I wish Julia would be more strict wrt type coercion of integer to float values.

I've once spent a day debugging the issue caused by the following line of code, where t1, t2 are floats and v is an array:

d = (t2 - t1) * length(v)

It should've been LinearAlgebra.norm():Float instead of length():Int. Had julia been stricter the code would have failed to run, saving me much time.