[maemre]

Static analysis is not just about inferring types and hot paths for optimization. For that kind of stuff, a dynamic analysis is most of the time way better (lots of JIT compilers with speculative optimizations prove this point). There is another goal where static analysis shines: verification. If I'm writing a somewhat critical application, I want to make sure that it behaves according to my intent in all cases. For example, if I'm writing an airplane sofware, I want to make sure that it will at least never invoke any undefined behavior (this is done in ASTREE project). Static analysis is a powerful tool to give such guarantees in many cases.

Some highly dynamic language features make analysis really imprecise or really hard (in terms of computation cost). There has been quite a lot of work on making static analyses that can handle such language features (for example control flow analysis helps analyzing code that uses dynamic dispatch or closures a lot but cost of the analysis is exponential in terms of the precision level most of the time). Sometimes people tackle analyzing highly dynamic languages like JavaScript but at a huge time cost in certain cases [1]. I'd prefer using a language designed with static analysis in mind if I were to prove certain properties about my code.

[1]: http://www.cs.ucsb.edu/~benh/research/papers/dewey15parallel...

[pron]

Right, but an important point here is "some highly dynamic language features make analysis really imprecise or really hard". Not all "late bindings" are born equal. JavaScript's dispatch is very different from Java's from a static analysis perspective, the latter being almost indistinguishable from pattern matching.

[maemre]

I totally agree with that. When I mentioned dynamic dispatch, I had something like Scheme or JavaScript in my mind. Dynamic dispatch in those languages require a more subtle analysis to obtain precise results. I wish I could edit my parent comment to clarify my example.

To clarify my position (hence my parent comment's point) on this general matter, I'm OK with any language feature that is amenable to static analysis in a practical amount of time. This puts me closer to PL conservatives in Steve Yegge's spectrum.

[zimablue]

I think the standard counterargument is that if unit testing is ubiquitous, we're already admitting static analysis is at best very incomplete.

[pron]

I think some guys actually proved that in the 1930s. But so what? Program analysis and testing complement one another.