What if the main VM already has optimised sorting algorithms like most do? Then you would have to come up with another strawman argument.
If you aren't happy with the performance of the existing sorting library functions, then that is when you should investigate alternatives or writing your own.
No - there are interviews out there with members of the actual Twitter development team that specifically state they ditched Ruby because it couldn't scale.
It was nothing to do with the database. Revisionist history much?
Let me guess, you use Ruby and don't like the fact that it might not be the be-all and end-all?