00:00:00 --- log: started forth/09.10.21 00:44:34 --- quit: ygrek (Remote closed the connection) 01:19:15 --- quit: nighty_ (Client Quit) 01:54:07 --- join: kar8nga (n=kar8nga@jol13-1-82-66-176-74.fbx.proxad.net) joined #forth 02:22:14 --- join: H4ns (n=Hans@87.187.174.241) joined #forth 02:56:50 --- join: Al2O3 (n=Al2O3@h130.240.40.69.dynamic.ip.windstream.net) joined #forth 03:01:26 --- join: ygrek (i=user@gateway/gpg-tor/key-0x708D5A0C) joined #forth 03:41:11 --- join: neceve (n=ncv@unaffiliated/neceve) joined #forth 03:42:38 --- quit: Al2O3 () 03:46:11 --- join: gnomon (n=gnomon@CPE001d60dffa5c-CM000f9f776f96.cpe.net.cable.rogers.com) joined #forth 04:08:27 --- quit: kar8nga (Remote closed the connection) 04:10:47 --- join: TR2N (i=email@89-180-159-143.net.novis.pt) joined #forth 04:17:53 --- quit: gnomon_ (Read error: 110 (Connection timed out)) 05:34:05 --- quit: schme (Read error: 54 (Connection reset by peer)) 05:37:08 --- join: schme (n=marcus@c83-249-82-26.bredband.comhem.se) joined #forth 05:46:46 --- join: tathi (n=josh@dsl-216-227-91-166.fairpoint.net) joined #forth 05:57:29 --- join: gogonkt_ (n=info@218.13.62.189) joined #forth 06:01:01 --- quit: gogonkt (Connection timed out) 06:06:43 --- nick: gogonkt_ -> gogonkt 06:12:23 --- join: gnomon_ (n=gnomon@CPE001d60dffa5c-CM000f9f776f96.cpe.net.cable.rogers.com) joined #forth 06:39:17 --- join: qFox (n=C00K13S@5356B263.cable.casema.nl) joined #forth 06:42:47 --- quit: gnomon (Read error: 110 (Connection timed out)) 06:46:00 --- nick: gnomon_ -> gnomon 06:48:50 --- join: chturne (n=chturne@host86-164-146-150.range86-164.btcentralplus.com) joined #forth 07:09:32 --- join: Al2O3 (n=Al2O3@71.31.155.59) joined #forth 07:39:13 --- join: Rugxulo (n=chatzill@adsl-065-013-115-246.sip.mob.bellsouth.net) joined #forth 07:39:55 someone want to explain direct / indirect / subroutine threading to me? 07:40:32 Rugxulo: Do you have any knowledge of assembler? doesn't matter on which platform, just a generic understanding. 07:40:40 yes 07:40:47 Rugxulo, you could do worse than reading http://www.bradrodriguez.com/papers/moving1.htm 07:41:03 That article isn't 100% accurate, but it's a very good starting point. 07:41:13 gnomon: you beat me to it, I was just looking for that link :) 07:41:42 Firefox awesomebar for the win! :) 07:41:59 Being at the wrong computer for the lose :( 07:43:32 "At least one 32-bit Forth has been written for the 8086/8088." ... pardon my ignorance, but which one? 07:47:46 --- join: kar8nga (n=kar8nga@jol13-1-82-66-176-74.fbx.proxad.net) joined #forth 07:49:42 eh? I thought ITC *was* the smallest method for code size, meh 07:50:24 no 07:50:31 theres also token threading and thats smaller 07:50:37 but also alot slower 07:51:22 direct threading is also smaller than indirect 07:51:31 its also much faster on the x86 platform 07:51:43 fastest would be sub threading tho 07:52:04 but i personally dont think thats a forth threading mechanism at all :) 07:53:48 so IsForth uses what then? 07:53:53 direct 07:53:59 BTW, you misspelled "poioson" ;-) 07:54:06 poison ? 07:54:10 on the website 07:54:13 where? 07:54:17 link? 07:54:27 Pick yer poioson! 07:54:32 oooh lol 07:54:33 http://www.isforth.com/download.php 07:54:38 damn heh 07:55:33 i saw you say that in here the other day but didnt understand what you were refering to and you had already left heh 07:55:34 fixt 08:01:37 yep, it's fixed, but "yer" ain't a word :-)) 08:01:59 it is so too 08:02:16 if "irregardless" is a word then so is "yer" 08:02:43 I think Merriam (or whoever) accepts "doh" as a word now too, as well as "frenemies", "mouses", etc. 08:03:03 heh 08:03:34 so what your saying is that the english language is being butchered even more than the forth language 08:03:41 yes 08:03:53 BTW, what does that "323" at the far bottom left (web site) mean? 08:04:07 English = huge hodgepodge (like C++) 08:04:19 323 == 17 * 19 08:04:28 uh ... 08:04:37 yeah, that explains everything :-P 08:04:47 bottom left of what website? 08:05:03 isforth.com/downloads.php 08:05:05 wow that hit count went up fast! 08:05:38 refresh doesn't change it, so it must be smart ;-) 08:05:40 i reset it the other day 08:05:43 its semi smart 08:05:55 Aww, that was my nickname in high school. 08:06:01 hemidemisemi-smart? ;-) 08:06:04 when I look at a page i see a count for each page individually 08:06:15 you just see a total hits no matter what page you look at 08:07:31 isn't "copyright (c)" redundant? 08:07:36 besides, it's 2009 now ;-) 08:07:38 no 08:07:39 well 08:07:48 actually stating copyright is now redundant 08:07:57 used to be you HAD to state it for it to BE copyrighted 08:09:05 cp rite | ed - 08:09:30 Bad command or file name ... 08:10:06 * Rugxulo bets Win7 will still include debug and edlin ... 08:10:31 sorry for OT, just bored ;-) 08:10:38 in here there is no OT 08:10:56 i basically only ever enforced one RULE as it were in here 08:11:03 anyone can chat about anything they want to in here 08:11:04 * gnomon sighs 08:11:12 debug was excluded in Vista. 08:11:13 i refuse to get anally retentive about chat content in here 08:13:15 --- quit: ahelon () 08:13:25 gnomon, no it wasn't, not on mine (32-bit) 08:13:36 no OT, just indirect threads ;-) 08:13:52 Rugxulo, egads! You're right! 08:13:53 gnomon: maybe for your 64-bit it was (since it's a DOS! app) 08:14:05 --- join: gogonkt_ (n=info@218.13.62.189) joined #forth 08:14:13 --- quit: gogonkt (Remote closed the connection) 08:14:40 in direct threading is only useful in a harvard architecture and even THEN its not required 08:14:49 i have a direct threaded 8051 forth for example 08:15:12 --- nick: gogonkt_ -> gogonkt 08:18:04 so here's a dumb question: is IsForth coded for speed or size (mostly)? 08:18:15 readability :) 08:18:17 * Rugxulo feels like he already asked that before 08:18:24 ty tathi :) 08:18:30 tho i dont always succeed there 08:18:32 any time :) 08:18:46 you use NASM, right? 08:18:58 until i have finished my own built in assembler yes 08:19:04 -Ox for the win ;-) 08:19:21 yes i hate taht. X is not a digit in hex 08:19:32 i actually prefer a386's way of expressing numbers 08:19:35 123 decimal 08:19:38 "-Ox" is a new option meaning "highest necessary" 08:19:39 0123 hex 08:19:47 a386 is weird that way 08:20:00 im a registered user of it :) 08:20:05 ditto 08:20:08 the dos version of isforth is assembled with it 08:20:08 although I never use it 08:20:19 yeah, you said that, but it's unreleased, right? 08:20:29 for now 08:20:45 pro: A86/A386 is fast, small, .OBJ support (doubt needed by you) 08:21:11 con: weird macro syntax, no SSE2+, silly .ERR and .SYM file quirks, etc. 08:21:46 the macro language is a little difficult to understand but once you get to grips with it it is orders of magnitude more powerful that the bullshit that nasm has 08:22:02 when i assemble my dos version of isforth my vocabularies are FULLY HASHED 08:22:16 well, honestly macros kinda make me hurl, they are so ugly, misused to make HLLs programmers feel at home (and write crappy code) 08:22:26 taht means taht at assembly time i calculate the hash value of the word name and link the word to the correct vocabulary thread 08:22:42 thats how they are usually used yes 08:23:09 actually, there's another assembler I like a fair bit (although I'm mostly agnostic): Octasm 08:23:13 sor rather... how they are usually abused 08:23:40 my assembler preferences are... in this order.... devpac (for the amiga), a386 then nasm 08:23:48 originally written in A86 but quite different (and has both DOS and OctaOS version), feels almost Forth-like with it's multiple-instructions-per-line 08:24:07 I like FASM, NASM/YASM, Octasm 08:24:39 i see yasm as being pointless 08:25:03 no, but NASM has improved a lot making some of YASM's original features moot 08:25:30 YASM is faster, C89 (I think?), better debug info, GAS parser, etc. 08:25:36 oh, and now TASM parser 08:26:01 no .OBJ though, plus NASM itself is now BSD licensed and supports x86-64 08:26:09 ya 08:28:02 I think A386 actually assembles itself, but since it's closed source, we can't look :-P 08:28:08 it does 08:28:12 but FASM and Octasm (and others) are self-hosting also 08:28:20 but EJI writes all of his code in 100% upper case (ugh) lol 08:28:35 well, he's old school (to say the least) 08:28:38 he is also no longer activly developing it 08:28:42 at least it isn't AT&T :-x 08:28:48 obviously ... last updated in 2000 08:29:12 its a dos application. i asked him if he would create a linux version of it and he declined 08:29:28 well I could've guessed that much ;-) 08:29:42 he should open source it and let ME do it :) 08:29:44 he had a very very wimpy Win32 example package (AWIN) though 08:30:04 I know it's not exactly the same, but you could use FASM 08:30:04 i was never intereted in that 08:30:11 and I was? :-) 08:30:23 no. fasm is not a386 08:30:33 no .OBJ, but if you only want a Linux version, that doesn't matter anyways ;-) 08:30:39 fasm is also ONE source file of 28945629847562896592789487562983465234 totally uncommented lines 08:30:45 I know 08:30:50 not worth using 08:30:56 it's definitely worth using 08:31:12 no. anyone that codes like that has no fucking idea how good software is produced 08:31:16 not worth using 08:31:19 sure he does, he's smart 08:31:28 it runs on like 10 OSes! ;-) 08:31:40 who gives a damn. its source puke 08:31:49 but he understands it, at least 08:32:04 its worse than 99% of the C code i have had the missfortune to have to look at 08:32:11 still updated, IDEs, fast, optimized, etc. 08:32:19 heh, no way, ever seen IOCCC? 08:32:28 nowhere near that bad 08:32:30 nty lol 08:33:32 I'm a neophyte, but was wondering whether in time, something like this: :QUADRATIC ( A B C X -- N ) >R ROT R@ * ROT + R> * + ; would be obvious, as experianced Forth programmers, do you still have to punch those operations in and make sense of it, or has your experiance let you see how all that works right away? 08:33:48 FASM assembles itself in like half a second, NASM takes a minute or two ... 08:33:48 *experianced 08:33:59 wha 08:34:04 *experienced 08:34:47 Rugxulo, nasm assembles isforth pretty fast tho... isforths sources are small 08:34:57 nasm is compiled, not assembled 08:35:00 I know 08:35:13 but even without -O, NASM is slower than YASM (well, at least on some of my really old machines) 08:35:25 compiling C is a majory complex issue and takes inordinate amounts of time 08:35:44 I think I'd have more trouble deciphering NASM's sources than FASM's, honestly 08:35:49 if openoffice were coded in forth for isforth it would take about half a second to compile the whole thing 08:36:04 neither are commented. both fail 08:36:08 egads, I think OO probably takes hours and hours ... 08:36:16 it does 08:36:27 blargh 08:36:50 chturne: I suck at Forth, don't really know it, but as with everything, you get better over time (is that a good enough answer?) 08:37:16 there is no such thing as "learn X in Y time" 08:37:16 I hate hate *hate* slow / kludgy / easy-to-break C/C++ compilations!! 08:37:21 specially if Y is 24 hours 08:37:46 heh, well you could learn the basics of Brainf*** in 24 hours, but you probably wouldn't have the experience to write anything yet 08:38:08 brainfuck was never intended to be USED 08:38:14 it was a proof of concept 08:38:27 why would anyone even want to use it 08:38:28 and misused ever since ;-) 08:38:33 but... whitespace is worse 08:38:34 fun? challenge? :-) 08:38:39 also funny 08:39:02 Rugxulo: Heh, I've just discovered Forth, and am absolutely loving it :) 08:39:05 Befunge is fun, ETA is nifty, etc. 08:39:20 a lot of the esoteric languages resemble Forth, actually 08:39:27 forth encourages knee jerk reactions to it in one direction or the other :) 08:39:56 chturne: I can follow that fairly easily...but I usually try to avoid dealing with that many parameters at once, if possible. 08:40:26 tathi: So, what would you do, factor it? or maybe add a few line breaks and comment? 08:40:34 I'm just looking for some style help :) 08:40:48 chturne, dont worry too much about "style" at this point :) 08:40:57 Heh, ok. 08:40:59 nobody can TEACH you style. you have to discover your own 08:41:01 chturne: whatever style you prefer is what you should use 08:41:20 I guess when I'm more experianced I will grab some Forth listing and learn the style from that. 08:41:25 else you'll go crazy worrying about whitespace, naming convention, etc. ... all the pointless stuff 08:41:27 just try to avoid having to pass too many parameters to a single function and having to juggle too many of them at the same time 08:41:32 thats a SOFT rule tho 08:41:37 depends on how it was used...I might be tempted to put the parameters in a structure, and passing the address to that. 08:41:41 I don't like 'developing' my own style though, I prefer to stick to the languages' own convention, if any. 08:41:56 chturne: even C doesn't have only one convention, though 08:42:01 It's OK to do that sort of stuff sometimes though 08:42:05 chturne, forth has no syntax is something alot of people say 08:42:11 really it has user defined syntax 08:42:11 else nobody'd need "indent" 08:42:18 which also means it has user defined style 08:42:18 --- join: DrunkTomato (n=DEDULO@ext-gw.wellcom.tomsk.ru) joined #forth 08:42:30 nobody should EVER use indent 08:42:33 its simple 08:42:43 indent using the space bar. never indent more than TWO spaces 08:42:56 now all the problems associated with code over here 08:42:58 I find the lowest level often has a bunch of stack manipulation, and worry more about the readability of the higher level stuff. 08:42:58 go away 08:43:02 some people use like 9 or 10 tabs in a row (ick) 08:43:10 Well, I'm using gforth's emacs package, which does alot of the indenting and such automatically. 08:43:35 chturne, heh i wont use any editor that tries to DO things for me 08:43:49 I440r: I find such features highly convinient. 08:43:51 thats why i cant use netbeans. there is NO way to turn off smart indenting 08:44:02 they never indent right. :) 08:44:13 but thats a classic example of "personal style" 08:44:17 I440r: you can turn it off 08:44:23 in Emacs, I mean 08:44:41 i use joe to edit all my sources. its the lesser of all evils with regards to unix editors 08:45:02 it sitll falls WAY short of absolute perfection which ive only ever seen in one editor 08:45:05 codewright 08:45:17 chturne: if you don't mind changing the order, you could do : quadratic ( c b a x -- n ) dup >r * + r> * + ; 08:45:46 tathi why do you use dup >r instead of dup>r 08:46:15 dup >r is very inefficient 08:46:23 because I don't know what system chturne is using, and some don't have dup>r 08:46:33 I440r: so is the P4, but I'm using it now ;-) 08:46:34 specially if the forth doesnt cache top of stack 08:46:50 tathi: Thanks. But from the users point of view (yes, thats only me :P), it's a lot less convienient remembering that stack order than the original I think. 08:46:50 tathi: those forths fail :P 08:47:03 I440r: besides, how often does the difference even matter? 08:47:10 push ebx. push ebx to return stack. pop ebx 08:47:15 assuming top of stack is cached 08:47:38 if he can run Emacs, it's already a given that his machine is "fast enough" 08:47:39 chturne: sure. It all depends on the context. 08:47:40 well. dup >r is two tokens. that makes it larger and slower 08:47:43 Well maybe not /a lot/, I'm clutching at straws here. 08:47:58 Rugxulo: Ouch! 08:48:02 dup would be a pop ebx, push ebx, push ebx in a system that didnt cache top of stack 08:48:02 and BTW, Joe is indeed nice, try Jupp sometime 08:48:11 followed by a pop ebx, push ebx. push ebx to return stack 08:48:14 very inefficient 08:48:22 chturne: ouch? seriously, Emacs is fairly big! 08:48:43 emacs is kind of a crappy operating system and it lacks a decent editor ? 08:48:54 tho. im told it has a nice built in tetris game 08:49:04 which is a game i cant stand 08:49:04 even the old 586 let push/pop pair, so it's probably not that bad 08:49:05 Just kidding, I know it is. My machine takes less than a second to load an emacs window under X11 (with GTK) so I don't care. 08:49:14 tetris is overrated / overimplemented 08:49:29 and they usually get it wrong. :) 08:49:41 i looked at emacs for the amiga once. 40 menu items, each with 30 or 40 items in them. some of which have 30 or 40 sub menu items nested to the umpteenth level 08:49:49 that told me how useful emacs was 08:49:59 Emacs is useful ... but so are other (smaller) editors 08:50:22 heck, if I had it installed here, I'd be in ERC instead of Chatzilla ;-) 08:50:24 dup>r on a sub threaded forth that cached top of scak would be a single opcode 08:50:29 in isforth its 08:50:32 code dup>r 08:50:34 Emacs Calc and org-mode are just irresistable for me now though, I'm stuck to Emacs in some ways because I can't be bothered to learn about another editor 08:50:37 push ebx to return stack 08:50:38 next 08:50:54 way more efficient than dup >r 08:50:59 never used Calc, org-mode *sounds* good but seems confusing to me 08:51:29 Rugxulo: Then with respect, you're over complicating it, I can't imagine a simpler interface to such a system 08:51:42 i think there isnt a single editor for *nix thats worth a damn 08:51:48 it supports so much, I mean, it's a little weird 08:51:59 ALL operations within an editor should be able to be accomplished with at most TWO keypresses 08:52:06 that includes shifts, alts or controls 08:52:13 cant do that in *nix 08:52:16 Maybe I'm biased, I've been using for ages now, so the keybinding are etched to my brain :D 08:52:41 i would LOVE to se a unix version of the old dos multi edit version 3.01 08:52:49 after that version me got kind of crappy 08:52:57 never tried that one 08:53:18 multi edit was written in its own macro language. the entire editor was really justa macro compiler 08:53:30 I liked multi edit 08:53:32 you never got the source code for the editor tho. just the compiled stuff 08:53:41 ya. me too 08:53:59 ( By the way guys, were you talking about Tomasz Grysztar when you were saying how the author of FASM was old school, and wrote a huge software system with no comments? ) 08:54:07 they still make a windows version of it (ick) 08:54:09 no, Eric Isaacson 08:54:12 Ah 08:54:26 eric isaacson wrote a86 not fasm 08:54:31 he was one of (or maybe the first?) to write an x86 assembler 08:54:45 eric isaacson wrote intels first x86 assembler 08:54:45 something weird like that, I forget exactly 08:54:50 Sorry, I started reading that half way through your conversation :P 08:54:53 he was one of the two ppl that created it 08:55:11 but yeah, FASM itself has like no comments either, so ... ;-) 08:55:33 im pretty damned certain that the source code for a386 is fully commented 08:55:49 yeah right, for whom? 08:55:54 but i can pretty much guarantee that the entire codebase is 100% uppercase other than in strings to be displayed 08:56:02 for HIM 08:56:22 no, but I bet he used a ton of weird macros too 08:56:32 yea im betting he did too 08:57:14 both are blisteringly fast, though 08:57:21 I'm a young programmer, and think I prefer CAPS :P At least in Forth thats the case. 08:57:41 no caps makes your source files look like a wall of text 08:57:55 Haha 08:58:01 ITS LIKE THIS, ITS JUST SEEN AS REALLY BAD FORM TO CHAT ON IRC IN ALL UPPER CASE 08:58:05 well, some people prefer "MOV EAX, OFFSET CS:[MYDATA] ; blah" and some don't ;-) 08:58:18 theres another thing i dont likle 08:58:33 mov ..... 08:58:39 Oh, ASM too.. I've been learning ASM recently and much prefer to read it in CAPS, hex numbers look much nicer to my eyes in CAPS 08:58:46 moveax,..... 08:58:50 use SPACS 08:59:07 * Rugxulo agrees 08:59:09 having operands to opcodes all align in a column is dumb 08:59:21 because your disassociating the operands with the opcodes they belong to 08:59:27 and associating them with each other 08:59:32 mov eax, 1234 08:59:34 push ebx 08:59:52 mov [ebx+2*edx], 1234 09:00:10 read each opcode as a single entity. you have a nice eailly readable column 09:00:13 i could 09:00:16 write every 09:00:30 sentence like 09:00:30 this 09:00:38 and it would be VERY difficult to read in a nice flow 09:00:38 my biggest pet peeve is unreadable macros 09:00:41 same same for your source files 09:00:55 have you looked at the macros in isforth? 09:00:58 are they unreadable? 09:01:02 not really, no 09:01:06 (haven't looked, I mean) 09:01:20 whaty makes any code unreadable is when its uncommented or badly formatted or worse... BOTH 09:01:38 TASM code is usually the worst 09:02:17 well. intels first assembler required you to tab over your operands because it fully expected operands to be in a very specific column 09:02:25 which is why most assembler coders today DO that 09:02:29 bad bad bad 09:02:40 major legacy FAIL. 09:02:53 Fortran-y, fun fun fun 09:03:04 but not as bad as the legacy fail you see in the arm assembler which requires you still have an END directive at the end of the source file 09:03:20 arm make awesome processors but they dont know a single thing about how to produce good code 09:03:29 their macro language is pretty much useless too 09:04:23 even the gas macro language is better than the arm dev suites assemblers macro language 09:04:28 lol 09:05:11 AT&T is abhorrent, and people *still* prefer it (though Intel-style has been supported for years ...) 09:05:54 movl $4, %eax 09:05:58 (yeah, much easier to read, NOT) 09:07:32 lol 09:08:05 i hate gas. i hate all those visually cluttering % symbols too 09:08:57 .intel_syntax noprefix ftw!! 09:09:28 even in forth assembler should look like ASSEMBLER. not like forth 09:09:48 and x86 should look like x86. not like 68k 09:10:12 heh, tell that to Herbert Kleebauer 09:10:20 who? 09:11:06 some guy on alt.lang.asm 09:11:14 craziest syntax ever 09:11:29 worse than HLA? 09:12:11 yes 09:13:18 http://groups.google.com/group/alt.lang.asm/browse_frm/thread/868f3ad543cbb538?hl=en# 09:13:36 i rarely go to alt.lang.asm or CLF either 09:13:43 clf is a waste of time 09:14:16 nothing but "deliberate idiots" asking stupid questions and other people posting nothing but benchmarks or results of benchmark tests 09:14:26 99.99% noise 09:14:34 00.01% signal 09:17:09 well, alt.lang.asm isn't moderated but comp.lang.asm.x86 is 09:17:38 :) 09:17:46 which is obvious if you read any of the thousands (!) of flames to/from Randy Hyde or Betov or ... 09:18:41 why would i bother? :) 09:19:00 dunno, Brainf*** not challenging enough I guess ;-) 09:19:31 bf is less of a challenge to me than C is 09:19:38 reading it i mean 09:20:24 I440r: I finally figured out why we disagree on tabs :) 09:20:30 why? lol 09:20:35 btw. IM RIGHT grrrr 09:20:40 I440r: our preferred commenting style. :) 09:20:51 I440r: you prefer commenting on the right, I prefer commenting above. 09:20:57 did you read the isforth page on commenting style? 09:21:16 it explains why i believe code left, comments right is FAR superior to interleaving code wtih comments 09:21:19 Your preferred style requires absolute positioning, therefore requires either eequal tab sizes or, better yet, spaces for everything. 09:21:33 I440r: Yup. I still disagree though :P but that's taste, really. 09:21:37 you can read my code as a single column (which has long been known to be easier to read) 09:21:45 and my comments wont get in the way 09:22:18 the chinese have it right with their text orientation. a tall column is easier to read fast than a wide row 09:22:30 For regular text, yeah. 09:22:41 how is source code NOT regular text? 09:22:47 it still applies. 09:22:56 Chinese isn't too hard to learn, too many glyphs 09:22:57 I don't see source as regular text ... 09:23:21 your brain can scan down a tall thin column far quicker and by reading it quicker you are far more likely to remember it 09:23:27 I know it's a big thing wherein source is supposed to resemble regular text to encourage legibility and everything 09:23:36 chinese writing is not easy for sure :) 09:23:43 i was talking about their orientation of the text 09:24:01 if readability was everything, why not use COBOL? :-)) 09:24:19 exactly. for a programming language to resemble more a natural language than say... pure maths... makes it far easier to read 09:24:25 C/C++ is definitely not readable 09:24:27 but, to be honest, i prefer source to be readable source, not readable text. I guess this is part of the reason why I'm not too much into Forth anymore. :P 09:24:40 Rugxulo, making code readable is probably THE single most important thing a software engineer can do 09:24:53 i say making it readable is FAR more important than even making the code work 09:25:04 not if it's already stable, robust, bug-free 09:25:09 then it doesn't matter as much 09:25:21 erm. i would call a MAJOR bullshit on that 09:25:31 even if i dont need to "fix" it or "modify it" ever again 09:25:32 if no one has to (or wants to) maintain it, readability isn't important 09:25:36 i STILL might need to understand it 09:25:50 if its not readable its not WORTH maintaining 09:25:52 period 09:26:03 depends on the app ;-) 09:26:08 nope 09:26:48 In my experience, it's also not necessarily what a piece of code does, but how it fits into the overall picture. 09:27:02 arke correct 09:27:13 I can deal with code being a bit ugly, as long as its clear where it fits in. 09:27:17 but. when i write a function FOO. i describe what the fucntion foo does 09:27:23 GCC and Nethack are probably very hard to maintain / read / understand, but people do it anyways 09:27:32 when i reference it i explain its "big picture" purpose 09:27:54