Obviously, “shiny new languages” are going to surface all the time. And when they do, you need to make it your business to understand them... at least, to have “kicked the tires” a time or three. But you know, a shiny new language is not going to cause millions of lines of existing mission-critical code to just go “four paws up.” It doesn't happen that way, and it never will, because existing software that works well just might be the most valuable thing that there is.

Software never gets “rewritten,” let alone in some sexy new language. When a company selects a language to write their stuff in, they inevitably stick with it. That's the only strategy that is economically feasible. And it's also a pretty good strategy.

Programmers don't look at a library of source-code and give proper consideration to the fact that this is a hand-built resource in which several million dollars have been invested, and which enables this business to function. “Rewrite all this in Ruby?” Are you nuts?!?!