"Perl 6 is the long-awaited redesign and reimplementation of the popular and venerable Perl programming language. It's not out yet - nor is there an official release date - but the design and implementations make continual progress. Innumerable programmers, hackers, system administrators, hobbyists, and dabblers write Perl 5 quite successfully. The language doesn't have the marketing budget of large consulting companies, hardware manufacturers, or tool vendors pushing it, yet people still use it to get their jobs done. Why argue with that success? Why redesign a language that's working for so many people and in so many domains? Sure, Perl 5 has some warts, but it does a lot of things very well."
Post a Comment
Member since:
2005-07-09
Member since:2005-07-07
Member since:2005-07-08
Member since:2005-07-06
Sorry, but some people can't stand seeing ( and ) all over the place.
While LISP has the features, most find it quite ugly and hard to follow (harder than Perl).
Besides, what's wrong with Perl6 doing all this stuff? LISP did some of it first. So what? Perl6 is not LISP, very clearly.
Member since:2005-07-09
Member since:
2005-11-11
perl6 has been about promise but little else. i have been following development of this project for some time as an observer. the absence of many longtime core perl luminaries from this project is troubling - the "new generation" who signed on to make this project happen have not been able to deliver, and appear not to care. i have watched annoying bickering on lists, name-calling, people being jerks, and in the background, almost no progress. so much of perl6 will be new that no one is going to commit to the full stack until there has been a serious bake-in period, and even a full 1.0 release is at least two years away.
in the meantime ruby is delivering on most of the promises of perl6 at the language level, even if its performance is not great. these problems would not be solved by parrot in any case - the perl6 team is being hopelessly naive about the premise that one VM can accelerate all languages.
Member since:2005-09-26
Member since:2005-11-11
>> being able to mix different libraries from different languages will be a more important feature.
how????? you are presume that these languages type the same way, have the same call semantics, return types etc etc etc...even if this can be abstracted in the vm, it makes no sense at the language level. this is but one pipe dream of the parrot team, and it is why eventually they will give up and just use pugs or get parrot to do all it is ever tasked with doing - run perl6.
Member since:2006-01-01
Member since:2006-01-16
Lot's of useful CPython libs are written partially in C (like numeric python, pygame and so on); they rely heavily on CPython features and are absolutely not portable. There is no way you can get those libs to run on Parrot without a full rewrite, the Python community isn't going to do that and if somebody in the Perl community has to do it why not do it in Perl?
The same should hold true for other major languages like Ruby.
Member since:
2005-08-22
Member since:2005-11-11
i have been using pugs since its first release. yes it is nice but pugs != perl6. the more time they spend on this platform, the less time they are spending developing the platform they actually intend to release.
my prediction is that within a year they will see this and simply make pugs their release code, realizing that the very small group of people who still care abou perl6 will just make due with it.
Member since:2005-07-06
Member since:
2006-01-11
* Either they should enhance Perl5 in a compatible way : Faster, better libraries...
* Either the creators of Perl realize that Perl is becoming obsolete, comparing with Python & Ruby, and a new language could be created from ground up for fun and profit. Perl 6 introduces such dramatic changes that one wonders why don't they fully abandon the Perl syntax ( and its syntaxic noise ... )
The goals behind Parrot seems equally strange. The idea that a register based VM could change significantly the performance seems naive when modern VMs use JIT compilers and adapts to the real number of the host CPU's registers. That VM is also too low level to permit advanced optimisations ( see LLVM ... ).