00:00:00 --- log: started forth/05.06.28 00:59:10 you know, creating words that make words is really great. 01:19:26 --- join: virl (Phantasus@chello062178085149.1.12.vie.surfer.at) joined #forth 01:29:12 :) 01:30:07 I wish there were some better string handling words in isforth though 01:30:31 because I have shit like this right now: 01:30:31 create $buf 100 allot 01:30:31 create $colon ," :" 01:30:31 create $cap ," cap" 01:30:41 01:30:42 : name-capability? \ Generate a capability name for the current service. 01:30:43 curser @ s>str $buf str+ >r 01:30:43 $colon count r> str+ >r 01:30:43 $cap count r> str+ 01:30:43 $buf swap over - 01:30:43 ; 01:30:54 with 01:30:56 : str+ 2dup + >r swap cmove r> ; \ Append a counted string to 01:30:56 \ an address. No bounds checking yet. 01:30:56 : s>str s>d (d.) ; \ Convert a number to a string in the 01:30:56 \ current base. Maybe this should always 01:30:58 \ use hex, since that's what the standard 01:30:59 \ follows. 01:32:14 01:32:30 I think the real problem is I'm trying to use the header database as the database for my own application 01:32:45 and I should just create my own header. 02:01:59 naah 02:02:08 I440r just doesnt like using strings, it seems 02:02:19 :) 02:02:29 yah 02:02:36 what I'm doing is this: 02:02:41 $0C pid \ Engine RPM 02:02:42 { 02:02:42 rpm @ dup 02:02:42 hi 2 << -> 02:02:42 lo 2 << -> 02:02:42 } 02:02:50 well, prepend that with this: 02:02:55 $1 service 02:02:58 compiles a word called: 02:03:20 : 1:12 rpm @ dup hi 2 << -> lo 2 << -> ; 02:03:49 but it makes pid all complex to name the word,.. so I'm wondering if I shouldn't just have my own headers in memory that don't use strings. 02:03:58 and just a looking up of 1, and then 12. 02:04:06 and store an address into the dictionary 02:08:22 but that means I can't use things like 'words' to find definitions. 02:17:23 lol! Dude, what are you writing? 02:19:03 Dude... 02:21:07 arke: an OBD-II diagnostics system 02:21:41 whats that? 02:21:52 ...oh, isn't that one of those engine senors? 02:21:57 sensors* 02:22:44 OBD-II is like 4-5 standards rolled into one.. most is network interfacing, and packet information, and there's some policy stuff in there as well 02:22:51 but yah, engine sensors are part of it 02:23:39 sweet 02:23:44 im writing an engine simulator 02:24:01 its hard... :)# 02:24:12 well, I wrote one before but it wasn't a "true" one 02:24:25 this one will take into account piston inertia, camshaft stuff, etc. 02:24:33 and yes, it will be freakin' _hard_ 02:24:59 ah 02:25:02 good stuff 02:25:11 I have to write something similar to that for misfire detection 02:25:30 which uses kind of an engine simulator to figure out when the engine is operating outside of parameters 02:25:54 lets collaborate my friend! 02:26:09 I scratched all my code 2 days ago and am currently figuring out how to start over 02:27:22 Well, I definetely have to keep track of the current pressure inside the piston 02:27:25 which I didnt before 02:28:02 (the reaon I 02:28:21 (the reason I've been writing engine simulators is because I wanna come up with a certain "ultimate" engine) 02:28:44 think electronically controlled, boxer, 2-cylinder, 2-stroke, diesel 02:29:01 maybe without the electronically controlled, but the other stuff definetely 02:29:45 err, thats 4 cylinder, not 2 cylinder 02:29:59 nod. 02:30:10 sounds cool 02:30:16 yup 02:30:33 vibration wouldnt be too much trouble, since it'd be like an 8-cylinder 4-stroke 02:30:46 power stroke every 1 stroke, you see? 02:31:11 since its a diesel, theres no ugly mucking around with burning oil either 02:31:31 also, have you heard of the miller cycle? 02:31:34 http://en.wikipedia.org/wiki/Miller_cycle 02:32:14 I was thinking adding a centrifugal clutch which engages or disengages a miller intake system fed by a supercharger 02:32:30 yah 02:32:34 interesting 02:32:50 I'm a big rotary engine fan,. they're what I work on mostly 02:32:54 after like 2000 rpm, the piston does the compression, but before that, the supercharger those 02:33:08 yeah, I love rotaries too, but I also love diesel piston engines 02:33:32 just so much elegant than a gasoline piston 02:33:44 yah 02:34:12 anyway, another thought I had was a small air tank which serves as a compressed reservoir 02:34:58 air is stored from it automatically and retrieved at very very low RPM to reduce "supercharger lag" 02:35:09 ah 02:35:23 which really isnt a problem with superchargers at all 02:35:41 however, with superchargers, low-end torque always suffers a tiny bit 02:36:12 but the supercharger needs to do alot of work exactly then for the miller cycle to work 02:36:25 so the air tank is there to help it out 02:36:38 well, this is just an idea of course. 02:38:21 sorry for the rant :) 02:38:23 haha 02:38:29 nah, don't worry about it, I'm just thinking about the setup 02:38:34 what's your goal? 02:38:45 like, just to fuck around with the engineering? 02:39:11 for fun, really. 02:39:26 but also because I think it'd be perfect for a small car or motorcycle 02:39:34 nod. 02:39:34 and insanely good MPG too 02:40:49 Only problem with the boxer arrangement I can think of right now is that a Diesel will need a long stroke 02:40:57 and that will make it quite wide, ya see? :) 02:41:12 yah 02:41:21 the pancake configuration also makes them hard to work on 02:41:34 True. 02:42:12 But, it reduces the vibration to a point of needing no flywheel 02:42:51 The little "weight" from a clutch/transmission would be all ever needed 02:47:14 I just had an idea. No idea if it'd even work, but... 02:47:54 I call it sequential feeding :) 02:48:14 In a bank, the first cylinder gets a normal fuel, air mixture 02:48:51 the second cylinder receives the exhaust from the first with very little fuel added 06:55:40 hrmm, they do that in some turbine systems or superchargers 06:55:56 but i think that it'd just hurt performance in a piston system 06:56:14 the point is emissions 06:56:30 theres some really strict emissions shit on diesels 06:56:35 it would help 06:56:52 i think 06:56:52 :) 06:56:56 but why not just have a supercharger? 06:57:33 it's just a turbine after the piston system. you can burn up the excess fuel there and get increased performance 06:57:34 there is 06:57:56 thats a turbocharger, not supercharger 06:57:57 :) 06:58:09 *charger 06:58:10 =p 06:58:14 :) 07:17:36 --- join: PoppaVic (~pete@0-1pool65-11.nas22.chicago4.il.us.da.qwest.net) joined #forth 07:18:10 Mornin' 07:27:41 --- join: sproingie (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 07:43:39 --- join: ianp (~ian@inpuj.com) joined #forth 08:36:22 --- nick: Raystm2 -> nanstm 09:09:02 --- join: Herkamire (~jason@c-24-218-95-147.hsd1.ma.comcast.net) joined #forth 09:09:03 --- mode: ChanServ set +o Herkamire 09:09:15 Lo, herk 09:09:52 hi PoppaVic 09:11:19 I still wish to hell the utf8 wide-char support was cleaner *sigh* 09:11:33 I'm tempted to try to deploy it. 09:11:34 on what? 09:11:38 in C 09:11:55 For this whole mess I'm baking 09:12:15 you wish the utf8 wide-char support in C was cleaner? 09:13:07 yeah, it'd be wonderful if we had a compiler-directive to make it default and switch to asci as easily. 09:13:37 I'm cogitating name-fields/structs 09:14:01 oh 09:14:30 I can't see it a Bad Thing(tm) - but it COULD be a Good Thing(tm) 09:14:49 the issues arise when I try to recall whom does what to whome and how. 09:16:58 But, hell I still embrace ascii - so that is the least of my worries (except for modularizing shit related to strings) 09:18:51 I've pretty well concluded I need (at least) one more rewrite... Ideas keep leaking in and alternatives or misconceptions keep cropping up. 09:19:51 i didn't know there was any wchar support for utf8 09:20:02 being variable sized and all 09:20:04 that's what wchar_t is 09:20:17 i thought wchar_t was utf16 09:20:20 at least, in the gcc universe (afaik) 09:20:38 I could be wrong - it happens. 09:20:50 But, we usually speak of utf8 in ##C 09:21:00 i would guess it's ucs2 or utf16 09:21:31 I didn't even know C had any syntax for utf 09:21:38 i'm rather sure it doesnt 09:21:40 hey I could be wrong - maybe folks using wchar_t ARE using utf16 09:21:54 see man wprintf 09:22:06 and man 5 utf8 09:23:22 it's in section 7 for me 09:23:34 anyway, I've made some progress (if goingforward means having to go back 4 files) 09:24:38 I believe I am "down" a few structs and funcs, but aswirl all the conceptuals, I seem to be progressing. 09:28:58 looks like wchar_t is ucs-4 09:29:15 at least in gcc. i think it's implementation-defined 09:29:17 ugh - could be, I read it the other way 09:29:28 according to bits/wchar.h: 09:29:34 #define __WCHAR_MIN (-2147483647 - 1) 09:29:34 #define __WCHAR_MAX (2147483647) 09:29:36 yeah, it really deserves to be a module. 09:30:18 (I hate those sorta' defines) 09:30:27 utf8 is a real bitch for C 09:30:43 although there's some advantage to using it -- you're forced to use a real string type more or less 09:31:28 some wouldnt call that advantageous, but the code should be clearer 09:32:10 well, except for unix98 doing its best to use obtuse names like "mbsrtowcs" 09:32:23 yeah, I know 09:32:34 it's been the bane of C for more than 20 years 09:33:12 these days i don't take any language seriously that doesn't have a real string type 09:33:13 ..damn nearly puts me in mind of dos/doze stupidities 09:33:15 or at least a library 09:33:37 that's nice - for new lang ;-) 09:34:41 string types? 09:34:54 with C, it just emphasizes the simple fact the the comittee-commies only care about perverting semantics/syntax and libc is still the kitchen-sink 09:35:03 virl, sure 09:35:34 libc is shameful 09:35:37 a "string" is just another type - a superset of char which is small int. 09:35:41 glibc doubly so 09:36:06 no, a string is a pretty abstract thing 09:36:15 sproingie: I'd love to defend them, but I have to agree in essence: they need to codify it better and do major breakout 09:36:22 it's a sequence of characters that might have any of a zillion encodings 09:36:42 I was defining string ala' defaults 09:37:25 possibly the worst part about libc is how it freely mixes ansi/posix/unix98/sus with arbitrary extensions 09:37:35 don't take methere, please 09:37:45 the commies piss me off 09:38:11 what would you prefer? 09:38:16 it's like learning fig-forth and then suffering f79, f83, ans and f83/fpc and shit 09:38:40 C is driving itself into irrelevance anyway, it's purely in maintenance mode. C++ is doing some useful stuff these days 09:38:53 stuff from boost is getting into the c++ standard, fer example 09:38:54 virl: codify the semantics; codify the cpp; and then break out and codify libc 09:39:15 luckily, I wouldn't ever touch c++ again w/o a huge paycheck 09:39:32 i'll do it for a modest paycheck if that's what i gotta do 09:39:40 fuck that 09:39:56 my main problem with code jobs is that i can't really understand most peoples code until i rewrite it myself 09:40:06 C++ went to hell years ago, I can barely puzzle out the stuff I see anymore 09:40:11 so i'm in a job now where i write lots of code, but it's all mine 09:40:25 yeah c++'s template syntax is out of control 09:40:43 not just that - polymorphing, inheritance, privacy, etc 09:40:48 lots of it has to do with lack of a real type language with type constructors and whatnot 09:41:08 yes, I keep thinking about that 09:41:21 tis why I use vtables all over the place 09:41:22 you have constructors for instances, but not for types themselves 09:41:43 so you can't really do something like a Maybe type in C++ 09:42:17 well, there are multiple issues for 'types' - I've been cogitating the funcs and returns required, but I have only 20 years experience. 09:42:28 data Maybe a = Nothing | Just a deriving (Eq, Ord, Read, Show) 09:42:40 you know how much verbage that would take in C++? 09:43:02 ns: part is lexing, part is parsing, and the rest is func-ptrs 09:43:13 no idea, I no longer touch c++ 09:43:40 Frankly, I'd write in ANS forth before suffering C++ ever again. 09:43:47 i'm sure alexandrescu can prove that, being turing complete, c++ can express that type in only 300 lines of code 09:44:15 as I said and implied, I'd love a dialog on types and what support they'd need. 09:44:59 they need a different language. C++ isn't going to do anything radically new 09:45:04 Madworks idea of stack-node tags as func-ptrs makes more sense as vtable-ptrs, and the idea is intriguing 09:45:25 stack node tags as func ptrs ... ??? 09:45:26 Yep, and I eschew c++, live in C and like forth's 09:45:32 hehe 09:45:57 sproingie: yer too tied to yer universe of c++ and (whatever).. We've bespoken this shit before in here. 09:46:44 my universe is largely perl and python. forth is something i like for the combinator aspects 09:47:09 i don't even touch c++ in my job. shame, because i could have gotten promoted if i did 09:47:20 the idea is: I use my queue/stack module/code - and every node has a int tag (right now), but his idea of a func-ptr tag (or a vtable) has far greater implications. 09:48:19 I've had to fall back on ancient c++ class identifying ideas so far, but hey - I'dlove feedback and new ideas for it. 09:48:22 as in a type tag on all data? 09:48:33 i.e. manifest typing? 09:48:45 yep, every stack-entry is a node; the tag declares what it was 09:49:07 "manifest" - I like that ;-) 09:49:09 hm. well, if you want a fairly nice type system, you could do worse than factor's 09:49:37 it pretty much tags every piece of data as well 09:49:42 explain - bear in mind, I was a cook for 20 years - and I code because I like it ;-) 09:50:07 lot of overlap between coding and cooking 09:50:16 yep 09:50:24 ..so I keep telling newbies 09:50:40 I always found it nearly relaxing as reading a book 09:51:10 my cooking's like my code: i hack with just a smidge of theory basics. so i create good stuff small scale, but don't have the discipline for a big project 09:51:15 Leatherwork is more artsy, reloading and shooting are a bit more mechanical 09:54:06 The prob with most programming is: they just won't use structs, let alone func-ptrs. And, of course, overloading in C is verboten - while C++ is just nasty. 09:54:34 verboten? 09:54:42 In forthish stuff we see other issues, but they can boil down similarly 09:54:45 forbidden 09:54:54 PoppaVic, you are from germany right? 09:54:59 nope 09:55:07 Michigan, US, CONUS 09:55:36 but, both sides are euro several generations back, and 1/2 was german, yeah. 09:55:51 verboten's pretty common in english 09:56:05 virl: I just always liked reading WWII shitzu 09:56:25 ok 09:56:56 using the german carries a connotation of some stern-faced stormtrooper type saying "Nein! Das ist verboten!" 09:57:01 Anyway, yeah: I'm on the cusp of a "new language" whic sorta' melds older ideas; Just because everything is pissing me off. 09:57:07 right 09:57:50 Some things, gcc does right; others, forth did right.. Getting upside the middle and figuring out who and where is the issue. 09:57:53 or sergeant shulz at any rate 09:58:07 "I SEE _nothink!_" 09:58:19 (schultz) 09:58:27 i loved that show 09:58:35 yeah, it was classic 09:58:54 tv nowadays is almost depressing 09:59:11 one thing i'm seeing forth not helping me with is getting the context of what any individual word does 09:59:19 even CSI - which intrigues me - often depresses me 09:59:35 sproingie: no 'see' word? 09:59:47 i mean i see words beginning with "swap dup" and i get a big ? over my head 09:59:57 I think the kids nowadays call it "reflective" 10:00:01 ahhhhh 10:00:14 well, the gymnastics are hereditary 10:00:19 i know what those words do, it's just not obvious what it's doing it do 10:00:37 I can use 'em, but I like the { A B C } locals 10:00:45 i think it's actually a fundamental expressiveness problem with combinator syntax 10:00:57 yes. I want to resolve that all 10:01:16 i'm pondering a language that can switch between applicative and concatenative 10:01:19 USING them? fine - FORCING me to use them? Not Good. 10:01:28 haskell is almost that language actually, but it has so many other restrictions 10:01:32 ok, you lost me: exemplars? 10:02:05 applicative is where the args to a function are coupled with the application of it: foo(bar, baz), or (foo bar baz) 10:02:11 ok 10:02:17 concatenative is, well, forth. bar baz foo 10:02:21 ahhhh 10:02:28 it's not obvious how much foo consumes 10:02:37 damned schools - creating terms as they go the fuck along! 10:02:45 right 10:02:52 but you can do tricky things with it that you can't do easily with applicative 10:03:05 sure, it's like vaargs 10:03:07 tradeoffs. but parts of expressiveness really suffers 10:03:23 well with varargs, they're still bounded by delimiters 10:03:52 haskell has point-free programming which makes it really easy to compose functions as combinators 10:03:52 No. In theory, you can call foo(...) forever, until you screw the stack-access 10:03:58 but i like a language with state 10:04:13 I think the "state" is overrated 10:04:20 forth is all about state 10:04:39 I use stateish func-ptrs and do NOT let the state be accessible. 10:05:00 forth actually is an odd mix 10:05:06 indeed 10:05:17 tis why I am coding this mess 10:05:27 on one hand it's very stateful, but if you confine everything to the stack, there's no state side effects to speak of 10:05:40 there has to be a level of trust in the programmer being sane, and also a level of checking 10:06:00 on the third hand (if you're one of those types), using the stack for all state is just as bad as a bunch of globals, but more confusing 10:06:07 in the interpreter, sure - we need to loop on failoure+reset 10:06:25 upper, lower-hand; gripping hand ;-) 10:06:34 because if you use the stack for lots of state, you're really just juggling it all around in a structure that isn't designed for it 10:06:42 Right, I hate globals with a deep passion. 10:07:06 i think haskell's state transformer monads are actually the right answer, but the syntax for them is just hideous 10:07:24 I *know* we need a few global vars/structs in major programs, but MODULES should be stackframed. 10:08:37 well, again: haskell is beyond me as well as a "monad"; But we already know that any forthish system is going to preproc/compile/run - and never more than one per. 10:09:00 haskell is largely beyond me too 10:09:02 the "global state" is the controller 10:10:07 i like what the mozart-oz guys say about state: functional purity is overrated, there's plenty of provable strong theory around state as well, and some problems need state to express clearly 10:10:19 i gotta get their book. supposed to be the next SICP 10:11:20 re.. 10:11:57 functions/stackframe are fine; State is also fine - but the controls need to be out of reach physically: make a call change the foo 10:11:59 some todays security problems with some protocols result into their stateless behavior. 10:12:17 statelessness is irksome, yeah 10:12:55 i prefer "change the foo == make the call" 10:13:18 aka condition variables 10:13:24 html for example: as-is, fine - request/receive; something tcp/ip BEYOND that is needed. 10:13:37 tcp/ip is very stateful 10:13:42 tcp is anyway 10:13:43 I disagree: I do NOT want putzii mucking directly on my vars 10:14:08 I want the putzii to call myfunc that may or may NOT let them alter the state 10:14:08 well, assigning a condition var *is* calling a function, basically 10:14:17 it's just syntax sugar 10:14:29 sup guys 10:14:48 Folks may holler "semantics" - instead of "syntax". 10:14:48 the neat thing about c++ is that you can veto an assignment. of course it's so easy to get around, that it's never done 10:15:39 in C++ you NEED the endruns, because the asswipes generally have forgotten public AND "protected" and make the universe "private" 10:16:50 and, of course - we can get into polymorphism and heirarchial-inheritance - which I won't. 10:17:03 * sproingie looks at gforth's glocals.fs ... doesn't quite look like an easy port to rf 10:17:30 you mean the { A B C } stuff? 10:17:34 yep 10:18:00 yah, it's messy and not complete in the least - it's a Forth-Dimensions adaptation. 10:18:22 the fact that rf doesn't appear to have "postpone" makes it really difficult 10:18:27 No reason for suffering that shit 10:18:45 Not sure HOW to replace it, but I will need to do so. 10:19:28 yipes, i gotta get going. 10:19:29 By allowing for no return-stack and a datastack plus "any stack", it might be doable. 10:19:32 * sproingie waves 10:19:34 * sproingie & 10:19:36 laters, sproingie 10:20:23 The idea of a "stackframe" can become a real whore, but I think we need it. 10:23:37 back in a bit 10:23:39 --- quit: PoppaVic ("Pulls the pin...") 10:30:36 --- join: PoppaVic (~pete@0-2pool238-100.nas24.chicago4.il.us.da.qwest.net) joined #forth 10:30:58 flubber 10:36:44 hmm.. lexxing (parse) is in an interesting issue 10:40:34 --- quit: PoppaVic ("Pulls the pin...") 10:41:28 no one uses forth for games, depressing. http://groups.google.at/group/comp.lang.forth/browse_thread/thread/ca4d522a119e86b/5133fa7536e9f5a9?q=forth+games&rnum=2&hl=de#5133fa7536e9f5a9 10:46:49 --- join: PoppaVic (~pete@0-1pool46-10.nas30.chicago4.il.us.da.qwest.net) joined #forth 10:47:26 Hmm, I wish madwork was around.. I got ideas to bounce. 11:16:31 --- quit: PoppaVic ("Pulls the pin...") 12:22:28 --- join: snoopy_1711 (snoopy_161@dsl-084-058-140-059.arcor-ip.net) joined #forth 12:30:55 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 12:31:17 --- nick: snoopy_1711 -> Snoopy42 12:43:30 --- join: tathi (~josh@tathi.bronze.supporter.pdpc) joined #forth 13:01:11 is there a light weight forth system with a very advanced compiler(big optimizations)? 13:02:58 iforth ? 13:03:06 note.. NOT isforth, but iforth 13:03:12 :) 13:03:13 but that costs $99 usd 13:07:58 damn, should I ever create a fast and optimized forth, I think I'll take money for that. its frustrating. 13:09:11 :) 13:09:40 I've got some ideas...but I'm still doing design. 13:10:06 compiler optimizations are a waste of time 13:10:14 just code it right in the first place and you dont need any 13:10:48 ]it doesnt NEED to be 1 biollionth of a second faster 13:10:58 billionth even 13:11:02 for some things it does. 13:11:06 nope 13:11:17 like, ray tracing, for instance. 13:11:42 you cant do real time ray tracing 13:11:46 when you're doing complex calculations for a million pixels or so, little differences can really add up. 13:11:50 you can do a close approximation of it however 13:12:10 but you STILl dont need your compiler butchering your algorithms behind your back 13:12:16 no, but you can cut your rendering time by a lot when you're doing complex stuff. 13:12:41 optimizations are NOT suposed to "butcher your algorithms" 13:13:02 The whole POINT is to do transforms which DO NOT AFFECT the computations other than speed-wise. 13:13:18 any time you destroy the 1:1 corelation between source and object you have butchered it 13:14:02 bah. 13:14:07 :) 13:14:45 I mean, all the x86 processors work differently. So they execute your code in a different fashion, but the results are still the same. 13:15:02 Optimization is supposed to be the same sort of thing, only in software rather than hardware. 13:15:03 I440r: yoh 13:15:17 hi 13:15:53 I440r: so I've patched my copy of isforth to let me create word headers easier, from existing strings 13:16:48 yea tathi told me about your refactoring :) 13:16:54 ah cool 13:17:28 I440r: I'm kind of in a weird spot.. I need to create strings on the fly for my names, but doing the string creation is really awkard 13:17:33 awkward 14:26:56 hi everyone 14:27:29 sup crc 14:29:04 --- nick: nanstm -> tiff 14:31:28 not much 14:31:35 I'm still working on the handbook 14:32:35 hi 14:55:33 * I440r just got a fresh order of green coffee beans and is roasting coffee 14:55:42 had to SUFFER store bought coffee 14:55:42 ick 14:59:24 hmm, yummy :) 16:49:05 --- quit: tathi ("leaving") 17:01:27 --- join: asymptote (~dmesg@220.muma.nsvl.chcgil24.dsl.att.net) joined #forth 17:17:07 --- quit: asymptote ("Free the mallocs!") 17:33:57 --- join: blockhead (~bob@dialin-744-tnt.nyc.bestweb.net) joined #forth 17:39:55 * crc is finished the first version of RetroForth's Handbook :) 17:53:18 * sproingie wakes 18:19:16 --- nick: tiff -> Raystm2 19:18:00 --- join: KB1FYR (~Alex@d-66-63-85-222.suscom-maine.net) joined #forth 19:35:01 --- part: blockhead left #forth 20:56:51 --- quit: danniken (Read error: 54 (Connection reset by peer)) 21:17:30 --- quit: alexander_ (Remote closed the connection) 22:45:40 --- quit: virl (Read error: 104 (Connection reset by peer)) 22:47:03 --- join: virl (Phantasus@chello062178085149.1.12.vie.surfer.at) joined #forth 23:24:25 --- quit: sproingie ("Konversation terminated!") 23:59:59 --- log: ended forth/05.06.28