> I agree. I'll make a patch for the warning, just so that we can see what it would look like.

Here, belatedly. It includes the lambda string change. The warning includes `defun` and `defmacro` but not plain lambdas.

Removing doc strings from ambiguous lambdas didn't break anything, as expected. I think the warning makes sense, because it does expose some mistakes where doc strings were returned but shouldn't be, or returned strings also exposed as (nonsensical) doc strings.

There are quite a few cl-defmethod forms with this syndrome but those definitions never were ambiguous (the strings  are always a return value, never doc string).