Parrot 2.0.0 "Inevitable" Released

http://use.perl.org/~chromatic/journal/40114?from=rss&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+PlanetPerl+%28Planet+Perl%29&utm_content=Google+Reader

by fogus • 17 years ago

26 11 17 years ago

use.perl.org

btilly4• 17 years ago

I'll believe that it will support "all dynamic programming languages" when they support Python.

Note that I used to believe in the project. I donated money to help them. Then I started to wonder what was going on there. Then for me the clincher was listening to Sam Ruby's presentation at OSCON a few years ago about his experience in trying to get Python running. From then on I've regarded Parrot as a fun project for those involved which I won't take seriously. Since then they've cleaned up development, they are making progress towards a far less ambitious Perl 6 than they set out to create, and I wish them well. But I personally doubt it will ever become widely used as a production platform, and I guarantee that it will never be widely viewed as the prime platform for running a wide variety of dynamic languages. If for no other reason than the fact that they don't have the people and resources to match what is being done on other platforms. (Have you paid attention to what Google has done for JavaScript execution?)

Just consider the facts. Python is a well defined target with multiple ports to different virtual machines. It is basically a perfect example of a realistically complex language in wide use. If you look at his background (see http://en.wikipedia.org/wiki/Sam_Ruby), Sam Ruby is pretty much the ideal person to work on a cross-language virtual machine. There has been literally no person that I've heard of in the Parrot project before or since who has as much relevant experience. And he failed.

The reasons why he failed are complex. But the biggest were all related to specific people problems (luckily mostly fixed since), and a design myopia that believes that Perl 6 subsumes all other dynamic languages, and therefore internal conventions that support Perl 6 efficiently will support everything else well (emphatically not true). Half those problems are gone. The people issues have improved, the design myopia has not.

Wake me up when they get serious about supporting Python. Until then any claims of "all dynamic languages" should come with huge disclaimers.

chromatic4• 17 years ago

As I've posted elsewhere, the project to support Python 3 on Parrot is Pynie. Its repository is: http://bitbucket.org/allison/pynie/

You can see the entire list of languages supported and in development on Parrot at: http://www.parrot.org/languages

btilly4• 17 years ago

And I've responded elsewhere at http://use.perl.org/comments.pl?sid=44429&threshold=0.... For those who don't want to follow the link the short version is that I am unconvinced and still think the claim is false advertising.

pmarin• 17 years ago

Most of the parrot languages in http://www.parrot.org/languages are sadly outdated or unmaintained. Many of them even not work in the current version of Parrot.

guelo• 17 years ago

From the release announcement: "Parrot is a virtual machine aimed at running all dynamic languages."

Did you miss the word "aimed"?

garnet7• 17 years ago

> Wake me up when they get serious about supporting Python.

Looks like Allison will be giving a talk on Pynie at PyCon: http://us.pycon.org/2010/conference/schedule/event/88/

And, regarding difficulty in implementing a given language, hasn't the PCT ( http://docs.parrot.org/parrot/latest/html/docs/book/pct/ch03... ) made that a lot easier of late?

Parrot 2.0.0 "Inevitable" Released

Related Stories