3 ms·
It's a silly idea. Fortunately, none of the people actually involved in progressing perl5 or perl6 have anything to do with this moonshot.
by mst 8y ago
It's a silly idea.
Fortunately, none of the people actually involved in progressing perl5 or perl6 have anything to do with this moonshot.
- rurban 8y agoIt's a silly idea only as afterthought. perl6 is still hooked onto it, Larry Wall constantly confirms the validity. They went through some successful VM's already, so why not. Esp. since the rakudo/nqp layer needs a lot of work. - They have three layers in it's main implementations now: moar (prev. parrot) - nqp - rakudo, with jvm or v8 as other backends for nqp, and tvmjit as possible replacements for the moar/nqp layer, and p2 and niecza as standalone impl's. perl5 being destructed is lucky to have cperl as failover replacement. Pluggable only on the XS API and CPAN layer.