It's a very practical ML-family language. It runs on Mac, Windows, and Linux. It doesn't really sacrifice anything, either. The last thing I delivered with it was a network health utility, which did UDP and TCP sockets and platform API calls very cleanly. It's really not a toy language. Distribution is cool too, because you can build for a system with the runtime installed or build a single-file executable. My suggestion: build a utility program with it for your own purposes and if you're productive with it.