This is actually one of the first really positive reactions to Perl 6 I've seen so far. It seems quite nice indeed, but I have yet to really try it for myself.
From what I have seen, "Perl 6 is a more practical Haskell" is reasonably fair, though elides a lot of things. And that is actually the most interesting question in my mind; if Perl 5 is a Swiss Army Chainsaw, Perl 6 is a multidimensional Swiss Army Meta-Chainsaw. I question whether there's really that many people who can safely wield such a device. But I'm glad someone's trying it, the answer is worth knowing.
And being the first language with arguably a more fungible grammar than Lisp will be interesting too. So many wonderful things can be done with that.... and so many more so very, very terrible things. It will certainly be interesting.
I question whether there's really that many people who can safely wield such a device.
Remember that much of the motivation behind Perl 6 design decisions is to improve the defaults such that the multidimensional artifacts can stay safely stowed from people who and in programs which don't need them.
I'm talking about the still-hypothetical "usable public release" for both of them. Despite the variance in calendar ages, both are really only now starting to come to some level of practical usability, it seems to me.
Not porn. Its a static yellow HTML page with black text with the author talking about his experiments with Perl 6, Haskell and the similarities between them.
This is always one of my motivations for learning a new language.
In my field of business (development / .NET), there is no way C# is going away as the language of choice, so learning new languages will very rarely allow me to use those fully in the line of duty.
However, learning different languages, addressing problems differently, allowing different kind of constructs with different levels of flexibility often reveals other solutions to problems and allows you think about things outside your normal day-to-day language's constraints.
Often the result of learning new languages is that I see concepts and constructs which are neat and useful and which I like to the point that I will backport it to C# if it can be done in a non-intrusive and not too cludgy way.
I expect this to yield true for many other programmers as well.
That gets trotted out as the counter argument (even with vb .net vs c# where the 2 languages are very similar), but it's moot. Code reuse doesn't happen without clean interfaces (which reduces the problem of different languages) because few programmers will read your code even if it's in the language they want.
That code reuse doesn't happen is really a myth and where I work, things gets reused a lot.
That said, what language to use boils down more to code-maintenance than code-reuse since on .NET code reuse can (and will) happen transparently across CLR languages.
I work for a software vendor and having high quality, maintainable code which others can work with is a requirement for everything you do. Using some "exotic" language would not be fully compliant with that requirement.
This is why hosted languages are awesome. You can sneak in your $AWESOME_LANGUAGE and have it interact with the legacy application. Granted, this won't work in every situation, but a .NET shop should have a lot of IRON options.
