If they would keep putting the extra effort in debugging the Erlang VM C code, the benefits of this debugging would be available to the entire community of deve

If they would keep putting the extra effort in debugging the Erlang VM C code, the benefits of this debugging would be available to the entire community of developers using Erlang. Now they are in their small corner of the world and every bug they squash is just a bug of their project. Overall it's worse for "the world" and better for them. This makes sense for commercial software but it's weird when you think about an open-source project. Just imagine if all the effort in porting Ruby/Lisp to Java/C++ or similar translations for big projects would have been invested in improving the VMs for these dynamic languages (yeah, most of these rewritten projects didn't have experienced C guys, but the OP certainly has and is one). Is bug-fixing and hacking on compilers and VMs really that hard? (I never approached these problems in practice because I've never had to scale something that hard and squeeze that last drop of performance, so I'm really asking the question, it's not rhetorical).