I wonder whether there's some synergy with the discussion around conditional conformances on opaque result types. If we go with one of the proposed syntaxes for in-line conditional conformances there, maybe we could also use it to denote conditional default implementations in protocols?
3 Likes