00:00:00 --- log: started forth/05.07.21 03:30:34 --- quit: alexander_ ("Leaving") 03:31:40 --- join: Robert__ (~snofs@c-f778e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 03:32:18 --- nick: Robert__ -> Robert 05:39:50 --- join: PoppaVic (~pete@0-2pool238-98.nas24.chicago4.il.us.da.qwest.net) joined #forth 05:40:04 good morning 05:40:31 Mornin' 06:11:04 --- nick: Raystm2 -> nanstm 06:11:22 Hi. 07:50:47 --- join: sproingie (foobar@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 08:15:43 --- part: arke left #forth 08:47:54 --- join: JasonWoof (~jason@Herkamire.student.supporter.pdpc) joined #forth 08:47:54 --- mode: ChanServ set +o JasonWoof 08:48:04 hiya, herk 08:50:00 mornin' :) 08:50:56 Hi 08:54:29 --- join: tathi (~josh@tathi.bronze.supporter.pdpc) joined #forth 08:54:35 Hi 08:54:49 Hi 08:55:22 Anyone experienced with error-correcting codes? 08:55:27 herk? have you a nice image of the ppc register-mess? I'm still trying to "visualize" it like I can the z80/x86s 08:55:40 hi tathi 08:55:43 mess? 08:55:45 Hi PoppaVic 08:55:54 yeah, I wouldn't call it a mess. 08:56:01 32 general-purpose registers 08:56:08 there's 32 int registers, 32 float registers, and if it's new enough 32 vector registers 08:56:12 the link register saves the return address 08:56:26 well, to my head - looking at -S output - it's noxious.. I just never seem to locate a nice little image to latch unto 08:56:31 and the condition register is 8 4-bit fields 08:56:46 try one of motorola or ibm's manuals 08:56:52 they usually have a bit of a picture. 08:57:21 oh, yeah, -S output just has numbers, so you have to know which operands are registers and which are actually numbers. 08:58:47 oh no... 08:58:58 heh - now, wait a sec there! 08:59:12 Yer telling me they have names AND numbers?? 08:59:28 that's standard RISC stuff. 08:59:35 * PoppaVic good lord, no wonder I get confuzzled 08:59:41 oh, sorry, not names. 08:59:49 your basic registers are r0 through r31 09:00:05 well, I saw that sorta' thing, and also 'W' and such 09:00:32 W ? 09:00:59 hmm, you're running OS X. 09:01:05 so apple probably cleaned a bunch of that stuff up... 09:01:05 I dunno, man - I've been spinning on this thing for days.. Lord knows how many urls I've visited 09:01:25 hang on a sec, let me find my link to the spec. 09:01:33 that would be great, thanks 09:02:08 btw, if anyone wants/cares - I finally nabbed libffi 2.0 source - that apparently builds 09:03:48 --- join: snoopy_16 (snoopy_161@dsl-084-058-172-103.arcor-ip.net) joined #forth 09:04:51 http://www-128.ibm.com/developerworks/eserver/articles/archguide.html 09:04:56 get Book I 09:04:57 loading... 09:04:59 see page 17 09:05:29 dunno how much that helps, but that's the register set. 09:05:34 Is it then that IBM actually makes the ppc from some other product of theirs? Clues seem to so indicate 09:06:24 err...IBM had the POWER architecture (Performance Optimization With Enhanced RISC) 09:06:41 ahhh 09:07:03 POWER was originally a multi-chip thingy. 09:07:05 I usually see notes with autoshit now that lump it with some RS6000 or sumpin' 09:07:12 yeah. 09:07:31 in 1991 (?) IBM got together with Motorola and Apple to design a new architecture (PowerPC) 09:07:47 ahh, I remember that now 09:08:03 It just never went much past apple 09:08:18 yup. it's fairly similar to POWER, but they named the instructions differently, left some things out, added some new things, etc. 09:08:40 ahhh - that might explain some confusion thru autoshit 09:08:45 right. originally Microsoft was going to release Windows for it, and the whole world was going to go PPC :) 09:09:01 yah, I remember all the hoopla 09:09:33 I think that was right arount 386->486/pentium time 09:09:35 I find it amusing that now that Apple is dropping PPC, Microsoft's XBox 3 (or whatever it is) is a PPC-based machine 09:09:47 * PoppaVic groans 09:09:50 yah, sounds about right 09:10:31 Heh. 09:10:33 Anyway, I've done a fair amount of PPC assembly -- if you have specific questions, let me know... 09:10:44 You're right, that IS amusing. 09:11:37 Thanks, tathi I will. I'm just about to where I need either asm-programs or c programs, because there is so DAMN little I can learn otherwise. I would think we could call on bios/kernel for info as well. 09:12:11 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 09:12:28 yeah, maybe. Sounds like you're trying to provide a lot more bare-metal info than autoconf gives you. 09:12:31 --- nick: snoopy_16 -> Snoopy42 09:14:54 well, it occurs to me that we need essentially a program or two, and maybe a header or two for any platform/kernel/cpu 09:19:50 program or two...like autoconf compiles little C programs to test for features? 09:19:58 yes. 09:20:59 We have to begin from someplace, and I'drather begin from ansi-C and let the compiler popen/system or whatever. But, it looks like a lot of stuff is just "because 2million lines of script sez so" 09:22:25 Do I sense some Bash-bashing? 09:23:48 hmm? Oh. You cannot rely on bash, or zsh, or even sh. Hell, last I heard - autoshit on doze relies on perl 09:25:10 Robert: however, even if I could rely on sh or bash, I'd not write a "script" per-se 09:25:26 autoshit already makes that huge, messy mistake 09:25:30 Nah, I'm not a fan of sh scripting either. 09:25:47 Usually use Ruby for those kind of tasks. 09:26:16 I figure either specifying "cc foo...; ./foo" is acceptable, or the sortest damned makefile possible. 09:26:47 prolly gcc -ansi foo.c.... 09:27:34 Because, near as I can tell, gcc sees more platforms than anything other than the autoshit 09:28:28 The question then becomes: what can we learn from C-source/calls and or versus asm-modules? 09:29:01 brb... 09:29:04 --- quit: PoppaVic ("Pulls the pin...") 09:32:25 --- join: PoppaVic (~pete@0-2pool238-226.nas24.chicago4.il.us.da.qwest.net) joined #forth 09:34:06 OK, now my idiot isp will leave me alone 09:34:19 can I ask what you want to use this system for that you can't do with autoconf? 09:34:32 I can't even tolerate autoshit 09:35:04 so what do you use now? 09:35:12 what I'm looking for is a cleaner, simpler alternative to gnu autotools that doesn't rely on several very different interpreters and shells 09:35:57 I don't use anything "now".. I tried autoshit on a couple projects, worked great - until folks on macs or 86s FROM my mac tried them. 09:36:12 Nowadays I just write as simple a makefile as I can 09:37:20 hell, everytime I thought I had the generic inputs done, they'dchange the language in the next "upgrade". It was too much to suffer 09:37:35 And do you run into problems with a generic makefile? 09:37:58 usually, no - folks do when they use a weird alternative to gmake/gnu-make 09:38:31 But, I still believe the problems go deeper than the makefile 09:39:42 tathi: for example, I can envision when and where to use inline-asm in C - but most folks can't envision a problem anymore than some newbies can envision why asm itself is "restrictive" 09:40:28 I'm guessing it's a problem that is generally the fault of books and classrooms. 09:41:28 And, of course, an oft-cited solution to a lot if things is "well, look at libelf" (which raises another layer of confusion) 09:44:55 ANyway, yeah.. I despise autoshit; I believe the problem is systemic and pandemic; and just keep tinkering. As mentioned before as well, I want to write a libTIL to support a portable-shell for this an other projects. 09:47:00 define libTIL 09:54:09 basically a forth virtual machine as a library. 09:54:18 something like that, yeah. 09:55:20 but, getting under even that - before it - we need to ascertain cpu/bios/os=platform, and then dirs, files, etc.. I just dislike the current approaches ala' bandaids 09:55:39 /=/-/ 09:56:58 --- quit: onetom (Read error: 60 (Operation timed out)) 09:57:44 so.. we build a minimum of code, to generate further info, to generate more, to generate - what? a dbase? prolly, all sorts of things 10:14:49 why is asm restrictive? 10:15:27 JasonWoof: well, as I meant it (and most don't see it), it means writing for a single cpu 10:15:35 sometimes a cpu and os 10:15:39 agreed 10:15:41 or bios 10:16:00 anyway, it's not "wrong", but some newbies can't understand that 10:16:41 JasonWoof: I believe it MIGHT be because some languages allow directly mixing others into the stew 10:17:18 I think newbies can understand "if you write in asm, it works only on this kind of computer, if you write in C, then with a (hopefully small) amount of headaches, you can get it to run on many machines.... so long as you only use libs that work on those many machines" 10:17:34 not sure, of course... But, I remember writing .asm files Long Ago, and NOT mixing inline asm for many years 10:18:00 I haven't ever mixed C and asm 10:18:03 CPU-dependant code isn't that much of a problem.. if you look at the programs that need to be fast, ports (of the optimized routines) usually appear for most common processors. 10:18:04 JasonWoof: I sorta' doubt they teach (or maybe HEAR that). 10:18:26 Of course, it's nothing you should write your UI code in... 10:18:38 But some people are overly phobic against it. 10:18:54 --- join: onetom (~tom@ns.dunasoft.com) joined #forth 10:19:06 phobic against what? asm? 10:19:09 I dunno.. After all these years I guess I just expected MORE from the professors, universities, Darpa and gurus. 10:19:16 JasonWoof: Yeah, using it. 10:19:41 ok, I'm less confused now that I looked back at who was saying what 10:19:47 hehe 10:19:49 Haha. 10:19:55 It happens 10:20:21 Robert: yeah, I'm not sure how newbies get it into their heads that asm is scarry 10:20:33 I can remember the days when I was "Oh, *C*, most cool! ALL my code is now PORTABLE!" 10:20:38 maybe it's the teachers, maybe it's x86 asm (which I don't know at all) or maybe just talking to C people 10:21:17 I guess - and this is just me - tinkering/learning asm is just easier with a forth 10:21:35 JasonWoof: Most people simply seem to lack a basic understanding of how computers work. 10:21:49 Robert: right :) 10:21:51 JasonWoof: Judging from 4 years in assembly language support channels. 10:21:55 But, reconciling assorted code/vm/systems/libs/source can be a real PITA 10:22:03 which is why I think asm would make a fine first language... you could just teach about how the computers work 10:22:06 JasonWoof: Where "most people" I mean the ones trying to learn asm. 10:22:25 Er, -I or +with 10:22:55 well, I made EMU happy and took their entry-programming course tostop arguing, and playing "pretend asm" was not remotely entertaining, but I saw their point. 10:23:02 I had no idea how the computer worked until I learned asm 10:23:05 JasonWoof: That's probably a pretty good idea. Should teach people to appreciate the highlevelness of other languages. ;) 10:23:19 I knew C... it has it's own sort of vague model of how computers work... 10:23:28 yes 10:23:29 Same here, basically. 10:23:31 as does forth 10:23:44 Yeah, I'd recommend learning asm before Forth. 10:24:08 Before C? Tough question, but probably. There's no reason to know C if you don't want to get close to the machin. 10:24:09 forth has a nice simple idea of how the machine works 10:24:11 +e 10:24:14 Robert: that would be fine, but there is no "asm" there ARE a multitude of assemblers 10:24:16 it's just rare that you actually have such a machine :) 10:24:39 PoppaVic: That's not a very big problem. 10:24:55 Robert: on the contrary, it is a huge problem 10:24:57 Robert: oh, there's plenty of reasons to learn C... money, libs, friends 10:25:02 PoppaVic: Maybe for newbies trying to learn, but since they all generate the same code... 10:25:09 umm, no 10:25:27 PoppaVic: They should, unless they're buggy. 10:25:33 Some assemblers are, though ;) 10:25:34 Robert: are you somehow of the opinion that everyone is using the same mainframe, CPU's and assembler? 10:25:45 Huh? 10:25:47 I don't really know how big of a deal it is that you have to port asm 10:25:57 tathi ported isforth from x86 to ppc, and I don't think it took him that long 10:25:59 JasonWoof: can be irksome 10:26:05 With "different assemblers" I assumed you meant different assemblers for the some CPU. 10:26:15 same* 10:26:20 I know it can be irksome, but it cas also go smoothly 10:26:32 porting a C program to another platform can be a royal pain too 10:27:06 well, you have cpu/regs; then you have semantics/syntax; and then, of course, the bios stuff - yeah, given a mainframe or a lab of identical machines, they can all do the same chore. 10:28:10 I recall from my (ancient) days, getting access to the mainframe in a lab sucked, that pcs in a lab were nearly as bad, and I generally wrote code at home. 10:28:17 JasonWoof: Yeah... you should check the Voyager image archive. The data is compressed with a non-standard format, and the decompression program is written in VAX asm, C with VMS extentions, and Fortran (which f77 won't compile). 10:28:49 For a field study on portability. ;) 10:29:02 JasonWoof: yes, porting C is a major pita for several reasons, but I will never get the #C folks to agree ;-) 10:29:11 To their defense, I should say that it was just a matter of seeking out the VMS stuff and remove it. 10:29:40 I STILL say that they need to stop dicking the semantics and beat the living frogsnot outta' wtf "libc" is supposed to be 10:31:03 and, watch the screams and stares when I mention "libansic", "libposixc" and friends ;-) 10:31:36 Robert: hehe :) 10:32:01 sometimes I wonder if it's just better to write a really clean and simple implementation in asm or whatever platform specific stuff you want 10:32:11 and document it so well in terms of what stuff does, and how it works 10:32:11 sometimes I agree 10:32:15 then it can be easily ported 10:33:11 Same issues in C, all complicated by IPC/libs/filesystems/devices/permissions/users/groups/ffi, etc 10:34:16 Maybe it began when folks forgot that a 'kernel' and 'drivers' were just a bios-extension. I dunno 10:35:17 anyway, I'm going to call it a day, check on lunch, set up dinner and talk with youse all tomorrow - stay well. 10:35:22 --- quit: PoppaVic ("Pulls the pin...") 10:37:51 he's never going to get anywhere :) 10:40:09 Maybe not... I'm a bit amazed that the next best thing to two 30-line functions for my applications using some FFTs, is the considerably larger FFTW library. 10:40:59 Which contains some nicely optimized code, but TONS of glue to handle everything, and implement very general functions. 10:41:22 :) 10:42:26 hmm...tar.gz is 1.9 MB...yeah, that's a little bit bigger :) 11:18:21 --- join: qFox (~C00K13S@92pc222.sshunet.nl) joined #forth 11:18:39 Hi foxie. 11:18:49 hey mr robert sir 11:19:00 you wouldnt happen to have any experience with the nes cpu would you? 11:19:26 its a modified 6502 cpu, but i cant seem to find exactly what modifications it has. opcode specific 11:19:38 (not my reason for signing on, but hey :p) 11:20:00 Nope, sorry. 11:20:02 I don't get PoppaVics idea, when he needs a forth system why doesn't he take gforth and generate a specialized version out of it? 11:20:07 Only 6502 coding I ever done was on the C64 11:20:13 same thing 11:20:17 appearantly 11:20:41 What do you mean, the NES also uses a 6510? 11:21:00 nope, modified 6502 but that thing is appearantly very close to the 02 11:26:32 qFox: there's quite a few nes emulators 11:26:40 open source ones... 11:27:45 yeah but are they documented 11:31:00 who knows 11:31:10 seems like there's enough of them that one must be :) 11:31:25 unless they share the samem undocumented code base for the opcode stuff 11:32:56 well i've only went into it this morning so i'll read a little more on it first 11:33:15 i just knew that some here are cpu freaks, t'was worth a shot :p 11:49:45 hi guys 11:50:41 Hi slava 11:52:18 hi slava how is factor cooking? 11:52:39 got a new release ready, just waiting for the win32 port maintainer to fix a few things on that platform 11:53:15 your project isn't an one man project? wow... 11:53:34 Haha 11:55:37 Robert? 11:56:25 Just amused at your amazement over someone's programming language not being a one-man project. 11:56:41 ah, ok. 11:56:41 about* rather 11:56:53 Er 11:56:54 by* 11:57:05 its mostly one-man. nobody touches the runtime or compiler except for me 11:57:13 * Robert HATES prepositions. :) 12:00:10 when did you started with the language? 12:00:23 almost two years ago i started the java implementation 12:00:39 the native implementation is more recent 12:00:54 java implementation is defunct now 12:01:52 --- nick: nanstm -> tiff 12:10:23 --- nick: tiff -> nanstm 12:11:14 the java vm is some kind ugly 12:11:48 i prefer having my own runtime 12:17:44 what do you mean? 12:18:10 your own output runtime or your own vm runtime? 12:18:29 s/runtime/vm 12:18:34 whatever you want to call it 12:19:59 and how fast is it compared to eh perl5? =) 12:20:34 much, much faster 12:21:38 wow.. 12:22:27 I need to test it, but it had problems to compile on my machine until yet. 12:22:55 gcc 3.3 won't compile it, this is a documented gcc bug 12:23:26 gcc 3.3.5 12:23:57 get 3.4 or 4.0 12:25:35 pff.. 12:25:47 for the next release i will upload statically-linked linux binaries 12:25:55 since a lot of people have problems with gcc/SDL/whatever 12:31:09 --- quit: KB1FYR (Read error: 104 (Connection reset by peer)) 12:31:18 --- join: KB1FYR (~Alex@d-66-63-85-222.suscom-maine.net) joined #forth 12:32:06 what did you with factor? 12:32:16 what did you do with factor? 12:32:36 what do you mean? 12:34:36 which cool applications did you create with factor? =) 12:35:03 http server, gui toolkit, 3d plotting, mandelbrot fractal 12:35:25 the 'real' app will be a personal information manager (to do list, calendar, stuff like that) 12:35:26 coming soon 12:52:13 --- join: I440r_ (~mark4@rrcs-24-242-160-169.sw.biz.rr.com) joined #forth 12:59:40 what's your gui toolkit? 13:00:15 it borrows some concepts from clim 13:02:28 --- quit: Robert ("Bye") 13:03:58 http://factor.sourceforge.net/tutorial.png 13:06:53 with what do you do the gui? 13:07:05 sdl, xlib, opengl? 13:07:11 sdl and soon opengl 13:29:34 --- part: KB1FYR left #forth 13:53:45 why did you design factor? 13:54:07 because i saw a gap that needs to be filled in language ecology 13:54:45 --- join: Robert (~snofs@c-cf78e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 13:55:31 which gap? 13:55:50 dynamic language with postfix syntax 13:56:25 joy was a good first attempt, i took over where joy left off. joy didn't have much of a live environment or object system 14:03:51 --- quit: qFox ("this quit is sponsored by somebody!") 14:26:01 See how C++ corrupts the beautiful Forth concepts: http://www.allegro.cc/forums/view_thread.php?_id=511358 14:26:44 heh 14:27:05 tagged pointers are a better approach to what this guy is doing 14:27:33 Heh. 14:30:29 --- quit: crc () 14:30:44 --- join: crc (crc@pool-70-16-146-207.phil.east.verizon.net) joined #forth 14:33:26 this shit bag of a human isn't a real C++. 14:33:33 hi crc 14:35:23 hi virl 14:35:58 outch! 'I didn't think of the string only approach' I'd like to kill something. 14:38:04 --- mode: ChanServ set +o crc 14:38:37 forth-tcl? hehehe 14:38:44 achieving new lows in performance 14:40:55 * virl is a performance loving minimalism and then there comes some idiotic C++/Java guys. *urgs* 14:41:44 hmm parrot is slow, I thought it would be faster than perl5 but its slower.. 14:43:33 when this performance slow down will be in perl6, gforth will be faster than perl *g* 14:44:45 perl is very slow. gcc, sun's java vm, and sbcl are hundreds of times faster than perl 14:45:50 iirc, perl beats java in speed. ok ok only the bytecode version. 14:46:03 that's not true 14:46:20 sun's vm has a generational gc, and the compiler does runtime type feedback and polymorphic inline caching 14:46:24 maybe something crappy like kaffe is slower than perl 14:46:59 i wouldn't be surprised ;) 14:47:30 java's performance problem is with extremely bloated and messy libraries 14:47:32 no the official java vm, ok maybe time delivers values which are not correct, but when not it's true perl is faster than java. 14:47:50 the vm itself is extremely fast, faster than gcc on a very large number of numerical benchmarks 14:47:54 don't use 'time' 14:47:58 I only tested something very very simple, ehm 'HelloWorld' 14:47:59 it will include startup time which can be 1-2 seconds 14:48:05 then you're just being dumb 14:48:56 try writing a numerical benchmark like a discrete cosine transform 14:48:56 why dumb? 14:49:01 java will probably beat perl by an order of magnitude 14:49:08 because 'hello world' is not a benchmark 14:49:14 and you're including startup time in the test 14:49:26 it's a startup speed benchmark 14:49:30 ok 14:49:48 java's startup speed is not very good, but once it starts up the compiler optimizes numerics very well 14:50:03 and PICs optimize OOP, i will be using a similar technique, but a bit more static in factor 14:50:19 not very good? it's the worst thing I have ever seen. 14:51:12 i'm not disputing the slow startup time 14:51:15 but the compiler is really very good 14:52:12 and I think it's slow crap 14:54:02 well its probably 20 times faster at least than any free forth on numerical code 14:54:06 so what are free forths? slow as molasse? 14:55:18 btw, steel bank common lisp starts instantly, and its compiler is almost as good as sun's 14:58:18 hotspot was some jit shit, iirc. 14:58:48 i've studied the technology behind hotspot and self (hotspot's precedecessor, a smalltalk-like language) extensively 14:58:54 yes, its "jit shit" 14:59:55 fine, that you have studied it. hopefully you do something which shows, that forth isn't dead. 15:00:05 forth is dead, except in embedded 15:00:21 That's a life, too. 15:00:43 Hey, as a geek I know everything about a scaled-down life! ;) 15:02:35 and java isn't faster than gcc code 15:02:46 it depends. 15:03:28 oh, really.. 15:03:54 did you ever do proper benchmarks, or are you just making assumptions? 15:04:27 no, I'm just reading 'the great programming language shootout' 15:04:37 that's bullshit 15:04:44 well, some of the benchmarks are valid 15:04:46 The one with the "DO NOT TAKE THESE RESULTS SERIOUSLY" disclaimer? 15:04:59 hehe 15:06:17 I don't see such a disclaimer on this site. 15:06:51 Hm, maybe that didn't survive the maintainer/host shift, if I'm even thinking about the same thing. 15:06:58 But I think I am. 15:07:19 He still got a point though, benchmarks aren't to be taken TOO seriously. 15:07:50 i should publish some benchmarks comparing perl 6 with gcc, except the gcc times include the startup time for the linux kernel. 15:08:01 people like virl will fall for it ;) 15:08:12 "omfg perl6 is 1337 fast!!!111" 15:08:15 That's a somewhat unfair comparsion, though ;) 15:08:16 Java is crap and this will never change. too slow, too idiotic, too complex, too bloated, too much buzzwords, too annoying compilers, too annoying file formats. 15:08:28 virl, i don't like java either 15:08:40 but i've been doing java on and off for 9 years, so my reasons for not liking it are valid, unlike yours 15:08:42 you just read slashdot 15:09:09 i think java is better than perl, php, ruby, python or any of the other 'scripting' languages 15:09:22 "better" in which way(s)? 15:09:29 pretty much every way 15:09:52 except slashdot popularity, i admit 15:09:52 I prefer Ruby to Java when it comes to text processing/general prototyping or scripting. 15:10:09 well i was talking about application development 15:10:11 Anything which isn't speed critical. 15:10:12 first, I programmed in this shit-bag, it's one of the frustrating languages out there. 15:10:41 when i program something, my first choice is factor or common lisp 15:10:50 then java 15:11:09 Yes... frustrating is the word. Incredibly long class/method names, a bunch of stupidities like the Integer/Float classes... 15:11:12 hehe 15:11:54 I mean, those things are basically just a lack of syntatic sugar. 15:12:07 actually i haven't written any java code in quite a while 15:12:07 Robert, the core semantics are broken too, syntax sugar or not 15:12:07 But the overall taste of the language becomes overly bitter. 15:12:07 Robert, the very fact that primitive type exists, no closures, etc. 15:12:07 it has too long development cycles for that it uses bytecode as it's base. 15:12:38 Robert, have you ever played with squeak? 15:12:57 slava: Well, even if it's broken, it could be fixed by simply allowing manipulation with the wrapper classes (which I heard they did in some later version) 15:13:02 Nope. 15:13:10 squeak is cool 15:13:17 I don't like java coders, and the people who created it should die. 15:13:26 I saw Ruby being described as Smalltalk with human-readable syntax, so I gave it a try ;) 15:13:36 ruby really sucks compared to smalltalk ;) 15:13:45 slava: Why is that? 15:14:01 ruby has a funky syntax with lots of special cases that really don't add anything 15:14:07 also the implementation is not as good as squeak 15:14:09 ruby is gay, somehow it beats python in this point. 15:14:10 virl: Hmm... check mindprod.com to change your attitude against Java _coders_. ;) 15:14:23 virl, you are very ignorant 15:14:46 java coders are mindless stupid coder noobs which don't know the machine under their butt. 15:14:48 most java coders have to use java, because of their employer, etc 15:15:06 virl, and you're not a mindless stupid coder noob? what have you coded? 15:15:21 slava: I admit the implementation could be better, and so could the documentation... but most of these special cases can be ignored, and even if they don't really add anything, they don't significantly make the programs/environment larger. 15:15:26 Or slower. 15:15:51 Robert, try squeak; the development environment is much nicer than ruby's interpreter + text files style. of course its a matter of taste 15:16:17 can i ask a question ? 15:16:21 why even bother ? 15:16:22 I440r, of course you can 15:16:26 and that are the other part of coders, the one who need to use Java because of their employer, for me that are no Java coders. 15:16:26 why ever bother what 15:16:37 i mean seriously. why bother with smalltalk, ruby OR squeak 15:16:53 because they're 10000x more productive than forth to code in :p 15:17:06 lol 15:17:19 squeak has a cool graphics library 15:17:21 It either works well or not at all. 15:17:26 Nice. 15:17:30 so thats because someone added on 15:17:38 squeak, only a nice toy. 15:17:40 nobody added one to forth (yet) :) 15:17:42 What I like about Ruby is the I/O and string processing libs. 15:17:49 Very nice for quick hacks. 15:17:50 I440r, well there's something for you to do :) 15:17:57 its on my todo 15:18:02 cool 15:18:08 but right now im learning arm asm 15:18:15 and coding an arm forth 15:18:17 slowly 15:18:19 neat 15:18:22 with bad code lol 15:18:31 do you code arm ? 15:18:37 hey, what is your design approach? 15:18:41 not yet, although planning on learning arm 15:18:56 arm is the only asm i ever saw that totally screws with my head :) 15:19:00 direct threaded 15:19:02 of corse 15:19:10 subroutine threading is simpler 15:19:19 i find it to not be 15:19:27 I440r doesn't believe in stc :( 15:19:33 hi cr 15:19:35 crc* 15:19:40 hi slava 15:19:52 oh, no use the ficl fast (tm) switch threading! 15:19:55 hi crc, i had a very very very very important question to ask you the other day 15:20:08 ask it! 15:20:09 wtf is switch threading 15:20:12 erm i forget what it was 15:20:14 lol 15:20:17 heh 15:20:19