00:00:00 --- log: started forth/14.09.29 00:05:07 --- join: proteusguy (~proteusgu@183.89.94.164) joined #forth 00:22:28 --- quit: bluekelp (Ping timeout: 272 seconds) 00:25:59 --- join: bluekelp (~bluekelp@2001:470:1:41:795d:fd25:a6db:7aa7) joined #forth 00:25:59 --- mode: ChanServ set +v bluekelp 01:14:55 --- join: true-grue (~grue@95-28-181-143.broadband.corbina.ru) joined #forth 02:14:33 --- join: Mat3 (~Mat@91.65.76.247) joined #forth 02:14:40 hi all 03:24:41 --- quit: Mat3 (Quit: Verlassend) 03:30:19 --- join: Zarutian (~zarutian@46.22.110.168) joined #forth 03:50:17 Zarutian: go view the channel log, if you have time :) 04:27:29 ttmrichter: I know of this model too, when I first read something about linguistics and try to learn the meaning of this terms. this model is more conventional in a linguist's piont of view. and maybe when used in human languages, is model more specific. while you can see how my definitions are more simple, and they can apply to more general things. 04:28:11 ttmrichter: anyway, you see, it is why I wish to re-define them -- for I understand them differently. just as in math and physics, when one understands something differently from the books he readed, he goes to re-define them and to form new theory (or system, or structure) to help himself understand things. and then, is the new theory helpful? it is better or worse? more clearly or not? he (or others) just goes to 04:28:11 refine the theory again and again. 04:31:43 --- quit: xyh (Remote host closed the connection) 04:45:37 --- quit: proteusguy (Remote host closed the connection) 04:58:57 --- join: nighty^ (~nighty@hokuriku.rural-networks.com) joined #forth 05:52:22 --- quit: DGASAU (Ping timeout: 272 seconds) 06:09:45 --- join: DGASAU (~user@p50993595.dip0.t-ipconnect.de) joined #forth 07:28:20 --- join: karswell` (~user@87.112.211.95) joined #forth 07:29:52 --- quit: karswell (Ping timeout: 240 seconds) 09:15:23 --- quit: Zarutian (Quit: Zarutian) 09:17:36 --- join: proteusguy (~proteusgu@ppp-110-168-229-154.revip5.asianet.co.th) joined #forth 09:36:39 --- join: dys (~user@2a01:1e8:e100:8296:21a:4dff:fe4e:273a) joined #forth 09:37:56 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 09:38:32 --- quit: Zarutian (Read error: Connection reset by peer) 09:39:15 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 09:59:06 --- join: xyh (~xieyuheng@2001:250:3002:5550:6ea1:cc0f:bcb2:b187) joined #forth 10:23:31 --- quit: joneshf-laptop (Ping timeout: 272 seconds) 10:33:35 --- part: xyh left #forth 10:40:24 --- join: joneshf-laptop (~joneshf@98.208.35.89) joined #forth 11:10:29 --- quit: bjorkintosh (Ping timeout: 260 seconds) 11:15:33 --- join: Mat3 (~Mat@91.65.76.247) joined #forth 11:31:45 --- quit: Zarutian (Quit: Zarutian) 11:33:54 --- quit: Mat3 (Quit: Verlassend) 12:07:03 --- join: Mat3 (~Mat@91.65.76.247) joined #forth 12:51:05 --- quit: Mat3 (*.net *.split) 12:58:27 --- join: Mat3 (~Mat@91.65.76.247) joined #forth 13:11:49 tangentstorm: ping 13:12:30 heya Mat3 ! :) 13:13:05 been a while since i was awake when you showed up :) 13:20:52 hello tangentstorm 13:21:01 how's it going? 13:22:59 I finished my AOT compiler backend for Nim (former Nimrod) some time ago and start now porting a simple C compiler with this compiler as backend so Nim get native-code compilation support 13:23:36 cool 13:23:46 what does nim's compiler produce? 13:23:55 compiles to C 13:24:14 nice 13:24:29 is it published online somewhere? 13:24:34 * tangentstorm should really take another look at nim 13:25:09 my compiler environment have the advantage to support debugging at runtime for example (beside working bare metal at wish) 13:26:19 and i assume it's very very fast too :) 13:26:26 however I still plan to finish my retro inspired concatenative language Saiwa in between 13:26:43 ^indeed 13:28:13 * Mat3 reading the dsPIC programming manual beside 13:31:04 what happen to your projects ? 13:31:04 i haven't had a lot of time to work on my main projects for a while. i was doing a lot of client work. right now i'm working on a poker program. 13:32:14 i have started to expose some of my pascal code as a python module 13:33:56 (my terminal library). now that i know how to do it, i'm hoping to do the same for my little forthlike language. 13:35:07 well, from Python to Nim is just a small step, specially for someone who prefers a Pascal-esque syntax in combination with static type systems :) 13:36:13 yeah, i've used it before. i liked it. 13:36:23 you have written your own Forth inspired language ? 13:36:23 did you already port ngaro to nim? 13:37:14 not really. right now it's just a bunch of functions with a forth-like syntax. 13:37:15 no, I've ported my own one instead (it's now part of the compiler as immediate language) 13:37:22 but it has a really really cooll interactive shell. 13:37:28 * tangentstorm looks for screenshot. 13:38:05 http://imgur.com/gallery/zMvxYZ0 13:38:58 it's hard to tell how it works from just the screenshot, but basically there's a little popup window and as you enter commands, they go onto one of two user stacks 13:39:40 in the screenshot, you can see i'm typing '2 2 +'. if i hit enter, the number 4 appears on the line above. 13:39:56 there are keyboard commands to directly push and pop as you type 13:40:03 the interface is looking nice (a bit like the colorforth one) 13:40:28 thank you. it's all text terminal based 13:41:12 by the way, I ported the SDL based terminal code with a lot of code rewriting to Nim 13:41:47 oh yeah? 13:42:17 this was needed because I found a cheap way for possible hardware implementations 13:43:14 my console libraries right now do pretty much everything except actually display the screen. i should hook what i have up to the sdl terminal. 13:43:35 that would be nice 13:49:42 * tangentstorm resolves do spend some more time with nim when he has the chance. 13:52:21 * Mat3 thinks porting retor to Nim would offer some code simplifications 13:52:32 ^retor=retro 13:52:49 --- quit: true-grue (Read error: Connection reset by peer) 13:55:49 --- quit: nighty^ (Quit: Disappears in a puff of smoke) 14:15:49 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 14:38:50 --- join: xyh (~xieyuheng@2001:250:3002:5550:6ea1:cc0f:bcb2:b187) joined #forth 14:40:53 tangentstorm: what forthlike language you have in mind ? where are the ``functions with forth-like syntax.'' 14:54:34 hello xyh 14:54:42 yeah, that means different things to different people 14:55:05 + DocPlatypus 14:55:07 I mean, Factor is different enough from Forth to me that I'd call it something else other than Forth-like 14:55:19 I mean the inspiration is there but seriously, arrays on the stack? 14:55:33 and I think everything has a type in Factor as well 14:56:38 yes, Factor is strongly typed if I remembered correctly 14:56:52 and that's what ruins it for me 14:57:05 I'd rather the compiler/interpreter stay out of the way 14:57:22 if I want to add the number 7 to the letter "A" I know what I want 14:57:54 same with bitwise operations on booleans 14:58:32 a type system has mainly advantages for code generation so I suspect there where introduced in Factor because of that reason 14:58:53 okay, real life example in Forth 14:59:19 some of these words aren't standard Forth but it's obvious what they *should* do 15:00:00 : umd+ ( d1 d2 -- t ) 4dup d+ 2dup 2>r 2-rot dumax du< 1 and 2r> rot ; 15:00:20 the "1 and" takes the place of "if 1 else 0 then" 15:00:55 and I think I can get rid of the dumax somehow in there, haven't rewritten all of these yet 15:01:05 one can views two values in the stack such as :: (* value, type-tag -- *), as a one typed value, then one does not have to encoding type to (for example) the 64-bits fixnum 15:02:41 in some languages types make sense 15:03:02 type is important when you have to tell the differences between address and value 15:03:05 but having a separate floating point stack is as far into it as I would want anything Forth-like to get 15:03:44 I do not use Factor myself 15:03:52 can't blame you there 15:04:19 DocPlatypus: that's an obvious and nice example for the disadvantages of type systems in general but please have in mind that Factor features quotations and stack combinators for which a type system ease optimizations at code generation for sure 15:05:46 --- join: vektorman (~quassel@78.83.25.116) joined #forth 15:06:48 I think it is that Factor wants a GC for implementing list, then it has to tell the differences between address and value (to implement a GC), then it need type 15:07:38 xyh: i was just talking about the little mini forth shell thing i made for minneron. i wouldn't even call it a language at this point, though i'll probably borrow a bunch of control structures from retro. 15:08:29 right now, there's just a stack and the ability to invoke a handful of predefined functions using rpn syntax. 15:08:42 you can also put strings and numbers on the stack 15:09:05 there's no control flow and no way to edit the dictionary without recompiling yet. 15:09:27 no way to even define new routines at runtime. 15:10:12 anyhow, I need implicit type tagging for my own language because this 1. ease mapping stacks to a set of registers (specially for x86 where register access is somewhat non orthogonal) 2. detetcting logical errors at typing 15:10:31 ^detecting 15:13:25 a type system is also of advantage for memory access because the storage width is always known so pointer arithmetic can be optimized (which is impossible with Forth where register cached values can't be assumed to be pointers or numerical values) 15:14:28 okay, who knows about greenarraychips.com and their product(s)? 15:14:39 do you necessarily have to use colorForth to program these? 15:14:59 --- quit: xyh (Remote host closed the connection) 15:16:51 --- join: xyh (~xieyuheng@2001:250:3002:5550:6ea1:cc0f:bcb2:b187) joined #forth 15:17:14 I read about a programming environment based on Phyton some time ago, otherwise the chip seem to be programmable though running an eForth based interpreter 15:17:24 tangentstorm: about the design of an UI of a forth-like language, I am thinking of adding some windows (can be used optionally) to show what values are in the two stack at run time :) 15:18:17 xyh: Just sudied your home page beside and found your bare-metal code quite interesting 15:18:27 ^studied 15:19:42 Mat3: :) it is based on the code of baremetal-OS, but I re-write all of them 15:20:44 Mat3: if you have fasm and qemu. just give it a test :) it is very easy 15:21:12 for modern CPU's with AMD64 ISA you can use safety 2 MB pages, this would simplifiy your code and reduce the memory demand 15:21:30 I give it a try 15:22:27 why have you leave out the SMP initiation code ? 15:23:40 I just have implemented a new reader of numbers, just like smalltalk's, you can type ``2#-11111111'' or ``2#-1111_1111'' and ``.hex'' will print ``-ff'' for you 15:25:01 that's nice 15:25:03 just ``base#digit'' where base can even be 100, but in digit you can only use 36 chars as number 15:26:28 and you can use ``_'' to improve readability :) 15:31:53 re the above type discussion: there are often more than these three things conflated together into the word type. 15:32:50 1) bit/byte/cell layout. Something I usuall call record layout. (How much memory is this kind of thing going to take) 15:33:38 2) assertations/checkable characteristics that subsets these kind of things from other things 15:34:14 3) what is allowed to be performed to the thing(s) 15:34:43 and 3 is where I really start to dislike types 15:35:06 because sometimes it does make sense to add the number 7 to the letter "A" 15:35:36 actually I think Forth's # word called by . relies specifically on this 15:36:01 it depends on what you are doing. If you are combining huge libraries of code that nonone has the nenna (a bit like mana but just for bother) to look over. 15:38:08 or if you are adding two numbers together but one is representation of degrees while the other is radians around an vector then catching such stuff can be valueable 15:38:24 : # base @ ud/mod rot 9 over < if [ char A char 9 - 1- ] literal + then [char] 0 + hold ; 15:38:28 another advantage for non typed languages lays in the possibility for mixed-type calculations 15:38:30 this is exactly what Gforth does 15:38:53 how ever the chars, ints and such of Java/C/C++ and such are just example of the 1) I mentioned 15:39:20 yes, sometimes it's a PITA to have to store small numbers in a 64-bit cell because we're on a 64-bit machine 15:39:59 and perhaps one of Gforth's worst shortcomings... it can't run the 16- or 32-bit kernels on a 64-bit or larger machine for various reasons 15:40:18 need some sleep, ciao 15:40:27 --- quit: Mat3 (Quit: Verlassend) 15:41:39 then you have example of 2) just for pointers/reference (and the pointers/references themselfs but not the thing they point to) such as linear ownership of datastructures, that they are pointing into GC'ed memory or manually allocated and such. 15:43:31 Zarutian: if one just use two value in the stack as one typed value, then the ``bit/byte/cell layout'' just becomes ``two cells'', uses more spaces, but the operations of type can be very flexible. the type-tag of (* value, type-tag *) can even be an index of a hash-table of strings (as symbols) 15:44:32 xyh: I havent even gotten to the static versus dynamic checking yet 15:46:24 you probably have heard the term leaky abstractions? Forth isnt as much leaky as being transparent from the top to the bottom 15:47:53 Zarutian: you mean ``two cells'' abstraction is vulnerable ? 15:59:59 --- quit: xyh (Quit: 再见 :)) 16:26:16 --- join: xyh (~xieyuheng@2001:250:3002:5550:6ea1:cc0f:bcb2:b187) joined #forth 16:26:59 Zarutian: I have just readed :: http://www.joelonsoftware.com/articles/LeakyAbstractions.html 16:27:02 :P 17:13:58 --- quit: vektorman (Remote host closed the connection) 17:31:04 --- quit: joneshf-laptop (Ping timeout: 244 seconds) 17:33:04 --- quit: xyh (Ping timeout: 260 seconds) 17:38:28 --- join: joneshf-laptop (~joneshf@98.208.35.89) joined #forth 17:40:49 for when xyh return: I mean that Forth is unabashed in recognizing the leaky abstractions and ?trasperancifing? them. 17:58:35 DocPlatypus: There's two things I can answer about Greenarrays. 17:58:52 First, no, as of fairly recently, you don't need to use Color Forth to program the chips. 17:59:29 Second, you're not going to be able to program the chips anyway. Any company that has a four month turnaround on commercial queries is not a company that is going to last much longer. 18:05:01 ttmrichter: I have had about four hour turnaround for semi-ccommercial queries to ATMEL (long back though) 18:06:44 ttmrichter: okay, what do you recommend as an alternative? 18:06:54 GA144 is probably overkill for my needs anyway 18:09:08 ideally something I can write my own Forth for instead of having to use the one that the vendor drops in my lap 18:09:48 DocPlatypus: PIC32? (which is a MIPS one) 18:10:15 DocPlatypus: some ARM M4 core that seems really popular nowdays 18:10:21 ok 18:11:05 for what I'm thinking about doing I may need two... based on what I've read about DMD-era pinball games I think a separate CPU does the display versus the game logic 18:11:12 the game logic CPU wouldn't necessarily need to be that beefy 18:11:30 but for the type of display I am doing, that one would be 18:12:50 DMD-era? 18:13:03 this isnt some physical pinball machine? 18:24:39 DocPlatypus: In the space the GA144 occupies, there's not a lot. I think the Propeller is the only other small multi-core chip taht could do the trick. 18:24:59 Zarutian: this is a phyiscal pinball game. 18:25:04 Honestly if you want a Forth core you might be best off making one on your own with an FPGA. 18:25:05 DMD era refers to games made after about 1990 18:25:20 ok... pinball history lesson 18:26:19 In about 1977 or so manufacturers started making the switch to solid state. Early solid state games had only numeric displays (usually 7 segment but Gottlieb used a different type that had center segments and allowed the "1" to be in the middle and not on the side) 18:26:38 ttmrichter: have you looked at those FPGAs that ATMEL made (and still makes)? From what I gather the bitstream isnt obfusicated or propertarieized. 18:27:00 Atmel? Or Actel? 18:27:07 ttmrichter: Atmel 18:27:23 In about 1986 or so came the transition to alphanumeric displays. Then in 1990 the first dot matrix display games came out. These were still physical pinball games, the display was a dot matrix display. 18:27:27 Didn't know they made FPGAs. 18:28:05 Now, there were a couple of oddball video game/pinball hybrids... Baby Pac-Man, Caveman, Granny and the Gators 18:28:06 ttmrichter: they dont make a lot or varied but it is aviable afaict 18:28:20 the first two in 1981, the last in 1984? 18:28:21 Hmmm... I can only get components, not boards. 18:28:50 Caveman was in a regular pinball cabinet and kept score on a regular numeric display, Baby Pac-Man used a CRT for the maze and for scores 18:29:01 ttmrichter: well, it looked to me too that one has to rig something up oneself. 18:29:21 anyway... except for the Pinball 2000 platform every game until Jersey Jack's Wizard of Oz has used a dot matrix display 18:29:31 (after 1990) 18:29:40 DocPlatypus: please continue. We ttmrichter arent ignoring you as IRC has not crosstalk problem like audio conversation. 18:29:52 I've said everything I think 18:30:02 dot matrix display in colour or just grayscale? 18:30:27 dot matrix in grayscale. Stern is just now upgrading to color. there are aftermarket color DMDs for a lot of older titles though 18:30:30 Wow! They're expensive too! 18:30:44 now JJP is using a full 26" or so LCD display 18:30:47 full color 18:31:18 it would be kind of cool in a way to shut off the pinball part of the machine and watch TV between games 18:31:55 ttmrichter: what I want to know is if the logic/gate/lut density is comparable to those of Xilinx, Actel and other big names in FPGAs and CPLDs. 18:32:14 The biggest one they make has 50K gates. 18:32:25 yes, let me repeat that so it's not lost on everyone. until recently pinball games used the same monochrome dot matrix plasma displays for most of two decades 18:32:39 That's for a LQFP144. 18:32:39 ttmrichter: what I like is the orthogonality of the cells 18:32:51 mainly because Stern had no real competition. but that's another rant for another day 18:33:33 btw... I'm talking about "new Stern"... the original Stern went out of business, and Gary Stern went on to run Data East's new pinball division which later became Sega which then became Stern Pinball, Inc. 18:33:40 DocPlatypus: maybe also because they had lot of dot matrix plasma displays versus amount of pinball machines they could sell ratio? 18:34:06 Zarutian: that could be it. and why screw up the cost and economy of scale when nobody else is making the damn things? 18:34:32 actually in 2008 the economy almost took Stern right out 18:35:11 DocPlatypus: I heard that the most sought after component/subassembly of pinball machines and video game cabinets are the coin acceptors. 18:35:32 Zarutian: maybe by route/location operators 18:35:58 a new trend, though, is membership-based arcades where the games are on free play 18:36:15 at least how I read the law in Texas, the game is legally not a coin operated amusement device if set on free play 18:37:47 I never understood why those things couldnt just accept something like 1wire esque medalion or such, in a slot. Then the cabinets could offer save games, item storage and limited value storageþ 18:37:52 s/þ/./ 18:37:53 no, the most sought after components are "unobtanium" chips 18:38:15 Votrax SC-01(A) and other ICs no longer being made 18:38:19 what are those? old rom chips or sound chips that noone knows how to make any more? 18:38:34 ever play Q*Bert? 18:38:43 Votrax? sought after by chiptunists and such? 18:39:02 the sound of Q*Bert swearing (and Ugg, Wrong Way, Slick, etc) is made by feeding random phoenome data to a Votrax SC-01 18:39:10 it's a speech synth 18:39:31 the Black Hole pinball among others used it for actual speech as well 18:39:50 DocPlatypus: I dont think so. The only arcade here in town/island is also a bowling alley and their selection wasnt/isnt much. 18:40:10 well it's a new trend, it's not happening everywhere 18:40:21 some places are "pinball deserts" still 18:40:25 DocPlatypus: most video cabinet games I am familar is through MAME 18:40:35 pinball deserts? 18:40:52 yes there is also a PinMAME though it's less useful because you need a playfield to play 18:41:04 Zarutian: Places where pinball is basically not played. 18:41:09 desert meaning no pinball games for X miles... usually ~50 or more 18:41:18 so what is so special about the Votrax chip? 18:41:39 they quit making it but the games were designed for it. there is one guy that still has them and they sell for ~$70 each. 18:42:20 has it been decapped and visual6502-ized? 18:42:48 I think technically enough info is available to emulate it 18:42:58 and emulate it well? 18:43:59 I can actually understand why "just emulate it" might not be a viable answer. 18:44:03 It's irrational, but there. 18:44:11 I want a PDP-8/E or a PDP-8/M. 18:44:23 I haven't looked into it. for running in the games themselves, a board would have to be made to plug into the original Votrax's IC socket 18:44:23 The real thing, not an emulation or an FPGA clone. 18:44:38 I mean emulate it so well that it gets the quirks right 18:44:40 DocPlatypus: Making a board to do that would be the trivial part. 18:44:58 as long as there are still some Votrax chips out there to replace the ones that croak, the incentive really isn't there 18:45:23 well, why not decap the ones the croaked. 18:45:32 I think the patents have expired. only issue might be the copyrights on chip masks 18:45:35 Zarutian: Even if it's a perfect emulation ... it's still an emulation. If someone gave me a perfect emulation of every quirk and fubar of the PDP-8, I'd still want a real machine. 18:46:20 in the case of the Votrax, I wouldn't care as long as I heard Q*Bert "swear" as I remembered it, or "do you dare to enter the Black Hole?" etc 18:46:24 ttmrichter: would molecularly nanotechnologically assembled PDP-8 do or is it deeper than. 18:46:37 that. 18:46:44 If it were a component-by-component rebuild that wouldn't bother me at all. 18:47:15 even if the components are manifactured nowdays? 18:47:17 As long as I've got the living, breathing machine, complete with its digital soul, hulking over in the corner mocking me with its slowness and its limits. 18:47:27 Yep. That would suit me fine. 18:47:40 Build it from original blueprints but with components made yesterday? No problems. 18:47:50 (Good luck on getting core memory made, though.:D) 18:48:19 core memory? isnt that basically grid of those ferrit magnetic cores? 18:48:38 Tiny magnetic donuts threaded together crossways in a grid, yes. 18:48:57 those cores that sometimes show up in various electronics and cost around a provibial nickel? 18:49:39 well, they will sure ship the cores and the wires but it is up to you to thread the thing together 18:49:45 They show up in eBay at points, but I'm not so sure about the nickel part. 18:50:00 Right. Because actually assembling those is the expensive part. :) 18:50:10 It takes incredible amounts of manual labour to build even a small core. 18:50:56 I am talking about the rings. I have seen those things in damn near any electronics device I have taken the cover off. (Usually as ferrit core for some inductor) 18:51:01 Quickly eyeballing eBay, you're looking at 2-300 dollars for a bank of core. 18:51:17 The cores in core memory are about 0.5mm across. 18:51:22 how much do the materials cost ya rekon? 18:51:39 Materials costs are negligible. 18:52:05 http://www.ebay.com/itm/Vintage-DEC-Digital-pdp8A-pdp8-pdp-8A-16k-x-12-Iron-Core-Memory-H219B-G850-/290977043330?pt=LH_DefaultDomain_0&hash=item43bf955382 18:52:30 perhaps some hackerspaces can show people how to make the core memories and can then sell them for much more. 18:52:39 Hmmm... Not seeing the actual core matrix there. 18:53:33 So that's $200 for a board which may or may not work manufactured almost fifty years ago. 18:54:28 Not including shipping. The shipping would be more than the board. 18:55:02 There's an actual PDP-8: http://www.ebay.com/itm/Digital-DEC-PDP-8-L-Classic-Minicomputer-w-core-memory-vintage-PDP8-/141321724323?pt=US_Vintage_Computers_Mainframes&hash=item20e76e61a3 18:55:07 Only $4000! 18:55:19 As-is. Probably untested. 18:56:11 Anyway, I'm off to torture some students. 18:56:19 humm... reminds me that Landsnet (power distributer in my country) is upgrading some of their control computers from old DECs (not PDPs though) 18:57:02 and I am off to bed 18:57:23 wow. the DECs are old, I think the Therac-25 units were controlled by PDP-8 computers 18:57:26 --- quit: Zarutian (Quit: Zarutian) 18:57:36 that was a good 30-ish years ago 19:51:44 --- nick: aksatac___ -> aksatac 19:59:23 --- quit: proteusguy (Ping timeout: 258 seconds) 20:06:19 --- quit: DGASAU (Remote host closed the connection) 20:07:28 --- join: DGASAU (~user@p50993595.dip0.t-ipconnect.de) joined #forth 20:12:33 --- join: proteusguy (~proteusgu@ppp-110-168-229-72.revip5.asianet.co.th) joined #forth 23:22:00 --- quit: proteusguy (Remote host closed the connection) 23:48:20 --- join: proteusguy (~proteusgu@183.89.94.164) joined #forth 23:59:59 --- log: ended forth/14.09.29