(And yes,I realize that's what this project is trying to fix)
NQP is pleasant to work in. Perl 6 regexes are wonderful(although I haven't had time to work on writing a language using NQP's support for them yet).
Perl 5? I'm finding myself increasingly attracted to it, in part due to being around some very influential Perl 5 people. I haven't been using it much given that most of my coding time is currently devoted to Parrot, and most of the rest is dabbling in C+Lua, but I'm a new developer who's interested, certainly, in joining the Perl community.
It seems to my admittedly inexperienced mind like most of Perl 5's flaws are fairly avoidable with the right modules from CPAN (or even just the right modules/pragmas that ship with Perl). Sure, there's a lot of bad Perl 5 code out there, but the only languages for which that isn't true are the languages for which there isn't a lot of code of any kind out there. Perl 5 is not nearly bad enough a language to deserve its reputation. It's not nearly my idea of a perfect language, but it's a language I like and that I can stand to use while I wait for Parrot and Rakudo(assuming no other Perl 6 implementation shows up out of nowhere and somehow gets to performant and spec-complete status quicker than Rakudo) to mature, or for another language I can love to arrive on the scene(a better Common Lisp would be one of these).
Tl;dr: to answer your question, I'm a new Perl developer. I'm not really starting any substantial projects in Perl 5 yet(although I am in NQP), but when I have more free time on my hands, I am confident that I will.
I do have extensive experience with PHP, some personal projects in C++ and have written some very useful (for me) Lisp code.
Perl seems to be a Lisp without macros (lisp-macros) with C-ish syntax. Actually, with lots of ready to use macros, but without the easy possibility of adding your own macros. And it is faster than PHP, Python, Ruby, etc. So it has been good so far.
To write web pages I'm using Mason.
I am genuinely curious where you got your ideas about on perl usage?
Regardless of growth or decline in any of Perl, Python, and Ruby, my opinion is they're so equivalent it's barely worth discussing them separately. Pick the one that has the libraries and frameworks you need. It so happens that http://search.cpan.org is the easiest way to check up on libraries for any of the languages.
Perl6, if they ever pull off a fast and stable release, is another matter. It's likely to be a game changer. I actually quite dislike this perl5i thing and the declarex syntax changing modules in general because they needlessly fragment the language.
It means that for basic stuff they are pretty much the same. Variables & data structures (scalars, arrays, mappings), creating and calling functions, creating and instantiating classes, looping, if/then/else, talking to the network, talking to the OS, talking to the db, etc. This basic stuff is in all these languages; the only real differences are in the syntax (and that's not even very different).
For many programming tasks, you're just doing these basics over and over again. Any language will work, so you may as well pick one that's got it where it counts -- modules, since that will make the most difference in how much extra code you have to write. And Perl wins that competition with the CPAN.
> Would they have gone off in search of the Holy Grail with Perl 6 ...
Thing is, it's not a Holy Grail. It's just the next revision of Perl. Take a look at some example Perl 6 code online; they've cleaned things up, fixed some boo-boos, and upon seeing it I expect you'll matter-of-factly say, "oh, yeah, still looks like Perl :)".
http://en.wikipedia.org/wiki/Perl6#Goals
Sure, at some high-level they're all the same, but the devil is in the details. Just saying they're all the same is specious.
If you have some specific questions, I suggest asking them at http://perl6.org/community/irc , which I've found to be very friendly and helpful.