But it's entirely consistent with the fact that std::unexpected<E>::value() returns val. If P2549 is accepted, this would make sense.

Let's put this change on hold until P2549 lands.

Actually, I made a suggestion to include it into the paper and we were unsure whether that would be a purely editorial issue or whether that would require additional review.

If not we can simply add those changes to the next revision of P2549, which would simplify things a bit

P2549 is currently pending LEWG electronic polling. I think we agree that renaming the exposition-only member is a purely editorial change, yet I would prefer to have this in the next update of the paper, before LWG starts reviewing it.

yet I would prefer to have this in the next update of the paper, before LWG starts reviewing it.

I see. I would include this in R1, which if EP passes will be published in June mailing. If this does not require additional review, then it seems straightforward to me that this is a natural additional driven-by fix.

Well, LWG will review it for consistency, but LEWG won't (and doesn't need to).

Abandoning this PR since it's superseded by #5692.