p2 was too much work to come up with a backcompat p5 library with all its problems. And I saw no easy way to add multithreading. Forking perl5 was easier in the short run.

In the long run switching out to a fast VM for tight loops like p2, rperl or spvm (formerly called sperl) is still being considered. Perl6 inline::perl5 is unfortunately not doing too good, and I see no way forward with their architecture.

perl11 is doing very well. Many perl6 features are already in cperl, many more not yet completed. The design is done and sound.