00:00:00 --- log: started retro/11.01.26 01:42:47 --- quit: aisa (Read error: Connection reset by peer) 01:42:47 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 01:42:48 --- nick: aisa_ -> aisa 04:06:29 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 04:06:30 --- quit: aisa (Read error: Connection reset by peer) 04:06:30 --- nick: aisa_ -> aisa 04:09:48 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 04:10:07 --- quit: aisa (Read error: Connection reset by peer) 04:10:08 --- nick: aisa_ -> aisa 04:27:29 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 04:27:30 --- quit: aisa (Read error: Connection reset by peer) 04:27:30 --- nick: aisa_ -> aisa 04:30:39 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 04:30:40 --- quit: aisa (Read error: Connection reset by peer) 04:30:40 --- nick: aisa_ -> aisa 05:17:51 --- join: Sgeo_ (~Sgeo@ool-18bf618a.dyn.optonline.net) joined #retro 05:20:37 --- quit: Sgeo (Ping timeout: 240 seconds) 07:05:33 a possible cleanup: http://paste.lisp.org/display/119054 07:09:08 hmm, i personally find the original clearer 07:11:46 ok, I'll leave the original in place; thanks 07:17:34 np 07:34:47 --- join: qbject (417dcfe2@gateway/web/freenode/ip.65.125.207.226) joined #retro 07:41:14 --- join: aisa_ (~aisa@c-68-35-167-179.hsd1.nm.comcast.net) joined #retro 07:41:32 --- quit: aisa (Read error: Connection reset by peer) 07:41:33 --- nick: aisa_ -> aisa 08:29:41 --- nick: qbject -> [qbject] 10:30:47 --- nick: [qbject] -> qbject 11:03:35 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #retro 11:05:13 Hi, roarde. Are you the source of the syntax highlighting that crc tweeted about this morning? 11:05:33 qbject: that would be cfa 11:05:57 qbject: roarde is working on the jpico highlighter 11:05:58 nope. I'm still stuck. I'm working on joe, that was vim. cfa said he'd deliver yesterday, and there it is 11:06:27 I said I'd deliver (indistinguishable) and well . . . 11:06:30 Ah, gotcha. 11:06:48 Congrats to cfa then. 11:06:57 yep 11:07:22 I've only played with joe a little. Why did you decide to focus your efforts there? 11:08:23 crc liked nano, I grew tired of vi and was familiar with wordstar. Nano doesn't have macros, joe does, and can be called as jpico or jstar. 11:09:05 the current joe highlighter is in the repo, as well as the old nano one, and cfa's vim one 11:09:34 joe highlighter do trailing space? 11:11:01 * crc doesn't recall 11:11:29 comes with unstable, or where is it/are they? 11:11:33 the box I'm logged into presently doesn't have syntax highlighting enabled in joe 11:11:43 in the retro11 repo 11:11:50 thx 11:12:02 http://rx-core.org/rx.fossil/dir?ci=66e0437eee7155b0&name=extras 11:16:28 I'll do top-of-the-list stuff. Trailing space, tabs, then what other 2 or 3 things? 11:17:47 those, and highlighting the combinators would be pretty good 11:18:36 first, I'd have to know a combinator if I saw one :D 11:19:51 dip, sip, take, curry, if, ifTrue, ifFalse, when, times, iter, iterd, while, bi, bi*, bi@, tri, tri*, tri@, cons, each@ 11:21:15 Is there any chance at all that the first character of a word will ever be newline? 11:22:35 and if all newlines are CRLF, should a macro correct it to a copy, or highlight? 11:22:58 (I assume you're at work, I'll be around to wait or look at logs later) 11:24:08 nm q #2: obviously highlight CRLF as bad and have macro avaiable manually to fix it 11:24:45 no chance 11:25:34 tab, cr, lf, and space are not going to be recognized as part of a word name under any normal circumstances 11:40:35 * cfa unidles 11:40:37 hi all 11:42:22 --- join: roarde1 (~roarde@adsl-074-247-008-005.sip.ags.bellsouth.net) joined #retro 11:42:32 --- quit: roarde (Ping timeout: 255 seconds) 11:42:41 --- nick: roarde1 -> roarde 11:42:51 --- quit: roarde (Changing host) 11:42:51 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #retro 11:46:09 Hi cfa 11:46:18 hey qbject 12:03:27 Heard you rigged up vim syntax highlighting. Very cool. 12:13:33 thanks... but it's just a modification of my syntax file for retro 10 which was, in turn, a modification of the standard ANS syntax file bundled with vim. (as such the file's quite messy.) 12:14:53 works well for me though. if you use it you'll likely find that some core words are missing and other words that are no longer core (due to deprecation) might still get highlighted. just let me know and i'll patch accordingly. 12:16:52 (there should actually be a slightly better version in core soonish) 12:22:45 I'm actually wondering if I can implement an editor inside retro and get it working without highlighting. 12:23:10 I'm sort of a quixotic bootstrap junkie that way. 12:23:49 That would be my learning-retro project. 12:23:58 there's already a block editor implementation 12:24:07 you're thinking of a file editor though, i expect? 12:24:53 Yes. File/text. Something that would eventually form a comfortable environment for natural-language writing at some point in the future. 12:25:15 why not take a look at how the block editor is written as a starting point? 12:25:41 Sounds good. 12:25:45 http://rx-core.org/dev/playground.fossil/artifact?name=ae8e2736d12b9eecd799bfd38e7ebfee1faa77b3 12:25:53 But I'm guessing it doesn't do highlighting? 12:25:58 nope 12:26:38 Is colored text outside ngaro's skillset? 12:27:23 no, though i'd imagine that certain VM implementations mightn't support it 12:27:31 the standard C ones does, though 12:27:38 the editor employs colour for the status line, for example 12:27:58 Oh cool. 12:28:33 look at the opening definitions in the editor' chain -- textColor, statusColor and uiColor 12:29:01 they use words from the console' vocab 12:31:09 Oh yeah. Right there. 13:00:57 cfa: May I ask what you do with Retro? 13:01:36 i'm just experimenting in/with it 13:02:29 though i'm thinking of writing another (simple) game in retro as a proof of concept 13:04:16 What was the first? 13:05:38 http://rx-core.org/dev/playground.fossil/dir?ci=270831b8b8130f90&name=hangman 13:05:55 oops 13:06:22 --- quit: roarde (Quit: Leaving.) 13:06:24 http://rx-core.org/dev/playground.fossil/dir?name=hangman 13:07:02 that was originally for retro 10 13:11:45 Ahh. And this is graphical, not textmode? 13:13:25 ascii graphics 13:13:40 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #retro 13:15:51 --- join: Mat2 (5b4085ac@gateway/web/freenode/ip.91.64.133.172) joined #retro 13:15:59 Hi @ all 13:16:02 hi Mat2 13:16:08 hi, Mat 13:16:13 Hi Mat2 13:16:20 and hey roarde 13:16:51 'lo, cfa 13:17:23 colors (and cursor positioning) are supported on any vm running under a VT100/ANSI compatible terminal 13:17:50 I can't find the edlin vm. 13:18:52 no editor has a retro vm implementation (except perhaps emacs0 13:19:19 edlin ? 13:19:23 crc you mean no common editor has been ported to run inside the vm? 13:19:37 wasn't sure I said that out loud 13:20:44 crc or no common editor can reach into the vm's storage and edit the things found there? 13:20:50 qbject: no, I was facetiously suggesting a vm written in edlin script 13:21:12 but come to think of it, dos batch could probably be done and should also work on windows 13:22:45 I was starting to write an editor in lua with cat before I realized that lua couldn't possibly self-host and scrapped it. 13:24:25 off to bed, night all 13:24:31 * cfa idles 13:24:33 ciao 13:24:35 Once I deliver some joe stuff, I'll be starting on some of my plans, the first being an editor 13:24:59 and I'm starting to get a bit of a handle on that collaborate-in-small-bits thing 13:25:33 If you can say more about that, I'd be interested. 13:25:56 crc: i've begin porting retro over to my vm and hope it's ok if I ask questions here time by time 13:26:00 (Sorry for asking such general questions. Just trying to get a feel for how this community works.) 13:26:18 Mat2: sure 13:26:52 So far, a bulletin board with each forum belonging to a user, each post a file list, each list linked with a master list 13:27:11 files can be url, comment, file . . . 13:27:14 qbject: there was an implementation of the ngaro vm written in emacs lisp once 13:27:37 add that to two irc channels and some nifty client scripting 13:29:07 one day I'd like to update it to work with recent images... 13:29:11 so, one finds a "file" of interest, grabs it, enters #irc1. Script enters #irc2 and queries for the file's pseudo-nick. 13:30:28 crc: you mean you could run the vm inside emacs? Crazy. 13:30:43 roarde: Sounds amazing, though I'm not sure I follow. 13:31:05 I don't have enough of it to explain it at all well. 13:31:23 That's alright. 13:31:35 qbject: correct 13:31:53 I probably need to actually start hacking on the language to make sense of it anyway. Too busy reading right now. 13:32:00 qbject: not so crazy because it would be an easy step for integration of a forth system into AutoCAD 13:32:43 Mat2: now that's a very interesting idea. 13:34:48 (Mat2: I met a Lisp hacker last week who asked me specifically about applying Lisp in a CAD environment. Without AutoCAD-specific experience I wasn't sure what to tell him.) 13:35:21 I don't C. What's good reading or study to learn about building a new ngaro implementation? 13:35:42 And all new work is with ngaro3, right? 13:37:09 roarde: I'm interested in building a physical machine, not a vm. Reading old books on FORTH and stack machines because they're somewhat outside my previous learning. 13:38:39 roarde: what do you mean ? 13:39:34 Stack Machines (Koopman), Starting FORTH (Brodie), and Intro. to Elementary Computer & Compiler Design (Steele) if you're interested. 13:39:52 qbject: http://www.jwdt.com/~paysan/b16.html 13:40:10 I want to learn about making a ngaro vm in another programming language. The best source for ngaro information is retro.c+core.rx, but I don't use or understand C. 13:40:38 I'll skip learning it if I can, but will probably need to. Would like to delay that, though. 13:41:02 roarde: the python implementation is also feature complete if that helps 13:41:24 thanks, crc, but I read through the vm list. nothing there for me. 13:41:35 will be using the python vm, tho 13:42:23 roarde: hmm, I know what you describe (C is a realy ugly and mad language) but all you need to understand it is this small book: The C programming language 13:42:41 Looking at the top of core.rx can give me enough of a start to get a handle on the C source. 13:42:50 Mat2: Very nice. I ended up here because of this: http://excamera.com/sphinx/fpga-j1.html 13:44:05 thats a really simple and useful cpu design, yes 13:46:45 Simple and useful are all I need to see beauty. 13:47:42 bbiab; heading home (roads are messy due to snow) 13:48:40 careful, crc, there may be southerners 13:49:03 qbject: There is also the 4stack design which is a VLIW cpu with 4 independent stacks 13:52:28 roarde: what are southerners ? 13:53:02 some kind of para-military ? 13:53:02 People from the southern part of US, especially southeast. 13:53:39 I was in Atlanta, Georgia during a recent, rare winter storm; grew up there. Hooo boy! 13:54:01 Can't drive on an inch of snow, much less ice. 13:54:11 ah ok 13:54:20 I don't know what's more dangerous. Southerners in the snow, or folks from upstate New York when they think Winter's over. 13:54:38 crc probably has to contend with both. 13:55:00 Actually, it's New Yorkers come south in spring when "it's only raining". 13:55:24 Out-of-control wet pollen is unbelievably slick and invisible. 13:55:41 yikes. 13:56:19 pollen is generally much coarser in the southeast 13:56:25 I just remember living in Rochester, or visiting Syracuse in May, and feeling like I had accidentally exited onto a racetrack. 13:57:52 qbject: have ideas on how you'll implement a display? 14:00:22 First off, my core intention is to try and build a self-hosting minicontroller. I say mini- instead of micro- because I'm going to TRY and build it without an FPGA. 14:01:17 woohoo! 14:01:26 And the controller will be set up to talk to others of its kind over a parallel bus. 14:01:47 you mean a TTL cpu ? 14:01:53 That's right. 14:02:04 what chip families you planning to use, or perhaps discretes? 14:02:05 ah. 14:02:06 Or rather, cmos implementations of formerly-TTL designs. 14:02:56 I'm starting by looking at cmos 74xx equivalents. 14:03:05 I'll let you know as I get deeper into it. 14:03:22 so, perhaps implement ngaro (if you choose retro) in TTL, or build something else and write any vm necessary? 14:05:22 I'm tempted to say ngaro-in-TTL, since so much work has already been done there. 14:05:41 But I feel like I should understand more before saying for sure. 14:05:45 Anyway, 14:06:26 --- nick: Sgeo_ -> Sgeo 14:06:45 the controller design and its software elements will be (I assume) broadly useful. 14:07:19 that'll be a lot of board, tho 14:07:20 my personal goal is to be able to daisy chain a bunch of them and use them as interfaces for the various elements of a personal computing device. 14:07:42 It may be, yeah. We'll see. 14:08:11 you're familiar with GreenArrays? Daisy-chaining suggested this. 14:09:13 I am, and their products look fantastic. However, they're proprietary and even if they weren't you'd need your own fab to replace parts. 14:09:28 Lately I have an itch about things that can't be repaired. 14:09:47 What's to repair? 14:11:20 On the other hand, what kind of terms are you considering for sharing your work? 14:12:01 BSD license 14:13:11 :D ( while waiting for your response, you could probably hear me chanting from there: "don't let it start with 'g'") 14:13:26 I'dve let it go if it did, tho. 14:14:24 If I'd been sensitized I probably would have, but my design goals preclude the kind of control that GPLers would like to have. 14:14:37 So I don't even think about it. 14:16:24 Until recently, I didn't even consider attribution. But no one should sell freely sharable work as their own, so I've come around. 14:16:45 Exactly. 14:16:45 Sell, sure. As their own, no. 14:17:35 Plus, there's a historical usefulness. 14:17:42 yep 14:18:18 If, in 30 years, someone needs to figure out how something came to exist in order to service or even document it, they need to know who was involved. 14:19:32 gotta be a way to get an acceptable result without laying out all that board, tho 14:19:43 Probably is. 14:20:13 Like I said, I'm just starting to dive in. 14:20:49 The folks who make the Ben Nanonote are working on another project with a Xilinx FPGA that they're pretty excited about. 14:24:48 I have to idle an do paperwork for a while. 14:25:06 Likewise. Thanks for chatting. 14:25:20 --- nick: roarde -> roarde_ 14:25:42 ciao, time for some sleep 14:25:58 cya, Mat2 14:26:17 --- quit: Mat2 (Quit: Page closed) 14:30:29 --- join: crc_ (~quassel@li125-93.members.linode.com) joined #retro 14:32:41 --- quit: crc_ (Remote host closed the connection) 14:50:21 --- quit: qbject () 16:05:04 fina;;y home 16:05:52 any close calls? 16:06:01 --- nick: roarde_ -> roarde 16:07:05 no, but the roads are really bad. snow, freezing rain, then sleet 16:07:10 now snowing again 16:10:00 has your area cut back on road clearing like many have this year? 16:10:40 no, but the side streets won't be cleared until later 16:10:49 the major roads aren't too bad currently 16:11:43 side streets'll thaw a little tomorrow, then refreeze tomorrow night 16:12:44 --- join: qbject (~qbject@209.234.140.58) joined #retro 16:13:19 wb, qbject 16:13:29 Thankee. Likewise. 16:17:55 --- quit: qbject (Ping timeout: 260 seconds) 16:19:51 crc: are the vm's in the 11 repo ngaro3? 16:20:03 yes 16:22:42 --- join: qbject (~qbject@209.234.140.58) joined #retro 16:23:44 roarde: I realized I never answered your earlier question. ASSUMING that I can get the size of these boards down to a reasonable size I can devote one to each peripheral, including one to the display. A pair of hardware stacks would provide pretty fast access back and forth to a framebuffer, and then could pop to the display as needed. 16:26:00 On my "later" list is a display protocol that would allow the physical display to use a much-reduced bandwidth to the host. More of the work would be done by the display itself. 16:28:53 That's essentially what I'd like to accomplish, I'm just trying to get consistent user-programmed hardware as close to weird proprietary hardware as possible. 16:29:20 ya know, an old video game board would already have much of what you need 16:29:40 galaga, pac-man, or ms-pac would be ideal 16:29:53 oops - can't get the galage vidram anymore 16:29:58 galaga 16:30:49 You mean logic, or display hardware specifically? 16:30:57 existing slow chips are easy to pull and replace 16:31:56 both. Enough existing lands for variety, and room to modify. And most of the vid section's there, you'd just need to mod for the output type you want. 16:32:08 and replace those slow chips, I assume 16:32:56 That's a good thought. I was think that the best approach would be to prototype with big TTL chips, then miniaturize with wee little cmos clones and see how fast it would go. 16:32:59 remove the z-80's and tie in there. 16:33:52 Heh. If I were going to use a microprocessor, Z80 all the way. 16:34:50 this'd be so easy, cheap, and efficient with ga4's, but ok, I won't mention it again :) 16:36:11 I don't mind you trying to convince me, but how common are those chips? How likely is it that they'll still be available in 10 years? 16:37:32 the F18 core's been around quite a while, so if not ga4, then something very similar will be available 16:38:05 right now, it's hard to get hands on ga4, but I really believe that'll change 16:39:02 emphasis is on ga144 for now, and it's nice and worth it, but it's more of an investment in time and funds 16:39:19 --- join: AndChat- (~qb@252.sub-174-254-2.myvzw.com) joined #retro 16:40:36 (Wick on a train is a neat idea but...) 16:41:00 --- quit: AndChat- (Client Quit) 16:41:33 --- quit: qbject (Ping timeout: 255 seconds) 16:42:33 --- join: qbject (~qb@252.sub-174-254-2.myvzw.com) joined #retro 16:42:48 There we go, I think. 16:43:05 good luck 16:44:03 --- quit: qbject (Client Quit) 16:44:42 --- join: qbject (~qbject@209.234.140.58) joined #retro 16:45:07 * qbject checks the log... 16:46:49 What about straight f18s? 16:47:12 I'd rather see a physical ngaro of some sort, but unless it's used enough to warrant real fab, not sure it's worth it. 16:47:33 don't know of anyone still supplying single f18s 16:47:34 Agreed. 16:49:49 I have no illusions of pervasive adoption of what I'd like to create, but I'd like it to be unique in its longevity. That means making sure that as many options exist to build it as possible, regardless of whether I go out of business, or the companies fabbing the silicon I use. 16:49:51 but I've got this idea that dedicated gates can be repurposed and reused. When very busy it can be a long wait for a gate, but 16:50:23 when less busy, it's pipelined, even less busy it's somewhat parallel 16:51:22 seeing your views on hardware, I pronounce you home. Retro's the same thing in soft bits. 16:52:00 * qbject grins 16:52:27 That's a pretty fantastic welcome. Thanks much. (See, aisa? Told you this was the place.) 16:52:50 Now if I only knew what I was talking about :P 16:52:51 :-D 16:53:36 Nah. I haven't learned to use it, but I've been watching with interest more than long enough to know how it trends and some of why. 16:54:48 Well, roarde, this is important. I don't know what I'm talking about either. I had a lot more success in a machine shop than I did in a CS classroom. If we can come to understand this, it means that there's hope for someone to figure out what we've done down the road. 16:55:18 k, listen up. We're invaluable because we don't know. 16:55:37 Now you're getting it. 16:56:23 Anyone who's done something long enough to be good enough to teach has been at it to long to introduce that something effectively. 16:56:43 Looking around, this holds especially true for forths, etc. 16:56:57 People who are supposed to have it together are throwing away clock cycles, hardware, and joules upon joules of energy as if there's no end. 16:57:35 I'm thinking of it another way, though your thought is true. 16:58:13 I'm journaling what and how I'm learning, and will write it formally when I know enough to do so. 16:59:09 Ah, cool. 16:59:09 I'm journalling publicly as I go so that the entire process of learning is documented firsthand. 16:59:25 After that, someone else will have to pick it up, as I'll be too far along to have the new eyes that're needed to see what I or some other new one could really use. 17:00:04 Mine would mislead more than inform right now, so I'll wait. 17:00:16 I'm a better editor than writer. 17:00:52 Well, I am too. And as I said, better machinist than programmer. But I see your point. 17:01:41 Speaking of writing, many of these same letters I'm casting out could be rearranged as part of, say, a highlight or macro file. 17:02:14 Understood. Talk with you later. :) 17:02:22 hope so 17:02:49 I'm off to read. 'Night all. 17:02:55 co'o 17:03:21 --- quit: qbject (Quit: Leaving) 19:29:23 --- quit: roarde (Quit: Leaving.) 21:47:14 --- quit: aisa (Quit: aisa) 23:59:59 --- log: ended retro/11.01.26