00:00:00 --- log: started forth/04.11.04 00:02:14 well i'm going to bed. nite 00:18:36 --- join: warpzero (~warpzero@dsl.103.mt.onewest.net) joined #forth 00:29:34 IdleRPG@freenode: qFox was accepted into Pi Beta Phi! 00:29:36 ^^ 00:59:36 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 01:10:57 --- quit: mur (Read error: 110 (Connection timed out)) 01:28:17 --- join: dostoyevsky (sck@buug.mind.de) joined #forth 01:34:42 --- part: dostoyevsky left #forth 01:34:55 --- join: dostoyevsky (sck@buug.mind.de) joined #forth 01:42:39 --- join: mur (~mur@kyberias.uiah.fi) joined #forth 01:54:02 --- quit: mur_ (Read error: 110 (Connection timed out)) 01:57:38 --- quit: warpzero ("Tried to warn you about Chino and Daddy Gee, but I can't seem to get to you through the U.S. Mail.") 02:09:23 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth 02:14:51 --- join: allemann454 (~allemann4@pD9E1FF39.dip.t-dialin.net) joined #forth 02:17:16 --- part: allemann454 left #forth 02:20:32 --- quit: mur (Read error: 110 (Connection timed out)) 02:58:18 --- join: wincent (~wincent@void-109.pmnet.uni-oldenburg.de) joined #forth 03:22:02 --- join: warpzero (~warpzero@dsl.103.mt.onewest.net) joined #forth 03:44:38 --- quit: mur_ (Remote closed the connection) 03:45:13 --- join: mur (~mur@mgw2.uiah.fi) joined #forth 04:03:22 --- join: crc (crc@37-pool1.ras11.nynyc-t.alerondial.net) joined #forth 04:27:04 --- quit: Serg[ICQ] ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") 05:02:51 --- quit: fridge (Remote closed the connection) 05:20:00 --- quit: crc ("Time for bed... Goodnight!") 05:21:13 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth 05:31:32 --- quit: mur (Read error: 110 (Connection timed out)) 05:31:45 --- nick: Klaw` -> Klaw 05:44:35 --- join: fridge (~fridge@dsl-220-253-64-46.NSW.netspace.net.au) joined #forth 05:53:02 --- join: Serg_penguin (~z@212.34.52.140) joined #forth 06:07:13 --- quit: Klaw (Read error: 110 (Connection timed out)) 06:12:47 --- quit: hefner ("foo") 06:26:46 --- join: Klaw (~anonymous@c-67-171-207-91.client.comcast.net) joined #forth 07:03:04 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 07:03:05 --- mode: ChanServ set +o Herkamire 07:16:47 --- join: Robert (~pink@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 07:23:19 hi 07:24:06 Hi 07:24:25 Did you invite a writer from over there? ;) 07:29:32 ??? 07:29:42 Robert: i can't grok what u say ;))) 07:33:06 bye, time 2 go home ! 07:33:10 --- quit: Serg_penguin () 07:50:11 --- quit: I440r ("Leaving") 11:06:02 Robert! back? :p 11:06:37 i optimized a few things in kuf. its running at about double the speed i think :P 11:07:09 Hi 11:07:14 That's cool. :) 11:07:50 I worked a little on G, wrote a really nice FASM macro to do compile-time hash table creation. 11:08:01 Now I'm thinking about how to design the dictionaries. 11:08:12 during optimization i did screw up for, i thought it was >r 1 r> ,so i just changed it to 1 swap. took me about a few hours to figure out it was >r 1 >r instead :p 11:08:28 and last time i booted it had a fubar constant, but thats way at the end so easily fixed, i think :p 11:08:50 i hate hashtables. make it hard to debug 11:09:10 Oh, how? 11:11:08 Also thinking about how to store source code and comments. I can't see how people take decompilers seriously. 11:11:13 i could never decipher your hashtables back 11:11:20 (didnt you use them in F as well?) 11:11:26 Yep, used them in F. 11:40:54 --- join: ows (~ows@a81-84-114-211.netcabo.pt) joined #forth 11:43:05 i dont remember details, but i do remember i bumped into them a few times in the beginnen, and hating them every time 11:46:13 Too bad, qFox. 11:46:46 You'd better get used to them if you want better performance than Quake III on a 386. 11:47:20 >.< 11:52:19 personally, i'd be pretty impressed if you got q3 to run on a 386 11:52:40 that would mean you installed some kind of windows (and not 3.11) on it 11:52:49 or at least emulated it, which is twice as bad and slow 11:52:55 and THEN were able to boot up q3 11:53:02 yep. impressive. just like my forth 11:53:02 ;) 11:53:04 hashtables for dictionary lookup? 11:53:12 yeh 11:53:53 but dictionary lookup doesnt need to be speedy, does it? 11:54:03 Sure it does. 11:54:08 i mean, tick type of words are usually only ran in compilation mode 11:54:11 Depends. 11:54:23 Yeah, but you do want a fast compiler. 11:54:24 There's EVALUATE 11:54:42 imo, compilation executed words, dont really REQUIRE to be super optimized, sure its nice and all, but if it works, its fine 11:55:10 Guess it just depends on how much you have to compile. 11:55:16 Exactly. 11:55:18 Or how often. 11:55:26 :p 11:55:44 you know damn well what i mean. 11:55:46 But I can't see any reason to make it slower than necessary. 11:56:05 qFox: Really, I used non-hashed dictionaries and a rather slow compiler in tforth. 11:56:11 i dont see how making it complicated by hashtablelookup makes it worth that 11:56:17 qFox: It takes a while to compile on the 386 11:56:30 oh i'm sure it does. 11:56:40 but i dont use a 386 as a test machine 11:57:26 as long as i cant code enough asm to be able to code my compiler in asm, i dont think there's a real point to doing things very speedy 11:57:34 for me. 11:57:39 Ah, right. 11:57:46 I'm talking about MY code 11:57:51 i wasnt :p 11:57:56 well ok i was 11:58:04 And I'm, as you very well know, a stupid asmhead far, far from reality 11:58:09 when i said i noticed the hashtables and hit some walls because of them 11:58:14 but i broke them all! 11:58:15 :p 12:23:04 Robert 12:23:13 i created a variable in F like: 0 var fox-state 12:23:40 now i have a word that does something like this> : word [ ' fox-state 5 + ] ... ; 12:23:49 this word works fine for base 12:23:53 but not for state 12:24:01 might thee know why? 12:24:14 Hmm... 12:24:14 its exactly the same, except for the variables 12:25:24 its the reason why constant (and create and probably a whole bunch of other words...) is fubar :) 12:25:38 when the word gets executed, the whole thing hangs 12:25:56 oh 12:25:58 bah 12:26:00 No idea. :) 12:26:03 nevermind, its not exactly the same 12:26:07 : fox-[state] [ ' fox-state 5 + ] fox-unlit ; 12:26:07 : fox-(base) [ ' base 5 + ] literal fox-unlit ; 12:26:27 Hehehe :) 12:26:48 I was just going to ask what "..." was, but decided not to in order to avoid insulting your intelligence. ;= 12:26:53 :p 12:27:12 well, fox-unlit moves the TOS of F stack to the top of kuF stack 12:28:53 Mmmm... dirty hacks. 12:29:00 actually not a hack 12:29:25 just _the_ way to work with the kuF stack in F 12:29:33 its how core words get their value 12:29:53 : fox-@ ( adr -- n ) fox-lit @ fox-unlit ; 12:29:53 : fox-! ( n adr -- ) fox-lit2 ! ; 12:30:34 I mean, F is ugly as it is, without kuF on top of it... Together they're like a naked RMS...with a goatse picture printed on his ass...fucking a goose. 12:30:42 hehe 12:31:02 its not my fault F is coded so... well just so 12:31:03 :p 12:31:32 ;D 12:31:33 Nah. 12:31:37 I'm RMS and you're the goose. 12:31:49 ... i resent that analogy :\ 12:31:55 XD 12:32:00 Want to switch position? 12:32:07 neither. 12:32:11 Awww. 12:32:36 need to bugtraq my next bug 12:32:36 :p 12:32:49 ;-D 12:32:59 kuf is going haywire while loading 12:33:46 when its done, and i can finally see my statusbar at the new optimized speed, i'll show you it :) 12:33:55 its kinda ugly, and kinda sweet at the same time ;) 12:34:00 Hehe 12:34:19 but how do you create an editor for the same of anything? :\ 12:34:35 i keep coming back to the awefull fact that you have to keep copying data 12:34:42 for anything the user does.. 12:34:52 especially stuff like delete and inserting data 12:35:00 isnt that just hell to do? and sloooow? :( 12:35:06 or am i missing some hack? 12:35:35 You mean in the Editor? 12:35:44 in a... 12:35:50 You can, for example, keep a list of lines. 12:36:03 yeah thought about that 12:36:14 but then there's still the copying part 12:36:22 linked list of crlf'ed lines 12:36:25 (right?) 12:36:43 crlf is for CP/M users. 12:36:49 But yeah, something like that. 12:36:50 or whatever. 12:36:53 but still 12:37:00 when somebody edits the 2nd char on a 200 char line 12:37:07 Heh. 12:37:16 you still have to move the 198 other chars one ahead, dont you? 12:37:19 That's just 50 words, shouldn't take long to copy. 12:37:21 --- quit: ows (Success) 12:37:22 and fix the gui... 12:37:25 Better than 64kB 12:37:38 true 12:37:46 but its the only way? 12:38:07 or rather, its _the_ way? 12:38:23 i'm sure there are other, more unconvential ways to do it, but ... 12:38:24 You COULD keep a list of all characters, but that would be very stupid. 12:38:33 like i said... 12:38:34 90% of the time you're adding stuff to the end of the line anyway. 12:38:53 stupid, and inefficient, and slow loading 12:38:59 and hell with large files :\ 12:39:10 Not sure about how well it would work with a list of lines. 12:40:09 think i found my point of error. 12:41:01 * Robert needs to write a better memory allocator. 12:41:20 eh. sometimes i wonder why the fuck these errors keep causing a infinite loop of printing stuff :s 12:41:26 this time its literal 12:41:35 going to do a wild guess at stackfuckup 12:41:44 : penposx++ ( -- ) [ ' penposx >dfa @ ] literal inc ; 12:44:32 Looks like something that would fuck up, yes. :) 12:44:41 ... why 12:44:46 it worked before 12:44:57 ' penposx should return an address, or error 12:45:06 >dfa should keep an item on the stack 12:45:08 @ as well 12:45:10 so... 12:47:38 it appears that ' is .... malfunctioning 12:47:39 :p 12:47:58 Hehe, I remember having a lot of trouble with that 12:48:11 Especially when compiling, interpreting, postponing etc.. 12:48:43 its the first time actually 12:48:57 and i blame assumption 12:48:59 :p 12:49:32 i asumed that my in kuF defined word (word) did the same as the one i defined for kuF in F 12:49:39 but i guess its not... 12:50:31 Heh :) 12:54:46 ok.. debugging is going to be kinda hard 12:54:59 ' gets defined before dup and over do ;) 12:55:17 (at first it uses a word coded in F) 12:55:38 or well.. i'm pretty sure i coded a tick in kuF somewhere : 12:55:38 \ 12:59:29 Hihi 13:01:07 ehm. it seems as if .... [ doesnt change the compilation mode to interpretation?? :| 13:01:12 wotta 13:01:21 XD 13:03:03 yep. its simply compiling those words 13:03:06 grrr 13:17:41 huh.... something is very odd now :p 13:17:57 Aww, poor fox. What now? 13:20:15 more debug frustration 13:20:15 :p 13:20:27 but i am a little clueless about this one :\ 13:21:21 :-/ 13:26:40 --- join: Raystm2 (Rastm2@AC9BA7D9.ipt.aol.com) joined #forth 13:27:13 Hi Ray 13:27:33 Hi Robert : You are well today? 13:27:44 A colorforth question: does it have namespaces/user dictionaries? 13:27:49 Yeah, I'm fine 13:28:01 Over there, OK? 13:29:24 Robert if mem serves me well there is the one two-part dictionary with room for 256 macros and 512 defs. 13:29:30 two part because 13:29:41 only 512? 13:29:43 :\ 13:29:47 one part is a jumptable of names 13:29:48 oh per dict you mean? 13:30:02 (one dictionary total ) 13:30:08 right 13:30:36 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 13:30:39 and those names are the same distance into the jumptable of code. 13:31:13 when you call a name it figures out how deep it is into the code dictionary and executes that address 13:31:25 does that make sence. 13:31:59 qFox - remember that the idea was to keep track of the signals in chipchucks chips 13:32:22 he was writing sourceless code and started to loose trac 13:32:33 hence colorforth was borne 13:33:02 but others are trying to write colorforth in colorforth and such limits will be naturaly over come 13:33:25 as too are wellness, we are fine :) 13:33:31 Ah, OK. 13:33:45 I was just thinking a little about how to organize this, and needed some inspiration. 13:33:48 Thanks 13:34:03 hope I helped. would you care to elaborte? 13:34:13 elaborate even :) 13:34:21 hm. when i type state 5x it's going haywire. interesting :p 13:34:50 actually... when i type 5 5x it does too 13:34:52 I don't think I'll use the colorforth model, but instead have a stack of dictionaries that I'll search through. New words are added to the top of it. 13:35:09 this sounds like a stackmodelfuckup somewhere now ... 13:35:14 That is, to the dictionary on top 13:36:37 Robert : for information hideing? 13:36:38 qFox: 5x? 13:37:00 Raystm2: Yeah... That and (I hope) some compile-time speedup. 13:37:00 yessir 13:37:01 qFox: Ive missed the context. 13:37:14 oh. bugs in my forth after a major optimization attempt :p 13:37:52 --- quit: swsch_ ("Leaving") 13:37:52 qFox: thanx -- understood now -- thought you ment 5x def in colorforth there for a minute :) 13:38:14 nope 13:38:37 Robert: colorforth uses 'mark' and 'macro' for information hideing. 13:39:37 I believe that the reason colorforth is limited in dictionary size is because of it's position in the image. 13:39:45 hm. isnt dsp supposed to move around when stacks increase? :p 13:40:07 Sean Pringle solved this by giving it its one space on the other side of a memory boundary somehow. 13:40:57 sort of breaking the rules of protection mode. 13:42:32 I'm sure your correct qFox: as I tend to know more about colorforth then forth itself. 13:42:44 Hm, OK. I read about colorforth's dictionary model (well, some of it) today. 13:42:56 I really don't know either well enough :) 13:43:03 oh no, my comment was aimed at my dsp not moving... :) 13:43:13 ah :) 13:43:32 i've tracked the bug down to a different type of bug now 13:43:35 its not 5 stack items 13:43:38 its 5... things 13:44:01 Robert : info is growing ( slowly ) best to search throu the mailing list with google sometimes. 13:44:04 oh and dsp is not supposed to move, dsp@ is ;) 13:44:49 thanx for clearification qFox :) 13:45:05 hrm. but. damnit. what the hell is this bug 13:48:07 I'm implementing support for objects into the Forth compiler, I bet some of the puritans are going to hang me for that. :) 13:50:43 >_< 13:50:45 :p 13:50:56 Robert should be pretty cool , OOP and functional have there uses and are amazing together I'm sure. 13:51:43 --- join: Sonarman (~matt@adsl-64-169-94-169.dsl.snfc21.pacbell.net) joined #forth 13:51:46 ok at this point i'm just wondering whether i'm not overstepping some boundry of some kind 13:51:49 Hi Sonarman 13:51:52 someone asked me yesterday - mapping programs or listing programs - I have to answer both.:) 13:51:57 hi Sonarman :) 13:52:05 Raystm2: What do you mean? 13:52:09 hi 13:52:32 like ( getting two examples please hold) .............. 13:53:05 OK, sure 13:53:25 maybe i should trash it all and start over 13:53:26 :p 13:53:35 mapping http://freemind.sourceforge.net/wiki/index.php/Main_Page 13:53:41 * Robert checks 13:53:43 or just go back to my previous version actually 13:54:16 javascript error ;) 13:54:18 qFox: I'll hurry on with G so that you can port it. >:) 13:55:20 haha 13:55:27 yeah perhaps thats best 13:55:47 Raystm2: Ah, I see 13:55:55 listing or out lineing http://john.redmood.com/organizers.html 13:56:24 these are taken to be opposites yet both are valid together -- like oop and function :) 13:56:39 Python is both 13:56:50 some would not agree :) 13:57:00 from both camps 13:57:17 lisp is both too . 13:57:29 And forth is as usual a little bit of everything, with some people bitching about how unforthish it is. ;) 13:57:33 going to ditch this version and back to a previous one and do things step by step 13:57:35 ... tomorrow 13:57:44 qFox: :) 13:57:52 qFox : ack :) 13:57:52 I'm going to see if I can get the compiler to work. 13:57:59 all i did was optimize anyways 13:58:05 Will probably take a little while. 13:58:13 Then I'll have to write all system drivers in Forth. 13:58:22 Which will be pretty interesting I think 13:58:28 i just need the bare system i'm using in F 13:58:40 and bios :\ 13:58:42 qFox: Good news then, G will be much more minimalistic. 13:58:50 The core that is., 13:59:01 oh G may be bloated, as long as i dont need to load unused words ;) 13:59:08 I often wondered why there is no forth driver to drive all linux drivers out of the linux box = am I just neive ? 13:59:28 all i really need though is a way to edit my own source into the boot, like with F 13:59:39 ok, preferably easier then that, but hey ;) 13:59:49 Raystm2: What do you mean, that people should rewrite Linux drivers in Forth? 14:00:14 wish i could get bochs to boot from the image or a real floppy :( 14:00:21 well -- until then -- I mean why is there no forth layer ontop of the work already done i c? 14:00:31 i = in 14:00:33 qFox: Oh, any idea why that doesn't work? 14:00:40 because bochs hates me 14:00:51 Raystm2: What would the Forth layer do? 14:00:56 qFox: :/ 14:01:13 i dont think it boots from a physical floppy 14:01:21 and it for some reason wouldnt accept the bin as an image 14:01:25 so i dont know 14:01:53 ohwell. going to get some sleep. later 14:02:40 --- quit: qFox ("this quit is sponsored by somebody!") 14:02:57 Robert : the forth layer would drive drivers already written in c for linux as a crutch -- i guess -- since the machine seems to be built for the c lang ultimaltely now that things are going in circles in development :) 14:03:03 Night :) 14:03:03 Hmm 14:03:03 I wonder if the F boot loader lacks the boot signature. 14:03:15 night Robert :) 14:03:26 Hm, nope. 14:03:32 Raystm2: Ah, no, that was to qFox 14:03:37 I won't go to bed yet 14:03:39 okay 14:04:32 I think using a Forth layer for the C drivers would complicate matters. I used to do a lot of stuff like that in the old system, that's why it's not very good. Too little integration 14:06:15 okay == but for the uninitiated -- myself -- I could benifit -- yet I see how completely complicated such a system would be, in forth were supposed to learn our hardware :) 14:07:32 I guess that's what gforth is after all -- right? 14:11:09 You mean a ton of forth definitions on top of another OS kernel? :) I guess so 14:11:56 yeah 14:12:18 so it's been done to death there :) 14:12:32 and I asked and received :) 14:26:33 i found what I was looking for. Coroutines in Forth: http://home.hccnet.nl/a.w.m.van.der.horst/forthlecture6.html 14:26:56 oh cool hi OrngeTide. 14:29:22 i wish i could come up with a better way to organize links on my page. if anyone has some ideas please share 14:29:39 if your familiar with the folks in #c4th-ot -- I learned recently that van der Horst lives very close to Tim Nietz - they just learned this themselves after years of contact on lists. and now they visit one another too :) 14:29:52 neat 14:30:00 --- quit: tathi ("leaving") 14:38:21 --- join: madwork__ (~madgarden@derby.metrics.com) joined #forth 14:38:21 --- quit: madwork_ (Read error: 104 (Connection reset by peer)) 14:45:04 --- quit: wincent (Remote closed the connection) 14:47:48 --- join: madwork_ (~madgarden@derby.metrics.com) joined #forth 14:47:48 --- quit: madwork__ (Read error: 104 (Connection reset by peer)) 14:53:28 --- quit: Raystm2 ("User pushed the X - because it's Xtra, baby") 15:26:54 --- join: madwork__ (~madgarden@derby.metrics.com) joined #forth 15:26:58 --- quit: madwork_ (Read error: 104 (Connection reset by peer)) 15:30:23 --- join: madwork_ (~madgarden@derby.metrics.com) joined #forth 15:30:23 --- quit: madwork__ (Read error: 104 (Connection reset by peer)) 15:49:16 --- join: crc (crc@241-pool1.ras11.nynyc-t.alerondial.net) joined #forth 15:58:18 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 15:58:24 Hi 15:59:11 Hi Robert 16:05:16 --- join: crc_ (crc@74-pool2.ras11.nynyc-t.alerondial.net) joined #forth 16:05:17 --- quit: crc (Read error: 104 (Connection reset by peer)) 16:05:31 --- nick: crc_ -> crc 16:30:24 --- quit: tathi ("leaving") 16:56:15 heh... last drive in the ailing raid failed.... Uh, my DNS was through that box... Now no DNS.. Just IRC :) 16:57:35 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 17:12:41 --- quit: crc ("Time for bed... Goodnight!") 17:18:20 --- join: hefner (~hefner@pool-141-157-7-146.balt.east.verizon.net) joined #forth 17:26:58 --- quit: cmeme (No route to host) 17:28:16 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 17:29:10 --- join: Hyrax__ (~das@64.238.237.215) joined #forth 17:39:06 --- quit: Hyrax_ (Read error: 104 (Connection reset by peer)) 17:39:14 Hyrax__, for dns you could use http://www.everydns.net/ .. it's free. I use it as a secondary for all my domains. 17:43:11 --- quit: tathi ("bed") 17:49:23 --- join: Hyrax___ (~das@adsl-64-219-100-33.dsl.lgvwtx.swbell.net) joined #forth 17:49:28 --- nick: Hyrax___ -> Hyrax 17:49:38 --- quit: OrngeTide ("going home") 17:52:00 --- quit: Hyrax (Client Quit) 17:52:46 --- join: Hyrax (~das@adsl-64-219-100-33.dsl.lgvwtx.swbell.net) joined #forth 17:57:13 --- quit: Hyrax__ (Read error: 110 (Connection timed out)) 20:46:29 --- quit: warpzero ("Tried to warn you about Chino and Daddy Gee, but I can't seem to get to you through the U.S. Mail.") 21:07:07 --- quit: hefner ("wub wub wub wub wub wub wub wub") 21:53:59 --- join: wandelf (~Olorin@h125.51.40.69.ip.alltel.net) joined #forth 21:54:09 Dwight, are you there? 21:56:40 he was a min a go 22:00:41 maybe he went to bed 22:08:35 --- quit: Sonarman ("leaving") 22:36:37 --- join: Serg_penguin (~z@212.34.52.140) joined #forth 22:44:25 --- quit: cmeme (Broken pipe) 22:45:30 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 23:07:47 --- join: Raystm2 (Rastm2@AC97BE18.ipt.aol.com) joined #forth 23:08:31 Robert : looks like I was wrong about 1 dictionary in colorforth. seems that macro and forth are separate dictionarys :) 23:08:48 --- quit: skylan (Read error: 104 (Connection reset by peer)) 23:13:10 --- join: skylan (~sjh@vickesh01-4739.tbaytel.net) joined #forth 23:28:00 --- quit: Herkamire ("bed") 23:44:03 --- part: wandelf left #forth 23:59:59 --- log: ended forth/04.11.04