00:00:00 --- log: started forth/03.12.11 01:44:07 --- join: adu (~andrew@adsl-64-142-12-128.sonic.net) joined #forth 01:52:54 --- part: adu left #forth 02:44:43 --- join: karingo (karingo@7.portland-01-03rs.or.dial-access.att.net) joined #forth 02:45:47 --- nick: arke -> rk 02:54:19 --- quit: karingo () 02:55:20 --- join: arke (~chris@ca-cmrilo-cuda1-c3b-124.vnnyca.adelphia.net) joined #forth 03:06:11 --- quit: rk ("leaving") 03:11:38 --- quit: arke (Remote closed the connection) 03:14:30 --- join: arke (~chris@ca-cmrilo-cuda1-c3b-124.vnnyca.adelphia.net) joined #forth 04:45:29 --- join: schihei (~schihei@p5085D4C0.dip.t-dialin.net) joined #forth 04:55:19 --- quit: hovil ("Leaving") 05:06:27 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 05:22:11 --- quit: I440r ("work -->") 05:55:34 --- quit: tathi ("leaving") 06:26:16 --- join: I440r (~mark4@saturn.vcsd.com) joined #forth 06:58:39 greets mark 07:01:18 --- quit: clog (^C) 07:01:18 --- log: stopped forth/03.12.11 07:01:36 --- log: started forth/03.12.11 07:01:36 --- join: clog (nef@bespin.org) joined #forth 07:01:36 --- topic: 'where people get together to talk about CVT, stirling engines, tesla turbines, data structure and algorithm design, Ham Radio, and occasionally Forth' 07:01:36 --- topic: set by arke on [Sat Dec 06 18:36:21 2003] 07:01:36 --- names: list (clog I440r schihei arke ianp Herkamire TreyB MysticOne skylan wUoNrFk chandler warp0x00 onetom oooo_ mur zardon jc @ChanServ Robert slava dpb9cpu) 07:08:28 * arke is away: I'm busy 07:20:52 terve arke 07:20:57 hey everyone 07:29:56 'morning mur 07:35:31 evnign herke 07:37:44 --- quit: clog (^C) 07:37:44 --- log: stopped forth/03.12.11 07:37:52 --- log: started forth/03.12.11 07:37:52 --- join: clog (nef@bespin.org) joined #forth 07:37:52 --- topic: 'where people get together to talk about CVT, stirling engines, tesla turbines, data structure and algorithm design, Ham Radio, and occasionally Forth' 07:37:52 --- topic: set by arke on [Sat Dec 06 18:36:21 2003] 07:37:52 --- names: list (clog I440r schihei arke ianp Herkamire dpb9cpu slava Robert jc zardon TreyB MysticOne skylan wUoNrFk chandler warp0x00 @ChanServ onetom mur oooo_) 07:42:51 --- quit: Robert (Remote closed the connection) 07:42:56 --- join: Robert (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 07:53:20 --- join: Speuler (~User@node-d-9180.a2000.nl) joined #forth 07:53:30 'morning 07:58:08 hi! 07:58:11 ltns! 08:00:52 ah, the_mark 08:01:11 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 08:01:15 how's it going ? 08:01:26 job satisfaction ? 08:04:28 job almost done :) 08:04:42 lovin it, wanna stay L/ 08:12:29 --- quit: schihei (Client Quit) 08:19:21 --- quit: Speuler ("Leaving") 08:28:41 --- join: Serg_Penguin (~z@212.34.52.140) joined #forth 09:05:52 --- join: aktnot (ident@181.80-202-66.nextgentel.com) joined #forth 09:17:11 --- quit: tathi ("leaving") 09:17:20 lol aktnot :) 09:17:25 gotta love that movie 09:17:35 5th element? 09:17:39 yea :) 09:17:45 I'm a huge fan 09:17:49 mila jovavich or however you spell her name 09:17:54 drool! 09:17:59 totally 09:18:49 its a very very cleverly done movie the way one scene leads into the next at a totally different location.... 09:18:52 love that heh 09:19:21 The best example of that was in the original Highlander movie, when they pan up from the parking deck into old Scotland. 09:21:24 It's odd that we were talking about Cygnal the other day. We got a call from our client, who wants to run 700 more of the old units that use the Cygnal part. In the second revision that didn't have accelerometers, we replaced the Cygnal with a MSP430 for cost and manufacturability reasons. Now we get to go buy 700 F005's. 09:21:52 :) 09:22:01 what is this device for exactly ? 09:24:06 It's part of a research program paid for by the National Highway Traffic Safety Administration (NHTSA). Georgia Tech won the study. The box has a GPS, cellphones, and accelerometers. When the car is started, the unit records the trip, a GPS record every second. These are uploaded either automatically when a certain percentage of disk capacity is reached, or if it's triggered by an SMS message. They overlay this on to maps, and do studies of 09:24:06 driving speeds compared to published speeds, and weather conditions and such. 09:24:18 Cygnal is now cilicon laboratories 09:25:17 The accelerometers watch for a crash condition. When this occurs, the unit connects a PSAP (public service access point) server, and uploads a crash record. This gets forwarded to a human operator, who can then place a call back to the cellphone in the unit, which is waiting for a call. 09:26:18 When a call comes it during this window, the speaker and mic are activated, so that the PSAP operator can communicate with the driver. The idea is that based on the crash data, emergency services can be dispatched tailored to the crash. If the car is upside down, the odds of injuries are much higher than a low speed impact. 09:26:40 The unit cannot call 911 automatically, as automatic 911 equipment is illegal. So it has to go through the PSAP server and dispatch system. 09:27:18 i woould never drive a car that had this sort of device built in as standared 09:27:25 Me neither. 09:27:33 Yet, you very well may be now, and not know it. 09:27:35 ill fucking walk 3000 miles before i let big brother track my every fsckiung move 09:27:43 ontrack 09:27:47 i wont have a car with that 09:27:47 No. 09:27:49 EVER 09:28:31 Many of the vehcicle computers already record the last 15 or 60 seconds of vehicle data, including throttle position, speed, brake pedal position, gear, seatbelts, etc. This has already caused some lawsuits. 09:29:03 People who bought some of these cars have had the data used against them. This amounts to self-incrimination, and also invasion of privacy, since the driver was never informed that his car was spying on him. 09:29:27 exactly 09:29:47 OnStar is something to worry about, but they only know where you are when the poll you, not second to second. 09:30:11 But if the police want to know if where your vehicle is, they can subpeona OnStar to poll it on a regular interval. 09:31:32 The insurance companies want to provide cost-per-mile insurance. You, person X, may drive fewer miles than person Y, but you may drive through a higher risk zone. Right now, driver Y is charged more, since rates are based on home zipcode, and distance driven to workplace. With the C-P-M insurance, you'd pay fair value. But they need hardware to prove that you *say* you drive only backroads, but are really taking the interstate. 09:32:05 Sooner or later, probably in the next 10 years, it's quite likely you won't be able to buy a vehicle without this, and you won't be able to disable it. 09:32:22 i wont buy a car like that 09:32:26 ill drive myu 65 stang 09:32:31 PERIOD 09:32:45 Yea, and they'll prorate your insurance accordingly, driving an unmonitored vehicle. 09:32:49 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 09:32:49 --- mode: ChanServ set +o kc5tja 09:32:52 They'll make it so expensive, you won't drive at all. 09:34:16 We had a unit in a car go dead a few months ago. One of the installers went to replace the unit, and found the wire bundle had been cut. Seems the 17 year old son decided his parents were spying on him. 09:34:36 lol 09:34:36 Right now, all the participants are volunteers. His parents own the car, and had volunteered for the study. 09:34:53 THERE is a use for these devices and it IS being used in this way 09:35:07 but it should be parental decision, not government 09:35:19 Had that been my kid (assuming I'd even participate in such a study), that kid would lose his driving priviledges permanently. 09:36:31 It's like putting APRS w/ GPS in a car. Why the hell do I want to broadcast to the entire amateur radio community and whomever else is monitoring where I am, and how fast I'm going? 09:36:51 If I'm working SAR or special events, that's fine. But day to day? You gotta be out of your damned mind. 09:37:03 aha - i got it!@ 09:37:15 devise some electronic counter measuers 09:37:26 dont disable the transmit - just disable the recieve! 09:37:34 jam it! 09:38:02 What about the flipside? The car won't start UNLESS it can hear the cell tower or satellite? 09:38:44 There's dozens of scenarios for transmit inhibit, receive inhibit, data falsification, etc. But once you've got something integrated into the engine computer, it's damn hard to work around. 09:38:46 erm they couldnt implement that, theres going to be times when the car CANT see tower or satellite and to disable the car could cause loss of life 09:38:52 that would cause all sorts of suits 09:38:57 it's kc5tja! 09:39:01 right there! 09:39:03 GPS moved, but the VSS line isn't pulsing? Or vice versa? 09:39:04 * MysticOne points 09:39:25 No I'm not. :) 09:39:33 I'm just a figment of your imagination. 09:39:43 oh... 09:39:48 he has an imatination ? 09:39:51 :) 09:40:01 nono, he's the real deal ... just not 09:40:11 See, I'm zen in a way... 09:40:18 that's cool 09:40:48 Zen And The Art Of kc5tja Repair 09:41:00 I like to think of myself in the same camp as Llewellan, in Ozy & Millie. :) 09:41:00 whatever 09:42:05 Anyway... 09:42:23 Should I work on qm today? Or FS/Forth? qm? FS/Forth? ....too many decisions... 09:42:27 FS/qm! 09:43:14 heh 09:43:27 brb - lunch 09:44:37 get FS/forth working under linux, then write qm in that :) 09:45:18 Herkamire: Nahh, qm has to be written in C more for public acceptance than anything else. 09:45:22 Otherwise, I'd love to. 09:46:14 Though, I've often considered building an emacs killer in Forth. 09:46:55 Well, I suppose qm kinda counts as an emacs killer. :D 09:46:57 We'll see. 09:47:08 It's possible I might consider re-writing qm in Forth at a later date. 09:47:12 Anything that kills emacs is good. 09:47:27 Including vi, ed, sed, cat... front panel toggle switches... 09:47:38 Nothing wrong with Emacs except that it ships with a whole suite of applications that 99% of its users never use, nor want. 09:47:59 A 30MB text editor isn't my idea of ``fun to maintain.'' 09:48:25 Swap a Forth interpeter into it. Stop writing lisp macros. 09:48:34 Nothing wrong with Lisp either. 09:48:44 Just don't ship frigging web browsers and e-mail and usenet news clients with it. 09:48:48 (note the plurality of those!) 09:48:57 Oh, and the myriad of *GAMES* with it. 09:49:05 Good grief. 09:49:13 it's an environment 09:49:16 Yep. 09:49:19 And I like environments. 09:49:36 Probably got it's own TCP/IP stack and filesystem, just in case you have to bring it up out of ROM. 09:49:43 What I don't like is the Microsoft-esque technique of pre-bundling of every damn piece of software ever written for it since time immemorial. 09:49:51 you just want to be able to install emacs without the browsers and other extranious *#&$@? 09:50:07 jc: It does have its own filesystem implementation; proof: it can traverse .tar.gz files like they were subdirectories. 09:50:21 Herkamire: I think emacs would be a fairly nice system then. 09:51:35 I'm not opposed to people using file systems, or web browsers, et. al. within Emacs. Just don't ship them as core components of the environment. 09:53:29 agreed 09:53:43 *cough*Windows*cough* 09:56:13 * kc5tja nods 10:03:16 Windows only needs to ship with one application. Freecell. 10:09:33 --- join: aktnot_ (ident@181.80-202-66.nextgentel.com) joined #forth 10:24:21 --- quit: Serg_Penguin () 10:28:34 --- quit: aktnot (Read error: 110 (Connection timed out)) 11:26:49 --- join: schihei (~schihei@p5085DA9C.dip.t-dialin.net) joined #forth 11:54:03 --- quit: schihei (Client Quit) 12:02:14 --- join: schihei (~schihei@p5085DC1C.dip.t-dialin.net) joined #forth 12:21:45 --- quit: schihei (Client Quit) 13:03:01 * arke is back (gone 05:54:32) 13:03:15 hey kc5tja 13:03:20 re 13:03:42 whatn up? 13:03:59 Listening to some Genesis. 13:04:05 im slowly gettiing used to dvorak 13:04:06 :) 13:04:15 ``whatn'' -- I noticed. :) 13:04:26 :) 13:04:53 ive been doing good, i think 13:05:03 Looks good on my end. 13:05:16 whats the order of the keys on that keyboard and IS it a better system ? 13:05:31 On what keyboard? 13:05:35 the locations of the non-alphanumerics still get me 13:05:48 dvorak. is that a better layout ? 13:05:58 Depends. 13:06:10 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 13:06:13 I440r: aoeu vs. asdf 13:06:21 For flat keyboards, it has a slight speed advantage, but ergonomic keyboards level that playing field for QWERTY keyboards. 13:06:28 I440r: ',.pyf vs. qwerty 13:06:46 For ergonomics, however, I don't think Dvorak can be beat. 13:06:55 Dvorak does have much less finger movement than QWERTY. 13:07:02 that is true 13:07:08 kc5tja: i hate those curved keyboads 13:07:15 how about an erganomic dvorak ??? 13:07:19 would that be even better ? 13:07:21 arke: I love them. My wrists feel much better. 13:07:27 i hate those "erganomic" keyboards tho 13:07:40 I440r: I tried Dvorak on my ergo, and I really didn't sense any advantage. 13:08:02 kc5tja: for me, its not the wrists but the back hand 13:08:03 k 13:08:13 neway back to werk 13:08:15 arke: Back-hand? 13:08:21 kc5tja: no advantage over qwerty on an ergonomic keyboard? 13:08:30 tathi: I didn't feel any. 13:08:37 speed or comfort? or both? 13:08:44 And I already type 90wpm+ on QWERTY, so speed wasn't my primary concern. 13:08:48 kc5tja: the bones between fingers & wrists 13:08:56 arke: Ahh 13:09:15 :) 13:09:27 * arke is away: games 13:11:14 ahh, ok then. When I learned dvorak I could only type about 30wpm. 13:11:41 Yeah, I grew up with QWERTY. I started programming at the age of 4 (though I didn't realize it back then) 13:11:44 Within two days I was doing dvorak at over 40. 13:11:49 :) 13:11:56 tathi: Two days? I find that hard to believe. :) 13:12:14 I've been using Dvorak for much longer than two days, and I'm 1/10th the speed I am at QWERTY. 13:12:35 well, I was just learning to touch type period. 13:13:01 so what were you programming at age 4? 13:13:45 (so I didn't really have the qwerty motor skills getting in my way) 13:13:53 I asked my dad about "The Game of Life" (hey, a 4 year old and a game -- who can resist?), which was written in TRS-80 BASIC. He taught me how to enter the program listing into the computer. 13:14:04 (from a magazine listing) 13:14:06 cool :) 13:14:32 heh, classic 13:14:50 i have some of those Tandy Adventure 'magazines' 13:15:27 kc5tja: and I had been playing piano for 7 years when I learned to touch type, so touch typing felt really simple by comparison. 13:15:59 tathi: I can't play a musical instrument to save my life. 13:16:07 I have no sense of rhythm. 13:16:17 Though I do have a guitar that I play/dabble with now and again. 13:17:20 :) 13:18:53 * kc5tja is having fun comparign Lamb Lies Down on Broadway to its live version. 13:19:06 Interesting to see how their playing style differs from their recording style. 13:33:32 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576567.sympatico.ca) joined #forth 13:33:44 kc5tja: I wish you were sane and liked Dream Theater :( 13:34:20 Dream Theater is just too damned hard for me to like. 13:34:49 * MysticOne cries 13:34:59 http://squidfighter.sourceforge.net/fargoal/ 13:35:01 :D 13:37:11 * kc5tja consoles MysticOne -- I like the Chocobo Song! 13:37:30 madgarden: Heh. I immediately recognized it as a Rogue-like game as soon as I saw the screen shot. :) 13:38:19 * MysticOne DCCs kc5tja a walker 13:38:23 hehehehe 13:38:46 ;) Remake of an old C64 classic. 13:43:16 * Robert is playing Super Mario. 14:05:10 What I miss is a remake of QIX. I loved that game. 14:30:05 --- quit: dpb9cpu (Read error: 104 (Connection reset by peer)) 14:38:13 * MysticOne grumbles about the slowness of perl 14:40:30 performing one query on a database with 1.3 million records, and then throwing the results into a while loop that checks each of those for a 1 or a 0 ... 14:40:36 it takes about a minute on a 1GHz machine 14:40:54 asm 14:41:36 Better SQL queries 14:41:48 jc: I can't narrow it down any more I don't think :) 14:41:58 SELECT success, rtt FROM results; 14:42:06 success is an integer, rtt is a float 14:42:29 then I check to see if success was 1 or 0 and auto-increment the appropriate variable 14:42:38 and just add all the rtts together 14:42:41 The sluggishness is largely in SQL's querying, I suspect. 14:42:59 select count(*) from results where success=1 14:43:01 select count(*) from results where success=0 14:43:06 nope, it's all perl 14:43:13 jc: hmmm... 14:43:15 select sum(rtt) from results where success=1 or success=0 14:43:23 hmmmm... 14:43:25 --- join: dnm (~dnm@bluesky.eff.org) joined #forth 14:43:30 And be sure success is an index field. 14:43:34 it isn't 14:43:54 If you you're allowed to add an index, I would. If not, not much you can do. 14:43:56 It should still work even without it being an index field. 14:44:08 It will work, you'll just get better performance with it indexed. 14:44:13 Yup. 14:44:16 yeah 14:44:21 But even with it non-indexed, it should be faster than Perl. :D 14:44:31 Since all the work is being done on the database side. 14:44:34 hmmm... 14:44:35 yeah 14:44:43 can I do select count(*) from results 14:44:47 and get the total number? 14:44:53 Sure 14:45:04 hmmm... 14:45:06 lemme try :) 14:45:39 Perl may or may not really be the slow part there. It's likely the repeated calls into DBI. 14:46:01 Not to say that it couldn't be faster if it were written in something else. 14:46:11 But the bright side is... AT LEAST IT'S NOT PYTHON. 14:47:00 Oh great, a Python hater. 14:47:03 nono 14:47:06 that's *one* call into the DBI 14:47:06 As if Ruby is any better. 14:47:11 it's perl after that 14:47:16 because it gets all the results and puts them in a hash 14:47:21 and then going through the hash 14:48:01 hmmm... 14:48:06 well just puting that in there as a query isn't working 14:48:12 what do I need to do it make that work in perl's mysql DBI? 14:48:26 phone 14:49:04 I just love all the python haters. They're all so jealous. 14:49:10 ? 14:49:29 ? ? 14:49:46 what are you talking about? 14:49:59 But the bright side is... AT LEAST IT'S NOT PYTHON. 14:50:23 As if he had something to prove. 14:50:32 I can prive I hate pyhton. 14:50:36 hehehe 14:50:37 prove 14:50:48 jc: so any idea what the syntax would be putting this query into perl's DBI object? 14:50:49 Whatever, dude. 14:50:58 like ... $dbh->prepare("SELECT count (*) FROM results"); doesn't work 14:51:16 Perl sucks ass, and that's all I need to know. Haven't touched a line of Perl in almost 8 years now. Thank goodness. 14:51:25 * MysticOne touches kc5tja with two lines of perl 14:52:38 That being said, I'll openly admit that Python is not the fastest language in the world. I also would claim that Python would be slower than Perl in this application. 14:54:43 did he poof? or still on the phone maybe? 14:55:00 Probably the phone. 14:55:09 I wonder why this query is wrong ... 14:55:34 I've never seen count() or sum() forms used, so I can't help there. I knew they existed, but didn't know how they were used. 14:56:53 the query looks ok to me. 14:57:07 Is it possible that mySQL doesn't support that form? 14:57:09 I use count(*) 14:57:27 ianp: With mySQL? 14:57:39 iut is SQL92 I believe 14:57:46 oh, it's count(*) 14:57:47 iut? 14:57:47 i dont use mysql 14:57:51 iut=it 14:57:57 Ahh 14:57:59 but it only returns 1 14:58:06 MysticOne: well, theres one row then 14:58:13 there's not 14:58:17 there's about 1.3 million :) 14:58:58 use postgres 14:59:01 hehe 14:59:08 * MysticOne smacks ianp for being l4m3 14:59:21 that would be counter-productive 14:59:21 :) 14:59:32 * kc5tja likes Postgres too. 14:59:47 I do too 14:59:52 but there's no point in changing it now :) 15:00:43 why not 15:01:00 ianp: because I have about 40 things using mysql 15:01:41 --- quit: I440r ("home -->") 15:04:58 well, it seems to be mysql actually taking about 50% of the time doing stuff ... and perl 50% of the time figuring out what mysql told it 15:05:48 I hear posgres is nicer 15:06:02 ianp: what do you do for a living? 15:07:24 Postgres is faster, cleaner, and more 'correct' than mySQL from a technological standpoint. 15:10:50 * dnm agrees with kc5tja in theory, but MySQL is quite a bit faster in many respects for certain use cases. 15:11:04 here's how I look at it ... 15:11:10 mysql was here almost a year ago when I started working here 15:11:25 no transactions.... 15:11:31 thought i know there is solutions to that 15:12:58 so, as the year has passed, many things are in mysql 15:13:02 lots and lots of things 15:13:13 changing to pgsql now would be stupid just because "it's nicer" 15:13:32 MysticOne: I'm not advocating you should change. 15:13:38 * kc5tja is just agreeing with ianp. :) 15:13:45 kc5tja: just pisses me off when people say shit like that 15:13:46 because it's stupid 15:13:50 "Yeah, I drive a Ford" 15:13:53 "pfft, get a Chevy" 15:13:58 or "I'm using Linux" 15:14:03 "it sucks, use BSD" 15:14:10 I'm advocating that you should play my roguelike: http://squidfighter.sourceforge.net/fargoal/ 15:14:20 madgarden: might give that a shot later :) 15:14:35 And woo! At least my scripting system is faster than Python. 15:14:38 madgarden: That's YOUR game? I had no idea. :) 15:14:48 Heh, yea. ;) Me and a friend in Germany made it this summer. 15:14:53 Took 3 weeks. 15:14:55 madgarden: It's not hard to beat Python in speed. :) 15:15:14 Even early Lisp interpreters are faster than Python. :) 15:15:25 kc5tja: Well, probably not. One of my main goals was to match Lua for speed. 15:15:38 Most of Python's time is spent incrementing and decrementing reference counts. That's one thing I dislike about Python's evolution -- they haven't renovated its memory subsystem yet. 15:15:43 Figured out the SQL yet, MysticOne? 15:15:51 jc: it doesn't work ... just returns 1 15:16:15 * kc5tja never could figure out the point to Lua. 15:16:19 madgarden: think it'll compile on OS X? :) 15:16:19 When you do the select, Perls DBI will return 1 row. That row->[0] will contain the count. 15:16:32 jc: ahhhh ... 15:16:41 jc: the DBI says to use $sth->rows() on it 15:16:46 which returns 1 15:16:53 lemme try that again then 15:17:00 kc5tja: Lua just seems like a syntax variation of X language. 15:17:09 MysticOne: Yep, it builds on OS X. 15:17:18 madgarden: How about Linux? :) 15:17:21 Might need to fiddle a bit, but I know we've got such a build. 15:17:24 (since OS X is basically BSD) 15:17:29 k5tja: Yep, builds on Linux just fine. 15:17:32 Trying to find ya a qucik example. 15:17:42 My programmer comrade runs Linux. 15:18:01 madgarden: I only saw a .zip file, so I assumed it was only for Windows. 15:18:21 madgarden: w00t! 15:18:23 DUDE, got some serious wind blowing outside. 15:18:43 Well, it comes with a Windows executable, but it will build for you just fine. You just need to have Allegro installed. 15:18:49 jc: so how am I supposed to pull the number out? 15:19:00 * kc5tja would like to see a variation of the K language, personally. I like the idea of vectors being a central concept in the language. 15:19:03 Where is the serious wind blowing at? 15:19:15 Here, it's just getting damn chilly at the moment. 15:19:18 madgarden: What is Allegro? 15:19:40 Allegro is a cross-platform game library. 15:19:46 madgarden: San Diego, California 15:19:52 http://alleg.sourceforge.net/ 15:20:04 $cursor = $database->prepare ('SELECT COUNT(*) FROM Parts GROUP By Device'); 15:20:04 $cursor->execute (); 15:20:04 my $countDevices = $cursor->rows (); 15:20:15 I'll hold on that. I have something like three or four cross-platform game libraries on my box right at the moment. 15:20:20 So, at least it's WARM wind then! >B[ 15:20:26 madgarden: No. 15:20:26 :) 15:20:34 madgarden: Well, warm with respect to NY state, maybe. 15:20:44 I'm in Ontario, so it's a bit chilly. 15:20:47 madgarden: But not warm for our area. It's been wet and rainy all day today. 15:20:51 And what other cross platform libraries? 15:20:57 SDL? 15:21:02 jc: that returns 1 15:21:13 SDL, pygame, and something else I don't remember the name of. 15:21:52 --- join: thin (thin@bespin.org) joined #forth 15:21:53 pygame at least depends on SDL, but still... 15:22:10 Paste your query in here. 15:22:57 $sql = $dbh->prepare("SELECT COUNT(*) FROM results"); 15:22:58 $sql->execute(); 15:23:07 my $cow = $sql->rows(); 15:23:07 die $cow; 15:23:32 What do you see when you run the SQL query from inside the mysql shell? 15:23:34 Well, Allegro is a pretty nice little package. Higher-level than SDL. 15:24:02 It may actually come with your distro. 15:24:35 madgarden: I use Slackware; unlike other distro's, it firmly believes in not bundling all of the Internet with the distribution. 15:24:52 1159144 15:25:04 (close enough to 1.3 million... was thinking it was more) :) 15:25:40 MysticOne: LIAR! What a n00b! You can't be trusted. Come back when you know how to count, m0r0!\| ;D 15:25:41 j/k 15:25:46 hehehe 15:26:13 Holy moley -- it's almost 3:30 here, and I haven't gotten a thing done on FS/Forth or qm. 15:26:25 * kc5tja did find a QIX clone though. :D 15:26:28 But I digress. 15:26:47 $database = DBI->connect ('DBI:mysql:MSP430:localhost', 'notme', 'foorbarbaz', { RaiseError => 1}) or die $!; 15:26:55 Do you have a RaiseError in your connect statement? 15:27:01 yeppers 15:27:04 and no errors 15:27:09 MysticOne: I can forward you a Mac build if you like. 15:27:14 madgarden: sure 15:27:17 madgarden: how big? 15:27:22 About 1M. 15:27:32 wanna e-mail it to me? 15:27:34 Sure. 15:27:38 mysticone@linuxpowered.com 15:27:49 jc: so it's returning 1 row ... which is correct 15:27:51 ** SPAMBOT ACTIVATED ** 15:27:55 jc: I need to know how to get INTO that row 15:27:56 :O 15:27:57 madgarden: hahaha 15:28:21 I guess I'll go off and try and get some kind of additional work done on FS/Forth. 15:28:32 kc5tja: sure you will... sure... you ... wi ... ll 15:28:43 What's FS/Forth? 15:29:14 Well, you could try fetchrow(). But unless you have a downlevel version of DBI, it should work. 15:29:21 I use that for counting things fairly regularly. 15:29:31 but $rows only gets the number of rows 15:29:32 err 15:29:34 rows() 15:29:39 MysticOne: Sent. 15:29:46 I'm assuming it returns a hash with the row number in it, right? 15:30:07 No, $sql->rows() is a special case. Perldoc DBI talks about it. 15:30:34 You should be able to do my $row = $sth->fetchrow_arrayref (); print $row[0]; 15:31:54 yep, fetchrow works 15:32:08 anyway, thanks :) 15:32:10 I'm heading out! 15:32:11 back later :) 15:32:13 See ya 15:32:36 Looks like I just landed an addendum to a contact to add vehicle radar to this project we're doing. 15:32:42 Ciao. 15:32:47 Sadly, it's only collision avoidance radar, not fire-control. 15:33:25 --- quit: thin (".") 15:35:57 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth 15:36:06 madgarden: FS/Forth is a version of Forth that I'm writing. 15:36:19 What's the FS stand for? 15:36:56 Did a web search, and there were a bunch of hits for FS-Forth. 15:37:12 OK, found your page. 15:42:44 Just joined #Forth Portal, BTW. I suppose I could post up some relevent stuff. 15:45:43 FS stands for Falvo Software. 15:45:50 aka "Falvosoft". 15:46:06 It was a company I started as a teenager. 15:46:30 FS/Forth is the only software I've written or bothered to maintain that survived the shutdown of Falvosoft. 15:47:01 Ahh. 15:47:13 Hmm, does bespin have problems with Opera? 15:48:28 IE I can't turn off the pop-up Forum Help, for instance. 15:59:28 I don't know. 15:59:32 I don't use Opera. 15:59:40 You're the first user of the site with Opera as far as I know. 16:00:17 Ahh well. 16:10:26 madgarden: fargoal uses Allegro? 16:10:39 kc5tja: whats the difference between touch typing and not? 16:11:36 tathi: Yep. Familiar with it? 16:12:50 madgarden: Yup. I have spent many happy hours trying to get it to run on PPC. ;) 16:12:51 :( 16:15:22 Argh, troubles? 16:15:41 Dual processor? 16:15:52 yes 16:16:06 why, does it have known dual processor bugs? 16:16:13 Aha. I do believe there are some issues with dual processors. 16:16:55 Well, that could be it...though I'm not sure I had gotten an SMP kernel working at that point. 16:16:59 Hmm, I'll have to try that. 16:17:08 --- quit: dnm (Read error: 110 (Connection timed out)) 16:20:24 Yep, known issue: http://sourceforge.net/mailarchive/message.php?msg_id=6403907 16:22:38 that sounds a lot less severe than what I was running into (though he doesn't give much detail) 16:22:50 well, I'll try installing it again 16:24:07 --- quit: aktnot_ ("leaving") 16:31:21 yeah, that's what I thought -- it doesn't even build. 16:31:30 it's trying to do inline x86 assembly 16:31:42 even though it _says_ it's building "C only" 16:32:37 Hmm... 16:33:29 I'm afraid I don't know enough about the various ports. But, I wonder if there is a specific makefile for your platform. 16:34:08 yeah, I don't know much about allegro. Last time I tried messing with it, I got the impression that it basically had never outgrown its x86 roots 16:34:27 But Gentoo seems to think that it builds fine on Sparc. 16:35:31 I can go ask around in #allegro. 16:36:07 I could do that too... 16:36:19 Or....I could just convert your game to SDL 16:36:21 ;) 16:36:43 Yea, that wouldn't be too hard either. 16:36:52 I may look into SDL during my "vacation". 16:37:24 I just like Allegro because it lets me do easy stuff like "draw_sprite()" and "play_sample()". :) 16:37:52 yah. 16:38:06 with SDL you have to install the add-on libraries for that kind of stuff. 16:38:37 OK, since there's an OSX port, I guess PPC builds work. 16:38:42 What's your OS? 16:38:48 Linux 16:39:55 re SDL: Yea, or write your own. I like the idea of a game library rather than a multimedia layer. But, I could always make my own little lib on top of SDL. 16:40:33 most people just use SDL_image and SDL_mixer. 16:40:48 Anyway, I wonder if the OSX build can be easily adapted for Linux. 16:40:55 Isn't SDL GPL? 16:41:03 madgarden: yeah 16:41:10 don't know about _mixer, but SDL_image is just a bunch of little tiny wrappers for the various image libraries. 16:41:39 So, SDL games have to be open source? Or is it LGPL... 16:41:54 SDL_image handles blitting as well? Different colour depths? 16:42:40 I bet it's LGPL 16:42:48 Loki was selling games, after all... 16:43:18 yes, LGPL 16:43:35 I'd hope so. I'd kinda like to sell my stuff. 16:43:39 madgarden: yeah, load images, convert colour depths, and blit. 16:43:58 ANd it supports OpenGL out of the box, dunnit? 16:44:00 if you want scaling and rotating and stuff, you need SDL_gfx or something. 16:44:06 I think so. 16:44:29 Scaling schmaling. ;) My stuff is pretty retro. 16:45:02 The thing is, basically all SDL does is help you get a framebuffer and input. 16:46:04 Yea, good. I'll have to take a look anyway. 16:47:41 yeah, looks like you just set a flag in SDL_SetVideoMode that says you want an OpenGL window. 16:47:57 Nice! 16:51:24 they put a lot of thought into SDL. 16:51:40 does the stuff it needs to do, and otherwise pretty much stays out of your way :) 16:54:49 Now, stuff like SDL_image, is that a seperate distributable lib/dll? 16:57:54 yeah, probably. Don't know how that all works on Windows. 16:58:04 check libsdl.org maybe 16:58:39 Yep, it is. So I'll have about 20 extra DLLs in order to support everything I'd need. 16:58:56 Static libs would be nice. 16:59:02 :) 17:03:32 I'm back! yay me! 17:06:09 :) 17:08:40 Did you get my email? 17:08:48 it's so cute!! 17:08:52 :D 17:09:10 Yea, it was a fun little project. 17:09:42 I died :( 17:09:51 I'm not suprised! It gets quite difficult. 17:09:56 I was on the first level 17:10:04 Oh, heh! 17:12:34 i think the quickest way to learn dvorak is to write an isforth IPC inteerface :) 17:13:27 take some ibuprofen when you switch, to help with the eye strain, my advice 17:13:41 :) 17:13:44 thx 17:13:56 ugh 17:13:57 arke: Touch typing is when you can type with little errors (and know when you've made an error, though you don't necessarily need to go back and correct them) without looking at the keyboard. 17:14:14 neat :) 17:14:16 E.g., you do everything by muscle memory and touch. 17:14:21 :) 17:14:38 * arke loves isforth 17:15:41 No doubt; isforth sounds like a nice piece of work. 17:16:43 * kc5tja is thinking of making a digital bug keyer for ham radio use as a kit. 17:16:46 What, you look at the keyboard? 17:17:45 zardon: I wish your handle was "Zardoz." 17:19:08 heh, nope 17:19:19 that looks like a ripoff of that movie 17:19:51 Hehe. Well, that was my point. 17:20:06 thats like seeing handles like 'Linux' 17:20:09 arg@! 17:20:48 I've tried to watch Zardoz, twice, late at night on the Space channel, but each time I've drifted off to sleep. Kind of a freaky movie to fall asleep to, what with that big head floating around and all. 17:21:15 yeah, same 17:21:32 its not falling asleep, since im PST, but its a weird movie 17:22:06 Heh, yes. And I noticed they have it on sale at Sunrise records. I was tempted to pick it up. 17:39:47 --- quit: tathi ("leaving") 18:08:49 --- join: epon (~epon@12-215-90-180.client.mchsi.com) joined #forth 18:09:11 --- join: jdrake (~GodlessHo@CPE00045afdd0e8-CM014410113717.cpe.net.cable.rogers.com) joined #forth 18:20:40 hi 18:21:30 re jdrake 18:21:52 anyone here run gentoo? 18:22:33 nope 18:22:39 hmmm... 18:22:47 was just going to warn any gentoo users not to mirror their archive 18:22:49 have a LiveCD around here somewhere 18:22:59 apparently it's "wrong", "not allowed", and they'll call you names and kick you from #gentoo 18:23:45 so don't mirror it :) 18:32:27 ah, not even sure what archive you speak of, binaries? 18:32:45 like mirroring one of their FTP sites 18:35:41 "not allowed"? Isn't that a distinct violation of GPL? 18:35:59 no, you can d/l stuff 18:36:02 you just can't mirror it 18:36:04 it's server abuse then 18:36:05 :P 18:36:32 OK, so instead of mirroring their server, why not just download their stuff and put it up on your own site? :) 18:37:10 :) 18:38:20 yeah 18:38:23 but I mean, really 18:38:37 friend of mine got kicked from their channel for saying he was mirroring a public server 18:38:42 then they proceeded to call him names after he was gone 18:39:24 warning! gentoo community full of childish jerks! news at 11. 18:39:24 That's not very gentoomanly behaviour. 18:40:12 y0 18:40:12 y0 18:40:13 y0 18:40:13 y0 18:40:15 hehe 18:40:16 yay 18:40:16 yay 18:40:17 yay 18:40:19 yay 18:40:27 Just one more reason to go with Slackware. :) 18:40:33 MysticOne, x86 ? 18:40:58 kc5tja: preach it, brother 18:41:10 I440r: im writing is4th forkk/exec/IPC code :) 18:41:21 I440r: yeah 18:41:21 kc5tja: ! 18:41:22 arke i want to see that code 18:41:25 kc5tja: suse! 18:41:36 MysticOne, i was chattin with him at the time 18:41:42 i wasnt paying attention in channel 18:41:45 I440r: it was ridiculous 18:41:52 sounds like it 18:41:57 it IS a fucking public server 18:41:59 PERIOD 18:42:06 I440r: when iits done, you may add it to the src :) 18:42:15 i vote we all mirror it NOW 18:42:20 as I explained it to the gentoo guys, the server admins should rate limit if they don't want someone to dominate the server 18:42:24 that's how the real world does it 18:42:27 arke ill edit it first :) 18:42:33 but i need signals and ipc etc 18:42:45 fork is easy - exec not so 18:42:45 the /real/ world provides rsync for approved mirrors, and then encourages their use so they aren't bandwidth-saturated 18:42:58 I440r: you and yaur damn tabs 18:43:11 ill edit those out first 18:43:12 yeah 18:43:15 (or lack thereof) 18:43:24 tabs do NOT belong in a source file at all 18:43:31 heh 18:43:33 2 spaces per indent thankyou :P 18:43:42 yeah, i droped them from my latest WORD 18:43:51 AND YOU PROLLY USE EMACS TOO 18:43:58 also - ill do a global lowercase except betwen quotes on it 18:44:03 I use vim! 18:44:04 no. fuck that 18:44:08 i use joe lol 18:44:16 vi 4 lyf! 18:44:23 I440r: y'know, given that I don't like 2 spaces per tab, but prefer 4, we would be both happy if we just stuck to real honest-to-god tabs and set display prefs as appropriate 18:44:26 ! 18:44:35 arke: Suse is broken by design; they use RPMs. 18:44:52 kc5tja: bxastard. 18:45:04 kc5tja: i like it! 18:45:45 kc5tja: what does slacck use then? 18:45:47 chandler, tabs are the nmost fucked up thing you can put in a source file. specially when more than one person edits. if everyne used their own indent sizes but used spaces the soruces wouldnt look all fucked up 18:46:04 ! 18:46:05 hahaha 18:46:06 your logic escaped me 18:46:06 * kc5tja also uses vim; and I also use 2 spaces per indentation level. :D 18:46:06 haha 18:46:08 -haha 18:46:16 if everyone uses tabs, then the source looks consistent to everyone 18:46:27 no they wouldnt 18:46:29 if everyone uses their own space settings, then half the code is in 4 space and the other half in 2 space 18:46:35 * arke agrees with Chan 18:46:38 do you even know what you're talking about? 18:46:39 because not everyone likes indents of 8 chars 18:46:39 * arke agrees with chandler 18:46:41 tabs are the shiznit. 18:46:44 8 chars is WAY to much 18:46:44 I440r: tabs are /not/ 8 chars 18:46:46 chandler: No it doesn't. Because a 'tab stop' is an abstract idea. 18:46:47 don't you get it? 18:46:52 I use 4 char tab indents. 18:47:05 kc5tja: use tabs for indentation; spaces for alignment (and only ever align at the same tab level) 18:47:08 a SINGLE space is sufficient 18:47:15 I440r: tabs are a single character 18:47:22 ASCI 009, I believe 18:47:27 er, ASCII 18:47:28 of an undetermined size 18:47:32 I440r: you tell the oitor how to represent tabs 18:47:35 correct. /you/ determine it in your display prefs 18:47:39 thus, you can view the code at 2 chars 18:47:43 and I can view at 4 18:47:49 and it looks like the preferred way to both of us 18:48:04 yeah, 4 is my preferance 18:48:12 ok. then we both edit the same soruce files and when you look at code of yours that ive modified the soruces look totally fucked up 18:48:24 * arke likes 8 18:48:26 I440r: what? no, when I hit tab, a tab char is inserted 18:48:30 tabs do NOT belong in ANY source file. i wont accept any files that contain them 18:48:31 when you view it, it looks like two chars 18:48:48 chandler - dont send me any sources that you have edited please :) 18:48:51 gagh. understand what I'm saying before repeating your uninformed assertion 18:48:52 I440r: if you use tabs only they wont 18:49:00 my editor injects 4 spaces every time i hit TAB 18:49:02 I440r: do you even understand what I'm saying? 18:49:05 zardon: fix your editor 18:49:05 chandler: As long as you ensure that tabs are only in the beginning of a source line, then I can see your point. Otherwise... 18:49:08 no 18:49:23 chandler your saying your sources will look fucked up on anyone elses machine other than yours 18:49:26 kc5tja: yes. hit the tab key when at the beginning of a line; hit space otherwise 18:49:29 I440r: NO! 18:49:33 that is not what I'm saying 18:49:38 my tab key inserts 2 spaces 18:49:40 I am saying that they will look just like how you want it 18:49:46 and how I want it on mine 18:49:56 not if more than one person edits it wont 18:50:07 that statement is TOTALLY WRONG 18:50:17 If everyone uses non-space-inserting tabs it will. 18:50:23 IE "real" tabs. 18:50:24 thats the whole point. get more than one developer working on the same source files and it takes all of 2 minutes for the files to become totally fucked in the head 18:50:31 yeah, i don't know how you are getting that 18:50:34 --- join: Sonarman (~matt@adsl-64-169-94-20.dsl.snfc21.pacbell.net) joined #forth 18:50:43 um. your statement is wrong. Stop extrapolating from an incorrect statement 18:50:46 the editor interprets TABs as it wishes 18:50:47 I440r: I see what chandler is saying. As long as tabs are strictly kept at the beginning of a source line, then indentation can be used to indicate program structure according to user preferences. A tab that appears in the middle or end of a line would be a syntax error, however. 18:51:15 kc5 yes as long as he is the ONLY person working on the files his indentations will always be consistent 18:51:22 I440r: Not true. 18:51:35 I440r: with two developers working, if they both follow the convention, then they will be able to work together and see their own desired view on their own machines 18:51:45 because they can set their own display prefs for the tabs 18:51:47 I can take your 2-space-per-indent and convert it to 8-space-per-indent in VIM *very* easily, and convert it back, with or without the use of spaces. 18:52:04 chandler oooh well - the RULE for isforth is NO tabs ... take that as a HARD convention 18:52:10 re Sonarman 18:52:15 not just a rule of thumb that noboedy ever follws anyway 18:52:21 I440r: you are repeating a stupid statement derived from an incorrect assertion 18:52:33 I have tried to show you that your assertion is incorrect 18:52:36 you have refused to listen 18:52:37 why? 18:52:47 TAB holy war! 18:52:52 :D 18:53:00 chandler do me a favour - work a few years as a consultant on source files that have been arround for a few years and edited by 50 or 60 different people 18:53:03 madgarden: Actually, I think chandler's argument puts a decisive end to both sides of the holy war. 18:53:09 THEN tell me i dont know wht the fuck im talking about 18:53:15 until then shut the FUCK up 18:53:33 I440r: you are being illogical. I have worked for years on other peoples' source. 18:53:35 I440r: Well, I can claim experience in that area, and I can also claim how to correct it, using vim, in a few seconds. 18:53:49 : set noexpandtab 18:53:52 : retab! 18:53:52 kc5 thers a problem with that too 18:53:56 : set tabstop=2 18:53:58 : retab 18:53:59 done. :) 18:54:02 I've worked for the past 7 years on shared source code where everyone uses tabs, and there has not been one problem with fucked up sources. There'd be hell to pay if someone indented everything with spaces, though. 18:54:11 any time you try "correct" a source file the owners get anal on you about no longer being able to do a DIFF 18:54:26 yes - you can tell SOME diffs to ignore white space 18:54:36 but ive never seen any that did it right 18:55:26 :) 18:55:29 Anyway, I think you guys should all just relax and play Sword of Fargoal: http://squidfighter.sourceforge.net/fargoal/ 18:55:30 ;P 18:55:56 I440r: Well, two things. 1) each open source project needs to have a coding convention in place to ensure that diffs are properly maintained; 2) This doesn't affect the use of tabs when tabs are properly interpreted for what they are -- as control characters. 18:55:56 ooh, nifty 18:56:28 0x09090909 is interpreted the same by diff regardless of whether they're displayed as 2 or 8 characters, or 4 or 5 or 3 or whatever. 18:56:38 kc5 im not talking about open source stuff lol - im talking about almost EVERY single c source file ive had the misfortune to have to work on as a contractor 18:57:09 writing source is hard when you dont know the keyboard layout ^_^ 18:57:52 kc5 you use 8 space tab stops and ill use 4 - ill edit your sources and when you get them back they are going to look fscked up to you because my 8 spece indents will take two tabs. then we have fred in the corner with 3 space tabs who now has to hit tab space tab space or tab tab space space.... 18:57:57 I440r: You claimed that people would have issues with diff. I offered a solution to the problem. This is why coding conventions exist. 18:58:05 damned soon you have a totally unreadable blob 18:58:06 s;jtktx 18:58:13 I440r: what? no, your indents take up one tab and are displayed as 8 chars 18:58:15 I440r: No, you are sorely, sorely mistaken. 18:58:21 You do NOT use 2 tabs per indent. 18:58:22 kc5tja's indents take up one tab and display as 4 chars 18:58:31 Because my 8 character tab will look like 4-character tabs on YOUR computer. 18:58:34 That's the whole POINT. 18:58:43 I440r: YOU TELL THE EDITOR HOW TO DISPLAY THE TABS 18:58:44 each indent corresponds to one level of indentation, no matter what the display is 18:58:51 er, each tab 18:59:06 thus, each level of indentation corresponds to one character (one byte) in the source file 18:59:46 i KNOW this. grrr - i dont fuck up sources, ive just had to deal with so fucking many of them that i fucking refuse to use tabs or accept ANY fucking sources containing them 18:59:51 PERIOD 19:00:01 MY convention is NO FUCKING TABS 19:00:03 * kc5tja shrugs 19:00:09 OK, that's fine. 19:00:09 the editor will format it the way you wnt it 19:00:24 That's your convention. 19:00:28 you seem to still not understand how they work. so I can only consider your position misinformed 19:00:33 But don't complain when other people choose to use tabs. 19:00:45 use tabs for scope, use spaces to line things up after you've indented for the scope 19:00:56 Sonarman: yes, someone else who gets it 19:01:09 I440r: youve been dealing with people that use soft tabs then. which is stupid 19:01:28 they can do so in any source file they dont intend to send for inclusion within isforth 19:01:37 any file containing tabs will be consigned to /dev/null 19:02:02 ln -s /dev/null /dev/ 19:02:05 ln -s /dev/null /dev/hell 19:02:06 Seriously man... I think you guys should scrap it out! 19:02:07 :) 19:02:10 "This is my convention." "Grr, I know what I'm talking about; I've been doing this for FUCKING YEARS! This is my position." 19:02:14 doesn't sound very reasonable 19:03:25 --- join: rpc (rpc@global.whiteh8.net) joined #forth 19:03:34 chandler deal with it. either use spaces or dont send me any source files. you pick 19:03:47 . 19:03:53 I probably don't have anything to send you anyway, given that I'm on a sane architecture 19:03:57 I totally misread "you pick". :-) 19:03:59 any file contaiining tabs or that uses upper case characters outside of quotes will be deleted 19:04:05 heh 19:04:07 chandler: PowerPC? 19:04:09 kc5tja: yes 19:04:11 i have this urgee to givee up 19:04:17 but i wont 19:04:21 I440r: but just the same, I'm saying that your argument isn't very rational 19:04:33 arke: We all do. 19:04:36 chandler its a backlash agains the shit ive had to deal with 19:04:47 have you LOOKED at the sources for isforth at all ? 19:05:18 nah. I don't really have a use for it, sorry. I'd like to, if you want to make it multi-arch at some point 19:05:33 i DO want to. 19:05:40 * arke is back (gone 05:56:13) 19:05:43 its always been the plan. but!!! 19:05:50 wb arke 19:05:54 it would be TWO seperate and distinct source trees 19:06:10 wouldn't at least some part of it be shared? 19:06:12 HEATHEN! HOW DARE YOU INSULT THE WAYS OF THE INTEL GOD!!! YOU SHALL BE CONDEMNED TO LIVE A LIFE OF ETERNAL SERVITUDE TO MICROSOFT! 19:06:14 i do not, i will not EVER support conditional compilation 19:06:20 :) 19:06:44 I440r: you don't have any words which are not arch-dependant? 19:06:47 Shared sources between the two are not conditional compilation. 19:07:09 chandler anything in the kernel can be considered arch dependant 19:07:26 anything outside that MIGHT be too - non arch dependant could go in a COMMON directory 19:07:46 oh, sure. I was just wondering about making it /all/ seperate and distinct 19:08:16 well - by that i mean i wont allow the interleaving of 28349765298346598273648975 different sources all into the same source file 19:08:21 like in the linux kernel 19:08:30 and most other "portable" sources 19:08:45 not 19:08:56 oh sure. linux kernel is not exactly an example of good design 19:09:25 no - its a pile of unreadable shit at the source level - runs great tho :) 19:09:56 i would classify the linux kernel to be CLOSED source 19:11:18 chandler should look at the sources for isforth and THEN decide if my conventions are irrational. 19:13:16 * I440r should also look at the sources for isforh and do some more work on it.... 19:13:49 btw, when you say that you can be found in #forth on your page, you might try saying what your nick is 19:13:50 * kc5tja should actually get off his duff and actually work on FS/Forth's ability to emit an ELF binary. 19:14:33 kc5 looked at isforths fsave.f ? 19:15:01 i thunked i did 19:15:02 lol 19:15:03 ohy 19:15:06 in the web page ? 19:15:08 lol 19:17:57 No, I don't have isforth's source code. 19:18:27 deliberatly avoiding looking at them? - ive done that with other projects too so i dont get blamed for stealing ideas lol 19:18:59 Yes, but not for the reasons you state. 19:19:18 I don't want to pre-bias my implementation strategy. 19:19:28 right - that too :) 19:19:54 I figure once I have a fair crack at it myself, then I can look at what you have and see what I did wrong. 19:20:08 lol or what I did wrong hhe 19:20:19 im not "the" expert - im learning as i go too 19:28:22 :) I also like to try to come up with my own solution before I look at other peoples'. sometimes I come up with something I like better. 19:29:26 most times i do too - unless i can SEE theirs is better 19:29:33 Herkamire: how is your colour forth going? 19:30:04 color forth kernel sources are damned impossible to read 19:30:15 as in tight? 19:33:33 I440r: Funny, I find them overwhelmingly more readable than non-color sources. 19:33:58 im talking about the hand build opcodes 19:34:31 he doesnt even give the assembler as a comment grr 19:34:53 its just a huge blob of unreadable blah 19:35:07 I440r: Well, there's no law that says, "All ColorForth Implementations Hereby Shall Utilize Compiler Words to Directly Emit Machine Language." 19:35:20 I440r: You won't like FS/Forth then. 19:35:30 its his going back to sourceless coding again 19:36:48 kc5 thers another assembler by sone guy on clf i forget who - ive looked at it and its next to useless to me - i dont want to have to think opcode encodings for all x86 instrucitons just to code in assembler - what th efsck is the point in having an assembler if you have to construct instructions manualy 19:37:51 The idea is to minimize code footprint and implementation complexity. 19:38:23 I don't know precisely what you're talking about; I'm not sure what author you're referring to, nor have I seen any such assembler before. 19:38:29 But I'm sure that reduction of complexity is his goal. 19:40:19 im searching for it - he is a damned good coder and his assembler is highly regarded, i just cant use it :) 19:40:36 fasm ? 19:41:50 albert van der horst 19:42:00 http://home.hccnet.nl/a.w.m.van.der.horst/forthassembler.html 19:43:11 ES: MOV, X| T| DI'| [MEM +8* AX] FFFFF800 X, 19:43:18 all instructions are done like that - ugh 19:44:21 hmm 19:44:28 not exactly pretty 19:44:53 he admits as much 19:47:57 thats outright beasty, even for x86 19:49:16 yup 19:50:51 postfix asm is not that bad, certainly not worth that 19:51:00 Actually, he has a great idea for writing an assembler. 19:51:04 But I would do it a bit differently. 19:51:18 I'd take the RISC-like approach, where all the opcode "options" are encoded as part of the opcode's name itself. 19:51:45 well, RISCs often have extended opcodes 19:51:47 So a segment register to integer register move would be "mfs, ds| ax|" or something. 19:52:00 where mfs stands for Move From Segment-register 19:52:01 kc5 that would be prefereable to having to know every opcodes exact encodings 19:52:28 I also would have distinct "load" and "store" instructions, instead of a generic "mov" catch-all. 19:53:10 yeah, the mnemonic 'overloading' is really not needed 19:54:22 ldrb, al, $45 c, 19:54:29 ldrib, al, $45 c, 19:54:44 * kc5tja forgot the 'i' to indicate immediate values. :) 19:54:54 Or, better yet, 19:54:58 ldrib, al= $45 c, 19:55:12 LoaD Register Immediate Byte AL with the value $45. 19:56:51 It becomes a bit difficult to represent some of the more advanced addressing modes that way though. 19:57:06 E.g., how would one write MOV [EAX+ECX*4],EBX 19:57:37 It's clearly a store, so the first two letters have to be 'st'. 19:57:48 And its working with a 32-bit word, so the last letter has to be 'd'. 19:57:49 stwxm? 19:57:59 Store Word Indexed Multiple 19:58:13 or something :) 19:59:03 Maybe it'll work better if I have the type character at the beginning. 19:59:25 dstrr <- double-word store with register-register indirect 20:00:02 oh man, im lovin RISC atm 20:00:18 zardon: We don't have the luxury. :( 20:00:57 you must be able to pickup some cheap hw down there 20:01:30 zardon: None as cheap as x86. You can pick up server-grade motherboards for $80 here, with $150 to $200 CPUs to go in them. 20:01:47 yeah, true 20:01:52 Total system cost, icnluding case, mobo, power supply, keyboard and other peripherals is about $1000 for a SERIOUS-grade workstation. 20:01:53 but even used would be nice 20:02:06 that is cheap 20:02:25 I know because I built one for my roommate's recording gear. :D 20:03:00 Pentium IV 2.6GHz processor, 512MB of RAM, GeForce 2 chipset card (doesn't need the latest, but it needs to be fast enough to update the screen in real-time during recording), etc. 20:03:08 And 160GB of harddrive space. 20:03:12 nice 20:03:24 Too bad it has to run WinXP. :( 20:03:36 its amazing what current market share does to progress 20:04:09 What I find amazing is that I built this machine like it was nothing. And yet, it's more powerful than even the supercomputers of only 5 years ago. 20:04:24 yeah, probably 20:04:58 cheap hardware is good, but im willing to pay the 'RISC tax' 20:05:11 I would be, if I could afford it. :D 20:05:17 heh, yeah 20:05:26 used hardware, the way to go 20:05:27 * kc5tja was seriously considering getting an AmigaONE box, just so I can have the PowerPC processor, without the Apple tax. 20:05:32 old servers 20:05:53 i looked into those boards, they are not cheap 20:06:03 Nope, they're not. 20:06:08 600-800 Euros last i checked 20:06:12 But they're not too shabby, with an on-board Radeon 8500. :) 20:06:28 well, its PPC + OFW that i dig 20:06:55 zardon: You currently using a PowerPC-based system? 20:06:57 Or a MIPS? 20:07:01 PPC 20:07:04 Ahh 20:07:11 been wanting a MIPS 20:07:13 What model PPC? 20:07:22 IBM 750FX 20:07:27 @ 600 20:07:27 Nice 20:07:33 yeah, its a nice chip 20:07:36 Bet it'll out-do my 800MHz Athlon. 20:07:42 heh, maybe 20:07:46 no vector unit 20:07:59 I've got MMX on mine. 20:08:14 But except for Neverwinter Nights, I never use those instructions. :D 20:08:19 heh, yeah 20:08:25 i don't even use the FPU 20:08:31 in my code at least 20:08:34 Although, some output by gcc might use them, not sure. 20:09:06 zardon: Are you using an older Mac, or an RS/6000? 20:09:07 thats one of the arguments to Gentoo is it not? 20:09:17 May 2002 iBook 20:09:43 zardon: Yep. However, Slackware is just as good in this respect; so little binary archives exist that source is pretty much all you get. :) 20:10:45 i hear that x86 chips are horrible to optimize for 20:12:12 that since they are effectivly microcoded, that the instruction latency changes with each new revision 20:12:56 While there is some truth to that, it's not because of microcoding. 20:13:05 If you stick with simple, RISC-like instructions, you get RISC-like performance. 20:13:12 The problem is you really only have 7 general purpose registers. 20:13:42 yeah 20:13:56 Register windows and register-renaming can mitigate that in hardware to some degree, but having only 7 *visible* registers to the compiler is a huge bottleneck. 20:14:13 well, thats what Itanic is for :) 20:14:15 Athlon/64 fixes this having 16 registers. 20:14:38 yeah, i could get by on 16 20:15:03 Yep. 20:15:04 but when you are trying to do effective instruction scheduling, more is better 20:15:20 32 is excessive, I think. 20:15:31 16 is quite adequate for most instruction scheduling purposes. 20:15:55 it will do, for sure 20:16:37 Yep; this is the second step in the right direction for the x86-series of CPUs. 20:16:50 The first was the transition from 286 to 386. 20:16:56 i think killing x86 would be the best way 20:17:12 I certainly agree. 20:17:13 start freash 20:17:22 But market forces dictates that that'll never happen. 20:17:28 yeah 20:17:29 back 20:17:32 hello 20:17:51 how is your colour Forth doing? 20:17:51 zardon: my forth is going very slowly these last couple weeks. 20:17:54 ah 20:18:07 nearing completion? 20:18:19 no, although I have a cool demo 20:18:40 nice 20:18:47 I've been doing a lot of thinking about interface stuff, and trying to decide how I'm going to migrate to new ways of organizing data 20:19:08 You know, flat mode on a 386 is really nice. The register orthagonality isn't bad, you've got some decent instructions. It's when you start tossing in traps and gates, and real mode, that it all goes to hell. 20:19:46 the fact that it has modes is a little much i think 20:19:51 Yes 20:20:13 user/supervisor 20:20:16 zardon: thought you were talking to me for a sec. I was about to agree :) 20:20:22 oh 20:20:28 compile/interpret ? 20:21:08 yeah, i didn't like ColorForth at first 20:21:18 but i think it does have merrit 20:21:56 I really like seing complete useful systems 20:22:13 Once I found a machine I could get it run on (out of 5 or 6 I tried), I found the keyboard was a little too alien. 20:22:54 i've not actually run it 20:22:58 the idea i mean 20:23:40 I don't have any opinion about ColorForth, other than with the wierd keyboard and limited hardware it runs on, I couldn't do anything with it. 20:24:03 I don't know enought about Forth to have that kind of opinion :) 20:24:12 there is a version of colorforth that runs within M$ windows 20:24:21 that's how I tried it 20:24:30 and also a version that runs on Linux 20:24:32 and an x86 X version 20:24:54 cool 20:25:05 so I could still run it on my girlfriend's computer :) 20:25:28 Wait, something ain't right. 20:25:38 You're a computer geek, and you've got a girlfriend? 20:25:44 You're lyin' about sumthin'! 20:26:02 No, he's lyin' WITH sumone 20:26:29 heh 20:26:56 That's almost bash.org candidate material. 20:27:08 Sonarman: hehe 20:27:25 a good one too :) 20:27:35 she let me install linux on her machine after Windows died 20:27:44 Good lord, man, marry her! 20:27:49 heh 20:28:45 jc: not entirely out of the question. definatily not soon though 20:29:10 she even likes ion :) 20:29:22 Dayumn 20:29:26 Herkamire: Dude, if YOU don't want her, I'LL take her. :D 20:29:30 the VM? 20:29:32 WM 20:29:40 yeah. ion WM 20:29:44 nice 20:29:52 quite the catch is sounds 20:29:55 kc5tja: pfft, you couldn't handle a woman! 20:30:03 kc5tja: because if she's decent, she likes Dream Theater 20:30:13 what's dream theater? 20:30:19 MysticOne: You misspelt indecent. 20:30:21 Herkamire: best band EVER! 20:30:31 kc5tja: touche... 20:30:49 Penumbra is good too (I've only heard one of their songs, but it is GOOD) 20:30:50 I went to a Christmas party hosted by my boss at In-N-Out on Monday. 20:31:00 I was accosted because I didn't know of a band called "Hessy" or some such. 20:31:25 heh 20:31:47 I listen to DT like, 24/7 20:32:19 So I accosted her back with, "What? You don't know about Genesis? The world's first progressive rock band, that influenced everyone from Jethro Tull to Yes and beyond? You don't know about Jethro Tull? Nor Yes? Oh, what about Pink Floyd? Gosh, aren't YOU behind the times..." 20:32:45 Fink Ployd! 20:32:48 kc5tja: lol (re misspeld indecent) 20:33:03 Herkamire: he's mean to me because I'm mean to him 20:33:11 :) 20:33:22 An eye for an I, a tooth, for a tuthe.... 20:33:22 but in the end, we both know I'm right about everything 20:33:22 even if I'm not 20:33:34 * MysticOne creates a paradox and the world ceases to exist 20:33:41 It's not often I'm wrong about these kinds of things, but this time I'm right. 20:33:56 hehehe 20:34:08 hehe 20:34:15 I have a largish Jethro Tull collection 20:34:25 But not enough Pink Floyd. 20:35:13 I saw PF at Ga Tech a few years ago, for the Division Bell tour. In the rain. It was... OK. 20:37:44 :) 20:38:00 * arke is away: I'm busy 20:38:32 * arke is back (gone 00:00:02) 20:38:54 :) 20:38:56 :) 20:42:29 jc: Pink Floyd is an awesome band. However, their last tour was a bit of a wind-down. They stopped everything after that tour. 20:42:51 jc: Their earlier tours were incredible -- I wish I could have gone to one. 20:45:23 kc5tja: I still think they really went down hill after Roger Waters left 20:45:49 dvorakk you biiiiiitch 20:46:50 Yea, I agree. I've never cared for The Wall much, but A Nice Pair, DSOTM, Atom Heart Mother, that genre I always liked. Animals was good. Division Bell was pretty good. 20:47:05 I have a couple of albums by the other dude that was with them.... er, his name is... 20:47:11 * arke likes Dio 20:48:26 Damn. I seem to have a major mental volume offline... 20:49:21 David Gilmour. 20:49:45 red hot chili peppers 20:50:10 Dio is good. Not big on the RHCPs. 20:50:13 jc: I loved The Wall, though 20:50:22 Dream Theater! 20:50:23 yay! 20:50:40 A lot of people do. I just wasn't one of them. I think sitting through the movie 8 or 9 times, stoned as shit, didn't help. 20:50:42 dvorakk you biiiiiitch 20:50:42 dvorakk you biiiiiitch 20:50:42 dvorakk you biiiiiitch 20:50:43 dvorakk you biiiiiitch 20:50:43 dvorakk you biiiiiitch 20:50:43 dvorakk you biiiiiitch 20:51:23 I've been discovering Bluegrass, lately. Many, Ricky Skaggs, Bill Monroe... Those guys can crank. 20:51:40 s/Many/Man/ 20:51:53 jc: oooh, no... Bela Fleck and the Flecktones! 20:52:23 Haven't hit them yet, but I'm always looking for "new" stuff. I'll check that out. 20:52:37 jc: lemme see if I can send you an mp3 or something ... 20:52:45 That'd be cool. 20:52:53 jc: it's like, bluegrass/jazz and all altogether ... and Bela plays a mean banjo :) 20:53:56 Someone also gave me a couple blues CDs to listen to. Anthony Gomes, Lucinda Williams have been two I really like out of this batch. Got some Son House, but it's remakes of stuff Ive already got. Also got some Lightnin' Hopkins. 20:54:25 And as much as it embarrasses me to say this (from a political viewpoint), there's a couple Dixie Chicks songs that have some excellent bluegrass playing on them. 20:54:44 i like the Dixie Chicks 20:54:45 * MysticOne gasps! 20:54:51 jc: you're not a patriot! 20:54:55 * MysticOne lynches jc 20:54:59 j/k 20:55:00 :) 20:55:07 I said it was hard to admit... 20:55:28 But that dumb broad really needs to learn to keep her yap shut. 20:55:44 :) 20:55:47 * MysticOne is all for dissent 20:55:52 it's what formed this kuntry to being with 20:56:20 I'm for it to. But the troops are just doing their job, and one their bound to do. You don't diss the troops. 21:00:28 im gonna giv. uuuuuuuuuuuuuuuuuuuuuuup 21:00:47 jc: well ... depends 21:01:06 unless you people quickly convince me not to 21:01:25 arke: DON'T DO IT!!! 21:01:47 Sonarman: GIVE ME A REASOON TO 21:02:05 arke: be nice to dvorak 21:02:11 he can be your friend 21:03:34 ugh 21:03:44 hee hasnt been nice yeet 21:05:33 come oooooonnn 21:05:50 im about to geet the scissors 21:05:53 Maybe you need some incentive. 21:06:03 How about we break a finger everytime you mistype? 21:06:14 jc: I'll upload the song when I'm done listening to Dream Theater ... last song of the album, but, I can't bring myself to stop it to listen and find out which Bela song to send you :) 21:06:25 heh! k. 21:06:39 Throw me a good DT, too. I don't know them. 21:08:21 Can you throw me a good DT too? I don't know them either :) 21:08:39 aoeuaoeuaoeuaoeuaoeuaoeuaoeu 21:08:40 aoeuaoeuaoeua.ouaoeu 21:08:52 DVORAKK I HATE YOU 21:08:56 uuuuh 21:09:00 wow 21:09:24 someone kill arke 21:09:41 i just realized that im suddenly typing faster 21:09:46 :) 21:09:49 MysticOne: ... 21:10:00 you'r'e typing faster because you aren't making sense 21:10:01 that's why 21:10:01 :) 21:10:08 .... 21:10:12 bastard 21:10:20 hehehe 21:11:01 * arke codes 4th 21:11:33 cooding is hard when you dont know thee keeyboarcd 21:11:37 VOMTEIDxantqexu 21:11:59 why thee fuck is it doubling??? 21:12:02 aopoiaquinsahbeuisntaboe-nahxoek-nahoxei 21:13:11 * kc5tja is away: food 21:13:58 I recognize that language, arke. It's Navajo. 21:14:10 nahxoek mean "Great River", right? 21:14:21 no, it's nohoxei 21:14:21 ............... 21:14:21 .. 21:14:32 sorry, nahoxei 21:14:39 bastards 21:14:53 haha aarke 21:15:12 The best kind of fun to have is always at someone elses expense. 21:15:23 baostards 21:15:24 se sarryy 21:15:26 -anootb-anzotibanoi 21:15:35 htdhxtahxtahtthba teeuhbxatheutahxeuthaxxethxb thxaeot khx aotek taex taxue taxeetuxatgitooxtaxta hx ncag.optndahxkvthdxentxn.c.gipna'to.ifnccfpuxentgxinc;fincaofugis;dihubvthx k;ntqhjxkvt;h jtvxqj vt; jktv; qtjkxotn qgui ntuhx vot;xi 21:15:38 yay 21:15:47 now that's lojban 21:16:10 coding is hard when you dont know the keyboard 21:17:22 bastards 21:17:22 bastards 21:17:23 bastards 21:17:23 bastards 21:17:23 bastards 21:17:24 bastards 21:17:26 bastards 21:17:28 bastards 21:17:31 bastards 21:17:32 bastards 21:18:46 --- join: ss (~matt@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth 21:19:08 --- nick: ss -> fridge 21:21:11 . 21:21:12 bastards 21:28:38 who? 21:29:38 dvorak :) 21:30:52 how can a C or ASM prog find its own PID? 21:32:47 getpid() 21:33:57 is that a syscall? :) 21:34:07 Syscall 20 21:34:08 yes 21:34:08 yeah 21:34:13 thanks 21:34:17 on linux, it's 20 21:34:34 what base? 21:34:42 dec 21:35:02 all of which are belong to me 21:35:02 I have trouble thinking in decimal, except for paychecks. 21:35:26 AYBABTU is a meme that should never die. I run across that in my playlist every few weeks. 21:35:27 :) 21:36:24 Along with The Shover Robots. 21:37:37 I440r: syscall by number? 21:38:46 you use 21:39:29 ( number-of-arguments syscall# "name" ) 21:39:31 i think 21:40:56 er.... 21:42:59 --- quit: slava (Read error: 110 (Connection timed out)) 21:44:31 yeah 21:44:38 thanks 21:48:09 a callback fork interface ....... neat 22:08:38 --- quit: Herkamire ("Dvorak only has one K at the end") 22:10:53 haha :) 22:12:36 sonarman you been using isforth :) 22:13:08 arke man 2 some-syscall to learn how to implement specific syscalls :) 22:14:30 i've never used it and I don't know a whole lot about it; I just like reading the code :) 22:14:36 heh 22:14:41 why dont u use it ? 22:14:51 not on x86 here 22:14:51 syscall.h 22:15:08 err syscall.h is about as much use as a nun in an orgy 22:15:12 unistd.h has the numbers for all of the syscalls for me 22:15:25 unistd.h is the one 22:15:34 but there are syscall man pages that say nothing but 22:15:38 "you dont need to know this" 22:16:28 asm/unistd.h 22:17:15 sys/syscall.h on bsd :) 22:18:25 * kc5tja is back (gone 01:05:13) 22:18:33 omg hi rpc~ 22:19:26 howdy 22:21:52 got anything done on your forth? 22:22:10 Wow. I never had to /ignore someone so fast on an IRC channel as I did just now. 22:22:25 lol who was that kc ? 22:22:35 I440r: A person named Jymmm in #hamradio. 22:22:47 did anyone watch that battlestar galactica miniseries? 22:22:55 on SciFi? 22:22:58 erm jymmm - ive seen that nick before 22:23:00 yes 22:23:01 #debian ? 22:23:06 nope, only get Space :( 22:23:20 or by other means that is by now wildly distributed by now 22:23:22 zardon: A fellow Canadian! 22:23:23 No, it's only on #hamradio right now 22:23:27 no he is in #linpeople hhe 22:23:32 space is a pathetic excuse for a channel 22:23:34 i knew i knew him from somewhere@ 22:23:40 there are quite a few in here 22:23:50 this channel is growing 22:23:51 hey, I like Space... 22:23:54 Space is ok, but sortof impoverished 22:24:04 I440r: I just did /whois jymmm and it reports he's only in #hamradio. 22:24:08 most of the new shows suck anyways 22:24:16 re-runs is where its at 22:24:21 i just did one too and it reports him only in #linpeople lol 22:24:25 oh speaking of which 22:24:29 space has only second choice in programming to city, enterprise for example - city has it on wed, space has it the following sunday when everybody has already seen it 22:24:32 i need to .torrent the new battle star gallactica 22:24:33 --- [Jymmm] (~nobody@adsl-68-124-183-198.dsl.pltn13.pacbell.net) : nobody 22:24:33 --- [Jymmm] #linpeople 22:24:33 --- [Jymmm] irc.freenode.net :http://freenode.net/ 22:24:33 --- [Jymmm] is an identified user 22:24:33 --- [Jymmm] idle 00:00:07, signon: Thu Dec 11 13:44:03 22:24:34 --- [Jymmm] End of WHOIS list. 22:24:42 rpc :-) 22:24:46 I440r: Yeah, same guy. 22:24:48 It's comforting. Heh. I usually watch Dark Angel on Tuesday, and then fall asleep to Voyager or whatever movie comes on afterwards. 22:24:54 * kc5tja compared his hostmask. 22:25:01 yea thats why i pasted heh 22:25:14 there must be some weird irc magic going on here 22:25:20 So what's with a bunch of Canadians doing Forth in here? :D 22:25:28 yeah, but City has Muary or whatever that 'Who is the dady' show is 22:25:37 so, its a total turn off 22:25:59 the Vancourver City is cool enough i guess 22:26:04 Yeah, that's a good point: this channel is growing... That's good to see. 22:26:32 im afraid that Forth is quite counter-culture atm 22:27:02 unless you can fit Distributed, Object Oriented, and Framework into the name 22:27:08 So is avante guard jazz. 22:27:17 zardon: Which I can. :) 22:27:21 yeah, well pop culture is a drag 22:27:24 heh 22:27:25 DOOF Forth, heh. 22:27:31 oh man 22:27:38 :) 22:27:54 DOOFUS -- Distributed, Object Oriented Forth User System. 22:27:58 I just like it because I'm a minimalist sort of guy. 22:28:05 yeah 22:28:06 So am I. 22:28:20 Plus, I would like to build Linux applications quickly and easily, without having to worry about compilers, type checking, and the like. 22:28:26 There is a place and time for all that stuff. 22:28:28 And no syntax is good syntax! Mmm... syntax. 22:28:35 If you're not a minimalist, you shouldn't join #forth.. :) 22:28:39 --- quit: Sonarman ("Lost terminal") 22:28:42 But mostly, it hinders exploratory coding. 22:28:53 Robert: QRP forever! 22:28:55 well, the fact that a person can implement their own version of the language is a major bonous 22:28:58 kc5tja: lol, yes. 22:29:28 "Yeah, just finishing up my C++ compiler :)))" 22:29:29 I dunno... I explored a lot of languages and ideas the summer before last, and Forth was the only think that blew me away. The others were just goofy syntax variations. 22:29:41 And what I find intriguing is, despite the fact that 100 people have 100 different versions of Forth, all ever so slightly incompatible with each other, it's amazing how incredibly compatible they are. 22:29:56 i am certainly not a minimalist but see an interesting compiler possibilities 22:30:11 * kc5tja likes Lisp/Scheme too. 22:30:16 I don't have the attention span to be anything but a minimalist. ;) 22:30:22 But Forth is easier for me to code and be able to create raw binaries from. 22:30:24 heh 22:30:29 madgarden: Hehe, I guess that's a good reason :) 22:30:44 Plus, it's so damn easy to use for scripting. 22:30:47 well, the fact that you can see exactly what code is being emited by your actions 22:30:49 Screw Lua. 22:31:35 * kc5tja never saw the point of Lua. 22:31:36 Exactly, it's so easy to implement. 22:31:54 Ruby, yes. Python, yes. Scheme, yes. Lua .... it's a bit schitzophrenic. 22:32:11 But, everyone seems to love it. At least in the game circles. 22:32:27 my favorite WM uses it for scripting and configuration 22:32:42 fridge: Sawfish? I thought that used scheme. 22:32:50 ion 22:32:51 Maybe I should look into using the scwm. :) 22:33:01 er i didnt like it 22:33:02 You must have a later version of ion then. 22:33:13 i am learning forth at the same time I am looking to implement a similar style system (based heavily on forth documents I see, but with some enhancements where I see them fitting in to my Master Plan (tm)) 22:33:18 since using windowmaker i havent found a single other windowmanager that i liked 22:33:34 sawfish uses lisp i thought 22:33:42 kc5tja: I upgraded in order to take advantage of float frames 22:33:43 I440r: WMaker still suffers from overlapping window syndrome. 22:33:52 kc5 define that ? 22:33:57 zardoz: I believe he said "ion". 22:34:00 overlapping windows never was a problem for me.... 22:34:02 I440r: I don't like overlapping windows. 22:34:08 i just shade em or minimize 22:34:25 A window manager should manage my windows for me, not me for it. 22:34:42 manage = draw 22:35:06 but floating windows are anoying 22:37:38 kc5 i think theres a wahy to stop windowmaker doing that but i wouldnt swear to it, i never bother with it 22:39:02 --- join: thin (thin@bespin.org) joined #forth 22:40:32 I440r: For when I really need the function keys on my keyboard, I'll switch to WindowMaker. 22:40:40 I do like WindowMaker. 22:40:44 But I find ion to be better for me. 22:40:52 re thin 22:41:06 sup 22:41:16 yo thin! 22:41:25 i just had a panic attack about you lol 22:41:44 i looked at how long it was since you had been in here while logged in/auth'd 22:41:51 heh 22:42:01 and its been weeks. i had to ask how long you can do that before the channel becomes deregged 22:42:03 never im told 22:42:23 i auth'd recently 22:42:40 the channel info says like 2 or 3 weeks ago 22:43:23 ah on the access list 22:43:51 so i440r, how is isforth coming along? 22:44:18 well i am plucking up the courage to do some more work on her 22:44:35 this contract is coming to a close soon - not sure about any next contracts 22:44:56 i gotta get back into the swing of things, force myself to start and get the ball rolling 22:45:12 once i got some momentum itll be all down hill coding till i burn out again :) 22:45:17 back into the jobless mad coding momentum ? 22:45:24 dunno 22:45:31 hope not 22:45:42 theres a possible contract in ohio for me... 22:45:52 forth or asm? 22:45:56 prolly c 22:46:22 are you good at c? 22:46:38 only if im paid good 22:46:45 heh 22:47:39 how much forth coding is there going around these days for companies that are looking for new systems, not maintaining old ones? 22:48:21 maintaining old forth is a hell of alot easier than maintaining old c... 22:48:25 http://www.beckysg1.34sp.com/images/swenema.gif 22:48:35 the work im doing now is for an existing product but the code is ALL new 22:48:39 i doubt anybody here knows the answer to that question 22:51:42 * kc5tja is doing a bit of exploration in creating an ELF header now. 22:55:16 oh kc5tja or anybody else here - physics question - A guy at work (retail store - Canadian Tire) seems to think that two things will drop at different speeds due to density. I think the guy is a looney bin, and said I was always taught that environment aside, two objects will fall at the same rate regardless of weight (i assumed density wasn't a factor in this either). Am I right as I suspect, or is he not quite the 22:55:16 resumed? 22:56:11 the only reason they could drop at different speeds is due to drag 22:56:25 or difference in WEIGHT 22:56:26 "in the absence of atmosphere" 22:56:40 erm wait 22:56:41 no 22:56:46 mass, weight 22:56:50 density doesn't matter 22:56:52 weight wont affect it - but maybe density will 22:56:57 i ferget lol 22:57:05 if they're different sizes, they'll have different air drag 22:57:07 weight the force of gravity on the mass 22:57:16 right 22:57:20 it doesn't matter 22:57:29 no - i think weight does lol 22:57:36 everything will fall at the same rate given the same conditions & no atmosphere 22:57:39 heavier will drop faster wont it ? 22:58:16 if there is no gravity, it will have no weight, and won't fall 22:58:31 if it was in a vacuum on earth, it would accellerate downwards at 9.8m/s/s regardless of its density, size, weight, volume 22:58:45 fridge, merci 22:59:00 do you live in .qc ? 22:59:05 fridge, do we have any nice documentation that is small and printable that I can blow him away with? 22:59:28 no, we only have experiments from highschool burnt into my memory 22:59:30 =) 22:59:34 Two objects, regardless of mass OR density, in a vacuum, will drop at the same rate. 22:59:55 I qualify that with in a vacuum, because it eliminates the effects due to air resistance. 22:59:56 is it covered by newton? 23:00:21 It was discovered by both Newton, and in an indirect manner, Galileo. 23:00:36 Here's something: http://www.glenbrook.k12.il.us/gbssci/phys/Class/newtlaws/u2l3e.html 23:01:47 And since weight is a force, and F = mass times acceleration due to gravity, weight also is immaterial. 23:02:12 hmm, i need something that discounts density for sure, or else he will just bring that point up again 23:02:18 This sums it up pretty nicely: 23:02:20 If Newton's second law were applied to their falling motion, and if a free-body diagram were constructed, then it would be seen that the 10-kg rock would experiences a greater force of gravity. This greater force of gravity would have a direct effect upon the rock's acceleration; thus, based on force alone, it might be thought that the 10-kg rock would accelerate faster. But acceleration depends upon two factors: force and mass. T 23:02:50 boy this conversation is boring 23:02:51 Why do you need to take density out of the question? Density has no effect. 23:03:03 exactly 23:03:06 Density is just mass/volume. Volume surely doesn't affect anything. 23:03:13 kc5tja, he thinks it does, thus i need to prove him wrong 23:03:21 who thinks it does?? 23:03:22 madgarden, in his demented mind it might 23:03:26 oh 23:03:27 thin, guy I work with 23:03:48 I invite him to drop a cinder block along side a penny, and watch which falls first. 23:03:57 look, mass DOES make a difference, but it is imperceptible. but if you have a moon and a marble, the moon is definitely gonna fall faster 23:04:00 They'll both hit the ground at (roughly) the same time. 23:04:11 thin: Not true. 23:04:17 The moon will hit at the same time as the marble. 23:04:33 kc5 except the moon is moving further away heh 23:05:02 The reason is the smaller mass of the marble is consequently easier to accelerate, and so it accelerates `faster' than the moon. 23:05:26 "But acceleration depends upon two factors: force and mass. The 10-kg rock obviously has more mass (or inertia). This increased mass has an inverse effect upon the rock's acceleration." 23:05:33 But precisely because of its reduced mass, it's acceleration is counter-balanced, and ends up being the same as that of the moon's. 23:06:12 The force does depend on the mass of the object, but it also depends on the mass of the Earth too. 23:06:25 F = kmM/(r^2) 23:06:28 i was thinking of F = G ((m1*m2))/r^r) but yeah, that doesn't say anything about heavier masses falling faster 23:06:32 if an object of any size is falling towards another object an infinite distance away will it continue to accelerate or will it reach a max ? 23:06:45 where k is the gravitational constant, m is the mass of the smaller object, and M is the mass of the larger object. r is the distance between the two. 23:06:53 --- part: thin left #forth 23:07:00 The Earth doesn't just drag the moon towards itself, the moon is also trying to get closer to the earth. A smaller object has a much weaker gravity... Doesn't that mean that the moon will hit first? 23:07:11 Robert: No. 23:07:17 Because it's a ratio. 23:07:22 a = F / m 23:07:25 --- join: thin (thin@bespin.org) joined #forth 23:07:38 The law of gravitation states that if m is large, then it's F is going to be larger too. 23:07:48 robert the moon is moving away from the earth - not towards 23:07:56 Conversely, if another m is smaller by a factor K, then its F is also going to be a factor of K smaller too. 23:08:08 I440r: Hypothetically speaking here. 23:08:08 I440r: Heh 23:08:13 ok :) 23:08:16 Let's not get too detailed, or the point of the discussion will be lost. 23:08:49 well i hope someone is jdrake is reading the conversation 23:08:55 cuz i'm /clearing as quick as i can 23:09:04 bleh 23:09:04 thin: I'm sorry to disappoint. 23:09:07 :-) 23:09:23 nah, i don't have any interest in this stuff 23:09:33 even tho i do take physics at univeristy 23:09:51 kc5tja: What if there was a planet the size of the earth out there, falling towards us? Then it would attract the earth with the same force as the earth is attracting that planet, right? 23:10:06 robert: correct 23:10:23 kc5tja: And thus the relative movement would be twice the speed as when the earth is attracting a very small object, with insignificant gravity. 23:10:44 ? 23:11:00 Bigger masses are harder to accelerate, that's why they don't fall faster. 23:11:11 Sure. 23:11:18 That's all there is to it. 23:11:21 ugh 23:11:29 But if _both_ masses are accelerating each other... 23:11:37 no, the force is higher 23:11:37 Yea, same goes for your pebble. 23:11:38 change the topic already 23:11:39 not the acceleration 23:11:47 But the pebble is easier to move, less inertia. 23:11:51 the acceleration is the same, the force is higher for the heavier masses 23:12:03 Right. 23:12:04 Robert: No, not at all. 23:12:13 SWAP DROP DUP, dammit! 23:12:24 F = ma; if m is small, then for any given F it will result in a substantially larger 'a'. 23:12:34 nip dup <-- optimized :) 23:12:47 Heh! :) 23:13:00 >r drop @r r> 23:13:04 swap drop dup was the first thing I ever tested on my fledgeling forth-like 23:13:12 Hrm.. 23:13:16 change the @r cuz i forget which word it should be 23:13:22 I'll do some more counting on that, heh. 23:13:25 r@ 23:13:26 Heh, I don't allow access to the return stack. 23:13:28 But now, time for school. Bye. 23:13:37 madgarden why not ? 23:13:39 robert: grade 12/ 23:13:41 School at 2am? 23:13:43 its an almost indespensible tool 23:13:54 madgarden: yeah, those crazy sweden ppl get up at 2am 23:14:06 thats why they are all depressed! 23:14:07 lol 23:14:17 Well... my forth-like is written in C, intended for scripting. 23:14:31 If they messed with the return stack, they could crash it. I only allow certain things on it. 23:15:01 whats your next goal 23:15:08 you gonna code anything in forth? 23:15:18 They don't need it for temporary storage, since I have stacks as a basic type. Yes, I have types. 23:15:22 Who, me? 23:15:54 I440r: To get a static binary image, we need at least a .text segment in the ELF file, correct? 23:16:06 --- topic: set to 'where people get together to talk about CVT, stirling engines, tesla turbines, C, ham radio, and occasionally Forth | how come nobody actually codes in forth?' by thin 23:16:27 yes - you will also need some other items in there - i would need to refer to fsave.f to remember 23:16:31 --- topic: set to 'where people get together to talk about CVT, stirling engines, tesla turbines, ham radio, and occasionally Forth | how come nobody actually codes in forth?' by thin 23:16:43 thbers some very iffy shit regarding the way linux treats elf files... BUGGY in fact 23:16:47 Well, I know it needs a string table, or else the loader doesn't have a clue as to which is the .text section. :) 23:16:54 how come nobody actually codes in forth? 23:16:58 yes heh 23:17:11 Because there are no good Forth development environments. 23:17:12 lets discuss this 23:17:16 but theres more - theres one sectin thats legal to NOT have. but linux doesnt like it not being thre 23:17:22 kc5tja: true 23:17:28 but why aren't there? 23:17:34 I'm hoping FS/Forth will address this issue for me. 23:17:41 thin: The point of my system is to allow the user to create a thin domain language for their problem. So, I could make a Forth-like language, but most likely I'll make some high-level thing that would suit the scripting problem at hand. 23:17:41 Because nobody cares, and nobody wants one. 23:17:55 Forth programming is too ``hard,'' so they resort to pre-existing languages like C. 23:18:26 Also, nobody programs in Forthy because there aren't enough jobs that require it. 23:18:50 kc5tja: i assume that the development environment the iTv engineers got to learn on was an excellent environment 23:19:02 I don't know. 23:19:04 actually its the string section thats optional - you DONT need it according to the elf standard but linux is bitchy if its not three :) 23:19:12 I refuse to make a statement one way or the other on that topic. 23:19:21 whats happening with that btw ? 23:19:25 I440r: My standard says it has to be there. 23:20:05 iTV is gone. 23:20:09 Has been for a long time. 23:20:19 yes but what about their intelectual property rights 23:21:34 intellectual property... so somebody could own the thoughts in my head and demand royalties? 23:21:45 yes 23:22:20 i think that is an aburd way to think about property 23:22:47 it only serves the well off to make them even more well off 23:23:05 I440r: Why is access to the return stack an indespensible tool, other than for temporary storage and wonky execution flow? 23:23:15 kc5tja: have you ever encountered a nice stand alone development environment for any language, and when you played with it you learned tremendous amount very quickly, it may have seemed like a basic environment but it provided everything necessary to start.. i've encountered that, and that's where programming becomes a lot of fun 23:23:43 madgarden how about : foo .... if r>drop exit .... ; 23:23:56 ans doesnt allow that but THAT is a very usefull mechanism 23:23:58 also 23:24:06 : blah .... if >r then.... ; 23:24:25 Give me some examples of how they're used. 23:24:29 put a return address onto the return stack. return twice to get back to where you came from 23:24:31 thin: I'm sorry, I don't understand what you're getting at. 23:24:50 madgarden hang on ill paste a function from my 8051 assembler 23:25:08 FS/Forth and isForth are both designed to go straight from source to raw binary with a minimum of fuss, while still supporting interactive development. 23:25:19 That's why I'm researching generation of ELF files now. 23:25:47 : asm> 23:25:47 mn \ get address of last mneumonic word 23:25:47 r> !> mn \ save return address to next mneumonic 23:25:47 [ ' (asm>) >body ] \ exit to this when opcode is completed 23:25:47 literal >r 23:25:52 dup ?: \ is there a previous mneumonic? 23:25:54 >r \ interpret it 23:25:58 drop ; \ or... the next one is realy the first 23:26:09 this is THE most important word in the assembler, its how i keep INTEL syntax in forth 23:26:17 kc5tja: i was asking if you have encountered one of those nice stand alone development enviroments for programming (any language), something that inspires your vision of a development environment 23:26:35 thin: Yes. BASIC on the old 8-bit computers. 23:26:45 i was inspired by my encounter with MUF, even tho it wasn't a fully interactive environment 23:26:53 each assembler mneumonic calls this word. this word pulls the return address off he stack. pushes the previous return address back onto the return stack and saves the current return address away till next time 23:26:58 you can think of it as a delay line 23:27:01 xxx yyy zzz 23:27:14 xxx doesnt execute till yyy is called. yyy doesnt execute till zzz is called 23:27:53 also being able to temporarilly store things on the return stackk saves you from having to do weird parameter stack juggling 23:28:01 >r r> are very important imho 23:28:18 Hmm. 23:28:23 I'm making progress. If I try to run a blank ELF header, it reports, "Cannot allocate memory." :D 23:28:33 but it apparently does see it as a real ELF. 23:28:34 Well, like I said, I have stack types, so there's no need to mess with the return stack. 23:28:50 kc5 would it be ok for you to paste to me a DUMP of your elf header and program header ? 23:28:52 As for the syntax... couldn't you use EVALUATE to do the same thing? 23:29:23 evaluate is a horrible word 23:30:03 i wont support it... it effectivly moves where your interpreting from - so an evaluated word CANNOT parse the input stream 23:30:04 DUMB 23:30:20 evaluate is of limited use in my opinion (others may disagree) 23:30:20 kc5tja: i believe that the main keys to an excellent development environment is an excellent help system that is accessible from the current interface rather than opening up in a different window. i also think the vocabulary should be clean, perhaps broken up into multiple vocabularies each with a distinctive purpose.. ( i'm thinking primarily in terms of newbies encountering this development environment ) 23:30:23 Still... I think you don't need to mess with the return stack to do this. 23:30:30 I only have an ELF header; I do not have a program header yet. :) 23:30:40 aha thats the reason maybe ??? :) 23:30:48 may i see the elf header ? 23:31:11 Well, if you want INCLUDE, you need EVALUATE? 23:31:18 EVALUATE is like LOAD, or INCLUDE -- it changes the input source to the string. 23:31:21 err no 23:31:36 When the string is exhausted, it will change back to its previous input source. 23:31:42 I find EVALUATE to be utterly invaluable. 23:31:53 kc5 yes - its NOT include, it doesnt change the input stream file it changes the input BUFFER address. 23:32:01 You need to maintain a stack of input sources though, non? 23:32:06 No, it changes the whole farging input source. 23:32:50 madgarden: Yes, you ultimately do need to maintain a stack of input sources. 23:33:00 Most Forth implementations use the return stack for this purpose. 23:33:20 kc5 if i say " blah blah" evaluate in a : def the string leaves its address on the stack. evaluate makes THAT address the input stream... therefore the words "blah blah" CANNOT parse the input stream.... 23:33:39 that memory address remains the input stream till its exhausted 23:33:45 --- quit: thin ("Leaving") 23:33:47 I440r: Because the input stream is the string. 23:33:58 also, " blah blah" evaluate should be done : ... blah blah ... ; 23:33:58 Dude, I cannot have an INCLUDE'd file parse the input stream, so why should a string? 23:34:08 never do at run time what can be done at compile time 23:34:13 also - you cannot turnkey the above 23:34:14 I440r: Not if you're expanding a text substitution macro. 23:34:27 I440r: EVALUATE is compile-time stuff. 23:34:29 kc5 define text substitution macro ??? 23:35:15 A macro. 23:35:21 A normal, every-day, honest to goodness macro. 23:36:02 you mean colon definition macros ? 23:36:12 ( a short-circuit AND implementation; usage: expr && expr && expr ) 23:36:30 : && S" DUP 0= IF DROP EXIT THEN" EVALUATE ; IMMEDIATE 23:36:35 instead of doing : blah postpone a postpone b postpone c ..... adnausium .... ; immediate 23:36:41 : myCondition exprA && exprB && exprC ; 23:36:59 the above gets compiled to: 23:37:05 yes. 23:37:11 exprA DUP 0= IF DROP EXIT THEN exprB ...etc. 23:37:30 your method above is better than the bulshit postponing methid, you can do loops inside your macros 23:37:54 but - my macro colon definitions are compiled and inject their code into the definition being created when they are invoked 23:38:02 and the code is like 2 pages or something 23:38:18 also... when you are done NONE of the macros take up any space in your executable 23:38:19 I440r: What the heck did you just say? In English again please? 23:38:23 All colon definitions are compiled. 23:39:04 kc5.... no - not in the above, the BODY of your macro is a string which is evauluated. but because you are in compile mode when you evaluate the string is COMPILED 23:39:05 very clever 23:39:06 but 23:39:18 Ummm....sorry, but you're dead-pan wrong. 23:39:25 m: blah 100 0 do i noop . loop cr ." blah!" ;m 23:39:32 : foo ..... blah .... ; 23:39:45 all of the code in blah is injected into foo 23:39:48 not a call to blah 23:39:53 and later blah disappears 23:40:00 thats how i implemented this 23:40:03 : blah S" 100 0 do i noop . loop cr" evaluate S( ." blah!) evaluate ; 23:40:23 What you implemented is exactly how evaluate works, dude! 23:40:23 --- quit: warp0x00 (Connection timed out) 23:41:00 --- join: warp0x00 (~warpzero@dsl.142.mt.onewest.net) joined #forth 23:41:00 kc5 no - i dont change the input buffers address and i dont have strings of code inside my code 23:41:10 the contents of blah is COMPILED code 23:41:13 When EVALUATE is called by the IMMEDIATE word (e.g., purely, 100%, God-enforced compile time), it's execution semantics are IIDDEENNTTIICCAALL to file-inclusion. It's literally injecting the code in the string *SMACK* into the definition being compiled. 23:41:15 not a string representing the source 23:41:34 kc5 yes i understand that heh 23:41:49 Well, sorry, but I just don't see the advantage to your system. 23:41:54 I see it as unnecessary complexity. 23:42:09 its what i said above. you evaluate the string because your macro is immediate but you are evaluateing it inside a colon definition and are therefore compiling 23:42:27 actually my macro colon definitions mechanismn is quite simple 23:42:43 the advantage is that its less verbose 23:42:52 You cannot parameterize it either. 23:42:56 always a plus 23:43:00 say again ? 23:43:04 what do you mean 23:43:05 You can't parameterize it. 23:43:08 explain that 23:43:14 You can't pass parameters to the macro to change its behavior. 23:43:32 show me a simple contrived example of this... 23:43:56 : foo exprA [ bitwise: ] && exprB ; 23:44:09 where && is the macro, and bitwise: is some means of affecting it's behavior. 23:44:15 if " ,...." else " ...." then evauluate ? 23:44:41 No, more like templates. 23:44:52 Here's perhaps a better situation. 23:45:03 In ANSI Forth, you always have to execute THROW after a file I/O operation. 23:45:20 well i see your method as WAY better than the usual ans infinite number of consecutive postpones inside an immediate word 23:45:26 What if you didn't want to do this? What if you wanted to store its value or whatever? 23:45:36 but its still more verbose than my method - maybe more versatile but i can live with that 23:45:43 I440r: Actually, my solution was ANSI inspired -- it's perfectly legal ANSI Forth. 23:46:05 kc5 i didnt say it wasnt but MOST ans coders do 23:46:14 * kc5tja nods 23:46:18 : blah postpone x postpone y postpone z ..... ; immediate 23:46:24 which is fscking oogerlie 23:46:30 what would be a good recommendation for a forth environment for testing 23:46:31 yours is WAY better 23:46:35 Heh yea, that's what I do in my system, but then I don't have evaluate working yet. :P 23:47:02 jdrake: What are you interested in testing? 23:47:08 the language itself 23:47:10 i dont support th euse of postpone either 23:47:19 for learning exactly how things work and so on 23:47:22 under which OS ? 23:47:27 macosx 23:47:36 if linux my opinion would be biased heh 23:47:36 jdrake: I would probably go with gforth to start with. 23:47:43 gforth is prolly your ONLY choice 23:48:06 good enough 23:48:08 merci 23:48:35 This stinks. I need a section table too to differentiate code and string segments. 23:48:37 FICL should build there as well. 23:48:39 * kc5tja forgot about that. 23:48:48 yup 23:49:01 who did ficl ? 23:49:08 he came in here a few times, smart dood 23:49:18 |( ELF Header Generation ) empty | 23:49:19 | | 23:49:19 |variable elfHeader | 23:49:19 |variable elfFile | 23:49:19 | | 23:49:19 |: 2, , -2 allot ; | 23:49:21 |: file S" a.out" r/w bin create-file throw dup elfFile ! ; | 23:49:23 |: close elfFile @ close-file throw ; | 23:49:25 |: turnkey elfHeader @ dup here swap - file write-file close ; | 23:49:27 | | 23:49:29 |here elfHeader ! | 23:49:31 | | 23:49:35 |$7F c, 'E c, 'L c, 'F c, 1 c, 1 c, 1 c, 0 c, 0 dup , , | 23:49:37 |$0002 2, $0003 2, 1 , $10000 , 0 , 0 , 0 , 52 2, 0 , 0 , 0 2, | 23:49:39 | | 23:49:41 | | 23:50:00 you need to brandelf that header 23:50:03 That's what I have so far. 23:50:10 John Sadler, Larry Hastings are listed under the project at SF. 23:50:17 brandelf? 23:50:30 this part is really badly documented but after the 1, 1, 1, you need a 03 for linux or a 09 for freebsd 23:50:49 brandelf is a util in fbsd to say weather or not the elf is a linux executable or not 23:51:14 how big is your text section ? 23:51:38 I440r: Well, for FS/Forth's core environment, it'll almost certainly be less than 16KB to 32KB in size. 23:52:04 your entry point is $10000 ? 23:52:21 Yes. 23:52:39 I'm going to configure .text so that it loads at $10000. 23:52:48 your program header entry size is zero but thats right for now :) 23:52:59 you need a section header size 23:53:15 When I get those headers in place, I'll give them a size. 23:53:26 Like I said, very distinctly above, I only have an ELF header at the moment. :) 23:53:54 you also need to specify the section header string table entry index 23:53:59 yea i know heh 23:54:25 and a section header number (number of section headers) 23:54:40 I'm only going to have two sections; the string section and .text. 23:54:41 --- quit: jdrake ("Snak 4.9.7 IRC For Mac - http://www.snak.com (There are 2 version numbers left before 5.0.0!!)") 23:54:51 I'm going to embed the data stack into the .text section directly. 23:54:54 the program header is easy, it usually follows the above 23:54:55 --- [Jymmm] (~nobody@adsl-68-124-183-198.dsl.pltn13.pacbell.net) : nobody 23:54:55 --- [Jymmm] #linpeople 23:54:55 --- [Jymmm] irc.freenode.net :http://freenode.net/ 23:54:55 --- [Jymmm] is an identified user 23:54:56 --- [Jymmm] idle 00:00:07, signon: Thu Dec 11 13:44:03 23:54:58 --- [Jymmm] End of WHOIS list. 23:55:00 argh bad paste grrr 23:55:13 \ program header starts here 23:55:13 1 , \ p_type 23:55:13 0 , \ p_offset 23:55:14 $8048000 , \ p_vaddr 23:55:16 $8048000 , \ p_paddr 23:55:18 here -1 , \ p_filesz 23:55:20 -1 , \ p_memsz 23:55:22 7 , \ p_flags 23:55:24 $1000 , \ p_align 23:55:30 thats my program header 23:55:43 the -1 and -1 in there BOTH get the text section size 23:55:54 basically 23:56:27 * kc5tja thinks the program header is a useless header. It contains no useful information. 23:56:49 it contains lots of usefull information that exists elsewhere :P 23:57:04 Precisely my point. 23:57:09 or really isnt needed 23:57:20 well. the pflags is next to useless 23:57:26 i give 7 = +rwx 23:57:35 That was my intent too. 23:57:40 which linux quietly ignores and sets +rx 23:57:41 But those flags exists in the section header for .text too. 23:58:24 How did you get around that? 23:58:35 thers a syscall to change the permissions on a block of memory 23:58:40 sys mprotect 23:58:49 Ahh 23:58:53 I figured as such. 23:59:36 Maybe I should research splitting the data and code spaces in a future FS/Forth implementation. 23:59:48 :) 23:59:59 that would make fsforth faster at execution than isforth 23:59:59 --- log: ended forth/03.12.11