it’s only better than other programs when you can’t include assumptions about the domain.
Yes, and also when the domain-specific information doesn’t exist. We don’t yet precisely know how language works, for instance. If we did, we could code it, but we don’t.
There’s also the matter of having a large additive O(c) constant. That’s not as big an issue as some people make it out to be (it can be remedied by clever choice of programming language—provided you know what programming language to use).
If it’s that good why isn’t it used? (Serious question).
edit: Based on your other comment I guess that’s it’s only better than other programs when you can’t include assumptions about the domain.
Yes, and also when the domain-specific information doesn’t exist. We don’t yet precisely know how language works, for instance. If we did, we could code it, but we don’t.
There’s also the matter of having a large additive O(c) constant. That’s not as big an issue as some people make it out to be (it can be remedied by clever choice of programming language—provided you know what programming language to use).