> It doesn't mean it's a feature that has to be vigorously defended. I'm not sure anyone is defending it. At least, I'm not. Basically Elm's compromise is safe

> It doesn't mean it's a feature that has to be vigorously defended. I'm not sure anyone is defending it. At least, I'm not. Basically Elm's compromise is safety, and less runtime complexity then Haskell. The trade off there is boilerplate. As with most things, there is no free lunch. I've stopped using Elm precisely because of these issues, but I understand why Evan made the choices he did.