2023-02-23 00:18:18 You know, an interesting aspect of Lord of the Rings occurred to me. There really isn't a "hero." No one really "put evil down." Instead, evil ultimately imploded on itself. Over and over, it would up tripping over its own actions. 2023-02-23 00:18:54 Frodo even FAILED to have the final strength to toss the Ring into the Cracks of Doom. Instead, it was Gollum's botched attempt to take it from him that led to its destruction. 2023-02-23 00:19:28 The hobbits stood up to Saruman in the Shire, but they didn't end him - that happened because of how he treated his own one remaining ally. 2023-02-23 00:19:55 Frodo and Sam actually won through to Mount Doom because of the in-squabbling among the Orcs defending the perimeter. 2023-02-23 00:20:37 Merry and Pippen only met up with the Ents because Saruman had had his orcs take them and bring them back to Isengard. 2023-02-23 00:20:58 And it was those Ents that were key to taking Isengard. 2023-02-23 00:21:03 It's the same pattern, over and over. 2023-02-23 07:03:44 hey people if you have any suggestions on interesting/classic articles or posts about Forth, I would love to hear about them. 2023-02-23 08:15:01 tsatziq: If you google Jeff Fox / Ultratechnology you can find some interesting writings. Fox was... well, basically he was (is?) a Forth / Chuck Moore "fanboy" who worked with Chuck over a long period of years. You might accuse the material of being a little one sided - it definitely sings the hpraises of Forth, and I've felt oer time that it sometimes "conveniently neglects" the downsides of the various 2023-02-23 08:15:03 techniques espoused. And you don't always see those less than perfect aspects until you've actually tried out some of the ideas. 2023-02-23 08:15:54 Then there's a book that Chuck self-published somewhere along the way (early on, though0). 2023-02-23 08:15:55 http://forth.org/POL.pdf 2023-02-23 08:16:12 This is very, VERY early Forth, before it was everything that it is today. 2023-02-23 08:17:45 There are a few papers out there that describe Color Forth, and a couple that get into some of the details of Chuck's chip design system. 2023-02-23 08:18:28 Finally, there's some interesting info found by digging around starting here: 2023-02-23 08:18:30 http://www.etherforth.org/ 2023-02-23 08:19:54 Somewhere down that last rabbit hole is a nice description of a development system that uses a pair of GA-144 Green Arrays chips; one is the "dev system" and the other is the "application target." It describes how one of those GA-144s has been programmed to implement an interactive Forth development environment that lets you program the other one. 2023-02-23 08:28:09 It's an interesting exercise because even though the GA144 has 144 cores, each core is VERY simple, has a very small amount of RAM, and so on, and yet you somehow have to get the usual Forth functions (editor, compiler, etc.) arranged so that they can a) work and b) communicate. I've felt like in some ways it's more like designing a digital circuit than it is classic "programming." 2023-02-23 08:29:10 I also saw an interesting article comparing Forth and Lisp a couple of months back - once I sip a little more coffee I'll see if I can float that to the surface in my browser again. 2023-02-23 08:30:14 So, guys, I'm still diving down the file system hole. I'm starting to think, though, that the really "good" file systems are probably overkill for the sort of work I expect to be doing. 2023-02-23 08:31:02 Especially reading that article last night on BeTRFS (the b-epsilon tree stuff) got me to thinking that my work would likely just never "load" such a file system in a way that seriously brought out its advantages. 2023-02-23 08:32:21 Probably what would result in the simplest code for me, and completely "adequate" performance, would be to just keep files in linear block order and move stuff around to open up new space or remove old space. I'll likely never have enough data in my file system to make that not perform well. 2023-02-23 08:32:57 Also, with the amount RAM I have on my computer, it's highly likely that I'll forever be able to just read my whole damn file list into RAM all at the same time. 2023-02-23 08:33:19 Maybe if I start mucking about with videos that would cease to be the case, but short of that I doubt I'll ever exhaust my RAM. 2023-02-23 08:33:32 And that just totally changes the nature of what kind of capabilities have value add. 2023-02-23 08:34:26 so I've got to ask myself, am I doing this just to have the convenience of using named files in my Forth system? Or am I doing because I want to be able to say I've written a "real" file system? 2023-02-23 08:36:21 In this BeTRFS, the idea is you buffer writes in the internal nodes until you run out of space, and then you flush buffer info down to children. The benefit comes from that flush operation running on what was originally a large number of small writes. But if I never fill up the starting buffer space, then I never need to flush and it was all a moot point. 2023-02-23 08:57:46 It does look, though, like this is a great example of a place where "big-O" analysis really shows the way to something better. Sounds like for the right kind of applications this BeTRFS is just way, WAY better than alternative file systems, on almost all workloads. 2023-02-23 09:45:59 thank you! will take a closer look at your reply when i get home :) 2023-02-23 09:46:19 Have fun - I love all this stuff. :-)