<tp>
"f0-touch" is the project name and 879d8c3eb7 is the first 8 characters of the SCM hash
<tabemann>
I use conventional version numbers because I otherwise won't remember what is what
<tabemann>
and also so the user knows what's newest
<tp>
and it all winds up as f0-touch.879d8c3eb7.tar.gz
<tabemann>
and how significant each version is
<tp>
I plan to just have a changelog of the different hashes
<tp>
i can do arbitary version numbers but cant be bothered
<tp>
it's only a simple tag for me
<tp>
I'm such a lazy coder I should be a legend ...
<tabemann>
you're not a lazy documenter
<tp>
no, thats cause I'm not a programmer ;-)
<tp>
I actually like writing accurate and efficient doc
* tabemann
is the kind of coder who usually thinks that code with comments minus code is adequate documentation ;)
<tp>
code with comments minus code?
<tp>
all great programmers hate documentation
<tabemann>
i.e. documents which are just word names plus signatures plus the comment that goes along with such in the code
<tp>
thats not a criticism, only a observation
<tabemann>
*documentation
<tabemann>
I've tried to not be a great programmer as of late
<tp>
we are what we are
<tp>
hahah
<tp>
you have to remember this true story, once a autistic girl who couldnt speak picked up a ballpoint pen and started drawing the most incredible drawings
<tp>
they included a Cossak on a horse, with drawn sword in a charge
<tp>
the eyes of Cossack and horse were the most incredible part, they seem to stare out of the drawing at you
<tp>
a psychiatrist seeing the drawings took the girl on as a special effort, determined to teach this awesome artist to speak
<tp>
and in a few years the girl could speak about 10 words
<tp>
but she never drew any more pictures, she seemed to lose the ability with her newly acquired speech
<tabemann>
I think the thing is that programmers assume other people think like them
<tp>
I draw no conclusions but note that we are what we are, and trying to be good at what we are not always runs the rick of being mediocre at whet we once excelled
<tabemann>
so they thing just signatures and simple comments are sufficient, since how the code works ought to be self-evident
<tp>
we all assume everyone thinks like us
<tp>
I must say, I really hate the programmer slogan of "good code is self documenting"
<tp>
because I know it to be utterly untrue
<tabemann>
think of my *_guide.md files versus my *_words.md
<tp>
or rather only true in very limited cases
<tabemann>
my *_words.md files are so spare to almost be useless
<tp>
yeah, theyre the tabemann equivalent of doxygen but harder to write ;-)
<tabemann>
whereas my *_guid.md files, at least to me, seem at least somewhat sufficient
<tabemann>
*_guide.md
<tp>
tabemann, I also really think that your tarballs should include the .md files as .html files
<tp>
and do away with the .nd files
<tp>
.md files
<tabemann>
do you know of a good .md to .html converter?
<tp>
i dont even include a .md file as you can see
<crc>
tp: I'd prefer both html and markdown; no reason to force someone to use a browser to read the project notes
<tp>
lol, fossil isnt a .md parser
<tp>
crc and no reason to force people to navigate .md files with a text viewer
<tp>
but it's easy to include both I guess
<crc>
I don't have an html viewer on my openbsd box
<tp>
tho I doubt people will bother with md
<crc>
or my freebsd box
<tp>
crc then youre a rarity
<tp>
I use DILLO for all my internal html browsing
<crc>
no X on either system
<tp>
it's fast light, never crashes, and cant handle JS or anything like that
<tp>
crc, well youre a coding god in my opinion, I won't argue with you on anything like this
<tp>
crc but I only read your html doc myself
<crc>
I use a mostly use a gopher based interface to my docs or the markdown files directly
<crc>
html I pretty much only access via an ipad
* crc
wonders how many prefer html...
<tp>
zillions ...
<tp>
imho
<tp>
in fact my forth doc site is very popular simply because it's html I, and also because it's the *only* html doc site for Mecrisp-Stellaris Forth :)
<tp>
-I
<tp>
the author supplies a text file readme, and frankly altho well done, it's horrible to use
<tp>
many Mecrisp-Stellaris users just access my html Definition page
<tp>
because browsers make it really easy to read and search
<tp>
i personally like the simplicity of doc creation with markdown, but dont assume because I do that users will love to read it in html
<tp>
oops I mean "read it in markdown"
<crc>
it'd be interesting to find a way to know for sure
boru` has joined #forth
boru has quit [Disconnected by services]
boru` is now known as boru
<tp>
how about a Retro Doc page with markdown and another with the same as HTML ?
<tp>
and count the hits ?
<tp>
I'm willing to wager a brand new, unused, genuine factory STM32F051 chip that HTML will win by far
<tp>
:)
<crc>
I'm sure you're right
<crc>
tracking hits would necessitate adding support for that to my gopher and http servers, which I've avoided doing so far
<tp>
I'm confident I'm right but only hard data would tell
<tp>
My first rule is: "automate where ever possible"
* crc
makes a note to explore this after the upcoming release
<tp>
My second rule is: "use data not intuition"
<tp>
so many things I was *sure* would be awesome hits just dissapeared into obscurity with a damp fizzle
<tp>
ABBA said ' we had no idea which songs would be massive hits, and were just amazed when one was"
<tabemann>
back
<tabemann>
tp
<tabemann>
that markdown converter is broken
<tabemann>
it cuts the files off partway through
<tabemann>
okay, I figured out the problem
<tabemann>
it was reading the *~ files created by emacs - lol
dave0 has joined #forth
* crc
should start testing his md to HTML against markdown from other people
<tp>
it's amazing you know someone for years, and one day they come out of the closet and say "terry, Ive something to tell you ..."
<tabemann>
I installed emacs on my work machine today, because it provides the best support for macros of any editor I've used (I have heard notepad++ also has good macro support, but I'm not familiar with its macros, unlike emacs macros)
<tp>
and then the shock sets in ... christ, theyre a EMACS user !!!!!
<tabemann>
lolol
<tabemann>
for work, though, I normally use Visual Studio Code (the horror), IntelliJ (the horror), and vi
<tp>
tabemann, actually I like EMACS, I'm editor agnostic
<tp>
yes, with github and VS, microsoft is making great inroads onto the OSS community
<tabemann>
at home, though, my primary editor is emacs, except for what git brings up, which is nano, because I had emacs set as my EDITOR, but somehow an update broke that and set it to nano
<tp>
Im definitely not a NANO user
<tabemann>
tp: I remember an April fools post on Hackaday about how Linux distributions are dropping support for vi and emacs in favor of VS Code
<tp>
it's so sad to see Linux users dumbed down
<tabemann>
people weren't so sure that it was an April fools post
<tp>
heheh, I saw that and knew it was a aprils fool
<tp>
there are still a lot of Linux people who absolutely hate microsoft
<crc>
tp: My markdown doesn't support everything, just a subset; I'm unlikely to support everything (definitely not doing all the syntax variations; e.g., I only support the tax header types, and I won't be implementing all of the link forms)
<tabemann>
I keep on using Github because it's A) convenient and B) I'm not going to move my code, or keep my code in more than one place
<tp>
it's ok, I make no judgement you sellout sycophant
<tabemann>
lol
<tp>
;-)
<tabemann>
I do hear people are moving back to sourceforge though
<tabemann>
whereas I disliked sourceforge for getting all ad-infested
<tp>
i guess it was an alternative to selling out to microsoft ?
<tabemann>
of course there's also GitLabs
<tp>
they have to pay their bills somehow ?
<tabemann>
yeah
<tp>
I just use sourceforge cause it's easy and free
<crc>
I moved off of sourceforge pretty quickly. Used darcs for a few years, git and mercurial intermittently, fossil, then bzr for retro11, and now back to fossil
<tp>
and I can host my Mecrisp-Stellaris doc site there for free
<tp>
crc I used CVS, mercurial, bzr then fossil
<crc>
I dislike relying on third party infrastructure
* crc
has a really crude version control thing written in forth, but it's not ready to distribute
<tp>
I was spending $12 a month hosting my Mecrisp-Stellaris doc on digital ocean and it adds up after a few years
<tp>
brb
<crc>
I spend about $20/month on servers, but that's covered by sales of retro and a gopher client on iOS
<tabemann>
aigh! forgot to commit my binaries for zeptoforth from my most recent release (even though they are included in the .tar.gz file I released)
kmstout has quit [Ping timeout: 246 seconds]
kmstout has joined #forth
kmstout has quit [Client Quit]
rdrop-exit has joined #forth
<tabemann>
hey
<rdrop-exit>
hey tabemann
* tabemann
now has a script to convert all his Markdown docs to HTML
<tabemann>
so that when I create release .tar.gz docs it will have documentation in HTML rather than Markdown
<tp>
cool!
<rdrop-exit>
congrats
<tabemann>
I've automated the release process as much as reasonably feasible at this point
<tabemann>
i.e. building all the kernel binaries, building full binaries for selected platforms (because switching cables is necessary this is not fully automatable), and then producing a .tar.gz file
<rdrop-exit>
I like pdf docs
<tabemann>
uploading .tar.gz files to github and to hackaday is not presently automated
<tabemann>
pdf docs can be prettier, but they're harder to generate
<tabemann>
e.g. generating PDF from Markdown probably would involve converting Markdown to LaTeX, and that to PDF
* tabemann
remembers the days he made a habit of writing his assignments for school in LaTeX - lol
<tp>
tabemann, I'm still a big latex fan
<tp>
but latex is just not suitable for the web imho
<tabemann>
LaTeX is suitable when you want the equivalent of a paper hard copy of a document
<tp>
sure
<rdrop-exit>
yes, which is what I want, a document of similar quality to a book
<tp>
but then thats not the web, thats " the equivalent of a paper hard copy"
<tp>
rdrop-exit, and for that LaTeX utterly rules
<rdrop-exit>
so what if it's not the web?
<rdrop-exit>
yes
<tp>
so what if it's not a book ?
<tp>
:)
<tp>
some of us cant be bothered trailering 5 tonnes of books around
<rdrop-exit>
Web-based documentation usually suck
<tp>
feeding 100 million Silverfish
<tp>
damaging our wrists holding 10 lb books up in bed
<rdrop-exit>
kids think generating a html or md from their code is documentation, usually it's garbage
<tp>
everything sucks, it's only the matter of how much that differentiates them
<tp>
as always, it's it's the tool that get blamed when someone has no gift, in this case *writing*
<tp>
if you have no writing skill anything will look crap, including latex
<tabemann>
rdrop-exit, but that's the problem with basing documentation on code comments and signatures, not on HTML or Markdown versus LaTeX
<tp>
and if you do, you can write awesome doc with Microsoft Paint
<tabemann>
I bet you can have documenation that is just as shitty with LaTeX as with HTML or Markdown - after all, one can have automatic documentation generation tools that use the same source for all three
<rdrop-exit>
Appendices can be generated, but documentation is written
<tp>
sadly documentation is a dying art
<tp>
onve companies employed skilled professional documentation writers, now it's all doxygen
<tabemann>
yes, I remembe reading about that
<tabemann>
my company doesn't even have doxygen
<tp>
do they do any doc ?
<rdrop-exit>
just compare the docs written in the 70s, e.g. Bell Labs, to what kids consider documentation today
<tabemann>
at least 95% of the code at my work is completely undocumented
<tp>
yeah
<tp>
tabemann, so sad
<tabemann>
doxygen would be a huge step up
<tp>
or the nasa docs, look at the doc for the lunar lander guidance computer is beyond awesome
<tp>
it's many thick documents, all utterly precise
<rdrop-exit>
As long as you expect to generate your docs, your docs will be garbage. Only the appendices should be generated.
<tp>
and that decice has ZERO faults or bugs
<tp>
in 1969 they went to the moon with a hand assembled computer with NO MCU's in it, the code was written by one guy
<tabemann>
my work is focused on getting bugs fixed for the next monthly release pronto, not documenting their code
<tp>
and it was 100% bugfree, the astronauts survived
<tabemann>
they don't seem to realize how self-contradictory that is
<tp>
tabemann, a bit soul destroying ?
<tabemann>
tp: at my previous job, I worked on MRI machines, and their code was the same exact way
<tp>
now in 2020, Boeing cant even launch a capsule into the right Earth orbit without spending 4.8 Trillion $ to get a 'clock sync error'
<tp>
tabemann, wow, thats cool
<tabemann>
both this job and my previous job have requirements for sure, but not documentation
<rdrop-exit>
people just go through the motions, rather than trying to produce quality code and documentation
<rdrop-exit>
Good documentation takes work
<tp>
well it's all about profit nowdays I think
<tabemann>
yes
<rdrop-exit>
and thought
<tp>
it takes skill and many edits
<tp>
and a gifted writer
<tp>
and costs lot$
<tp>
theyre a dead breed, those doc people
<tp>
they understood every aspect, researched every detail
<tabemann>
people today seem to view requirements as a replacement for documentation
<rdrop-exit>
Using automation to generate more garbage faster is a step backwards
<tp>
you can dl the NASA Apollo computer doc, I recommend people have a glance at it, it is truly awesome
<rdrop-exit>
agreed
<tp>
well the nasty Coronavirus seems to hate all the new stuff, we may see may old skills come back again
<tp>
many
* tabemann
has spent the last week and a half working on a feature that consisted of one page of requirements; if one just glanced over the requirements one might miss the complexity of the task
<tp>
tabemann, ouch, thats a special skill
<tp>
tabemann, is it some type of torture, are you in the Matrix ?
<tabemann>
I hate short-but-really-involved requirements because people tend to underestimate how long it would take, as was the case here
<tp>
tabemann, is it a cunning deceit to make it appear simple and cheap while actually being the opposite
<tp>
?
<tabemann>
I told my boss I estimated it would take six days - he insisted I shorten my estimate, yet in reality it took eight days
<tp>
hahaah, CLASSIC!
<tabemann>
well more than that, actually, as one part is not complete
<tp>
what a world we live in, but nothing has changed
<tp>
this stuff has always been happening
<rdrop-exit>
Experts always make it look easy, that's always been the case
<rdrop-exit>
That's the point
<tabemann>
my boss always A) underestimates time needed for things and B) has no concept of "work-life balance" (two of those days were on the weekend)
<tp>
tabemann, your boss reminds me of Lex Luthor (gene hackman) in the film Superman, he sends his mad scientist (richard prior) to kill Superman, and when he fails (its Superman!) Lex says to him "I ask you to kill Superman ... and you can't even do that one simple thing"
<tp>
tabemann, usually people like your boss have no concept of ** YOUR ** "work-life balance"
<tp>
theirs is just fine
<tp>
rdrop-exit, so true
<tp>
my last partner was a writer, she had a degree in it and could crank out 10,000 words of utter genius in no time at all, and loved it
<tp>
she is moderately tech competent as well
<tabemann>
one thing, though, is on the other hand the more of an expert one is, the more limited one realizes one actually is
<tp>
playing Scrabble with her was like taking a cap gun to fight a thermonuclear weapon
<tabemann>
that's why I give my boss what sounds like overestimates - because I realize just how much time something will really take
<tabemann>
hence impostor syndrome
<tp>
she would get exasperated and say things like "tp, I'm going to get you out of my misery"
<tp>
tabemann, hence bonafide WISE person
<rdrop-exit>
my right hand at work was an excellent writer, she's now a big shot at twitter
<tabemann>
conversely, often the people who are most convinced they are right are actually those who are the most limited
<tp>
tabemann, your bosses underestimation just requires that you calibrate his unrealism factor higher than other more realistic people
<rdrop-exit>
I disagree tp
<tp>
howso rdrop-exit ?
<tp>
she would get exasperated and say things like "tp, I'm going to PUT you out of my misery"
<rdrop-exit>
tabemann's underlying problem is his boss doesn't defer to tabemann's expert opinion.
<tp>
edI agree 100%
<tp>
rdrop-exit, agree 100%
<tp>
hence tabemann must calibrate his boss for that by multiplying tabermans estimates by K
<tabemann>
even you stick an "r" in my name!
<tabemann>
why do people do that
<rdrop-exit>
That will never solve the underlying problem
<tabemann>
so many people have made my name taberman or tabeaman rather than tabemann
<tp>
dont look for the ocasional r tabemann, look at ALL the times I *didnt*
<tp>
i recommend changing your name to "T"
<tabemann>
I'm just amused by the fact that people always do that
<tp>
then youll rarely be annoyed
<tabemann>
why an "r" of all letters
<tp>
why not ?
<tp>
if it happens a lot there must be a reason ?
<rdrop-exit>
I gave you the reason
<tabemann>
probably because berman is a far more common last name than bemann
<tp>
probably
<tp>
rdrop-exit, ok
<rdrop-exit>
there's an easy solution to that problem, your first job is the only one you should ever apply for, you should get headhunted for the rest
<rdrop-exit>
If your boss went through the trouble of headhunting you then they have a vested interest in your credibility
<rdrop-exit>
That was how Silicon Valley worked in the 80s
<tabemann>
in this case my boss hired me specifically; he actually called up a headhunting firm and asked them to find me - he knew me from a previous job, and was impressed by what he saw there
<tabemann>
he seems to both like me as a programmer and personally, but puts me under a lot of pressure and does not really (as mentioned) respect my time estimates or my personal time
<tabemann>
the oftentimes wants both quality code and code that is implemented quickly, without understanding the contradiction between the two
<rdrop-exit>
I can't relate to the work-life balance issue, as that was not something anyone expected in the 80s in tech
<tabemann>
e.g. I every other weekend have my daughter over; but on countless weekends I have spent time working at least part of the time rather than spending time with my daughter (at least my parents are around to spend time with her)
<rdrop-exit>
but if he doesn't have confidence in you, then you should leave or make him irrelevant
<tabemann>
the problem with leaving is that the job is extremely convenient in many ways - it's close to both my house, my daughter's house, and my daughter's school - almost any other job would be further away from at least one of those three
<rdrop-exit>
I spent months at a time away from my family for work reasons
<rdrop-exit>
In my day you got work-like balance via early retirement
<rdrop-exit>
* work-life
<tabemann>
I've always had weekends free at my jobs in the past
<tabemann>
this is the first job I've worked at where I've regularly had to work weekends
<rdrop-exit>
My last job I took 9 days vacation total over 10 years
<rdrop-exit>
(9 days total, not 9 days a year)
<tabemann>
actually at my last job you were forced to have work-life balance, because it was hourly (very well-paid hourly of course), and they absolutely did not allow you to work more than 40 hours a week
<tabemann>
(in theory you were supposed to spend free time doing training materials, but I openly flaunted this no consequences whatsoever, particularly because the training was typically completely irrelevant to my work)
<tabemann>
*this with no
<rdrop-exit>
this work-life balance thing is a sign of the age
<tabemann>
you put it as if it were a bad thing
<rdrop-exit>
it just wasn't a thing, and now it is
<rdrop-exit>
it especially wasn't a thing in tech
<tabemann>
well I like spending time with my daughter, I like spending time on my projects, and I like going on vacation with her and my parents
<rdrop-exit>
of course you do
merkc0 has joined #forth
<rdrop-exit>
I have four kids and a grandaughter, I like spending time with them
<rdrop-exit>
I retired at 47, and have had a great work-life balance since then
<rdrop-exit>
like I say, it's a totally different work environment and culture today
<rdrop-exit>
than we had in the 80s
<tabemann>
okay, I'm gonna head off to sleep in a moment
<rdrop-exit>
back then it was much more "lead, follow, or get out of the way"
<rdrop-exit>
goodnight tabemann
<tp>
rdrop-exit, good advice, I agree
<tp>
a boss like that needs to be fired asap
<tp>
hahah
<rdrop-exit>
:)
<tp>
Ive walked into a few bosses office and said "youre fired"
<tp>
when they looked confused Id add, "so ill be leaving"
<rdrop-exit>
:))
<tp>
i was merciless on bosses, it's my tech nature
<tp>
if I disagreed with a instruction Id just say "no"
<rdrop-exit>
the world has changed so much, the new normal is very strange to me
<tp>
theyd say "but Im youre boss" and Id reply, I dont care, I have never seen you as my peer
<tp>
then Id add ' want me to leave right now this moment, I'm perfectly happy to go" ?
<rdrop-exit>
right, no time to lose on a losing situation
<tp>
and watch the gormeless jaw drop, all their bs was helpless against my shield of steel
<tp>
exactly
<tp>
no one ever took up that offer by the way, I really tried on occasion
<tp>
lol
<tp>
I'd say 'look it's really easy, all you have to say is "terry id like you to resign right this instant"
<tp>
it's hilarious watching the gears churn
<tp>
but of course once that happened, Id start making other plans
<tp>
believing that they would also
<rdrop-exit>
Kids live in a strange world today, and our world seems just as strange to them, the generation gap is still alive and well
<rdrop-exit>
it all started with the baby boom
<rdrop-exit>
before then, generation gaps weren't really a thing it seems
<rdrop-exit>
lunch is ready, catch you later, stay healthy
rdrop-exit has quit [Quit: Lost terminal]
<tp>
cya!
<tp>
yeah it's a stranger world now with Coronavirus
dddddd has quit [Ping timeout: 260 seconds]
deesix has quit [Ping timeout: 250 seconds]
deesix has joined #forth
nmz has quit [Ping timeout: 265 seconds]
nmz has joined #forth
gravicappa has joined #forth
mtsd has joined #forth
merkc0 has quit [Quit: Leaving]
jsoft has joined #forth
xek_ has quit [Ping timeout: 244 seconds]
DKordic has quit [Read error: Connection reset by peer]
mtsd_ has joined #forth
mtsd has quit [Ping timeout: 265 seconds]
mtsd_ is now known as mtsd
mtsd has quit [Quit: mtsd]
mtsd has joined #forth
DKordic has joined #forth
dys has joined #forth
reepca` has joined #forth
reepca has quit [Read error: Connection reset by peer]
xek has joined #forth
mtsd has quit [Ping timeout: 246 seconds]
<veltas>
tp: I've got my STM32 disco
<veltas>
It's so cool it has lights and everything
<veltas>
"Info : Unable to match requested speed 500 kHz, using 480 kHz"
<veltas>
Might have to use a non-terrible computer for this lol, not sure if that's my laptop being olde
<veltas>
tp: What should I be using to connect to this thing?
iyzsong has joined #forth
<veltas>
tp: Okay I have it working, turns out the correct thing to do was just go straight to Mecrisp Stellaris docs, go figure
mtsd has joined #forth
dave0 has quit [Quit: dave's not here]
dddddd has joined #forth
mtsd has quit [Quit: mtsd]
<tp>
veltas, "Info : Unable to match requested speed 500 kHz, using 480 kHz" ... thats cool, I get it also
<veltas>
I have given up trying to use that debug env
Zarutian_HTC| has quit [Ping timeout: 244 seconds]
Zarutian_HTC has joined #forth
jn__ has quit [Remote host closed the connection]
lonjil has quit [Quit: No Ping reply in 180 seconds.]
cheater has quit [Ping timeout: 256 seconds]
lonjil has joined #forth
Vedran has quit [*.net *.split]
jackdaniel has quit [*.net *.split]
nonlinear has quit [*.net *.split]
tolja has quit [*.net *.split]
ornxka has quit [*.net *.split]
klys has quit [*.net *.split]
fiddlerwoaroof has quit [*.net *.split]
bluekelp has quit [*.net *.split]
jn__ has joined #forth
ornxka has joined #forth
klys has joined #forth
fiddlerwoaroof has joined #forth
jackdaniel has joined #forth
Vedran has joined #forth
cheater has joined #forth
nonlinear has joined #forth
bluekelp has joined #forth
tolja has joined #forth
lonjil has quit [Ping timeout: 244 seconds]
gravicappa has quit [Ping timeout: 260 seconds]
lonjil has joined #forth
gravicappa has joined #forth
<veltas>
Luke...
<veltas>
use the forth
* tp
works on his tarball release IDE stuff
lonjil has quit [Quit: No Ping reply in 180 seconds.]
lonjil has joined #forth
cheater has quit [Ping timeout: 256 seconds]
cheater has joined #forth
gravicappa has quit [*.net *.split]
nonlinear has quit [*.net *.split]
tolja has quit [*.net *.split]
bluekelp has quit [*.net *.split]
jackdaniel has quit [*.net *.split]
Vedran has quit [*.net *.split]
nonlinear has joined #forth
bluekelp has joined #forth
gravicappa has joined #forth
tolja has joined #forth
jackdaniel has joined #forth
Vedran has joined #forth
dys has quit [Ping timeout: 260 seconds]
xek_ has joined #forth
xek has quit [Ping timeout: 260 seconds]
jsoft has quit [Ping timeout: 240 seconds]
WickedShell has joined #forth
xek_ has quit [Ping timeout: 246 seconds]
X-Scale` has joined #forth
X-Scale has quit [Ping timeout: 256 seconds]
X-Scale` is now known as X-Scale
dave0 has joined #forth
dddddd has quit [Ping timeout: 260 seconds]
jedb_ has joined #forth
jedb has quit [Read error: Connection reset by peer]
boru` has joined #forth
boru has quit [Disconnected by services]
boru` is now known as boru
gravicappa has quit [Ping timeout: 264 seconds]
X-Scale` has joined #forth
X-Scale has quit [Ping timeout: 256 seconds]
X-Scale` is now known as X-Scale
<tabemann>
hey guys
<tp>
hey tabemann
<tabemann>
trying to figure out what the heck to do next with zeptoforth
<tabemann>
intertask communication?
<tp>
woo
<tabemann>
but most intertask communication should be as simple as two variables, a value and a flag
<tp>
youre building a monster
<tabemann>
or if the value is supposed to be an address, just one variable (with zero as unset)
<tabemann>
oh, on my PC forths I wrote utterly elaborate intertask communication constructs
<tp>
yeah, PC forts grow outa control sometimes, look at gforth
<tp>
it's dictionary is intimidating
<tabemann>
hashforth is really complex
<tp>
I think CRC took the right direction
<tp>
tabemann, I did try and compile hashforth once or twice on FreeBSD a while back and it didnt complete
<tp>
so Ive never used it
<tabemann>
apparently some people have had issues with it
<tp>
in fact in my experience about 99% of pc forths are *all* buggy
<tabemann>
but as I don't use FreeBSD I have no means of debugging it for them
<tp>
of course
<tp>
i have no doubt youd fix it in 5 minutes
<tp>
I think it's the growing Linux differences to unix
remexre has quit [Quit: WeeChat 2.7.1]
<tabemann>
whereas zeptoforth's only dependency is gas
<tp>
and of course FreeBSD ha a tiny user base in comparison so it's not a criticism from me if something wont compile on FreeBSD thats made on Linux
<tp>
exactly. Zeptoforth compiles smooth and fast here
<tabemann>
all things considered, the zeptoforth kernel is tiny
<tabemann>
it's only "large" when compared to the flash on smaller cortex-m mcus
<tp>
tabemann, I've almost finished my latest round of IDE updates here so I'll have a quick look at Zeptoforth soon and see if I can add hardware handshaking
<tp>
tabemann, yeah it is pretty small
<tp>
have you seen that the Mecrisp-Stellaris kernel is only 8.5kB ?
<tp>
thats the absolutes smalles kernel only
<tp>
I've never tried it so I dont how functional ist is
<tp>
but 20kB is fine for me. I like all the utilities and 64kB flash is huge for embedded
<tp>
1MB is beyond humungously gigantic
<tp>
hey tabemann can you see why this wouldnt work ?
<tp>
case
<tp>
$C0 of 1 endof \ prgChg
<tp>
2 \ {'all others' default broken?}
<tp>
$F0 of rdrop exit endof \ SysEx
<tp>
$D0 of 1 endof \ chPress
<tp>
endcase ;
<tabemann>
the zeptoforth kernel for stm32f407 is 21184 byte
<tabemann>
simple
<tp>
a Mecrisp-Stellaris user said it his code above doesnt work and Ive no idea what the rdrop is doing
<tabemann>
2 is on top of the stack, which is dropped by endcase
<tabemann>
it has to be 2 swap
<tp>
"simple" bwahhh, you programmers are legends to us techs!
<tabemann>
rdrop exit pops the return address of the current word, assuming that one has not put anything on top of the return stack, and then branches to the next address on the return stack
<tabemann>
tp: well you almost certainly know far more about hardware than I ever will
<tp>
tabemann, well Ive spent a life doing hardware, imagine what youll be able to do when youre 66 ?
<tp>
tabemann, but why does he have to use rdrop exit here ?
remexre has joined #forth
<tp>
is it just a way to clear the return stack and jump out of the case ?
<tabemann>
no
<tabemann>
it's for returning to the caller of the calling word
<tp>
tabemann, and at least youre getting into some hardware in your 30's, I didnt start Forth until 2014 when I was 81