> The first is modular implicits: it feels so kludgy to have to type `a + b` for integers, `a +. b` for floats and `a ^ b` for string concatenation. I know it sounds like a small thing, but it makes the language feel inelegant, and aesthetics are important.

That's interesting, as being forced to use the same operator for addition and for concatenation (and what's more with the result of mixed types if allowed often dependent on the order of the parameters) has always seemed extremely inelegant to me.

Addition and concatenation are not the same thing. Why they should share a symbol when they don't share some properties integral to their nature (it's non-commutative) is beyond me, and I think it's caused a lot of bugs that didn't need to happen.

Not allowing the addition of floats and ints together is less of a fundamental issue, but it also helps quite a few problems. Al alternative solution, if you didn't want to reply on implicit coercion would be to explicitly convert one operand to the appropriate type so they match.

Or you can go whole hog and provide an entirely separate set of operators for different types, like Perl did (string concatenation is '.', and equivalents comparitors are eq,ne,gt,lt,ge,le). That works well in Perl's case, but that's mostly because for Scalars Perl really just wants to know whether you are treating it as a number or a string, so there's only two types to account for.