and a strip accross the top for the typical menus [file] [edit] or whatever 00:59:29 but on the 19" screen, i wouldn't want to tile the windows 00:59:38 they're in this sort of ad-hoc arrangement which is quite convenient 00:59:54 ahh, I never found the ad-hock thing convenient 01:00:13 I found that I wasted lots of annoying time moving and resizing things so they would be decently large 01:00:30 which I got thuroughly sick of 01:01:08 maybe if I had a 30" screen I could convince my programs to make their windows big enough to be useful and nice 01:01:24 and there'd still be room to splay them out such that I can see part of each and click to bring up the one I want 01:01:38 not sure I'd really like it though 01:01:49 I like having everything in the same size/place and having tabs to switch 01:03:35 I made a site with a 512x512px square and a button below it 01:03:45 this didn't fit on my friends laptop screen 01:03:52 because there was so much extra crap 01:03:57 I hit the maximize button 01:04:18 but somehow there was still less than 540px tall that I wanted 01:04:56 the dock, the statusbar, the bookmarks, the navigation bar, the window title bar, the apple menubar 01:05:06 heh 01:05:17 you can hide all that except for the menu bar and title bar 01:05:23 it was a pretty nice apple laptop 01:05:24 not old 01:05:54 yeah, but it hurts your usability to do so 01:08:26 with the setup I've got (ion) the content area on my "windows" is larger than you could get on a mac by turning all that stuff off 01:08:45 heh, on the other hand i don't spend time configuring this ion :) 01:09:17 right, that's the trouble 01:09:30 point and click interfaces are slow to use, but don't require you to know anything 01:12:09 well, actually, I think keyboard interfaces (even terminal interfaces like managing files with cp, mv, rm, etc) are easier to learn than say windoze 01:12:26 problem is that just about everybody is already trained in how to use mac/windoze 01:13:26 people aren't into training 01:13:31 and don't seem to care much about effeciency 01:14:01 I still have a hard time understanding how there are so many people who spend hours of every day in front of a computer, and still can't type worth a damn 01:14:43 heh 01:14:51 i can touch type but i'm not as fast as some people 01:14:53 I had a boss at a web development company who sat in front of a computer all day who whould actually mutter "now where's the W" 01:14:56 and that sort of thing 01:15:07 lol i'm not that bad 01:15:38 I don't care if you can do at fast, but I think you should be able to touch type 01:15:45 kinda like reading 01:15:57 I don't care if you can read particularly fast 01:16:08 I NOT READ GUD!!!11 01:16:11 but you aught to be able to do it without sounding all the words out 01:16:26 so-un-ding 01:16:26 to find out what words they are 01:17:02 God created the universe in 7 days, werty refactored it in 2 days and made it fit in 8Kb. 01:17:39 is it GPL? 01:18:54 source code is for luddites 01:19:20 hehe 01:19:46 I've heard about those machines where you flip the eight switches the way you want them, and then press the button to store it 01:20:04 sounds efficient 01:21:01 yeah, you could do 8KB in about 2 days 01:22:34 --- join: ygrek (i=user@gateway/tor/x-c86b3d65ea352d8a) joined #forth 01:46:20 --- join: skas_wk (n=skas@202-90-54-135.static.linearg.net) joined #forth 01:48:11 --- join: ygrek_ (i=user@gateway/tor/x-a959b140c015d445) joined #forth 01:57:45 --- quit: yumehito (Read error: 110 (Connection timed out)) 02:04:08 --- quit: ygrek_ (Remote closed the connection) 02:04:48 --- join: ygrek_ (i=user@gateway/tor/x-7772c8800184e20c) joined #forth 02:14:04 --- quit: ygrek (Remote closed the connection) 02:59:43 --- quit: JasonWoof ("off to bed") 03:14:48 --- join: segher (n=segher@dslb-084-056-194-208.pools.arcor-ip.net) joined #forth 04:33:18 --- part: skas_wk left #forth 04:34:30 --- join: skas_wk (n=skas@202-90-54-135.static.linearg.net) joined #forth 04:57:15 --- part: skas_wk left #forth 05:08:15 hm. what forth does #forth recommend? 05:14:54 Which platform(s) do you have in mind? 05:15:07 x86 linux 05:15:47 --- quit: ecraven ("bbl") 05:15:48 In that case, you'll probably want gforth. You can find it packaged for most distros. 05:16:03 gforth is a good start for every unix and unix-like 05:16:22 cheers 05:24:00 forth seems very cool 05:27:10 not only is it cool, it is actually very useful too :-) 05:28:49 segher, seems that way from the literature i've come across - a strong real-world focus to it, unlike say, smalltak 05:29:22 oh smalltalk is great in its own world 05:29:45 i suppose you mean, the problem with smalltalk is, it *has* a world of its own ;-) 05:29:54 heh, exactly 05:30:20 it was an academic research thing originally 05:30:49 yes, it is a nice idea 05:32:29 hm, why cant i push + by itself without something before it? 05:32:45 gforth says: *the terminal*:47: Stack underflow 05:34:52 you don't "push" + 05:34:56 + is a word that eats two stack items, and pushes back another 05:34:58 yeah 05:35:02 you need two items that you want to add on the stack first 05:35:04 so you do 1 2 + 05:36:24 ok, maybe this will make no sense - but can you use a word as an argument - sort of like higher order functions in lisp? 05:37:59 you can take a ref to a word by "ticking" it -- the result is called an "xt" (execution token), and you can run that by calling execute 05:37:59 so 1 2 ' + execute 05:38:13 within a definition, you need to use ['] instead of ' 05:39:01 so : three 1 2 ['] + execute ; 05:39:32 segher, ok, but i noticed that i can do: 1 2 3 + 05:39:36 why does this work? 05:39:48 does forth know how many things ive put on the stack 05:40:03 you push 1, push 2, push 3, add 2 and 3 together 05:40:07 type .s 05:40:34 it shows what's on your stack 05:40:34 very useful for debugging 05:40:44 ok i see it 05:41:02 so does it do 2+3=5 .. and then 1+5=6? 05:41:16 after clear 1 2 3 + it should show 1 5 05:41:16 no, just 2 + 3 05:41:31 if you want to add all three numbers, you need to do 1 2 + 3 + or 1 2 3 + + 05:41:52 this isn't lisp, and it's a stack, not a list 05:42:12 ok 05:42:17 sorry, i see that now 05:42:37 you only have one data stack so you really don't want words like + to add everything on it :-) 05:42:46 true 05:42:49 :) 05:42:54 did you learn about stack comments yet? 05:43:08 the (--) things? just came across them briefly 05:43:24 ( this is a comment) 05:43:34 note the whitespace after the ( 05:44:32 cool 05:44:34 the stack comments show ( what is on top of the stack before this word -- what is there after the word was executed ) 05:44:52 so for example, DROP would be ( x -- ) 05:45:14 there was "x" (i.e., anything) on top of the stack, and DROP throws it away 05:45:25 2DROP ( x1 x2 -- ) 05:45:35 2DROP throws away two stack entries 05:45:47 DUP ( x -- x x ) 05:45:59 DUP duplicates the top stack element 05:46:02 etc. 05:46:38 if you have ( a b c ) on the stack, and you call DUP , you'll be left with ( a b c c ) 05:47:27 so anything that's on the stack that's not included in the "before" comment will be left alone 05:48:01 so you could instead write DUP ( ... x -- ... x x ) 05:48:14 but common practice is not to do that 05:48:24 since you'd do it for basically every word :-) 05:48:44 ah, i think i understand that now :) 05:49:14 because we dont really touch the other elements 05:49:22 so no point really mentioning them 05:51:06 yeah 05:51:26 you don't care about them, and you don't want to care about them even 05:52:00 right 05:55:28 segher, thanks for the tutorial. im off :) 05:57:03 have fun 06:21:59 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 06:40:36 --- join: vatic (n=chatzill@pool-162-83-233-103.ny5030.east.verizon.net) joined #forth 06:44:22 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-136-199.pools.arcor-ip.net) joined #forth 06:46:10 --- join: segher_ (n=segher@dslb-084-056-161-036.pools.arcor-ip.net) joined #forth 06:50:16 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 06:50:26 --- nick: Snoopy_1711 -> Snoopy42 06:58:37 --- quit: segher (Read error: 110 (Connection timed out)) 07:09:06 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-106-127.pools.arcor-ip.net) joined #forth 07:15:08 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 07:15:21 --- nick: Snoopy_1711 -> Snoopy42 07:16:48 --- quit: ygrek_ () 07:17:05 --- join: ygrek (i=user@gateway/tor/x-b0cf51bfb642184f) joined #forth 07:21:11 --- quit: vatic ("*poof*") 07:34:14 --- join: Snoopy_1711 (n=snoopy_1@dslb-084-058-123-063.pools.arcor-ip.net) joined #forth 07:37:22 --- join: Quartus__ (n=Quartus_@209.167.5.2) joined #forth 07:41:35 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 07:41:58 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-150-167.pools.arcor-ip.net) joined #forth 07:47:10 --- quit: Snoopy_1711 (Read error: 145 (Connection timed out)) 08:14:01 --- join: madwork (n=foo@204.138.110.15) joined #forth 08:17:01 --- join: yumehito (n=yumehito@b-internet.87.103.254.70.snt.ru) joined #forth 08:20:28 --- nick: brx` -> brx 08:48:16 --- join: Crest (n=crest@p5489da26.dip.t-dialin.net) joined #forth 08:52:29 Hi. 09:33:43 --- join: JasonWoof (n=jason@c-71-192-30-169.hsd1.ma.comcast.net) joined #forth 09:33:44 --- mode: ChanServ set +o JasonWoof 09:42:42 --- nick: segher_ -> segher 10:21:09 --- join: yumehito_ (n=yumehito@87.103.254.70) joined #forth 10:42:28 --- quit: yumehito (Read error: 110 (Connection timed out)) 10:48:37 hey jw 10:48:47 hi :) 10:49:20 hey, do you know about memory mapping files on posix? 10:50:39 I'd like to have multiple processes accessing a file, and some of those overwriting parts of it 10:50:59 and I'll ocationally need to make the file larger 10:52:57 guess I don't need memory mapping 10:53:16 just open, seek, read/write, close 10:58:51 --- join: brx` (n=brx@p57A78193.dip0.t-ipconnect.de) joined #forth 11:01:41 oh, that's where I was thinking of memory-mapping. when I'm reading, I read small amounts of data from all over the file 11:02:43 If you have random-access read/write patterns, then you might benefit from memory mapping. For sequential access, not so much. 11:03:27 cool 11:04:01 I can map the file such that it won't read the whole file into memory right? 11:04:24 Correct. 11:05:58 I was thinking that's the whole point... :) 11:06:07 just making sure 11:06:14 I'm not real clear on this stuff 11:06:29 this should go pretty easy 11:06:34 I don't know if Win32's POSIX implementation supports mmap or not. 11:06:43 when I write, it's only to one or two places 11:06:56 oh, that's fine, this is for a CGI program 11:07:10 I coded a DB in sql 11:07:12 but it was too slow 11:07:12 No Problem (tm) 11:07:24 --- quit: brx (Read error: 110 (Connection timed out)) 11:07:27 so I recoded it for berkely db4 11:07:32 that was only twice as fast 11:07:52 some of the slowdown was probably PHP converting everything to binary and back 11:08:09 and somehow the db4 plugin for php doesn't work on my server 11:08:42 I've been thinking about how the database is used, and I think a flat array of structures in a file would work very well 11:09:16 I don't know that nodes will ever be removed 11:09:33 records are 26 bytes 11:09:41 accessed by number 11:10:06 The mmap'd file solution won't scale to multiple servers without a lot of pain on your part, but if that doesn't matter then you'll end up with something much faster going this route. 11:10:12 seems easy enough to do file_seek(fd, 28*id); read(... 11:10:39 ok 11:10:59 not sure what would work well for multiple servers 11:11:29 think this database is simple enough that I'm willing and able to recode it if nessesary 11:11:44 it wasn't that hard switching from sql to db4 11:12:10 That all depends on the frequency of updates and the penalty for inconsistency across servers over the synchronization period. 11:12:28 right 11:12:46 updates will be frequent, but inconsistency is hardly a problem 11:13:31 hmm 11:13:47 I guess adding nodes would be a bit sensative about synchronization 11:14:06 eh, I'll burn that bridge if/when I get there 11:14:10 :-) 11:14:25 nice to think a little bit about what to do if the site becomes popular 11:14:45 but now it's time to get practical and make the thing work well now :) 11:56:36 --- quit: ygrek () 12:15:12 --- join: yumehito (n=yumehito@b-internet.87.103.254.70.snt.ru) joined #forth 12:25:07 --- nick: brx` -> brx 12:26:42 --- quit: yumehito_ (Read error: 110 (Connection timed out)) 12:57:44 --- join: vatic (n=chatzill@pool-162-83-233-103.ny5030.east.verizon.net) joined #forth 13:44:25 --- quit: yumehito (Remote closed the connection) 13:45:06 --- join: yumehito (n=yumehito@b-internet.87.103.254.70.snt.ru) joined #forth 13:57:19 --- quit: yumehito (Remote closed the connection) 13:57:29 --- join: yumehito (n=yumehito@b-internet.87.103.254.70.snt.ru) joined #forth 14:01:24 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 14:01:41 --- join: Quartus__ (n=Quartus_@209.167.5.2) joined #forth 14:02:38 --- quit: yumehito (Remote closed the connection) 14:03:07 --- join: yumehito (n=yumehito@b-internet.87.103.254.70.snt.ru) joined #forth 14:24:46 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 14:30:34 --- quit: yumehito (Remote closed the connection) 14:34:03 --- join: ASau (n=user@iggsn1inet2.beelinegprs.ru) joined #forth 14:38:53 --- join: Quartus__ (n=Quartus_@209.167.5.1) joined #forth 14:42:15 Good evening, Neal. 14:46:03 hi 14:50:48 --- join: edrx (n=Eduardo@fosforo.k8.com.br) joined #forth 15:10:09 --- join: crest_ (n=crest@p5489F6CD.dip.t-dialin.net) joined #forth 15:16:28 --- quit: Snoopy42 () 15:18:44 --- quit: Crest (Read error: 110 (Connection timed out)) 15:22:53 --- join: Snoopy42 (i=snoopy_1@dslb-084-058-131-078.pools.arcor-ip.net) joined #forth 15:23:51 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 15:24:09 --- join: Quartus__ (n=Quartus_@209.167.5.2) joined #forth 15:46:56 --- quit: vatic (Read error: 110 (Connection timed out)) 16:09:49 --- join: vatic (n=chatzill@pool-162-83-233-103.ny5030.east.verizon.net) joined #forth 16:10:26 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-129-046.pools.arcor-ip.net) joined #forth 16:18:36 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 16:18:50 --- nick: Snoopy_1711 -> Snoopy42 16:20:18 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 16:20:24 --- quit: crc ("Leaving") 16:20:52 --- join: Quartus__ (n=Quartus_@209.167.5.1) joined #forth 16:38:10 --- join: ttuttle (n=tom@gentoo/contributor/ttuttle) joined #forth 16:38:15 Quartus: Hey! 16:38:21 hey ttuttle 16:38:24 Quartus: How's it going? 16:38:27 hey Quartus 16:38:32 slava: Hey. 16:38:33 Hey slava. 16:38:37 Not too bad. What's new? 16:38:43 Quartus: Not much. Have you ever used Gopher? 16:38:58 The antique pre-web information dealie? 16:39:02 Quartus: Yeah. 16:39:06 i have 16:39:07 Sure. 16:39:10 Quartus: Now available on your cell phone at http://www.ttuttle.net/gopher.pl! 16:39:16 Heh. 16:39:18 perl? yuck 16:39:25 slava: :-b 16:39:38 slava: It's essentially all parsing/formatting. Perl's good at that. 16:39:40 although i'm debugging some C right now 16:39:45 Quartus: Wanna Digg it for me? ;-) 16:39:53 I don't know anything about Digg. 16:39:58 Quartus: Ok. 16:40:06 i have some kind of integer signedness/overflow issue that wreaks havoc when i have a large data heap, larger than 1.5gb or so, which results in some addresses being larger than 2^31 16:40:13 hence becoming negative when (erronously?) cast to signed ints 16:40:27 I think I may have signed up once to reply to a digg, but that was many moons ago and I don't know if I actually did it or not. 16:41:01 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-157-160.pools.arcor-ip.net) joined #forth 16:44:19 i hate debugging C 16:44:32 it's a very opaque process 16:49:40 do you use debuggers? 16:49:46 yes, i use gdb 16:56:03 --- quit: Snoopy42 (Connection timed out) 16:56:24 --- nick: Snoopy_1711 -> Snoopy42 17:02:40 --- join: skas_wk (n=skas@121.127.198.32) joined #forth 17:02:48 --- quit: skas_wk (Read error: 104 (Connection reset by peer)) 17:03:02 --- join: skas_wk (n=skas@121.127.198.32) joined #forth 17:07:13 --- join: Snoopy_1711 (i=snoopy_1@dslb-084-058-180-212.pools.arcor-ip.net) joined #forth 17:25:39 --- quit: Snoopy42 (Read error: 110 (Connection timed out)) 17:25:48 Who put the "C" in CLF? 17:26:36 eh? 17:27:15 Lotta C talk there these days. I've found it interesting though... 17:27:41 ah. 17:28:44 John Doty should get the Dubya award for failing to smell that his pants are on fire... 17:28:44 I tested a snippet of C in regards to one of Doty's unsupported claims. 17:28:51 I saw. 17:29:28 Nice to get the hard facts. 17:29:55 he keeps banging his kettle about how standard forth is terrible, and nobody uses it. 17:30:54 not letting the facts interfere. 17:31:12 Well he's provoked Passaniti to go on at length about some good stuff... 17:31:26 yes, which is good. 17:34:25 --- quit: edrx (Remote closed the connection) 17:34:37 In typical form, he ignores any correction and just plows onward. Irritating, but I'm sure that's the point. 17:36:13 --- quit: Snoopy_1711 (Read error: 110 (Connection timed out)) 17:36:24 well i just woke up after my first night of learning forth - and all i dreamed was pushing stuff on the stack and then popping it :/ 17:38:10 heh 17:39:49 I don't think I've ever dreamed about a data structure 17:40:41 so where was you guys reading the interchange between Doty and Passaniti? 17:40:47 s/was/were 17:41:12 where what now? 17:41:31 yes, we were just talking about it 17:42:46 is that discussion available somewhere online 17:44:08 you were here, weren't you? 17:44:30 twenty minutes ago 17:47:36 woops. i'll scroll up 17:53:34 not much to read 17:55:05 --- quit: ttuttle ("leaving") 17:58:09 --- join: snoopy_1611 (i=snoopy_1@dslb-084-058-153-183.pools.arcor-ip.net) joined #forth 18:01:34 --- nick: snoopy_1611 -> Snoopy42 18:35:48 --- join: nighty^ (n=nighty@sushi.rural-networks.com) joined #forth 18:39:11 anyone around? 18:39:18 I am. 18:40:11 i have 11 22 33 in my stack. and then i do a swap. and a dup. to get 11 22 33 22; can i write a one liner to do this? 18:40:26 : foo swap dup ; 18:40:37 : foo over ; 18:40:45 or just 'over' 18:40:59 Successive refinement. :) 18:41:11 ok, so that is defining a new word, i think. but isnt something like 11 22 33 swap. dup. possible? 18:41:22 i know that doesnt work - i tried it 18:41:31 Hang on, you're mixed up. Or I am. 11 22 33 swap dup gives you 11 33 22 22 18:41:50 yes. sorry 18:42:00 So what are you starting with, what do you want to end up with? 18:42:44 11 22 33 -> 11 33 22 22 18:43:17 Ok. So define a new word: : foo swap dup ; then 11 22 33 foo will leave 11 33 22 22 on the stack 18:43:44 hmm! ok, i havent come across how to define new words .. but was looking forward to it 18:43:54 well I'm not clear on what you want to achieve. 18:43:58 that looks really simple (which is what i'd have liked) 18:44:34 at the moment i did, 11 22 33 RET then swap RET then dup RET then . RET 18:44:44 i was wanting to do this in one line 18:44:55 so do it on one line. 11 22 33 swap dup . . . . 18:45:06 hmm. really? 18:45:34 er. why does that work? 18:45:42 cool! 18:46:04 why would it fail? 18:47:19 no reason, now that i think about it 18:49:22 Quartus, i've done : foo swap dup ; .. and im trying to use it like 11 22 33 foo 18:49:31 ok 18:49:47 but it says the terminal*:170: Undefined word 18:49:47 11 22 33 foo 18:50:10 the line before it said : foo swap dup compiled 18:50:16 Then you didn't put the ; 18:51:05 ah! 18:51:19 so ; signals the end of the word defn? 18:51:39 Yes. 18:51:56 ok, got it. 18:53:31 --- quit: nighty^ (zelazny.freenode.net irc.freenode.net) 18:53:31 --- quit: skas_wk (zelazny.freenode.net irc.freenode.net) 18:53:31 --- quit: brx (zelazny.freenode.net irc.freenode.net) 18:53:31 --- quit: arke (zelazny.freenode.net irc.freenode.net) 18:55:44 --- join: arke_ (n=f2@bespin.org) joined #forth 18:57:55 --- join: brx (n=brx@p57A78193.dip0.t-ipconnect.de) joined #forth 19:02:26 --- join: arke (n=f2@bespin.org) joined #forth 19:02:26 --- join: nighty^ (n=nighty@sushi.rural-networks.com) joined #forth 19:02:26 --- join: skas_wk (n=skas@121.127.198.32) joined #forth 19:02:26 --- mode: irc.freenode.net set +o arke 19:02:56 --- join: nighty^^ (n=nighty@sushi.rural-networks.com) joined #forth 19:03:50 orrite so i see why i was wrong 19:04:13 i thought operators were just being pushed on the stack, and they werent evaluated until i popped 19:04:37 but obviously that is not right. the operations are done as soon as i hit RET 19:04:54 it all makes sense now :) 19:08:50 --- quit: arke (Connection timed out) 19:12:26 --- join: Tomasso (n=tomasso@OL156-71.fibertel.com.ar) joined #forth 19:12:45 --- part: Tomasso left #forth 19:16:22 --- quit: nighty^ (Connection timed out) 19:25:32 i cant seem to use 0sp? 19:26:35 *the terminal*:233: Undefined word 0sp 19:26:53 why do you think 0sp is a word you should be using? 19:27:17 http://www.softsynth.com/pforth/pf_tut.htm 19:27:24 Are you using pforth? 19:27:28 "Between examples, you will probably want to clear the stack. If you enter 0SP, pronounced "zero S P", then the stack will be cleared." 19:27:32 no, gforth 19:27:36 but it looked like a nice tutorial 19:27:36 Well... 19:41:16 --- quit: vatic (Remote closed the connection) 19:57:19 --- join: brx` (n=brx@p57A79B93.dip0.t-ipconnect.de) joined #forth 20:06:06 --- quit: brx (Read error: 110 (Connection timed out)) 20:26:14 --- quit: Quartus__ (Read error: 104 (Connection reset by peer)) 20:28:39 Hello 20:28:41 :) 20:29:14 hey 20:29:28 Quartus: wassup ? :) 20:29:36 not too much. You? 20:29:47 Quartus: oh been around the world a few times 20:29:55 Quartus: and in china 20:29:55 all the way around, eh?\ 20:30:00 Quartus: oh yes 20:30:04 Quartus: all the way 20:30:07 Good times. 20:30:14 Quartus: with 3 days notice 20:30:26 Some people enjoy that. 20:30:33 Quartus: always nice to be told you are going to china for 15 days with a 3 days notice 20:30:56 Quartus: well I enjoy this with more notice , but i have a family :) 20:31:03