(2023-04-23) "Yes - extensible is the key"
------------------------------------------
No.

It isn't.

How are named words in Forth-alikes different from line numbers in VTLs? I
mean, semantically, they are just points in memory where control is passed
to. Imagine being able to assign aliases to particular line numbers. Do they
suddenly make your language more "extensible"? Or is the whole idea of
"extensibility" just based upon syntax, with the main feature being your
custom words look exactly like predefined ones? If that's the case, that's
definitely convenient but kinda lame at the same time. In case you really
need that kind of convenience, you can use VTL as a sort of "meta-assembly"
that everything else compiles to. Kinda like what JS is these days.

Piling up libraries, which comes as a natural result of extensibility, is not
the answer. Composing your solution from tiny programs that each do one task
only is more like it. At least when you're able to organize proper IPC. If
you're limited to loading a single program at a time, you can recreate the
same behavior within it, but it doesn't have anything to do with
extensibility of the language itself.

In case you're wondering, I was responding to Ken Boak aka monsonite ([1])
who had created several stack-based programming languages that, among
others, inspired my creation of Equi.

--- Luxferre ---

[1]:
https://retrocomputingforum.com/t/keeping-things-simple-tiny-languages/1441/3

You are viewing proxied material from hoi.st. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.