bkuhn wrote:
> > Why should we center our entire design around C?
Adam Turoff <[EMAIL PROTECTED]> wrote:
> Because Perl is a write-once-run-anywhere platform, and C is the only
> viable way of maintaining Perl support on all of the platforms currently
> supported.
>
> Because most (all?) of the people interested in implementing Perl6 are
> extremely fluent in C.
>
> Because C doesn't have Java's unnecessary complexities which
> hindering large-scale projects.
These are all good reasons to implement perl6 in C, and I agree with them.
Let's implement perl6 in C!
However, my proposal is that we also design perl6 in a way such that is not
intimately dependent on any particular language.
> Because it's a Simple Matter Of Programming to reimplement a C program as
> a Java program, given enough tuits.
Of course it is. I wasn't trying to disprove Turing Completeness. ;)
However, polynomial time can be really, really big, and the job of writing a
polynomial time algorithm can be even longer. Why not make the job easier
for those of us who have to do it by holding off on making language-specific
design decisions at the higher levels?
> Because the Python folks didn't have a problem basing JPython off of
> CPython.
Actually, this one isn't a good comparison. Python is substantially easier
to parse, and, is a much simpler language. I like Perl because it is more
complicated, but that also comes with a burden that we design its
implementation more carefully.
Larry worked for a summer trying to mold the C implementation of Perl into a
JVM port. I have spent substantially more time than that on it than that.
It's a hard problem. I'll be happy to send you chapters of my Master's
Thesis that I am finishing up this month, that is making the detailed
arguments as to why it is a hard problem.
I believe the difficult that we've had porting perl5 to the JVM is a bug in
perl5's design. I am trying to encourage people to fix that bug in perl6.
--
Bradley M. Kuhn - http://www.ebb.org/bkuhn
PGP signature
