Open source is a chicken-and-egg problem. If you want success stories, look at c/c++/fortran/perl/python/R/julia, old and new. I don't have a solution. I just blame IBM and all others who monetized on APL in a short dozen years and didn't manage to hold their ground.
In an alternate universe we would see current fortran users writing APL.
More realistically, had anyone write an APL compiler that gets some applications to run on any of the top 50 supercomputers with even one percent of its peak performance, APL would still have a chance.