On Mon, Jan 25, 2021 at 11:37:25AM -0500, Dan Cross wrote:
On Mon, Jan 25, 2021 at 11:05 AM Larry McVoy
<lm(a)mcvoy.com> wrote:
On Mon, Jan 25, 2021 at 10:55:34AM -0500, Richard
Salz wrote:
Osterhout's Tk was beyond amazing.
Still is, really. So far as I know, nobody has come up with anything
better.
The Inferno operating system that was essentially a commercialization of
plan9, implemented Tk with the Limbo programming language (which in many
ways is perhaps the most direct ancestor of Go). That was neat to play
with. Too bad it didn't have a lot of success.
We did something similar, I hated Tcl so much I paid a friend to make
a compiler for a very C like language that compiled to Tcl byte codes.
It's really what I'd like to see C evolve to:
http://little-lang.org
It had no XDR
because it was "reader makes it right" and datatypes
were tagged.
That's the first I've heard of that and I really like it. Most of the
time, you are on a network of machines that are the same, so why have
a network byte order, reader makes it right will just work. Neat.
I guess I don't quite understand that. I can get how it works for simple
data types (integers, floating point numbers, perhaps strings) but it seems
like it breaks down pretty quickly for anything with a more complex
representation (structures with multiple members, for instance; how does
one deal with padding, etc?).
Yeah, good points and I suppose that is why Sun did network byte order.
It's still appealing to have reader make right if you can do it, though
with todays out of order CPUs providing a pretty high instructions
per cycle maybe it just doesn't matter. There is a paper waiting to
be written.
--lm