[Toybox] [PATCH] clear.c: Clear scrollback buffer on non-vte (gnome based) terminals

[enh]

On Tue, Mar 5, 2024 at 11:49 PM Rob Landley <rob at landley.net> wrote:
>
> On 3/5/24 18:31, enh via Toybox wrote:
> >> We have a 7-10 year support horizon, How many terminal escape protocols have been relevant
> >> in the last 10 years: One. The story is the same for UTF8 and LP64
>
> There's a certain amount of 80/20 going on.
>
> If you 80/20 twice you get 96%. (80% of the remaining 20%.)
>
> If you 80/20 THREE times you get 99.2%.
>
> Diminishing returns kick in real fast. I usually do one round and wait for
> people to show up with use cases.
>
> > (to be fair, toybox does support ILP32 too, not just LP64 --- it's
> > just weird stuff like Windows' LLP64 that's explicitly out of scope
> > aiui.)
>
> They made their bed, explicitly and intentionally:
>
>   https://devblogs.microsoft.com/oldnewthing/20050131-00/?p=36563
>
> And then they made Windows Subsystem for Linux. (Twice, apparently.)
>
> > i honestly still
> > don't believe hurd actually exists!
>
> It technically exists, it just doesn't WORK.
>
> > i've never seen it, whereas i've
> > actively used all the others you mention, plus the two i just
> > mentioned, and Tru64 too. if you want "obscure but definitely a real
> > thing", how about Plan 9?)
>
> Eh, that wasn't exactly obscure, it was just tied up in really stupid
> AT&T->Lucent licensing shenanigans until ~Y2K (so that nobody could SEE it
> without forking over thousands of dollars, rumors of greatness but nobody had
> personal experience), by which point it was about ten years moot.

my university used it --- one reason i have so much hate for AIX is
because i plan 9 was basically my first unix, and the alternatives
were slow and bloated. (AIX makes a lot more sense as a grown up who
understands that IBM's business is not products but consulting.) IRIX
was at least _very_ pretty, but i still wrote my own window manager
because the startup time of 4dwm was unacceptable on the Indy [though
i think the Indigo wasn't as bad --- "The Indy is the Indigo without
the 'go'"]. (see also
http://www.art.net/~hopkins/Don/unix-haters/tirix/embarrassing-memo.html
for more along those lines!)

but plan 9 is still the only system i've used where _all_ the source
was just there by default -- and in a manageable quantity! -- and you
could just edit and rebuild trivially. don't like something? change
it! share your changes with your friends if they agree. i can imagine
no better system to have learned unix on. (though to be clear: it
taught several "how not to do it" lessons too. fixed-length buffers
and arbitrary limits made it hard to love at times, and ironically
pushed people towards stuff like perl that didn't have an opinion on
what a "reasonable" line length was, etc.)

> But it got
> looted for a bunch of ideas like procfs, and Linux virtfs is a v9fs is 9p2000.L.
>
> https://landley.net/kdocs/ols/2010/ols2010-pages-109-120.pdf
>
> You want obscure, the Bell Labs guys kept doing Unix releases after v7, all the
> way through v10, before starting over with Plan 9. They just never got published
> outside of the labs (for the same reason Plan9 didn't, the commercialization
> drive). They finally got released to the public about the same time Plan 9 did:
>
> https://www.tuhs.org/Archive/Distributions/Research/
>
> By the way, while FIPS 151-2 was in force and posix compliance was a requirement
> to qualify for federal procurement contracts, EVERYBODY did a Unix. Apple did a
> unix for mac hardware in 1988:
>
> https://en.wikipedia.org/wiki/A/UX
>
> Dell did a Unix:
>
> https://gunkies.org/wiki/Dell_UNIX
>
> Commodore did a unix for the amiga:
>
> https://en.wikipedia.org/wiki/Amiga_Unix
>
> Microsoft lied and claimed that Windows NT was a unix:
>
> https://en.wikipedia.org/wiki/Microsoft_POSIX_subsystem
>
> The "unix wars" got REALLY WEIRD for a while there.
>
> Rob
>
> (P.S. When I say "weird" I mean https://en.wikipedia.org/wiki/Eunice_(software)
> which yes, I used as a teenager.)