Did you ever wonder, while implementing this system, whether future generations of Perl should allow for distributed computing in an integral way. In a similar way to how accessing modules is integral to the language, for example.

Sorry, I don't follow.

Accessing modules in Perl is a useful aspect of the language, but it's not necessarily integral to Perl itself.

My belief is that it would be great to have a Perlish language that is a GRID language as well as an ordinary stand-alone language.

Can you define these two language aspects?

The way I see it, we use language to tell the computer what we want it to do. So one piece of code chops a job up into smaller parts, from 3-4 to millions; and another piece of code runs the smaller piece and returns a result.

One way I can see that changes that viewpoint in a really interesting way would be for the computers to work something like a recycling plant -- large things are input, they are broken down, stripped, cleaned and reassembled into something new and useful.

One way I can see that changes that viewpoint in a really interesting way would be for the computers to work something like a recycling plant -- large things are input, they are broken down, stripped, cleaned and reassembled into something new and useful.

For a very straightforward and simple (yet useful) application of this idea see: mapreduce.

update: there's a module on backpan, but none on search.cpan.org.

"What should it profit a man, if he should win a flame war, yet lose his cool?"