[wfleming]

I think the takeaway of how to succeed with this kind of project is good, but a missing element for why so many kind of never take off that I haven’t seen mentioned yet is that a lot of times the compatability of these alternative implementations is in practice lower than claimed even for old language features. E.g. it’s very common for Ruby and Python apps to have some native C extensions somewhere in their dependencies, and AFAIK the major alternative implementations have never supported them. (They’ve tried sometimes, but it’s never really worked out for obvious technical reasons, and the alternative of expecting libraries to provide multiple implementations has a rocky history as well.)

Combine that with the fact these languages are often used to make CRUD websites where I/O is likely to be a bigger performance factor than CPU and the faster alternatives look a lot less appealing.