Wow, that sounds like fun. In my case, there's no way we'd be able to port over the existing functionality to a different language in any less than a few months - and that would be all the API developers working full time. We're not going to do this because it wouldn't be a wise decision, but still useful to hear other people's experiences. I think we're more likely to switch to a different language/ecosystem for any major new functionality. At this point, we have a massive Python application and Python is quite, well...slow :(