00:00:00 --- log: started forth/04.10.22 00:04:09 --- join: swsch (~stefan@p5091FEC0.dip.t-dialin.net) joined #forth 00:46:37 --- quit: ez4 ("Quitting!") 01:00:21 --- quit: crc ("I was using TinyIRC! Visit http://www.tinyirc.net/ for more information.") 01:04:21 * swsch is away: I'm busy 01:06:32 --- quit: warp0b00 (Read error: 54 (Connection reset by peer)) 01:14:02 --- join: crc (crc@49-pool1.ras11.nynyc-t.alerondial.net) joined #forth 01:16:39 * swsch is back (gone 00:12:18) 01:27:24 * swsch is away: found it, off with me! 02:08:00 --- quit: crc ("Time for bed... Goodnight!") 03:12:32 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth 03:24:02 --- quit: mur (Read error: 238 (Connection timed out)) 03:31:33 --- join: warp0x00 (~warpzero@mi053.dn176.umontana.edu) joined #forth 03:50:00 --- join: Baughn (~svein@sveinoa.stud.cs.uit.no) joined #forth 04:28:13 --- join: mur (~mur@smtp.uiah.fi) joined #forth 04:39:42 --- quit: mur_ (Read error: 110 (Connection timed out)) 04:48:46 --- join: wossname (~wossname@rn-v1w5a06.uwaterloo.ca) joined #forth 04:50:11 --- quit: ok (Read error: 242 (No route to host)) 05:30:19 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth 05:40:53 --- quit: mur (Read error: 110 (Connection timed out)) 05:41:30 --- quit: warp0x00 (Read error: 238 (Connection timed out)) 06:01:53 --- join: fudgez (~chatzilla@ipuser70.megapipe.net) joined #forth 06:02:16 can anybody recommend me an _easy_ no-frills forth? 06:10:24 --- join: wincent (~wincent@void-109.pmnet.uni-oldenburg.de) joined #forth 06:32:29 fudgez well.... 06:32:39 i would always recommend my own forht because im always here to heop 06:32:41 help 06:32:45 --- quit: Baughn ("Lost terminal") 06:32:48 tho - its not a minimalists forth 06:33:01 I am running win32forth and it look GRRRREAT, just a little old GUI (win95) 06:33:03 it IS very well commented and im starting to get with the documentation too 06:33:13 fudgez, do you have a linux box? 06:33:22 not at work 06:33:23 I440r: Yeah, the main reason to use isForth is that you're an IRC junkie ;) 06:33:35 and because i comment my code 06:33:40 dammit :P 06:34:12 fudgez: I made an isForth-inspired (OK, OK... basically a port but with my own bastardizations) DOS Forth. 06:34:33 In case there's no possibility to run a Linux one 06:34:45 Robert: I really need something very small, ansi compliant, and with helpful error messages 06:34:52 This isn't ANS 06:34:56 But it's small. 06:35:03 I think the kernel is ~3-4kB 06:35:13 robert i also ported isforth to dos :) 06:35:15 Depending on if you're compiling with the graphics library or not. 06:35:21 I440r: Oh, really? 06:35:28 unreleased. for my own use 06:35:38 Cool :) 06:35:42 16-bit? 06:35:45 I have "thinking in forth", I bought it 6 years ago second hand and never bothered to read it until I found it last night. What forth is most compliant with the book? 06:36:54 fudgez: The question is if you want something that's completely compilant. The hard part about Forth (writing good code) is not too implementation-critical. 06:37:20 fudgez: I used the pforth (among other things) to learn Forth, even though I never used pforth. 06:37:22 win32forth is probably the best environment for windows, when looking at gui and windows-programming-possibilities 06:37:29 it is, however, very bloated 06:37:33 Hehe. 06:37:54 oh, i should say best _free_ environment 06:38:03 i havent been able to get my hands on a copy of swift 06:38:09 or winforth 06:38:20 (winforth != win32forth) 06:38:43 I programmed in a forth eons ago (8 years ago?) where I had to load files by specifying the disk cylinder, sector and track! 06:39:03 i started.... robert how long am i programming forth now? :p 06:39:10 No idea. 06:39:22 i guess about half a year 06:39:27 it didn't even have a file system. You had to know the starting offset, and the size of the "byte chunk" you wanted to read. 06:39:31 qFox: I know how long I've been doing it by grepping IRC logs. 06:39:51 fudgez: A lot of forthers prefer that. 06:39:59 fudgez: Note that this isn't a SANE community. :) 06:40:07 same. i started when i joined, which i think was about half a year ago 06:40:10 or longer, dunno 06:40:25 sane? hu? 06:40:26 Just half a year? 06:40:28 Feels longer. 06:40:30 seems longer 06:40:31 i know 06:40:50 well, check ya logs for the first time i joined. that should be it :) 06:41:59 it might have been a year though 06:42:15 but i dont think i was working with forth at the time i was developing my msn client 06:42:25 which i did march/april this year 06:42:45 --- join: proteusguy (~proteusgu@24-159-109-218.cpe.ga.charter.com) joined #forth 06:45:50 I need to recover my IRC logs for the last year 06:45:57 One block of the bz2 got corrupted. :( 06:46:10 I need to find a way to extract the data from the remaining blocks. 06:48:18 atm i'm passively trying to figure out how i would program parallel 06:48:36 You mean parallell processor systems? 06:48:42 yep 06:49:00 So what have you found out? :) 06:49:08 but a way to efficiently program using parallel processors 06:49:17 nothing yet, its much harder then it seems 06:49:21 like i said, passively 06:49:23 I've been thinking about how impressive clusters you could build from Forth CPUs. 06:49:23 :) 06:49:34 is gforth any good? I know GNU languages (except C and C++) are usually lacking behind others. 06:49:35 Depends on the task I guess. 06:49:58 fudgez: gforth is big and bloated, but runs everywhere. 06:50:11 how about bigforth 06:50:14 fudgez: I don't use it, so I can't say more than that about it, 06:50:16 Never tried., 06:51:11 Robert: what lean ANSish forth would you recommend for win32/DOS? something preferably console based (no GUI) and has history 06:51:42 fudgez: For ANS I think I'd recommend gforth. 06:52:06 Basically because it's GNU (i.e. Good[TM]) and the only ANS environment I know of. 06:52:23 okay then, screw ANS :-) 06:52:45 :D 06:52:57 What will you use Forth for? 06:52:59 i kinda wish win32forth had a DOS console interface 06:53:11 I mean, sometimes ANS could be useful, but often it's not. 06:53:15 just to give you an analogy; I don't want the "Visual C++" of forth, give me the Turbo C. 06:53:21 I440r why? 06:53:27 Hehe! 06:53:28 Robert: right now just learning it for its own good 06:53:32 i'll take visual forth :( 06:53:38 because windows is a pile of (&876 06:53:46 dos is better than windows 06:53:48 fudgez: OK, then I'd really recommend using a tiny Forth you can understand completely. 06:53:55 fudgez: Do you know assembly language? 06:54:00 whether win32forth has a commandline interface, or a gui.... it works doesnt it... 06:54:02 ibm pc-dos 7 is better than windows-ANYTHING 06:54:07 plus win32forths gui is coded entirely in forth 06:54:23 Robert: for 3 real processors, and MIX 06:54:27 or wait, at least the editor is, i am not quite sure about the main interface, but i believe that too 06:54:31 the windows console looks like shit 06:54:32 period 06:54:39 but thats windows fault too 06:54:43 fudgez: Then I think you should consider learning Forth from that point of view. 06:54:55 fudgez: Which is what I did, and i have the impression it helped. 06:54:57 tinyForth then 06:55:18 fudgez: I started out by writing a Forth in assembly language. 06:55:31 fudgez: That way you learn the language AND the internals. 06:55:44 alright 06:55:54 Robert> oh i did figure something out. that i was mixing up two types of parallel processing ways 06:56:02 fudgez: Moving Forth and Build Your Own Forth both desribe this process. 06:56:08 describe* 06:56:10 I have implemented several tiny toy Lisps already, and one painful pascal 06:56:15 Hehe. :) 06:56:19 you cant be a good forth coder without knowledge of the internals of the language - the good thing is that forth ENCOURAGES you to know it :) 06:56:21 fudgez: A Forth is MUCH easier. 06:56:36 fudgez: Currently I'm implementing a 510 byte one for a contest. 06:56:43 woah 06:56:49 fudgez: And in <4kB you can fit a good one. 06:58:48 I'll see how small I can get this one... Will of course upload once it's done. 06:59:36 Robert> there's the parallel type where the processor is a static type of action/function, and the one where every processor is indeed a processor with variable functions 07:00:21 Okay.. what are the former used for? 07:00:43 well, neural networks use static function type 07:01:00 --- quit: tgunr1 ("Leaving") 07:01:25 and i'm currently being flooded with neural network stuff in my classes so it gets a bit confusing 07:01:32 Hehe 07:01:56 I read about that Forth CPU using 20mW when doing 2400 MIPS. 07:02:27 well i've so far concluded that parallel programming requires a completely different way of thinking 07:02:28 If you can afford 2kW you would get 24 TIPS. ;) 07:02:51 hehe 07:03:05 Or, wait, 240 07:03:14 2kWs? 07:03:25 No, 240 TIPS 07:03:32 yes but the 2kW 07:03:36 is 2kWs? 07:03:39 when running i mean 07:04:11 kWs? As in kJ? 07:04:20 kilowat per second 07:04:22 like kwh 07:04:31 you pay electricity in kWh 07:04:41 That's a measurement of energy. 07:04:45 yes... 07:04:49 :\ 07:04:57 And "kilowatt per second" would be the change of energy consumption per second 07:05:11 kWs = kJ, kWh = 3.6 MJ 07:05:23 ok you know what, nevermind 07:05:23 And I meant that its consumption would be 2kW 07:05:24 :p 07:05:43 Like a microwave or something. 07:05:53 oh but that aint much, is it.. 07:06:03 Not THAT much, no. 07:06:08 For a supercomputer. :) 07:06:13 aye 07:06:14 The problem is that the processors run on <2V 07:06:25 2kW / 2V = 1kA :D 07:06:33 THICK wires. ,) 07:06:35 well, you're redesigning a cpu anyhow so i dont see the problem there ;) 07:07:03 but anyways, the thing about parallel cpu's would be that they are weaker but more 07:07:25 anyways, imagine how you would code it 07:07:31 some things can be done parallel 07:07:41 but others simply have to be done in serial fasion :\ 07:08:08 initialization of variables can be done parallel, but looping for instance has to be processed serial 07:08:46 so i was thinking of a structure where there would be a "big" processor, and 2 smaller ones 07:09:03 and the big one would be the main one 07:09:10 and the other two were support cpu's 07:09:21 that you can call uppon from inside the code 07:09:49 (no i obviously never designed a cpu, or looked at the schematics of one ;) 07:10:05 :D 07:10:10 We love you anyway. 07:10:33 it would be great though 07:10:40 The algorithms sure become more interesting either way 07:11:00 complicated is the keyword there ;) 07:11:13 but whether the whole thing would be faster, i dont know 07:11:32 or useable 07:11:33 :p 07:11:54 Well... do you have any particular applications in mind? 07:11:59 nope 07:12:07 The obvious thing would of course be breaking ciphers.. 07:12:11 But that's pretty boring. 07:12:22 and so not my area of intrest anyways :p 07:12:51 alternatively 07:12:55 and probably easier to use 07:13:15 you could create a cpu that processed a execution token at once 07:13:25 (i think intel already has one of these btw) 07:13:44 and by that i mean that it takes an execution token, and processes all the opcodes at once 07:13:49 instead of left>right or right>left 07:14:17 and that the coder could choose to either fill in the position, or leave it empty (nop, w/e) 07:14:46 w/e ? 07:14:48 whatever 07:14:52 Ah 07:14:57 know thy slang. 07:15:01 Pffft 07:15:02 or mine at least ;) 07:15:39 Pretty interesting technique. 07:16:13 but i have no specific application for this in mind, nor do i know whether it would be usefull at all 07:16:38 I can imagine that it would help a lot in graphics processing. 07:17:03 possibily. but isnt there already a fpu that does calculations independently and parallel to the main cpu? 07:17:28 on "modern" videocards i mean... 07:17:48 and if not, may i ask wtf they have big ass coolers for? ;) 07:18:06 :D 07:18:22 I meant that you could do a lot of calculations in parallell 07:18:25 GPUs are pretty specialized high-thruput processors 07:18:46 And that things aren't very serial. 07:19:10 they can't do much but what they can they do _fast_ - with recent developments, shaders etc. reprogrammable render pipeline, you can do general purpose stuff with them, too 07:19:23 there's a library somewhere to push code to gpus to do FFT and various other operations 07:20:11 right 07:20:41 i'm gonna ask my friend to see whether he can pull off such a parallel forth cpu 07:21:38 Heh, doing SETI research on your graphics card. 07:21:47 holly smoke, the SEE command is cool :-) 07:22:00 you never knew see...? :p 07:22:06 lol Robert 07:22:09 seti should die 07:22:14 long live connect4! :D 07:22:21 fudgez: You WILL change your opinion on that when implementing your own Forth. 07:22:39 dont listen to him fudgez, see rox 07:22:42 :) 07:22:42 Robert: it is a disassembler ;-) 07:22:43 fudgez: Decompilation in Forth is much easier than with other languages, but still a pain in the ass. 07:23:04 Depending on how your Forth looks, of course. 07:23:09 it worked pretty nicely in my implementation actually 07:23:10 :) 07:23:20 hm why did i quit working on kuF again? 07:23:21 I wrote an i386 disassembler; if I had to look Appendix A again (the opcode table) I will vomit 07:23:35 oh speed issues 07:23:52 robert one of these days you will build me a kuF in asm ;) 07:24:16 Robert: if a disassembler is hard, you can cheat and store the original source-text somewhere ;-D 07:24:30 that way, you keep even comments, if you want 07:25:00 fudgez: Yeah, that's a better approach. 07:25:32 fudgez: Haha... NASM manual appendix A is one of the most used bookmarks. 07:25:40 Together with wikipedia and google. 07:25:42 my see word would simply read calls and do a name lookup 07:26:00 but it wasnt able to see the "end" of a word yet ;) 07:26:25 Hehe. 07:26:34 should not have been hard though 07:26:38 * Robert is in the process of planning another Forth system. 07:26:41 my end of a word was a 0 07:27:17 thing is that, from a decompiling point of view, you wouldnt immediately know whether it was a number as argument for another word, or simply the end of a word 07:27:30 (easily worked around though) 07:29:41 how did it work on runtime? 07:30:09 oh at runtime it would know 07:30:22 because if it encountered instruction 0, it would mean teh end of the word 07:30:52 but from a decompiling point of view, when looking up lit for instance, it doesnt actually run the word, so doesnt skip the argument 07:31:07 hence it cant know that the next number is indeed a number and not a instruction 07:31:20 (this would mean coding exceptions for these kinds of words) 07:31:48 yes 07:33:36 anyways, when comparing speed in a simple loop, my forth proved much slower then the forth underneath it 07:34:34 :D 07:34:39 partially due to me wanting to use as few primitives as possible ;) 07:34:58 I'll write an optimizing compiler and outperform myself! 07:35:04 :p 07:35:30 --- quit: fudgez (Read error: 232 (Connection reset by peer)) 07:37:16 i find optimization compilers for forth kinda leet 07:37:59 --- join: ez4 (~ez4@pcp01518726pcs.reding01.pa.comcast.net) joined #forth 07:38:09 Yeah. 07:38:18 I want to do some research of my own in that area. 07:51:08 --- quit: cmeme (Broken pipe) 07:52:27 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 08:29:36 --- join: ASau (~root@83.102.133.66) joined #forth 08:29:42 Dobryjj vecher! 08:35:28 --- join: TheBlueWizard (TheBlueWiz@modem-087.nyc-tc03a.fcc.net) joined #forth 08:35:28 --- mode: ChanServ set +o TheBlueWizard 08:36:57 God kvaell, TheBlueWizard! 08:37:05 Any news? 08:38:13 ASau hiya....a tiny bit of news for me personally, yeah, but nothing noteworthy to tell anyone about it 08:38:30 * TheBlueWizard wonders what "God kvaell" means 08:41:24 Swedish "good evening." 08:41:35 It looks Robert is off-line. 08:42:22 ah 08:42:35 I've made significant changes to my GNU SL related project but have not uploaded them. 08:43:03 SL? 08:43:20 Each time I recall I either want to make more changes or forget password. 08:43:25 Sci. library. 08:45:50 ah...hmm...I think there are numerous scientific libraries out there...so, what is distinctive about yours, if I may inquire? 08:46:24 --- nick: mur_ -> mur 08:46:45 Terve! 08:47:17 TheBlueWizard, I've started reworking random number generators. 08:47:37 http://retroforth.org/dev/asau IIRC. 08:48:13 mur terve :) 08:48:21 moi 08:58:43 ASau: Privet! 09:08:08 Privet, Robert! 09:08:24 Long time no see :) 09:09:02 I've got job. 09:09:34 And I had to pass exams in between. 09:10:00 gotta go....all bye 09:10:13 --- part: TheBlueWizard left #forth 09:32:07 --- join: OrngeTide (orange@rm-f.net) joined #forth 09:32:52 neat. my recruiter found me a job where knowledge of Forth is "recommended" 09:34:40 Congratulations! 09:34:52 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 09:34:52 --- mode: ChanServ set +o Herkamire 09:35:02 Dobryjj vecher, Herkamire! 09:36:02 'morning ASau 09:36:06 OrngeTide, have you got more info, what job it is? 09:36:17 Embedded? 09:51:49 --- join: ows (~ows@a81-84-114-35.netcabo.pt) joined #forth 09:52:34 Dobryjj vecher, ows! 10:03:43 --- quit: proteusguy ("The #python split is remarkably silly...") 10:13:02 ASau, solaris kernel developer. :) 10:13:25 i'm guessing I have to do a lot of hacking in OF and Fcode 10:15:48 Hm. That's promising. 10:16:03 If you're Forth hacker. 10:16:10 OrngeTide: hmm.. sounds interesting 10:17:33 yea. it's more interesting than what I'm doing right now. (QA for some linux kernel modules) 10:18:23 OrngeTide: well, I'm saying that as openfirmware implementor 10:18:36 oh. at apple or sun? 10:18:45 or.. openbios? 10:18:48 openbios 10:18:50 :) 10:19:20 I'm trying to talk my company into using openbios. right now they are using AMI and have all kinds of problems and have to hire a bios developer anyways. they will probably use linuxbios though. 10:19:39 openbios uses linuxbios as lowlevel init - it's just with a different payload 10:19:52 since IWill (the maker of other motherboard) has expressed interest in supporting LinuxBIOS commecially. 10:20:03 actually Stepan (who's here, too) works on both linuxbios and openbios 10:20:09 oh. neat. 10:20:34 we contracted some linuxbios developer to get better support for the opterons we use. I don't remember who though. 10:21:52 seems like the SUN-resident OF guy was moved to a different department - I'm currently trying to figure out who's responsible for firmware now.. 10:25:04 hrm. I see. 10:25:48 one of the Apple OF guys did a little presentation at SVFIG a few months ago. that was cool. they use gforth to metacompile thier OF. 10:26:37 they do? 10:28:22 :) 10:33:25 Well, I've to go. 10:33:28 Bye! 10:34:03 --- quit: ASau ("leaving") 10:34:16 re 10:35:47 Hi :) 10:52:16 stepan! 11:02:22 --- join: Raystm2 (Rastm2@AC97DDAF.ipt.aol.com) joined #forth 11:23:01 Stepan, hi 11:25:22 Hi Ray, and others 11:36:44 --- join: tgunr (~davec@vsat-148-63-4-108.c001.g4.mrt.starband.net) joined #forth 11:38:07 Hello OrngeTide and Robert! Hoping you are both well today and every day! ;) 11:38:15 hi all 11:38:30 Except I slept until 1pm, missed the lecture, etc. I'm fine 11:38:33 Hi slava 11:41:01 "Dear me: I met a real live female human today. Unfortunately, since she was protected by a CRT, I couldn't quite touch her. So I touched myself instead. WoooooOOO" 11:43:05 Haha 11:43:11 IRC... :) 11:55:28 slava : Both hands on the keyboard :) 11:56:42 Anybody tried coding my c4thchess into there colorforths and found no pieces on the board? 11:57:20 for future ref : you have to make your own icons in Icons :)\ 11:58:03 This will change when the program is updated for the tutorial -- sorry for confusion :) 12:05:43 wife needs phone - brb 12:05:50 --- quit: Raystm2 ("User pushed the X - because it's Xtra, baby") 12:43:09 --- quit: swsch ("Leaving") 12:44:49 --- join: Serg[GPRS]_ (~z@193.201.231.126) joined #forth 12:45:08 HI ! 12:45:45 --- join: swsch (~stefan@p5091FEC0.dip.t-dialin.net) joined #forth 12:47:23 --- quit: wossname () 13:03:21 hey serg 13:04:07 hi 13:04:27 * Serg[GPRS]_ is deep in PHP auth/security trubbles 13:08:01 --- quit: Serg[GPRS]_ () 13:22:57 Robert 13:23:03 Yes? 13:23:20 about the parallel instruction thing 13:23:41 i think it might be doable, as far as being functional 13:24:12 "it"..= 13:24:13 ? 13:24:46 the cpu that takes an execution token and processes all the instructions at once 13:25:07 Ah, right. 13:25:25 What exactly is an execution token? 13:26:08 the data in the cpu that its executing 13:26:09 ehm 13:26:22 like it reads in a x bit number (depending on the width of that cpu) 13:26:41 but an instruction is only so many bits, so multiple instructions fit inside the execution token (word) 13:27:45 appearantly mmx uses a similar thing though :p 13:29:28 SIMD? 13:29:33 simd. 14:13:41 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 14:13:42 nite 14:13:44 --- quit: qFox ("this quit is sponsored by somebody!") 14:47:01 * swsch is away: I'm busy 15:10:23 --- join: wossname (~wossname@rn-v1w5a06.uwaterloo.ca) joined #forth 15:12:45 hi hi 15:13:22 what do you want to forth today? 15:13:32 WHAT DO YOU WANT TO FORTH TODAY? :p 15:13:38 lets think of a coding challenge ;) 15:15:35 yeah 15:15:37 good idea 15:15:56 wossname and I have a secret FORTH-RELATED PROJECT. 15:16:16 really? arke and ray have a secret forth project too 15:17:11 this channel is full of secrets 15:17:43 theres a secret forth channel too thats more active than this channel 15:17:46 kind of funny 15:18:08 how about a forth program to generate an ascii maze with these chars: +-| 15:18:24 + ? 15:18:24 all passages one space wide 15:18:46 |----------| 15:18:49 does it have to be a solvable maze? 15:19:02 which characters herk? 15:19:05 - and | ? 15:19:09 whats the + for ? 15:19:25 futhin, secret forth channel? 15:21:30 +-+ +-+ 15:21:30 | | 15:21:30 +-+ + + 15:21:30 | | | 15:21:30 + +-+-+ 15:21:32 | | 15:21:35 +-+ +-+ 15:21:40 ok 15:21:46 could just use * for all 15:22:14 *** *** 15:22:14 * * 15:22:14 *** * * 15:22:14 * * * 15:22:14 * ***** 15:22:16 * * 15:22:18 *** *** 15:22:20 rule 1) nobody is allowed to find algorithms off the net or in books etc ;) 15:22:26 lame 15:22:28 must figure out the entire algorithm yourself ;) 15:22:32 futhin: good rule :) 15:22:46 --- join: Sonarman (~matt@ppp-66-124-255-73.dsl.snfc21.pacbell.net) joined #forth 15:23:07 so who's committing to herkamire's challenge? 15:23:10 me 15:23:25 what challenge? 15:23:47 Sonarman: http://tunes.org/~nef/logs/forth/ :P 15:23:57 Sonarman: write a forth program from scratch to create a random ascii maze 15:24:07 no looking up algorithms 15:24:09 futhin: bah :D 15:24:17 Herkamire: cool, thanks 15:24:28 preferably with exactly one solution 15:24:38 Herkamire: how long do you figure it should take a guy to code this up? 15:24:41 whoo.. mazes:) 15:24:57 i wrote a bunch of mazes. but I found that hilbert curves were cooler looking and less work 15:25:35 ... it's C code.. :( http://orangetide.com/programming/weirdcode/hilbert.c 15:25:45 futhin: dunno. I'm not good at estimating times for programming, and I'm not sure what it'll take to do this 15:25:51 i think the challenges should be limited to less than an hour of coding.. combined with no looking up algorithms.. 15:26:02 --- join: paintcan (~wossname@rn-v1w5a06.uwaterloo.ca) joined #forth 15:26:06 i dunno. then you might just be lucky to remember the algorithm 15:26:17 like I remember the maze algo well enough to write one without looking. 15:26:33 hehe 15:26:36 (hint: think flattened out tree) 15:26:49 actually to be honets, i'd rather work on my zoomable UI:) 15:27:01 i'd rather work on making a forth widget. 15:27:06 lol 15:28:03 hrm. does anyone know of something just like procmail that works on IMAP? i want to have rules to sort my mail like I do in Mozilla. but I want to use pine/mutt instead. 15:28:17 did anyone participate in that functional coding contest in august? 15:28:28