They're likely only serving previously accessible domains already in their index as wayback machine links, which is neat, but doesn't really solve the problem of indexing the wayback machine in a broader sense.
Would be a very nice feature to have indeed, though the data is a bit too inaccessible to index as far as I can tell (even though I've not given it any serious effort, so maybe it is?)
I kinda consider that a feature and not a bug. If it were easier to find all the really deep stuff in the Wayback Machine, people would be trying to censor it all the time. I like being able to spear-fish my way into the deep shit by finding layers of URI references in other archived pages.
Would be a very nice feature to have indeed, though the data is a bit too inaccessible to index as far as I can tell (even though I've not given it any serious effort, so maybe it is?)