I don't know what "compilable Perl" has to do with the CPAN. "Compilation" suggests some sort of execution and distribution strategy, while the CPAN is an archive of resuable code and the surrounding ecosystem of dependency management, documentation, bug tracking, history, annotation, and comprehensive testing around it.
I'm suggesting that if you measure the amount of code submitted to the CPAN, the number of authors, the frequency of updates, and the freshness of versions, you'll see that Perl 5 is far from stagnant.
You mean CPAN is far from stagnant. CPAN is nice, yes, but I don't think CPAN alone is enough to invite newcomers to Perl.
Perl 5 itself is stagnant. New features are not developed in Perl 5 but Perl 6. Perl 6 ist still not mature after several years of development. I tried it on Parrot yesterday, it ran way slower than Perl 5. Who ever would use it when even the Perl advocates don't recommend it? Perl 6 looked so promising. I am disappointed.
I'm sorry to say, but you're merely showing that you're not aware of core development of Perl. Not a mistake of your's, mind, marketing of perl core dev is terrible and i need to start fixing that, but there's a lot going on:
A link "Current Developments" in perl.org would be helpful. Thanks for the links but I'm not interested in minor changes of Perl 5 (Unicode etc.). I was very interested in the new features of Perl 6.
No matter, I lost my patience and left Perl years before and returned into the Lisp and Scheme world where I was before. Sometimes I still use Perl 5 but merely for small scripts.
Perl 6 got stuck in Parrot while Lisp enjoys a revival in Clojure, Qi/Shen and Racket Scheme which are all usable and under heavy development.
The new Lisp versions support multiprocessors and compile to JVM, Javascript (V8), Lisp, Scheme or C. Lisp and Scheme macros are still superior. In Qi/Shen I can even write attribute grammars in BNF style. I can compile my source to standalone native EXE files which are really fast. For that reason I prefer Scheme and Lisp over Python and Ruby. I don't need Perl 6 anyymore!
I'm suggesting that if you measure the amount of code submitted to the CPAN, the number of authors, the frequency of updates, and the freshness of versions, you'll see that Perl 5 is far from stagnant.