To avoid any uncertainty: the subject is a parody for similar
discussion on perl6-internals@ about Parrot.

What I want to say is that after N years of developing Perl 6 we do
not have a practical (P in Perl stands for Practical) tool which can
work in real life.

Even worse: today there is no way to speed up the interpreter - it is
good for parsing the language but is very slow for real applications.
(mod_pugs is not the case while there is no command-line tool).

We have Pugs to play with. We have no tool to run with.

Current Pugs have options to compile the programme into several
intermediate languages, for example PIR:

You can
    pugs -CPIR helloworld.pl > helloworld.pir

But you cannot later do this:
    pugs -BPIR helloworld.pir

Standalone parrot does not help.

What is nedded is a very simple step: complete the Pugs compiler so
that it could provide good PASM (PIR). Going this way we will achieve
the main goal: it will be possible to use hi-speed applications
written in Perl 6 today. No care the language itself is not fully

It will change the status of 'conception' and 'project'
to 'utilizing' and 'using'.

A bit of fun is that three years ago the situation was better: language
was more poor, the tools for more pure. In April of 2004 I made
a toy server with demos of how real Perl 6 works on a web-server. There
was a bit of examples with comments in Russian that were written on
real Perl 6 of that day, compiled into Parrot bytecode (.pbc) and run
on Apache under Parrot virtual machine. I cannot do the same with
today's instrumentary (or maybe I do not know how to?).

Thank you for understanding :-)

Andrew Shitov
[EMAIL PROTECTED] | http://www.shitov.ru

Reply via email to