00:00:00 --- log: started forth/19.04.20
00:23:11 --- join: Croran (~quassel@2601:601:1801:6dde:7d93:b956:e1a6:b755) joined #forth
00:58:17 <siraben> proteusguy: gforth
00:58:27 <siraben> Should work on my forth too
00:58:42 <proteusguy> :-)
00:59:27 <proteusguy> made some progress on one of my white papers over songkran. hope to get a draft done soon... as usual. :-) 
00:59:58 <siraben> Fun stuff. How's the break going?
01:00:10 <siraben> I went to Hua Hin over songkran
01:00:36 <proteusguy> it's over. haha was in Bali again. Back in BKK have to deploy a few projects in the next week or so.
01:02:13 * the_cuckoo says hi :)
01:02:44 <the_cuckoo> had a "fun" couple of weeks - had a stroke
01:03:20 <proteusguy> the_cuckoo, :-( how are you? still got symptoms?
01:03:51 <the_cuckoo> absolutely fine - at less risk than ever it would seem 
01:04:03 <proteusguy> goodness - that's damn lucky!
01:04:12 <the_cuckoo> turns out i've had a hole in my heart my whole life
01:05:04 <the_cuckoo> currently on some blood thinning medication which effectively means it just won't happen again
01:05:11 <proteusguy> get that fixed up?
01:05:24 <the_cuckoo> later in the year, yeah
01:05:29 <the_cuckoo> dunno exactly when
01:05:51 <proteusguy> good luck!
01:06:09 <the_cuckoo> cheers :)
01:07:00 <the_cuckoo> worst thing about it is that i can't drive for the next 6 months
01:07:13 <the_cuckoo> best about it is that i can't drive for the next 6 months
01:07:16 <proteusguy> where do you live? car town?
01:07:16 <proteusguy> haha
01:07:21 <the_cuckoo> hate driving :p
01:07:27 <proteusguy> I'm in Bangkok. No point in driving here.
01:07:34 <the_cuckoo> :)
01:07:47 <the_cuckoo> i live in town, work from home
01:07:54 <the_cuckoo> no need to drive anyway
01:08:01 <proteusguy> that's fortunate
01:08:10 <the_cuckoo> for sure
02:02:31 --- quit: ashirase (Ping timeout: 246 seconds)
02:08:57 --- join: ashirase (~ashirase@modemcable098.166-22-96.mc.videotron.ca) joined #forth
03:10:30 --- join: dddddd (~dddddd@unaffiliated/dddddd) joined #forth
04:54:58 <`presiden> proteusguy, having holiday in Bali? also welcome :)
04:57:03 <john_cephalopoda> the_cuckoo: Oh, I saw a story just like yours on imgur just a few days ago.
04:57:31 <john_cephalopoda> There's some special twisty thing that can be used to easily seal the hole.
07:02:56 --- quit: ashirase (Ping timeout: 246 seconds)
07:05:20 --- join: ashirase (~ashirase@modemcable098.166-22-96.mc.videotron.ca) joined #forth
07:12:07 --- join: X-Scale` (~ARM@46.50.6.99) joined #forth
07:12:30 --- quit: X-Scale (Ping timeout: 245 seconds)
07:12:49 --- nick: X-Scale` -> X-Scale
07:16:24 --- quit: dys (Ping timeout: 268 seconds)
07:23:53 <siraben> Wish there was a book on "data structures and algorithms in Forth"
07:24:37 <siraben> An imperative algorithm more or less looks the same in the mainstream imperative languages, C, C++, Java, what-have-you
07:24:44 <siraben> but translating it to Forth changes a lot.
07:50:38 <proteusguy> certainly an interesting idea. Chuck Moore would probably argue that it's adding a lot of unneeded complexity because in forth you'd strictly implement something that fit you exact needs and nothing more. The cost of generalizing the algorithms would be un-forth-like. But a book showing small example problems that happen to need something like common data structures and algorithms would probably be quite compelling.
07:52:12 <proteusguy> Probably show the sample problem. Then the initial implementation explaining the data and algorithms. THEN the fully refactored OMG it hardly looks the same final version optimized in a "proper" forth-like manner. Quite a challenging book to write.
07:55:53 --- join: PoppaVic (~PoppaVic@unaffiliated/poppavic) joined #forth
07:57:29 <the_cuckoo> john_cephalopoda: yeah - seems like a fairly standard procedure - the doc described it as an "umbrella-like thing" which is used join the two parts - am generally kinda squeamish, so trying not to consider the details for now - will worry about it nearer the time no doubt :)
08:11:12 <`presiden> what you call something that written in a "proper" forth-like (or other language) manner?
08:11:14 <`presiden> an idiomatic forth?
08:12:05 <PoppaVic> "forth"
08:12:48 <proteusguy> PoppaVic, agreed. ;-)
08:13:16 <PoppaVic> If you can READ it, you might add: "[forth] source"
08:20:29 --- join: rdrop-exit (~markwilli@112.201.166.63) joined #forth
08:22:38 <siraben> proteusguy:  Thinking Forth comes halfway there, but it's more about the philosophy of Forth than actually implementing things
08:23:26 <siraben> What might be hard is how algorithms usually require temporary variables and so on, which seem to be big no-nos in Forth
08:23:47 <proteusguy> siraben, yep - it's an excellent book for how to approach a problem using forth. but I think what you suggest is a book that applies that concept to real sample apps.
08:24:55 <siraben> I think using ANS Forth's optional Locals word set would help in these cases
08:25:04 <rdrop-exit> Meh
08:25:06 <siraben> proteusguy:  heh, "Practical Forth"
08:25:25 <proteusguy> few real programs avoid any use of temporaries. indeed - Chuck's latest forths always have this "a" register which is used specifically for that purpose. And there there are STILL a few local variables scattered about.
08:26:04 <rdrop-exit> The A register is a register
08:26:31 <proteusguy> the thing against temps/locals is when every function is some big thing with several bits of local data declared at the top like it was C.
08:26:41 <proteusguy> rdrop-exit, yep - so it's a very efficient temporary.
08:26:55 <siraben> Imagine writing quicksort without locals or temporary variables, dear me.
08:27:09 <siraben> If quicksort has ever been written in Forth that is
08:27:13 <rdrop-exit> It's a machine register
08:27:37 <proteusguy> yeah... so??? it's a cell.
08:27:54 <siraben> https://rosettacode.org/wiki/Sorting_algorithms/Quicksort#Forth
08:28:13 <siraben> seems to detect an "email" in the code, here it is again: https://rosettacode.org/mw/index.php?title=Sorting_algorithms/Quicksort&action=edit&section=40
08:28:21 <siraben> this one renders correctly
08:28:37 <siraben> That isn't so bad, no locals or temporaries needed
08:30:40 <rdrop-exit> You see no difference between a temporary memory variable and a register?
08:31:48 <proteusguy> unless a register gets clobbers by primitives in the language then it is simply a nice place to put a fast efficient temp object, no? what am I missing? that's what one does with registers.
08:32:11 <rdrop-exit> As you say, fast and efficient
08:32:34 <proteusguy> your point? or are we now in agreement?
08:32:58 <rdrop-exit> Are you locals fast and efficient?
08:33:03 <rdrop-exit> * your
08:33:13 <proteusguy> they are if they're in the "a" register. :-) hahaha
08:33:26 <proteusguy> which is why Chuck put it there.
08:33:30 <rdrop-exit> That's the difference, yes
08:33:42 <proteusguy> completely and precisely my original point.
08:34:21 <siraben> compilers (liveness analysis, register allocation etc.) tend to make temporary variables whole registers anyway
08:35:40 <rdrop-exit> So if you start using the ANS locals wordset you are at the mercy of your Forth's optmizer
08:35:53 * proteusguy has no intention of doing so.
08:36:21 <rdrop-exit> I applaud you for it
08:36:54 <proteusguy> heh
08:37:06 <rdrop-exit> IIRC Chuck's main use for the A register is streaming through memory
08:37:25 <siraben> Forth compilers hardly optimize anything, so there's that.
08:38:14 <proteusguy> Jeff Fox wrote about it and showed some examples as I recall. It's basically a scratch cell that saves a lot of stack movement. Streaming through memory is certainly a valid usecase.
08:38:39 <siraben> It's still O(1) access anyhow, to access data in a register or the stack or heck, even RAM (cached or uncached)
08:38:41 <siraben> There's bigger problems to solve like the time complexity of the whole algorithm
08:39:38 <rdrop-exit> Knuth & Co. to the rescue
08:40:13 <siraben> I should add that certain constant time optimizations DO pay off nicely, like making TOS a register
08:41:30 <rdrop-exit> Or design a stack machine with on-chip stacks
08:42:17 <siraben> Knuth said something about premature optimization, hm.
08:42:42 <siraben> "We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%"
08:42:43 <proteusguy> he was all for it, right? :-D
08:43:24 <rdrop-exit> I wouldn't call on-chip stacks a premature optimization :)
08:43:55 <proteusguy> well, if it is, it's one hell of a commitment to that optimization!
08:45:04 <siraben> Judging by the way he codes and writes books, he is anything but sloppy :)
08:47:41 <rdrop-exit> I was reminded today of the transputer with its ABC three item stacks
09:10:32 <rdrop-exit> Midnight, time for bed. Ciao :)
09:10:40 --- quit: rdrop-exit (Quit: Lost terminal)
09:31:55 <siraben> Approaching midnight here, bye all
09:43:53 <john_cephalopoda> Forth is a nice abstraction. It can run on a lot of machines
09:51:37 <john_cephalopoda> I have to write some more Forth again.
12:20:59 --- quit: gravicappa (Ping timeout: 250 seconds)
12:36:40 --- quit: Zarutian (Ping timeout: 255 seconds)
12:44:14 --- quit: dave0 (Quit: dave's not here)
12:57:05 --- join: Zarutian (~zarutian@173-133-17-89.fiber.hringdu.is) joined #forth
12:58:47 <WilhelmVonWeiner> ls
12:58:51 <WilhelmVonWeiner> sorry lol 
12:59:02 <WilhelmVonWeiner> haven't been on a computer in a couple days I think
13:00:19 <WilhelmVonWeiner> interesting discussion on the locals stack
13:00:34 <WilhelmVonWeiner> it's why I kinda like FIFO the more I tackle algorithmic problems with Forth
13:02:50 <WilhelmVonWeiner> I hope Chuck writes another book before he leaves us
13:06:27 <john_cephalopoda> Heh, reading through jonesforth right now, the first things are done exactly how I independently implemented them :þ
13:08:57 <WilhelmVonWeiner> I need to write an eForth.
13:16:28 <john_cephalopoda> WilhelmVonWeiner: Is there a specification of which words are part of eforth?
13:16:45 <WilhelmVonWeiner> whatever's in the eForth book, I suppose
13:17:57 <john_cephalopoda> Is it available somewhere?
13:19:53 <WilhelmVonWeiner> Amazon, for one.
13:21:16 <WilhelmVonWeiner> https://wiki.forth-ev.de/doku.php/projects:ting_s_electronic_forth_bookshelf
13:23:20 --- join: cnidario (~dont@cli-5b7ec41e.ast.adamo.es) joined #forth
14:30:09 --- quit: tabemann (Ping timeout: 240 seconds)
14:35:14 --- quit: cnidario (Read error: Connection reset by peer)
14:51:45 --- quit: djinni (Quit: Leaving)
14:55:29 --- join: djinni (~djinni@68.ip-149-56-14.net) joined #forth
14:56:05 --- quit: djinni (Remote host closed the connection)
14:59:55 --- join: djinni (~djinni@68.ip-149-56-14.net) joined #forth
15:02:55 --- quit: pierpal (Ping timeout: 255 seconds)
15:16:02 --- quit: Zarutian (Read error: Connection reset by peer)
15:16:04 --- join: Zarutian_2 (~zarutian@173-133-17-89.fiber.hringdu.is) joined #forth
15:55:06 <WilhelmVonWeiner> http://ix.io/1GKz
15:55:42 <WilhelmVonWeiner> siraben: trees similar to your implemntation, but different.
15:55:52 <PoppaVic> The eforth implementations will surely specify what they offer - it's always minimal
16:13:07 --- quit: john_cephalopoda (Ping timeout: 252 seconds)
16:26:47 --- join: john_cephalopoda (~john@unaffiliated/john-cephalopoda/x-6407167) joined #forth
16:29:51 --- quit: Keshl (Read error: Connection reset by peer)
16:30:10 --- join: Keshl (~Purple@207.44.70.214.res-cmts.gld.ptd.net) joined #forth
17:11:46 --- join: pierpal (~pierpal@host197-221-static.34-79-b.business.telecomitalia.it) joined #forth
17:21:42 --- join: tabemann (~tabemann@rrcs-162-155-170-75.central.biz.rr.com) joined #forth
17:43:45 <tabemann> I'm stumped
17:43:54 <tabemann> I try debugging my buffer code
17:44:14 <tabemann> I try putting a single-tasked .s in the expansion code
17:44:15 --- quit: pierpal (Ping timeout: 250 seconds)
17:44:22 <tabemann> it works fine for trivial examples
17:44:38 <`presiden> but
17:44:38 <tabemann> but when I try with a nontrivial example, it crashes
17:44:53 <`presiden> happy debugging
17:44:55 <tabemann> mind you these nontrivial examples mostly work without the .s
17:45:12 <tabemann> except for a couple nontrivial examples, which fail regardless
17:45:43 <tabemann> which are why I tried putting .s in there in the first place
17:46:56 <tabemann> mind you I'm not running plain .s but rather ['] .s execute-single-task-io, which ensures that TYPE during .s does not cause a task switch
19:01:57 <tabemann> I figured out why my sixels weren't displaying right
19:02:04 <tabemann> somehow nulls were getting in my output
19:02:27 <tabemann> but the way I as dumping the output (basically just replacing escapes with spaces and dumping it to standard output) didn't display the nulls
19:17:12 <tabemann> yes!
19:17:29 <tabemann> I got sixels working under hashforth with my generalized sixel-rendering code
19:17:33 <tabemann> unfortunately
19:18:01 <tabemann> the latest version of the hashforth userspace refuses to build newer versions of hashforth
19:18:13 <tabemann> so to build newer versions I need to use an older version
20:00:11 --- quit: tabemann (Read error: Connection reset by peer)
20:19:43 --- quit: dddddd (Remote host closed the connection)
20:26:48 --- join: tabemann (~tabemann@172-13-49-137.lightspeed.milwwi.sbcglobal.net) joined #forth
21:01:52 <proteusguy> I was never able to get a reply from Dr Ting when trying to buy something, alas.
21:30:49 --- quit: APic (Ping timeout: 255 seconds)
21:40:24 --- join: APic (apic@apic.name) joined #forth
22:39:50 --- join: gravicappa (~gravicapp@h109-187-43-215.dyn.bashtel.ru) joined #forth
23:12:09 --- join: rdrop-exit (~markwilli@112.201.166.63) joined #forth
23:13:22 <rdrop-exit> Greetings Forthists
23:59:59 --- log: ended forth/19.04.20