2023-08-01 04:18:10 <veltas      > What codebase is od part of?
2023-08-01 04:18:43 <veltas      > GNU coreutils apparently ... so part of your BSD's codebase if you're not on Linux I would guess
2023-08-01 04:44:07 <dnm         > KipIngram: will do!
2023-08-01 04:44:16 <dnm         > DKordic: thanks for the pointer!
2023-08-01 06:41:52 <veltas      > To my shame, I only found out about this today https://en.wikipedia.org/wiki/Shunting_yard_algorithm
2023-08-01 06:42:27 <veltas      > I had been using my own algorithm which uses more memory (although is still 'linear', but is slower, for sure)
2023-08-01 06:42:48 <veltas      > I'm going to have some fun with this on paper tonight I think
2023-08-01 06:43:32 <veltas      > The algorithm for those who can't be bothered to click allows you to produce a stream of RPN from an infix expression using a single 'operator stack' to keep track of state
2023-08-01 06:43:44 <veltas      > I'm sure it's familiar to many
2023-08-01 06:44:11 <veltas      > I wonder if that's used in Chuck's infix code
2023-08-01 08:59:47 <Zarutian_iPad> http://bitmath.blogspot.com/2023/07/propagating-bounds-through-bitwise.html
2023-08-01 09:42:04 <veltas      > 
ACTION quietly appreciates

2023-08-01 10:05:04 <KipIngram   > veltas: It'd be interesting to see how that compares to the method Chuck gave in his old book.
2023-08-01 10:05:29 <KipIngram   > He discusses processing infix expressions, because I think at the time he was a little insecure about the "non-standard aspects" of his new creation.
2023-08-01 10:05:38 <KipIngram   > He grew out of that.  :-)
2023-08-01 10:05:53 <KipIngram   > I look forward to taking a peek at that too.
2023-08-01 10:06:09 <KipIngram   > The general idea is discussed in the dragon book too, I think.
2023-08-01 10:06:50 <KipIngram   > Man, I overslept this morning and am still "fighting through the sluggish" at the moment.
2023-08-01 10:07:15 <KipIngram   > But the world has launched, so at the wheel I am.
2023-08-01 10:07:27 <KipIngram   > Well, not literally in a car, but I'm "on."
2023-08-01 10:08:07 <KipIngram   > But yeah, Chuck had a period during which he was a bit "apologist" about Forth's unusual / unfamiliar aspects.
2023-08-01 10:09:29 <KipIngram   > I think it's kind of natural - I remember being a lot less "sure of myself" when I was young too.  Getting a little experience under one's belt is pretty important.
2023-08-01 10:09:49 <KipIngram   > That old "wet behind the ears" period.
2023-08-01 13:34:23 <veltas      > KipIngram: Yeah I've read the code generation part of dragon book a whole bunch, but for some reason I've missed this
2023-08-01 13:34:55 <veltas      > Probably because it's somehow less interesting than playing with IR and assembly output optimisation
2023-08-01 14:31:28 <KipIngram   > I'm really bad about being steered by what I find interesting instead of what's "potentially beneficial to me."
2023-08-01 14:31:47 <KipIngram   > But... this is supposed to be leisure, so... I won't worry too much about it.
2023-08-01 14:32:00 <KipIngram   > I hold down a job, so that checks the pragmatism box.
2023-08-01 14:32:16 <KipIngram   > Life's supposed to be fun.
2023-08-01 14:42:21 <KipIngram   > So I played some more  yesterday with the MAX32655 board.  I convinced myself tha tyes, Segger Embedded Studio actually is "running the board" and not using some kind of a simulator (it does have one).  So far, though, I've only seen output come back from the board there within the studio's GUI - I haven't seen it come via a screen session console, which I'd like to achieve.
2023-08-01 14:43:41 <KipIngram   > But, it does seem to be flying the board, so that's good.  And I am finding resources here and there - assembly code that handles the board init, and so on.
2023-08-01 14:44:06 <KipIngram   > I really don't think much init should be required to just send stuff out over the UART.
2023-08-01 14:45:04 <KipIngram   > Looks like there are quite a few options on how one CAN use it - there are DMA channels provided that will move data back and forth between RAM buffers and the UART's fifos.  But I'm sure just a software controlled loop, plain vanilla, should work too.
2023-08-01 14:46:07 <KipIngram   > What's made me so happy here is that I didn't have to fight to get this - I just told it what chip I was using and RAN it - it "just worked."  And the studio even opened up for the first time with a hello world project already all set to go.
2023-08-01 14:47:11 <KipIngram   > I didn't have to use the Segger debugger I bought, either - it's also willing to work with the built in CMSIS-DAP interface that's part of the 32655 board.
2023-08-01 14:47:58 <KipIngram   > It pops up a dialog when it detects that and makes me promise I'm not pursuing any commercial purposes, but it will let me clear that prompt for the rest of the day so I only have to deal with it once a day.
2023-08-01 14:49:06 <KipIngram   > They want money for the studio if you're doing commercial work, and apparently they want EXTRA money if you want to use CMSIS-DAP for commercial work.
2023-08-01 14:49:32 <KipIngram   > I think I'd have to solder some  pins to the board to use the Segger debugger.  I got it mainly to try to revive that brok SAMD51 board.
2023-08-01 14:49:40 <KipIngram   > broken
2023-08-01 15:05:04 <KipIngram   > Guess next I ought to plug in the working SAMD51 board and see if hello world works there too.
2023-08-01 15:19:46 <KipIngram   > Oh, also, I took the intel hex output produced by the Segger studio when I built in release mode and dropped it onto the drive mounted by the board, and that worked.  It "did something" with the board, and for the first time I did NOT have a FAIL.TXT file appear in the drive with an error message.
2023-08-01 16:15:33 <veltas      > I don't know who told you life was meant to be fun, but that's all the more reason to do things you enjoy
2023-08-01 16:15:47 <veltas      > Especially if they're not hurting anyone and are inexpensive
2023-08-01 16:22:31 <KipIngram   > Among others my supervising professor expressed that on one occasion.
2023-08-01 16:22:46 <KipIngram   > In the context of British sports cars being fun to work on.
2023-08-01 16:22:55 <KipIngram   > Well, and drive.
2023-08-01 16:27:11 <gordonjcp   > I mean that depends on the car
2023-08-01 16:27:23 <KipIngram   > Clearly.  :-)
2023-08-01 16:27:26 <gordonjcp   > MGB GT, fun to work on, pretty okay to drive
2023-08-01 16:27:34 <KipIngram   > I don't remember which specific ones he'd discussed.
2023-08-01 16:27:41 <gordonjcp   > MGB GTV8, fun to work on, actually pretty scary to drive
2023-08-01 16:28:03 <KipIngram   > I liked that little car Patrick McGoohan drove in The Prisoner.
2023-08-01 16:28:10 <gordonjcp   > MGB GTV8 engine swapped with a GEMS 4.6 out of a Range Rover, running a high lift cam and Tornado chip
2023-08-01 16:28:26 <gordonjcp   > surprisingly good fun to work on, do not forget to bring a change of underwear
2023-08-01 16:28:36 <gordonjcp   > That's a Lotus 7
2023-08-01 16:28:50 <KipIngram   > Loved that opening sequence in that show.
2023-08-01 16:28:52 <gordonjcp   > there are loads of clones of them, including ones you build entirely from scratch
2023-08-01 16:29:04 <gordonjcp   > I know someone who has a Locost with a 1.3 Ford Crossflow engine in
2023-08-01 16:29:22 <gordonjcp   > but, again, some mad bastard has one with a 4.6 V8 out of a Range Rover in it
2023-08-01 16:29:29 <KipIngram   > Some of that era's British "thriller" shows were really good.
2023-08-01 16:29:38 <KipIngram   > I liked UFO as well.
2023-08-01 16:29:52 <gordonjcp   > which, given that it's a 1960s Buick smallblock at heart, is a surprisingly pokey engine
2023-08-01 16:30:09 <gordonjcp   > UFO was good
2023-08-01 16:30:23 <gordonjcp   > you might like Randall and Hopkirk (Deceased)
2023-08-01 16:30:36 <gordonjcp   > they did a remake of it in the 90s, which was surprisingly okay
2023-08-01 16:30:37 <KipIngram   > Oh, I don't know that one.
2023-08-01 16:30:50 <KipIngram   > By the same name?
2023-08-01 16:31:00 <gordonjcp   > KipIngram: get it, don't read anything about it, watch it, just go into it "cold"
2023-08-01 16:31:14 <KipIngram   > Ok, good to know.  I'll try it out.
2023-08-01 16:31:15 <gordonjcp   > The Persuaders is a good laugh
2023-08-01 16:31:41 <KipIngram   > On a wholly different front, I loved the more recent "Coupling."
2023-08-01 16:31:47 <KipIngram   > That one was a comedy.
2023-08-01 16:32:02 <KipIngram   > My wife and I thought it was fantastic.
2023-08-01 16:32:05 <gordonjcp   > Tony Curtis and Roger Moore as millionaire playboys in Deep Legal Shit, whose only chance to stay out of jail is to work as fixers for the local judge
2023-08-01 16:32:12 <gordonjcp   > I've seen bits of it, it's good
2023-08-01 16:32:15 <KipIngram   > Particularly the "Jeff" character.
2023-08-01 16:32:56 <KipIngram   > There was one Coupling episode called "The Giggle Loop" that is just thoroughly seared into our memories.
2023-08-01 16:33:10 <gordonjcp   > heh, I'll need to look out for it
2023-08-01 16:33:18 <gordonjcp   > oh!
2023-08-01 16:33:33 <KipIngram   > I really liked the time that show took to set jokes up - you'd wait the whole episode for the punch line, but man when it came it was worth it.
2023-08-01 16:33:44 <KipIngram   > So much better than the "instant snicker" most shows go for.
2023-08-01 16:33:47 <gordonjcp   > you will either need to hunt online DVD shops for it, or find some sort of torrent of bits that may arrange themselves appropriately
2023-08-01 16:33:52 <KipIngram   > Seemed "more intellectual" somehow.
2023-08-01 16:33:55 <gordonjcp   > get hold of a series called Tutti Frutti
2023-08-01 16:34:20 <gordonjcp   > it's from the 1980s, about an aging rock'n'roll band doing a comeback tour
2023-08-01 16:34:36 <KipIngram   > One moment - about to have to speak in this meeting.
2023-08-01 16:34:53 <KipIngram   > This is our director - guy makes me nervous like I'm still a damn teenager.
2023-08-01 16:35:11 <gordonjcp   > but, it's written by John Byrne, so it's all quite out there
2023-08-01 16:35:12 <gordonjcp   > righto
2023-08-01 16:43:10 <thrig       > it's from the 2020s and the same aging rock band is still rolling
2023-08-01 17:10:16 <KipIngram   > Back now.
2023-08-01 17:10:24 <KipIngram   > Always glad to get that one over with.
2023-08-01 17:10:30 <KipIngram   > Rest of the week will be easy now.
2023-08-01 17:11:06 <KipIngram   > Never know what to expect from this guy, and he's one of those old crusty stern types that can really rip you a new one if he decides to.
2023-08-01 17:27:54 <KipIngram   > In the instructions for making smithForth, step 3 says "Write the result into field p_filesz of dmqc.dmp."
2023-08-01 17:28:25 <KipIngram   > What does that mean?  I don't have a dmqc.dmp file, and don't see "dmqc" in SForth.dmp file.
2023-08-01 17:31:26 <KipIngram   > Oh, he has a script that does whatever it is.
2023-08-01 17:31:43 <KipIngram   > Ok, I immediately don't like smithForth's choices about "command line mechanics."
2023-08-01 17:31:58 <KipIngram   > Looks like it does NOT do the termios diddling you have to do to get the "normal" behavior.
2023-08-01 17:32:39 <KipIngram   > Also apparently it's all upper case.  :-|
2023-08-01 17:36:02 <KipIngram   > planckForth doesn't either - output comes on the line FOLLOWING my input instead of to the right on the same line.
2023-08-01 17:36:08 <KipIngram   > And neither actually says "ok."
2023-08-01 17:41:06 <veltas      > KipIngram: I think SmithForth is going for ANS or Forth200x, and it's simpler to just force uppercase for that
2023-08-01 17:41:26 <veltas      > Because uppercase must work, lowercase is optional
2023-08-01 17:42:11 <thrig       > these days lowercase is the norm, unlike the shouty 60s
2023-08-01 17:43:48 <veltas      > I don't know why, Forth just looks more 'correct' in uppercase to me
2023-08-01 17:46:13 <gordonjcp   > nothing looks correct in uppercase
2023-08-01 18:04:20 <KipIngram   > I'll have to side with gordonjcp on that one, but I recognize it's a matter of taste.
2023-08-01 18:04:39 <veltas      > Of course, and I'm not surprised if lots disagree on that
2023-08-01 18:05:52 <thrig       > DNs 0X20 iS ThE bEST caSE sYstEM to use
2023-08-01 18:09:59 <KipIngram   > I read somewhere a statement by Chuck that he'd wanted an "advisory" standard - one that just gave recommendations and best practice; I don't remember the phrase he used for it.  But he said others insisted on set of strict requirements.  I'm definitely in Chuck's camp on that.
2023-08-01 18:10:29 <KipIngram   > Being flexible and moldable is part of Forth's whole philosophy.The very concept of a strict standard is in opposition to that spirit.
2023-08-01 18:12:14 <veltas      > Having a standard is a good idea, you don't need to follow it.
2023-08-01 18:12:50 <veltas      > In that sense it's 'advisory'
2023-08-01 18:13:06 <veltas      > And the actual required parts are not that big
2023-08-01 18:13:17 <KipIngram   > Well, in that sense all standards are advisory.
2023-08-01 18:13:45 <veltas      > And most Forths can be adapted to the standard with a relatively small patch or library
2023-08-01 18:16:32 <veltas      > The big issue with the standard is its awful choices, which is symptomatic of most committees
2023-08-01 18:16:35 <KipIngram   > Hey, gordonjcp, is Randall and Hopkirk just one season?
2023-08-01 18:16:41 <veltas      > Good standards are the exception, not the rule
2023-08-01 18:17:19 <veltas      > For all its (many) flaws, I consider ANSI C to be a good standard
2023-08-01 18:26:36 <KipIngram   > The upper case doesn't bother me as much as the borked text layout.
2023-08-01 18:28:19 <thrig       > pasted python pasta, anyone?
2023-08-01 18:30:49 <veltas      > nn
2023-08-01 19:21:39 <KipIngram   > I used pforth in a consulting project once - put it on a network socket and used it as an interface to a pick and place machine.  The goal was to do minimal work on it, but I nonetheless spent a couple of hours rectifying that console behavior.  Probably didn't even matter given I was going to socket it, but I still fixed it before I started on that part.
2023-08-01 19:21:43 <KipIngram   > Kind of a matter of principle.