00:00:00 --- log: started forth/04.08.26 00:00:16 --- join: mur_ (~mur@uiah.fi) joined #forth 00:01:36 Not sure, but it's close enough. 00:01:44 You get the idea though. 00:02:11 * kc5tja_ is thinking of writing a POP3 client in GForth that uses the Forth command line interface extensively in a manner not unlike an AX.25 BBS. 00:02:38 --- quit: mur (Read error: 60 (Operation timed out)) 00:03:10 Crap. I need to get to bed. 00:03:11 Night. 00:03:17 --- quit: kc5tja_ ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 00:57:58 --- quit: mur_ (Remote closed the connection) 00:58:29 --- join: mur (~mur@smtp.uiah.fi) joined #forth 02:00:39 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 02:02:41 --- quit: mur (Read error: 60 (Operation timed out)) 02:08:02 * Tomasu is away: dlrrp 02:08:06 --- nick: Tomasu -> TomasuDlrrp 02:50:00 --- join: Topaz (jonny@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 02:56:17 --- quit: mur_ (Remote closed the connection) 02:57:01 --- join: mur (~mur@mgw2.uiah.fi) joined #forth 03:53:14 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 04:03:54 --- quit: mur (Read error: 110 (Connection timed out)) 04:20:42 --- join: crc (crc@66-pool1.ras11.nynyc-t.alerondial.net) joined #forth 04:21:26 --- quit: mur_ (Remote closed the connection) 04:21:58 --- join: mur (~mur@smtp.uiah.fi) joined #forth 04:36:41 * crc is happy -- RetroForth is far more useful now :-) 04:44:30 Does anyone know an easy way to create zero terminated strings? 05:21:31 --- quit: crc (Client Quit) 06:13:10 --- join: mur_ (~mur@uiah.fi) joined #forth 06:21:28 only the obvious... 06:21:32 oh 06:24:02 --- quit: mur (Read error: 110 (Connection timed out)) 06:45:18 --- quit: I440r (Nick collision from services.) 06:45:21 --- nick: I440r_ -> I440r 06:45:29 --- join: I440r_ (~mark4@216-110-82-59.gen.twtelecom.net) joined #forth 06:45:29 --- mode: ChanServ set +o I440r 06:45:35 --- nick: I440r -> I44-r_wrk 06:46:01 dammit lilo let me chane my freekin nickname grrrr 06:46:09 asshole 06:53:25 hm? 07:06:53 what would you call a word that would take an address and fetch the 32 bit data at that address and returnd an updated address and the data ? 07:07:03 : blah dup 4+ swap @ ; 07:07:07 what would you call that ? 07:09:22 W/ 8 07:20:03 --- quit: mur_ (Remote closed the connection) 07:20:35 --- join: mur (~mur@mgw2.uiah.fi) joined #forth 07:32:59 --- join: fridge (~fridge@dsl-220-253-73-190.NSW.netspace.net.au) joined #forth 07:42:07 @cell+ 07:42:23 @+ 07:42:25 or something 07:55:14 I44-r_wrk: are you makikng fun of the g-d thing 07:57:42 no i couldnt change my nick too many times i a row 07:57:45 --- nick: I44-r_wrk -> I440r 07:57:54 so i couldnt correct the nick 07:58:13 ya gotta wait 20 seconds and if you wait 19 and retry you have to wait TWENTY seconds over again 07:59:11 oh 07:59:14 well bye 07:59:16 --- 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.") 08:10:55 --- quit: qFox (orwell.freenode.net irc.freenode.net) 08:10:55 --- quit: arke (orwell.freenode.net irc.freenode.net) 08:10:55 --- quit: cmeme (orwell.freenode.net irc.freenode.net) 08:12:34 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth 08:12:34 --- join: arke (~chris@wbar8.lax1-4-11-100-111.dsl-verizon.net) joined #forth 08:12:34 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 08:34:53 --- join: warpzero (~warpzero@150.131.180.0) joined #forth 08:42:55 --- quit: qFox (Read error: 104 (Connection reset by peer)) 08:45:14 kickass! 08:45:22 unpluggd the sparc hard drive 08:45:24 still works 08:45:24 :) 08:45:31 * arke tries to guess for how long 08:45:40 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth 08:48:30 im back 08:59:48 whats kc5tja's page? 08:59:56 and why'd he remove it from topic? 09:02:47 nm on the what 09:09:14 falvotech.com I think 09:11:46 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth 09:13:30 Things will be coming around the bend sooner than you think! I was just in a GAP at mall in Virginia where I see an item or two coming off the racks. Unfortunately they were ladies' items, and while I normally don't think twice about purchasing women's undergarments, I was feeling a bit bashful at the time. I don't think Squant's primary odor was very helpful to those items as well, but then again, I didn't closely inspect them, because no on 09:13:40 http://www.negativland.com/squant/ 09:13:54 haw, probably very old, but i hadnt heard of it before 09:14:01 (not probably, it is very old) 09:14:33 --- join: Robert_ (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 09:20:18 --- quit: Robert (Read error: 60 (Operation timed out)) 09:21:56 --- nick: Robert_ -> Robert 09:24:10 --- quit: mur (Read error: 110 (Connection timed out)) 09:29:01 --- join: tgunr (~davec@vsat-148-63-4-107.c001.g4.mrt.starband.net) joined #forth 09:34:45 --- join: OrngeTide (orange@rm-f.net) joined #forth 09:49:39 * TomasuDlrrp is back (gone 07:41:37) 09:49:44 --- nick: TomasuDlrrp -> Tomasu 10:05:30 which fbsd include file lits all the syscall numbers ? 10:06:49 --- quit: fridge (Remote closed the connection) 10:06:55 --- join: fridge (~fridge@dsl-220-253-73-190.NSW.netspace.net.au) joined #forth 10:07:15 sys/syscall.h 10:08:00 yea 10:08:03 i fount it hehe 10:08:04 thanx 10:09:07 --- nick: qFox -> kuvos 10:09:29 --- nick: kuvos -> qFox 10:34:12 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 10:34:12 --- mode: ChanServ set +o Herkamire 10:34:55 OrngeTide: that tiny ppc board was cool! 10:35:19 I want something like that with display capabilities 10:35:21 Herkamire, yea. i wish you could just buy it. 10:35:35 i wish apple would just try to do a G5 cube. 10:36:24 what's funny about apple is they took ibm's dual-core processor and made it a single core. but then they discontinued selling single processor G5s. 10:36:31 OrngeTide: do you have linux installed on your ppc laptop? 10:36:36 yea. i'm running gentoo 10:36:48 cool :) me too 10:36:50 use OSX silly 10:36:57 (except I have a desktop) 10:37:01 windows! 10:37:03 it's sleeping in OSX right now. :) 10:37:06 ^^ 10:37:16 now that i've found out how to remap capslock to control in OSX I mostly use OSX. 10:37:20 I ran debian-ppc for a while 10:37:23 fridge: I much prefer linux to mac osx 10:37:28 until I realised that OSX was pretty good 10:38:07 mostly when i'm on OSX i'm running X11 and one giant xterm. 10:38:32 i don't use Expose and all that stuff. 10:38:48 mol works pretty well 10:39:06 I'd rather run that under linux, then try to get everything installed with fink 10:39:31 I don't have enough free space for fink 10:39:50 mol is really neat:) 10:39:51 depends on what you want to use the machine for 10:40:29 --- join: Sonarman (~matt@adsl-67-113-235-174.dsl.snfc21.pacbell.net) joined #forth 10:40:30 i need a new mac. either a G4 iBook or i'm going to wait for a G5 iMac. right now i have a G3 500 iBook .. the videocard on it is really old and slow :( 10:40:35 I just use my laptop for email/browsing/dvds 10:40:50 played with mops a little 10:40:52 my laptop is my primary dev box. 10:41:10 car's leaving. be back in 30 minutes 10:41:14 seeya 11:06:04 --- join: mur (~mur@smtp.uiah.fi) joined #forth 11:11:29 arke: you here? 11:13:57 OrngeTide: you run linux on the mac? 11:14:05 sure. 11:14:20 or macos? 11:14:33 i run osx and linux both. i dual boot 11:15:04 when i got my first mac there was no osx. and i liked linux a lot better than macos9, at least for devel. 11:15:25 isn't osx the one that is part unix? 11:15:33 darwin? 11:15:34 but i continue to run linux on my mac because osx isn't as customizable. 11:15:51 osx is freebsd + mach + special apple stuff 11:15:58 futhin: yep, bsd with a shiny new init 11:16:16 what about darwin? 11:16:25 osx is freebsd? 11:16:27 weird 11:16:31 futhin: based on it 11:16:32 darwin is osx - the gui 11:16:33 in part 11:16:41 OrngeTide: thats aqua 11:17:06 ahh 11:17:06 fridge, it's more than just aqua. but yea. 11:17:09 nevermind 11:17:15 osx minus the gui 11:17:22 ahh. yea. - is minus :) 11:17:24 makes more sense 11:17:46 if you want gui on darwin you just run X11 .. so in that sense darwin is more like another freebsd distro. :) 11:17:47 I thought you meant it like spiderman 3 - return of zordock 11:18:03 although darwin and osx both have openstep-like file system layout. 11:18:10 fridge, ahah. :) 11:18:37 --- quit: mur_ (Read error: 110 (Connection timed out)) 11:18:39 i wish keyboards had a dash and a minus key. sometimes it'd be nice to have 11:19:14 various things use -- as a dash 11:19:26 that's kind of a hack though. :( 11:19:44 there are unicode characters for dash. but i totally don't know how to do any useful unicode on my keyboard 11:20:01 unicode has like twenty different astrisks too 11:21:48 haha 11:21:54 * Tomasu is away: later 11:22:34 qFox: you here? 11:23:55 OrngeTide: at least they had the sense to reject klingon from unicode 11:24:11 not that klingon supporters didn't try to get it included 11:24:47 yes? 11:24:51 fridge, i think tengwar (tolkien/elven script) is in there. 11:25:01 there are spare pages where klingon is normally placed though. 11:26:13 * qFox pokes Sonarman 11:27:25 qFox: i still have the #ifdef code, in case you're still interested 11:27:29 sure 11:27:36 cant hurt to take a look ;p 11:27:49 what's a good pastebin? paste.lisp.org isn't working 11:27:57 eh i have a link somewhere 11:27:57 sec 11:28:20 found one 11:28:26 time to start re-organizing my favs 11:28:29 http://www.rafb.net/paste/ 11:28:32 big huge mess in there 11:28:38 hehe that's the one i found 11:28:43 javascript chan :p 11:28:54 amongst others obviously 11:29:08 i have mailed him to support forth, he said he would give it a look some day 11:29:21 that was quite a few some days ago though :) 11:29:21 forth is hard to syntax-highlight anyway 11:29:28 http://rafb.net/paste/results/9K1fro30.html 11:29:36 well, core words ansi/fig is do-able imo 11:29:40 yeah 11:29:43 numbers, etc 11:30:35 but will that work in interpretation mode as well? 11:30:55 sure 11:31:07 hm it looks good yeh 11:32:27 well, /me is away 11:34:34 --- quit: qFox ("meh ircii's netsplit detection appearantly sux ^^") 11:35:28 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 11:36:10 hey 11:36:15 eugh 11:36:20 hurry quota reset 11:37:08 ? 11:42:05 my internet connection is shaped 11:42:11 it resets tomorrow 11:42:17 and I'm getting frustrated 11:42:24 because everything is so slow >=( 12:05:24 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 12:06:47 --- quit: futhin ("leaving") 12:16:00 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth 12:23:55 yay :) I figured out how I'm going to export my source to ascii 12:26:42 Herkamire, cool. how? 12:30:56 oh btw, Sonarman> i think that code might bork up when the ifdef's are nested 12:31:26 which might be an acceptible flaw considering the circumstances, but it should be noted nevertheless 12:31:34 acceptable 12:32:38 qFox: ah yeah, good point 12:33:05 no, i think it will work 12:33:13 no, no 12:33:15 it won't 12:33:18 :) 12:33:22 no because the code will stop not parsing at the first encounter of undef, period 12:33:24 :) 12:33:27 yep 12:33:39 but like i said, thats acceptable since you cant compile branch targets 12:34:01 it could just count the #ifdefs it encounters 12:34:18 an #undef would decrement that count 12:36:16 hmm 12:36:21 yes possibly 12:36:41 i think you can safely use the return stack for that purpose 12:36:45 during compile time 12:39:00 or a variable 12:39:13 suer 12:39:24 sure. even the data stack 12:39:38 slava: I'm going to loop through the source, marking each dictionary entry that is used. 12:39:47 (printing the source words as I go) 12:40:23 Herkamire: won't the exported source be kind of unorganized, then? 12:40:38 or, nevermind 12:40:46 then, I'm going to loop through the dictionary (ignoring unmarked ones) printing the dictionary entries out, and if they point to data, printing the data too (in a seperate file) 12:41:30 Sonarman: the source words are organized in blocks. The dictionary isn't in any decernable order, except for the core words 12:42:36 the core words are first, and in a particular order (this order is expected by the asm kernel used in the initial bootstrap) 12:42:51 but it won't get reordered, because they are first, and are all used 12:42:57 ok 12:43:21 if I were printing dictionary entries as I looped through the source it wouldn't keep them in order, but I decided not to do it that way 12:43:46 part of my recent decision making was how to store data 12:43:54 I decided to stick it all in one file 12:46:55 how do you determine whether or not a dictionary entry is used? 12:47:40 just a name, then a tab, then the data (encoded so there's no characters outside space..~ 12:47:49 Sonarman: by looping through the source 12:48:04 if no source token points to the dictionary entry, then it is not used 12:48:35 oh, ok. sorry; i'd forgotten about the way you store your source :) 12:48:41 :) 12:48:54 I'm looping through all the source tokens anyway to print them 12:49:34 I'll just make a byte array to track which dictionary entries I've used 12:49:52 Herkamire, what does the syntax look like? 12:50:49 slava: there's hardly syntax... looks like this: http://herkamire.com/jason/herkforth_screenshot 12:51:17 Herkamire, i mean the ascii export 12:51:55 oh, pretty standard. characters :[]'& indicate color changes 12:52:27 looks like normal ansi forth code sorta, except that ' means constant, and & means variable 12:52:58 ok 12:53:06 you only have global variables? 12:53:11 so your code is not re-entrant? 12:53:52 no 12:54:04 everything is global 12:54:08 not re-entrant 12:55:12 once I get to multitasking each preemptive process will have to have it's own set of CFA fields for part of the dictionary entries, and maybe it's own copy of data 12:55:35 cool 12:55:54 haven't gotten to multitasking yet 12:56:14 early on, another process will just be another copy of the whole system 12:57:08 i have the opposite approach -- tasks are co-operative and very lightweight. a task is basically just a copy of the stacks that can be switched in 12:57:10 no per-task data 13:01:48 & rnd-seed 0 [ time rnd-seed ! 13:01:48 : ,rl9 tos-tos 9 0 31 _rlwinm ; 13:01:48 : -random rnd-seed @ $10450405 * 1+ dup rl9 rnd-seed ! ; 13:01:48 : random -random *hu ; 13:02:27 slava: what about variables? 13:02:46 slava: so each process is compiling it's own copy of words and allocating it's own variables on the heap? 13:03:35 no, everything is shared 13:03:40 --- quit: Topaz (Read error: 104 (Connection reset by peer)) 13:03:47 a process is just data/call/name/catch stacks 13:04:00 --- join: Topaz (jonny@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 13:04:03 what is $10450405? 13:04:07 ppc opcode? 13:04:14 it's a number in hex 13:04:58 how do you pick those numbers for RNGs? 13:05:12 someone reccomended them. I don't remember who 13:05:19 I don't know much about RNGs 13:05:44 I didn't need a particularly good one (and still don't) so I haven't looked into it much. I just took the first simple solution I saw. 13:06:20 slava: so to do multitasking you have to be careful to write re-entrant code 13:06:58 that sounds aweful 13:07:10 you would be forced to do everything on the stacks 13:11:14 no, its not aweful at all 13:11:31 my variables have scope 13:11:46 and can be nested into a tree of namespaces 13:12:04 words are organized into vocabularies, which are one branch of the namespace tree 13:32:37 what key do you hold down to get the graphical Openfirmware boot menu thing? 13:33:05 apple? 13:33:09 option? 13:33:27 opt 13:37:55 thanks 13:38:02 movie. bbiab 14:12:59 --- join: SDO (~SDO@67-23-111-213.clspco.adelphia.net) joined #forth 14:13:33 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth 14:20:20 --- quit: qFox ("meh ircii's netsplit detection appearantly sux ^^") 14:25:58 --- quit: mur (Read error: 110 (Connection timed out)) 14:31:01 --- quit: tathi ("food, car repair, moving furniture -- back later") 14:40:30 --- quit: SDO (Connection timed out) 14:53:31 --- quit: arke (Client Quit) 15:21:12 --- join: madwork_ (~madgarden@derby.metrics.com) joined #forth 15:37:39 --- quit: mur_ (Remote closed the connection) 15:38:13 --- join: mur (~mur@smtp.uiah.fi) joined #forth 16:02:56 --- join: arke (~arke@adsl-69-209-69-154.dsl.chcgil.ameritech.net) joined #forth 16:02:59 --- part: arke left #forth 16:03:11 --- join: arke (~arke@adsl-69-209-69-154.dsl.chcgil.ameritech.net) joined #forth 16:05:18 --- 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.") 16:15:32 arke: ping 16:15:48 Sonarman: pong 16:16:07 i think our version of goto that uses will work in isforth 16:16:31 too late 16:16:34 i modified isforth source 16:16:35 :) 16:16:41 n00b 16:16:43 :P 16:17:01 dgfdasgfasfvfraf et wr ew 16:17:18 me too 16:21:34 --- quit: OrngeTide ("i'm really gone now.") 16:27:13 --- join: TheBlueWizard (TheBlueWiz@207.111.96.110) joined #forth 16:27:13 --- mode: ChanServ set +o TheBlueWizard 16:29:16 Hi TheBlueWizard 16:29:17 :) 16:29:32 arke hiya 16:29:51 * TheBlueWizard is very tired, in many ways 16:34:34 --- join: doublec (~doublec@coretech.co.nz) joined #forth 16:37:35 --- quit: arke ("Lost terminal") 16:57:43 --- quit: Topaz (Remote closed the connection) 17:06:06 --- join: mur_ (~mur@mgw2.uiah.fi) joined #forth 17:17:43 --- quit: mur (Read error: 110 (Connection timed out)) 17:23:27 * Tomasu is away: away 17:39:06 --- join: mur (~mur@uiah.fi) joined #forth 17:40:14 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 17:50:03 --- quit: mur_ (Read error: 110 (Connection timed out)) 18:09:53 try this 18:09:53 : blah create , does> @ ; 18:09:53 9 blah foo 18:09:53 foo '0' + emit 18:09:55 oops. 18:09:57 if that works then cretae works perfectly and so does ;code 18:09:59 doesn't work in bash lol 18:10:01 lol 18:10:03 fix bash too then 18:23:40 --- join: arke (arke@adsl-69-209-69-154.dsl.chcgil.ameritech.net) joined #forth 18:24:51 hi all 18:26:00 arke hiya hiya 18:26:31 * TheBlueWizard is approximating the notion of "hiya again" 18:27:06 --- join: mur_ (~mur@smtp.uiah.fi) joined #forth 18:27:23 TheBlueWizard: got any forth project(s)? 18:31:16 had one....put on back burner for damn too long time :P 18:35:07 why in (@#*&$(@#*$&#$% do you have to wait to type your password after typing "su" 18:35:45 I'm running at enough hundreds of MHz that my computer shouldn't be dropping characters when I type 18:36:51 --- quit: mur (Read error: 110 (Connection timed out)) 18:41:35 Herkamire: i guess it's to prevent brute-force password-finders 18:42:25 although i don't have to wait 18:42:38 Herkamire: security reason 18:43:04 Herkamire: hack the source, recompile :) 18:43:31 I had a BLONDE moment last night 18:43:45 looking at the source code to isforth, trying to figure out how to hack myself around something im trying to do 18:43:48 until i realized 18:43:51 though it shouldn't drop characters once the prompt shows up (there should be a small delay between invoking and the prompt....can't see why the typing has to be s-l-o-w though) 18:43:59 hey, if I CAN SEE IT, I can MODIFY IT TOO!!!! Doh! 18:46:54 TheBlueWizard: it's teh initial delay that get's me mad 18:47:58 there's gotta be a better way to slow down brute-force password finders 18:48:25 Herkamire: sudo has no delay, and its better anyway 18:48:39 Herkamire: (like, instead of su; make install; exit; I do sudo make install 18:48:42 never liked sudo 18:49:20 why don't they have the delay when you hit enter (before it respends whether the password is correct or not) 18:49:49 if you know of a better, *workable* (at least from security standpoint) way, we'd like to see that! Remember, there are MANY smart guys out there....so, coming up with a better mousetrap is HARD 18:50:02 interesting to hear you asy it is to prevent attacks. I did not think it was intentional 18:51:07 how about this, the password checker will only check a password 3 times every 5 secconds (per user) 18:51:26 arke: if you are a member of sudo list, then of course it can be quickly done...but that creates a potential security hole 18:51:59 the delay is much grater for me when I have other stuff running 18:52:01 * arke frequently does sudo /sbin/init 6 and then enter my pasword immediately afer 18:53:04 I think I've gotten my whole password typed before the prompt showed up before 18:53:29 Herkamire: same here. 18:53:37 and my root password is pretty long 18:53:38 it's rediculous 18:56:33 sorry, but I don't see it as a security enhancement, that my password get's displayed in plain text on the terminal sometimes when I'm trying to su 18:57:04 indeeed. 18:57:12 which is why I use sudo 18:57:13 :) 18:57:31 sudo sh = su 18:59:16 remember, there are script utilities (such as expect) that can "talk" with the programs, so if a program is instanteously responsive, then the brute force cracker can try out many, many combinations real fast 18:59:55 so you must include that factor into consideration if you want to develop a better mousetrap 19:02:22 I'm not at all sure that delay is intentional 19:02:39 I know there's an intentional one on failed user/pass combination 19:03:56 correct...again, same reason 19:04:04 * Tomasu is back (gone 01:40:37) 19:05:54 TheBlueWizard: right. I just don't remember PAM or anything having a setting for any delay between username and password. 19:06:09 so I'm questioning whether it's intentional. 19:06:53 or if it's just that the whole PAM thing requires it to look through several gazillion config files and load a bunch of libraries before it can decide what to do. 19:09:09 if there is a delay between connecting to an ISP and getting login prompt, then that delay is already paid for, so the password prompt would be quick....makes sense? 19:09:27 yeah. 19:09:28 there has to be a delay somewhere.... 19:09:32 its the latter 19:09:42 that delay is only apparant on the FIRST use of pam 19:10:33 or the first use of pam in a while -- presumably it can get swapped out 19:11:10 --- join: warpzero (~warpzero@dsl.103.mt.onewest.net) joined #forth 19:14:51 I'm asuming that the delay is just the time it takes the program to initialize etc 19:15:13 and that the reason it misses chars is that it reads directly from the terminal, instead of reading the characters that are in the STDIN buffer 19:16:18 --- join: mur (~mur@mgw2.uiah.fi) joined #forth 19:18:16 Herkamire: that can be tested (lemme do that) 19:18:33 --- quit: mur_ (Read error: 60 (Operation timed out)) 19:19:33 --- join: Sonarman_ (~matt@adsl-64-169-94-65.dsl.snfc21.pacbell.net) joined #forth 19:19:38 Herkamire: well, it recognizes pipes :) 19:19:52 --- quit: Sonarman (Nick collision from services.) 19:19:57 --- nick: Sonarman_ -> Sonarman 19:21:39 arke: what do you mean? you can do: echo -e "mypass\nhalt" | su 19:23:39 mmmmmmm 19:23:52 I'm thinking about the code to clear out unused dictionary entries 19:24:01 I'm worried it will take a noticable amount of time 19:25:09 I want to run it automatically every time you save. 19:25:14 but I'd hate to have to wait for it 19:25:21 and I don't want to have to remember to do it either 19:26:21 mmm. maybe it woun't take a noticeable time 19:26:45 I have almost 700 words of source 19:26:52 7000 19:28:03 hmm... I guess it's only 3 instructions per source token 19:28:42 fetch(token), shift(to array index), store(true at index) 19:38:37 ok, question for all 19:39:05 hows this for a data object format: 19:39:17 pointer (in dictionary) points directly to data 19:39:30 in a string, it would point directly to the first character 19:39:46 before the pointer (lower in memory) would be metadata 19:39:56 such as the length, type, and whetever else 19:40:05 so the format would be: 19:40:53 [whatever structured metadata], length of metadata, data type, data length, <> data 19:46:45 data types? 19:47:27 yeah, like 1) string 2) branch table 3) source code etc 19:47:44 4) array of 32-bit integers 19:48:42 is the length of metadata field really necessary? 19:48:51 as far as I can see, the type will mostly be used to decide how to display and edit the data 19:49:00 Sonarman: I think so 19:49:07 since the type should determine the length of metadata 19:49:26 well, my idea with the count was that you could add to the metadata 19:49:37 without changing the type 19:49:47 ok 19:49:51 eg some strings could have metadata about what sort of character set 19:49:57 or sets 19:50:25 lucky you, you're designing new features, i'm trying to figure out why i get an incorrect result for this: 1591517158873146351817850880000000 32769 mod . :-) 19:51:30 lol 19:51:39 slava is it REALLY that important ? 19:52:14 I440r, i'd like it to be at least approximately correct :) 19:52:22 lol 19:52:33 I can't think of a time when I wanted a computer program to do something with an integer bigger than 4 billion 19:52:42 i can 19:52:59 Sonarman: cool :) what for? 19:53:00 computing trahectory to alpha centuri 19:53:14 I440r_: you'd want that to be an integer? 19:53:15 Herkamire: i haven't used it for anything cool :) 19:53:20 cryptography! 19:53:20 andn you only got ONE shot at it... and once your in motion you cant change direction :") 19:53:21 :) 19:53:51 slava - correct results are not required there either 19:53:52 just occasionally the result of an operation in a calculator program is greater than 2^32-1 :) 19:53:56 just consistent ones :) 19:54:14 calculating large prime numbers... 19:54:55 ok, use a HUGE sieve :) 19:55:30 http://www.eff.org/awards/award-prime-rules.html 19:55:42 :-) 19:55:57 rule #1: result must be a prime? :) 19:56:10 "NOTE: Leading zeroes don't count! " 19:56:59 lol 19:57:20 all I need to do is have factors lazy list of primes and ask for 10000000 lnth 19:57:26 and wait for awhile 19:57:36 a long while 19:57:39 fsvo "awhile" 19:57:49 ( 3 ) 10000000000000000000000000000000 1000000 /i . 19:57:49 10118843564710973484731656 19:57:53 *sigh* 19:58:05 and hope that the maths in factor works ;-) 19:58:11 i think the _medium and _large versions of the division and modulus is broken 20:00:01 slava: where'd you get arbitrary precision integer math stuff in C? 20:00:10 or is this all in java? 20:00:14 Herkamire, i'm using the scheme48 implementation. 20:00:26 Herkamire, no, java gives correct results :) as does scheme48 itself. so i broke something :) 20:00:50 slava: you ported factor to scheme? 20:01:21 Herkamire, scheme48's kernel is written in C :) 20:02:09 i fixed it 20:02:34 i accidentally changed one of the s48 typedefs from signed to unsigned. and amazingly, it "mostly worked". 20:02:44 lol 20:02:54 heh 20:03:06 doublec, oh and remember the out-of-memory error with ratios? 20:03:14 doublec, that was caused by the gcd algorithm going astray, from incorrect results 20:03:47 ahh, ok 20:04:00 all bye 20:04:06 heh, amazing. nice find. 20:04:23 --- part: TheBlueWizard left #forth 20:10:57 --- quit: tathi ("poof") 20:26:27 ^_^ ^_^ ^_^ ^_^ 20:27:13 arghblbl 20:50:38 hi arke 21:02:40