“An interesting thought, to be sure ... but ...” I think that the fundamental difference here is that “the new generation of monkeys” in this case never had to stop and consider the state of the current ones. No matter what software-tool we might be talking about, it is still ... a tool. Therefore, it is and always will be subordinate to whatever-it-is that it is being used to do. And, in order to gain market acceptance, it must present an opportunity-cost that is low enough to be no brainer.
The Perl-5 system is by no means the only interpreted-language system out there on the field, but it certainly is the only one that I have encountered which could have produced Moose or any of its lesser children, without introducing any changes to the core system. (Well, an exception could be drawn for Common LISP.) Nevertheless, IMHO, Moose did, and entirely without fanfare, introduce a “game-changing improvement” that has near-zero adoption cost.
In like manner, haXe demonstrated that you could achieve significant productivity-benefits, and work-product target expansion, without changing the target-languages, simply by changing the language that you write in.
Perl6 might have have had academically-valuable advantages, but it asks asked too much of its adopters. It has had too high of an opportunity-cost. Its day in the sun, like it or not, has now set. Better luck next language . . .
“Another suitcase, another hall . . . don’t ask, anymore.” – Evita
| |
