00:00:00 --- log: started retro/10.02.28 06:34:36 --- quit: luptenschteiner (Ping timeout: 240 seconds) 06:36:05 --- join: luptenschteiner (~User@unaffiliated/probonono) joined #retro 08:58:37 --- join: Raystm2 (~1808e8d4@gateway/web/freenode/x-zesfbrcjikfyxkos) joined #retro 08:58:48 :) 08:59:15 crc, I've been talking with folks that UNIX/linux but want to colorForth as well. 08:59:37 My suggestion is that, with the newer systems we can do like Greg Bailey... 09:00:10 When in Linux and wanting to edit-read or write colorForth, just about any forth can be used to accomplish this 09:00:29 one would require the knowledge of colorForth to provide such an editor. 09:00:49 I think I personally will use the things found in your systems to provide this editor. 09:01:05 The OTHER part of the problem is code execution in a HOST. 09:02:29 Native colorForth and it's Round-Robin tasker require that colorForth be running as if it were the operating system taking up much of the cpu time -- i am not saying this correctly. 09:04:49 But what I was thinking is that UNIX systems can provide virtual consols to applications, and I was thinking of using a vc as the colorForth ap and then just providing those things that the colorByte code portion of a colorForth need by the cunning usage of any and all libraries and tools that will help UNIX communicate thru the SERVE extension to the round-robin. 09:07:39 Windows and Native colorForth have environmental blocks that get loaded when colorForth recognises the host. I assume that the same can be provided for Linux. 09:10:01 I assume I can get a colorForth code executer out of the cunning usage of your own vm. 09:11:26 ANd then, rather then being like Greg and using a propriety system to integrate colorForth into the windows host, I can leave much of that open and attachable ( hooks?) so that Linux users can have flexibility when they integrate into the system. 09:11:51 visible open hackable hooks 09:12:10 SO, the upshot is... 09:12:43 I think colorForth is about to become a RetroForth-ish project, officially, from my perspective. 09:23:05 lol, I suppose anyANSIForth can just load a config file or programel or script that can provide for navigation by block and proper reproduction and construction of the bytecode. 09:25:13 hi Raystm2 09:25:24 I'll write that in G and then test in Swift and Win32... That will be easiest and then I can convert to Ngaro. 09:25:28 How are you? 09:25:48 doing pretty good 09:25:56 Family okay? 09:25:59 Raystm2: retro now has file i/o and sockets :) 09:26:03 family is good 09:26:13 OH hey, that _is_ good news! 09:26:15 tammie is tired a lot (due to give birth in 6 weeks) 09:26:42 WOAH I didn't know! congrats! Girl or the other kind? 09:26:49 another boy 09:26:54 very cool. 09:27:13 this makes _____ kids ( a number is required to proceed ) 09:27:24 3 boys 09:27:44 lol shows how much I know. thought you had a girl in there... 09:28:09 WELL good on you and Tammie! 09:28:20 Hope mom is doing well. 09:29:04 other than tiredness, she's doing ok 09:29:23 I'm begining to become sensitive to the fact that linux users would like to colorForth, as you can see. 09:29:35 Personally, my best idea is to build on your own work. 09:30:28 And there is many reasons, most of which is I consider you to be the one of the most important forth specialist on earth. I don't think very many people have as many hours coding it competently as you do. 09:31:20 My project can branch from success and continue in that vain. 09:32:17 And I can help to provide an open source -- one-size-fits-nearly-all -- solution to the community for continuity sake if nothing else. 09:32:57 But to ease this, I do think I will start with the ANSI converters. 09:33:33 I can use the great work of Quartus to do something similar and in reverse of what he did on the ANSIRetro. 09:35:16 I also think that much of the colorForth kernel sources in the past can just be lifted and moved to these solutional systems. 09:36:12 OH request, please... 09:36:59 Is it possible to link colorForthRay at the new retro page as some directories point to the ray.rx link in ref to cfRay and that points to retroForth page. 09:37:09 Or is there maybe a better solution? 09:37:21 Raystm2: I'll get that fixed 09:37:34 you are so kind :) ty :) 09:44:01 I suppose I'll start documenting the situation in rj's chat. 09:47:57 [crcx/retroforth] 1ef713: add initial library to simplify server creation 09:50:48 Raystm2: do you know vx32? 09:51:01 I wonder how difficult it would be to run colorForth on top of it 09:51:22 searching to be sure I know what you are refering... 09:51:49 http://swtch.com/~rsc/papers/vx32/ 09:52:50 afaik google's NaCl is quite similar, but I don't know NaCl 09:56:36 I suppose what is required is OS level and possibly not applications level. This does not pre-suppose that some applications level versions are also redeemable. 09:57:59 The program -- when native rightly assumes it is the system. So, it requires zero-level clearance of the system. 09:58:28 if there is a vm that does that, I would like to give it a read-thru. 09:59:22 sorry, zero-level as in RingZero 09:59:56 this has to do with protections and also, I believe how the system calls are operated. 10:00:22 It would be more appropriate to think like GRUB. 10:01:22 one wonders of GRUB allows for switching focus between the partitions... allows for communication between running individual hosts on same machine ala linux/cf 10:02:42 something like that might make a nice compromise. Doesn't help much if your CPU is not truly Intel or good clone, but. 10:05:38 In the past, I have colaborated with Tim Neitz and we have used his code to communicate each of us using a colorForth box running his software and connected to a linux box running internet services and LAN services to the colorForth boxs. 10:06:36 I am proposing just moving that idea to the same machine, being able to switch back and forth and also communicate back and forth or use the others services from either at the same instance in time. 10:07:10 the machine would litterally run a linux task followed by a colorForth task and always swapping. 10:07:15 just like colorForth 10:08:51 but you could selectively focus one or the other system as the currently displayed and controlled system. 10:08:56 [crcx/retroforth] cf1a24: rename "accept" to "incoming" in sockets library; ... 10:12:40 yiyus gosh that's it! 10:13:52 the vm needs to sit where vx32 sits in the system. maybe a version of vx can become the vm that provides room for linux to run but this version of vx is also a completely contained colorForth. 10:14:58 [crcx/retroforth] b05fff: add http/0.9 server to rxweb 10:15:12 colorForth could be part of a ring-0 accessable vm that provides room for linux to co-habitate and run in that vm. 10:15:34 http/0.9 is brutally minimal, but it works enough for current purposes 10:15:56 (plus, it's just 4 words, with the server library loaded) 10:16:06 :) sweet. 10:16:09 http://retroforth.com/paste/?id=2013 10:17:00 wow that's it 10:17:04 neat! 10:18:03 Raystm2: I'm sorry I don't know any more about colorForth or vx32, but I can tell you 9vx (the vx32 version of plan9) is great 10:18:07 lol an aside project -- can colorForth provide room for linux to run along side. 10:18:23 and it runs an (almost unmodified) plan9 kernel writen for the bare metal 10:19:29 kc5tja used to be a plan9 enthusiast and I only gave it a cursory glance. I shall look again per your recommendation. 10:20:33 i see a lot of similarities between colorforth and plan9 (from an historic point of view) 10:20:40 This is the info gathering time and not the time to be trashing possible options. I take plan9 and all under advisement. Thank you yiyus. :) 10:20:42 plan9 is to unix what colorforth is to forth 10:21:08 I did not know. 10:21:39 * Raystm2 linking to plan9 info... 10:22:27 created by the same people, but a fresh start (rejecting standards), a simplified but more powerful version 10:31:30 I have gotten off on the vm tangent again. Wondering if the colorForth SERVE Extended Round-robin Vector Extension can become the SERVE Extended Round-robin Virtual-machine Extension. 10:31:57 This would mean that one would boot linux in colorForth. 10:34:56 [crcx/retroforth] f0cc0c: fix compile error in sdl port (Marc) 10:34:58 or rather, that colorForth would boot first, and then executing some colorForth code would read in the linux boot disk and provide space for linux to load. 10:36:33 not the best idea as colorForth crash should be handled gracefully by linux. 10:37:35 I think I might like to try my hand at a python interpreter for colorForth bytecode as well. 10:37:46 that's instantly portable. 10:39:04 I get the impression that some of the tryers would like to use the arrayForth kernels in linux and this is why they are trying vms. 10:39:35 but colorForth doesn't like the vm idea. It doesn't want anything but ring0. 10:40:08 really, practical considerations of arrayForth in linux... 10:40:45 what one wants is access to the colorForth byte code to run the simulators and other code, and a way to connect to testbeds and chiparrays. 10:41:53 connecting to testbeds and chiparrays is much more flexible and better covered in linux packages then in Windows systems, and therefore I assume that a linux user does not really want to run the arrayForth windows programs. 10:42:33 They just want to drive those linux set ups with the bytecode. 10:43:56 so I'm thinking that being able to view read and write the byte code is one thing and then executing in conjunction with libraries of functionality pertaining to linux drivers of hardware is the requirement. 10:58:08 hi Raystm2, hi everyone 10:58:16 hi docl 10:58:28 hi crc 11:00:00 hi there docl, Hows you? 11:00:35 pretty good 11:00:56 * Raystm2 ran out of sweetener and the one spoon of sugar in each of the last two cups of coffee has me spazing out on guitar. :) 11:01:08 I've been watching youtube videos on Bio-Printing 11:01:42 is that a quazi-science or something real? I know absolutely nothing about it. 11:01:46 aka organ printing, tissue engineering, etc. 11:01:51 it's real! 11:01:54 OH 11:02:20 apparently they use actual inkjet printers to do this stuff 11:02:34 youmean like printing up an organ with chemistry and tools and getting real working stuff. 11:02:41 exactly 11:02:41 I saw some of that. 11:02:46 it's amazing! 11:02:46 pretty neat. 11:03:03 it's essentially organ cloning 11:03:26 ya. growing and harvesting. 11:03:28 and it will save millions of lives 11:03:44 no more need for donor organs 11:04:00 if we grow full bodies with out brains we'll have batteries with-out all of that nasty matrix problem 11:04:03 hehe :) 11:04:12 actually one technique does use donor organs, but uses a detergent to take the cells out so the collagen scaffold remains 11:04:25 neat. 11:05:18 it's like the obvious route to solving aging... of course we'd need to cure alzheimer's and such as well 11:05:44 possibly you would just add neuron stem cells every few years 11:06:12 let the brain replace itself slowly, without losing the information 11:06:48 anyway... it could add several decades for many of us even without that 11:06:56 diabetes is toast! 11:07:00 yay :) 11:07:18 bake me some of that diabetic toast too. 11:07:21 :) 11:07:26 * crc looks forward to a replacement pancreas one day :) 11:08:27 * Raystm2 is looking desperately at the possiblility of diabeties brought on by massive dehydration, and is hoping his own situation is at least partialy reversable. 11:08:55 http://www.youtube.com/watch?v=0r_TJd6O5M8 11:10:01 I'm thinking I should try to get into this... they need computer techs to design the printers and stuff 11:10:28 I saw that in your forum post on cryonics 11:11:29 docl, you hit it brother. 11:11:32 that's the plan. 11:11:40 to beable to create devices. 11:12:08 I've been creating my own products for market for 20 years for other people. my turn now! muwwhahhhaaa 11:16:11 personally, I want a phone that is my mobile computing platform, my remote control to everything in my life, my keyless entry to car and home, and a cigarette lighter to boot because, afterall, i'm a gentleman. 11:16:29 haha 11:17:44 basically, think "You can be only one" thing in your pocket. Especially with the new RonCo SwissArmy knife attatchment. 11:18:08 by Leatherman. 11:18:54 when your hunting knife gets gps, full stoneage vs spaceage integration. lol 11:19:19 well, in a few years your swis army knife can be made of nanobots that reshape into anything you want :) 11:20:16 I always assume that with nanobot tech there will also be bio-bot tech as well. 11:20:19 maybe wear it as a hat or an armband when not in use... inflates into a computer, bicycle, or phone 11:20:37 I like the transport idea. 11:20:56 even wheeled shoes. 11:21:56 [crcx/retroforth] 20c147: properly fix devices (my mistake) 11:24:59 I want a set of rocket shoes that levitate me a few inches from the ground and move me whatever direction I lean. kind of like a ginger scooter, but with compressed air instead of wheels. 11:26:17 it would be computer controlled to prevent me from tipping over 11:26:56 [crcx/retroforth] 9e39a4: fix copyright notes on graphical vm 11:28:39 it could use a cochlear operating gyro. 11:28:48 just like you do. 11:36:40 --- join: Mat2 (~5f590857@gateway/web/freenode/x-epuvuiqsvtjfacjq) joined #retro 11:36:51 good afternoon everyone 11:37:08 howdy Mat2, welcome back :) 11:37:25 hi Ray, what's going on ? 11:39:30 linux colorForth birth for me... 11:39:50 crc working on retro commits 11:40:01 hmm, I suppose no native version but one for a MISC style vm like ngaro ? 11:40:05 docl is fixing the world with nanobots. 11:40:57 right. the colorForth doesn't like to play hosted because it needs ring0 protections access to the machine and os's don't usually provide anything less then ring3 even to vms. 11:41:36 it needs operating systems level vm access and not necessarily applications level. 11:41:45 I have here a working parser and interpreter stub for a functional coloforth dialect in C++ 11:41:46 weird how most OS's can be hosted at the APs leve. 11:41:54 Sweet! 11:41:59 features? 11:42:23 multithreaded interpreter 11:42:38 word size agnostic 11:42:39 very nice. 11:42:48 cool cool. good thing 11:42:54 I hope to finish it next week 11:43:15 excellent! 11:45:07 another inspiration for it was muforth and colorless colorforth 11:46:07 (utf-8 for color-token encoding) 11:46:09 ooh! Terry's colorless. hmmm... 11:46:25 thanks Mat2 for reminding me. 11:47:03 that's because I need a simple way for accessing text files 11:47:27 I see. Great idea, I think. personally. 11:48:35 have you take a look about the rainbowforth version in javascript ? 11:49:04 it works completly independent from any host installation 11:49:13 Howerd Oakford has always included an mode in his colorForth editor called colorBlind mode, and this colorBlind mode is accomplished by instantly translating the old forth puntuation in place of and in conjunction with the colors. it ends up looking like standard forth. 11:49:35 exact my approach 11:50:00 rainbow is too slow for me to be interested in trying to develop code examples for it. but I keep a copy and plan to look it over in detail over the next few months. 11:51:50 afk brb. I need a meal. :) 11:52:13 okay 11:54:48 crc: the new words for file support working nice 11:58:34 hi Mat2, how are you doing? 11:58:48 have you already uploaded your code somewhere? 11:59:11 not yet as the interpreter isn't finished yet 11:59:23 Mat2: good 11:59:32 Mat2: I added sockets last night too 12:00:29 yiyus: i'm puzzled at current about the best way to store pointers of class methods 12:00:52 i don't really know c++ (just the most basic concepts) 12:01:01 i never liked it 12:01:23 yiyus: the current code is a dirty trick which only works with recent gcc compiler (that's my problem at moment) 12:02:35 i'm not a fan of it either but it have some advantages against plain C 12:02:47 (the yntax is a mess) 12:02:50 syntax 12:03:10 crc: fine, I will play a bit with it tomorrow 12:03:57 yiyus: for sure the source code will be uploaded next friday at last 12:04:16 yiyus: together with a colorforth interpreter as example 12:04:26 nice, very nice 12:05:01 althugh anyway i'm quite busy now, i already was in the middle of a go ngaro rewrite when crc started to add new features ;) 12:06:17 hmm, I think it would be a good idea to create some kind of funture planning for new releases, crc ? 12:06:50 so we can synchronize our working better 12:08:16 --- join: cfa (~cfa@unaffiliated/cfa) joined #retro 12:08:18 hi 12:08:40 hi cfa 12:08:48 hey Mat2 12:09:10 crc: what about some kind of planning for new releases ? 12:09:33 well, i just try to keep it in sync but i have about 0 users, so it is not really an issue 12:10:52 :D, I know one user, a collegue of mine 12:11:58 and you can't forsee where and why make use of your code 12:12:18 Mat2: glad to know, but... why? is he using any of the extra features? because it doesn't have much advantages over fast-console except concurrency 12:12:30 and my plan was to change the way it works 12:12:32 yes, that's the reason 12:13:17 i'm moving concurrency to an optional extension (at compile time, for the moment) 12:13:42 with the extensions system i'm introducing, it is easier adding instructions than using the ports 12:14:04 i'm not sure yet how well it will work, but we will see 12:14:16 ah, you will using a similar approach as pAvm ? 12:14:40 yes, i think so 12:15:29 the only problem i see is that i would prefer to keep instructions at minimum 12:15:42 but if you want to add functionality you have to add it somewhere 12:15:58 [crcx/retroforth] 640526: fix return codes for "send" 12:16:03 yes, but to be effective you need a well balanced instruction set 12:16:56 [crcx/retroforth] 1c8c79: add sockets to console-fast 12:18:42 the pAvm interpreter is quite simple, each fetched word is a pointer to a class method executing an immediate stream (that are the instruction sets) 12:19:14 max. 255 interpreters for instruction sets can be used at moment 12:19:59 what i'm doing is reserving the last byte for instructions, and the rest (from a 32bits int) as extension id 12:20:39 when an instruction is not found in the current loop, the right extension is searched and it takes execution 12:20:57 which can hold in its own loop or give back to the core one 12:21:41 but as you can see, this is far from optimal with many changes between instruction sets 12:23:02 that's why each interpreter engine in pAvm implementing an JIT compiler which compiles a instruction stream into direct threded code 12:23:05 threaded 12:23:26 executed is always the compiled stream 12:26:23 so basicaly a specific instruction stream is compiled into a thread which get an ID 12:29:44 this stream can be executed from the interpreter in stream mode (the thread is repeated conditionally), as parallel thread or serial (all other threads are blocked though execution) 12:31:23 the interpreter fetching a cell in memory and executing one specific engine method 8stream, thread, serial) which executing the precompiled thread which id is the immediate parameter (next cell in memory) 12:31:45 it's sinmply some kind of indirect threded interpreter 12:31:48 simply 12:31:52 threaded 12:33:17 this way the C++ compiler can compile the dispatch routine to an indirect load and branch instruction 12:33:36 (if one find the right compiler flags) 12:35:56 and if I found a standard conform way to get a pointer from a class method 12:37:19 good luck with that 12:39:18 if i'm finished I will convert to D ;) 12:51:43 time for lunch, brb 13:32:28 --- quit: ncv (Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090520, built on: 2009/06/08 23:52:48 UTC http://www.kvirc.net/) 13:42:12 ciao 13:42:16 --- quit: Mat2 (Quit: Page closed) 13:46:09 bbl, going out with the family for a couple hours 14:27:12 --- quit: retro-commit (Ping timeout: 265 seconds) 14:29:16 --- quit: crc (Ping timeout: 258 seconds) 14:41:24 --- quit: Raystm2 (Ping timeout: 252 seconds) 15:40:33 --- nick: luptenschteiner -> probonono 17:31:14 --- quit: virl (Remote host closed the connection) 17:52:31 --- join: crc (~charlesch@71.23.210.149) joined #retro 17:52:50 --- mode: ChanServ set +o crc 17:57:40 --- quit: crc (Ping timeout: 260 seconds) 18:05:36 --- join: crc (~charlesch@71.23.210.149) joined #retro 18:05:37 --- mode: ChanServ set +o crc 18:06:04 accidentally knocked over the router and disconnected the wimax modem :( 18:41:43 retro-fast now has sockets; and a bugfix to the hangman game is also pushed to the repo 21:14:42 --- join: crc_ (~charlesch@71.23.210.149) joined #retro 21:15:07 --- quit: crc (Ping timeout: 265 seconds) 21:22:37 --- quit: cfa (Quit: cfa) 21:49:06 --- join: crc (~charlesch@71.23.210.149) joined #retro 21:50:02 --- quit: crc_ (Ping timeout: 240 seconds) 22:32:33 --- quit: crc (Ping timeout: 264 seconds) 22:36:47 --- join: crc (~charlesch@71.23.210.149) joined #retro 22:45:35 --- join: crc_ (~charlesch@71.23.210.149) joined #retro 22:47:49 --- quit: crc (Ping timeout: 268 seconds) 23:43:23 --- quit: crc_ (Ping timeout: 245 seconds) 23:47:29 --- join: crc (~charlesch@71.23.210.149) joined #retro 23:59:59 --- log: ended retro/10.02.28