in reply to Perl 5 Optimizing Compiler, Part 9: RPerl.org & The Low-Magic Perl Commandments
Sorry to disappoint you, but I am dealing with several projects, consisting of several million lines of source-code in aggregate, which are in revenue-earning production right now. The clients in question are not the slightest bit interested in rewriting them from scratch to deal with your “Commandments,” which would be a multi-million dollar capital expense with no return on investment. (They prefer for the systems in question to continue to earn millions, and/or to continue to reliably support business processes that right-now do so. Perl-5, in this case quite literally, “moves the freight.”)
Please don’t continue to bother to trumpet the supposed virtues of an improvement that no one in the business world can ever afford/choose to make. Computer software technology does not exist in a vacuum; does not sit on a high mountain and chant, waiting for enlightened pilgrims to come to it.
Please find a way to provide, instead, even a 5% improvement (let’s start small and work up ...) on some key aspect of what Perl-5 does right now ... to be done in such a way that no source-code changes are mandated, and that the full backward-compatibility of all key modules (DBI, Template, JSON, to name a few) can be objectively proved in-advance.
You find yourself “a voice in the wilderness” because the business-risk of what you propose is incalculable while the return-on-investment is essentially zero. Yet, you continue to talk very loudly. Why not focus instead on some perhaps very-small but well-chose aspect of what the customer actually wants, and find a pragmatic way to do even that small thing? This is what will put your name in lights.