The biggest reason is .NET and all of the nice tooling that come along for the ride. Profilers, remote debugging with breakpoints on other machines, recording p

The biggest reason is .NET and all of the nice tooling that come along for the ride. Profilers, remote debugging with breakpoints on other machines, recording program state over execution, cross compilation to phones for free, a rich ecosystem of libraries, etc. There are other good reasons though. Modern day F# is (except for a very small number of type resolution foibles) 100% backwards compatible to the first release. This is a huge problem for Haskell in business. Also, Type Providers are totally awesome for scripting over structured data.