The Arc idea of using single-char operators (I see :.!+, any others?) as intra-symbol syntax is clever . I wonder if any other languages have done anything similar?

APL (and its derivatives) are an obvious use of that.

Note also C-style array access: a[i];

It's not new, just new in a lisp (and a good idea if you ask me).

Refinements in Rebol http://www.rebol.com/r3/docs/concepts/funcs-evaluating.html#...

You mean other lisps, or are you being sarcastic?

This seems pretty different to me than the use of single character operators in C-like languages. "odd+pr" has a completely different meaning than "odd + pr" (which might be 3 functions being passed into some other function). In the latter, "+" is the addition function. "odd+pr" reads more like a single token indicating a function. Kind of like the difference between morphology and syntax, where "(+ a b)" is a phrase and "odd+pr" a derived word.

So this dual level syntax is what I'm getting at. I know there is the expression/statement distinction in other languages, but this seems different. Perl sigils might be a closer idea, but those are more like case markers, I think. You can't use Perl sigils to combine multiple tokens into another token-ish thing, as far as I know.

Hope that made some sense.

The Rebol thing mentioned in another comment, by the way, does seem more like what I'm talking about.

other lisps have that too. other lisps use . and , with special meaning though...

Not an arc user, but just curious: with the 'atstrings' feature on, how is an actual '@' escaped?

@@