Re: [Gnu-arch-users] new language, arch, furth, etc.

[Tom Lord]

    cduffy> And RFC822 isn't [a language]?

    walters> It is a language.  But it is a very well known one that
    walters> is already in use in arch and works well.

No, it doesn't work well.   It's a constant annoyance and has gotten
in the way repeatedly.

Here's an example.  I will give you the following programming problem
to solve:

        Using all new code (nothing copied from arch), write a library
        routine that reads a log file, extracts a particular header,
        and gives back a data structure that is a "natural"
        representation of the data it contains.

        Your program must not "know" about any specific arch headers.

You can't solve that problem because the only way to get a passing
grade is to write a program that reads the header:

        New-patches: bling blang bloom

and returns an array of values:

        answer[0] = "bling";
        answer[1] = "blang";
        answer[0] = "bloom";

Your program must also read this header:

        Summary: i fixed something

and return a string:

        answer = "i fixed something";

It must read this header:

        Renamed-files: this that one other

and produce the answer:

        answer[0][0] = "this";
        answer[0][1] = "that";
        answer[1][0] = "one";
        answer[1][1] = "other";

If I relax the rules a bit and let your program "know" about specific
arch headers you can solve the problem.   But then what happens if I
define a _new_ header: you'll still have to update your program.


    >> I thought that Turing-complete was a quality that languages may or may
    >> not possess, not a prerequirement for being a language at all.

    > The point is that Tom's example did not demonstrate the need for a new
    > Turing-complete language.

Nor am I proposing that the version variable language be turning
complete: only that it be a subset of a (good) turing complete
language.

-t