00:37 <KipIngram> In case annyone is interested, my FINND implementation is here:

00:37 <KipIngram> https://pastebin.com/0nF7Cjti

00:37 <KipIngram> That handles searching a linked list of vocabularies, accessed through variable PATH.

00:39 <KipIngram> This is the first code I've written with stack frames and exception processing in mind from the start.

00:40 <KipIngram> The bulk of the code handles full navigation of the structures, without regard for finding a match. The success case is handled by exception.

00:47 <KipIngram> I'm happy with how that bit of development went - I wrote it offline over the course of a few days, and then plugged it in. I made one transcription error - I put a single return at the end of check instead of a double return.

00:48 <KipIngram> And there was a design error. I don't know if I mentioned it or not, but I had it in mind that I didn't need BLK and >IN because I could put that information in a stack frame. That's how I coded it. But that's no good (and I should have realized it wasn't) because INTERPRET is run inside that loop and has to be able to change the stack.

00:49 <KipIngram> When I thought of that I whacked myself in the head and changed it in a couple of minutes (put in BLK and >IN), and then it worked. :-)

04:05 <KipIngram> Fun fun - I logged into Pastebin tonight, for the first time in years. Quite a few years ago I spent some time thinking about building a computer, out of discrete chips. The idea was to have a guaranteed back-door free machine, that I knew the full workings of.

04:05 <KipIngram> I got a design done for the data stack, and it looks like I documented it on Pastebin:

04:05 <KipIngram> https://pastebin.com/nfLqb58q

04:05 <KipIngram> Has the pinout and the logic equations.

04:06 <KipIngram> And an instruction set encoding.

04:06 <KipIngram> I assume I felt at the time that those equations would implement that instruction set. I didn't slug through it tonight to confirm, though.

04:07 <KipIngram> Used 22V10 programmable aray logic.

04:33 <KipIngram> You guys might enjoy this:

04:33 <KipIngram> https://www.cs.princeton.edu/~mfreed/docs/cuckoo-eurosys14.pdf

04:33 <KipIngram> It decribes something called cuckoo hashing. Looks like it could potentially have application to Forth dictionary storage.

04:34 <KipIngram> As the hash table becomes full, inserts take longer and longer, but the lookup time is always bounded.

04:34 <KipIngram> With a four-way set associative model you wind up using about 95% of your hash table space.

04:36 <KipIngram> For a lookup, at most you would compute two hashes of the name and then potentially search a set of 8 items to find the one of interest.

08:34 <veltas> MrMobius: I don't have a strong opinion on this one (re ease of assembler). If you do what KipIngram does and write a subset of the language you'd want it to be easy to extend incase you want another instruction

08:35 <veltas> If you want the language to have more complicated syntax to make it more familiar, then why use Forth at all? This is the thing I can't get my head around.

08:35 <veltas> complicated/conventional

08:36 <veltas> With 6502 though writing syntax parsing is not too complicated, depending on what you're going for

08:37 <veltas> In my Forth the assembly instruction I use is CALL, and I use it while writing words to the dictionary, so it has to be something you can use within your code.

08:46 <veltas> Even commercially funded forths like SwiftForth use a compromise syntax

13:01 <veltas> This is interesting: https://github.com/s-macke/VoxelSpace

13:02 <veltas> Also they reverse-engineered a Forth game but that's less good because they converted it to C https://github.com/s-macke/starflight-reverse

13:05 <MrMobius> veltas, I think you're getting to the root of the problem re assemblers

13:07 <MrMobius> why use forth? 1. youre on an extremely constrained system, 2. you could be doing it in any other language, 3. it's a toy/hobby/for fun

13:07 <MrMobius> if its 2 then why use forth for this? if its 3, then go for it and use a janky syntax

13:07 <veltas> And if it's 1 use an existing assembler?

13:08 <MrMobius> as long as youre only writing it for fun or only going to assembly a few test or demo programs with it then it doesnt matter if the syntax is weird or non ideal

13:08 <veltas> It is quite powerful being able to assemble code in your Forth though

13:09 <MrMobius> it is powerful but that's different than writing thousands of lines into an assembly file and assembling that which is where you would want 2

13:09 <MrMobius> which also has the advantage of being able to assemble other people's files

13:20 <proteusguy> If you all wanna see the first public beta test of our new platform this coming Thursday (8p BKK GMT+7) check out this Fb event: https://www.facebook.com/events/209019941016891

14:21 <KipIngram> proteusguy: What kind of platform is this? I'm trying to fit togher "product" and "rock band" in my head.

14:24 <proteusguy> KipIngram, easier to show than tell I think... ;-) https://drive.google.com/file/d/1G2oJmEkLXj7I8FzVTaTtsNhkbf2K1lNP/view?usp=sharing

14:25 <proteusguy> This video shows our platform compared to the two "leading" platforms. ;-)

14:39 <KipIngram> Showing works - only took a few seconds to understand. That's cool.

14:43 <proteusguy> Got a more direct link to the YouTube event now... https://www.youtube.com/watch?v=LDZFybcPnco

14:50 <f-a> https://sun9-17.userapi.com/impg/R92P2ajgO4tJVSauSMidW_Xeyxo4XvCMAS1y_g/YpnOe51EhcU.jpg?size=1620x2160&quality=96&sign=9e0999d274f05c33b12320f736e332b9&type=album

14:50 <f-a> (my friend’s)

14:55 <proteusguy> +1 not even sure how many time's I've read that book by now... still have my original copy!

14:56 <KipIngram> :-) I've got an original copy of it somewhere upstairs.

14:56 <KipIngram> Also.

15:03 <proteusguy> Here's mine. And the torn paper was my bookmark with a list of local BBS' in Forth Smith, Arkansas. This would be circa-1984-86. https://ibb.co/HP74nHN

15:03 <proteusguy> Note the modem settings for 300 & 1200 baud... ;-)

15:16 <veltas> f-a: Nice

15:16 <veltas> And nice kitty

15:17 <veltas> I'm assuming that's a print-out rather than the original print?

15:18 <veltas> I seem to remember the original was in a blue cover

15:18 <f-a> print out

15:18 <f-a> I think he printed it on a needle printer

15:18 <veltas> An older gent at my office has a copy of Starting FORTH, which he almost gave to me before realising it might be worth something

15:19 <veltas> No idea though, probably not worth too much right?

15:19 <mark4> probably not

15:19 <mark4> my father had a first edition but it got runed in a flood

15:19 <veltas> It's valuable to the right buyer but how to set that up? It's the sort of thing that could auction on ebay for £10

15:19 <f-a> yeah do not think there is a market for Fort memorabilia xD

15:21 <veltas> There's a couple on UK eBay right now, one is buy-it-now for £35, other is auction with no bids asking £30

15:22 <veltas> It feels like Forth activity has spiked in the time I've known Forth, maybe they'll become valuable later?

15:23 <veltas> Nobody will truly appreciate Forth until all the older Forthers are long gone probably :P

15:23 <f-a> fuck people amassing stuff for ebay

15:29 <veltas> Well that can go both ways, end up buying stuff that loses value

15:31 <f-a> a part of me hopes so :P

15:33 <veltas> Also their push to buy up old stuff to sell it on ebay means they will likely preserve it better than people who simply don't care about their old stuff and are trying to get rid of it or will just let their family throw it all out when they die

15:34 <veltas> So it might help increase the amount of preserved older tech/books

15:36 <f-a> resellers are a pain in many other markets

15:36 <f-a> e.g. I like fountain pens

15:36 <f-a> stopped w/ those

15:39 <f-a> b/c resellers

15:39 <f-a> I will stop complaining, at least we have an online copy

18:56 <KipIngram> Well, I added a snapshot / recover snapshot phrase immediately after WARM. The snapshot is there just so the recover operation will have something to recover - the recover operation right there is the re-entry point after error handling. The snapshot word stashes the data stack pointer in a variable before making the image - the recover operation uses it to reset the data stack.

18:56 <KipIngram> After that recovery we have QUIT.

18:56 <KipIngram> And then QUIT calls SNAP just before running INTERPRET.

18:57 <KipIngram> I found that arrangement to be fully bulletproof error recovery last time.

18:57 <KipIngram> Basically if a INTERPRET detects an error of any kind, the whole line is ignored.

19:04 <KipIngram> I suppose obviously a really bad glitch could corrupt the snapshot, or the roadmap to it, but other than that...