You set the framework for the debate nicely, but miss one of the frames.  You agree that this is about setting up a "rules of the game" in advance so that people are playing on a level playing field, and that the fundamental prinicple is one-person, one-vote.  You therefore concede that any set of rules - defined in advance - could be reasonable, so long as they are equal in application to individuals and known ahead of time.  There is no reason in your framework that NRAV wouldn't be a reasonable rule in advance.  The problem for you is identification.  I think you'd support NRAV if you could (reasonably) guarantee the voter voted once and only once, and that you believe identification is a reasonable (though itself not perfect) rule to ensure that part of the framework.  You'd vote (I think) against THIS BILL because it expands NRAV without correspondingly expanding identification (though even "reason AV" as currently is has that hole in it too, and the Dems. have blocked legislation to cure that defect).

The philosophical reason for NRAV is that RAV requires a person to make a declaration that might be used against them, it treats the non-present differently and unequally, and it makes voting more difficult. Those rights and conveniences are not just for homeless people, although the left might focus on that.  There philosophical justification against all NRAV is "voting shouldn't be easier - it should be hard" (I don't buy that) and McMillin's argument that the voting booth and one day are special important elements of the process (I see something there).  Amash has made a decision that the quantity of liberty and equality achieved by NRAV outweighs what he perceives to be minor quantity in fraud (less equality on a different measure since that cuts one-person-one-vote) and fairness.

To be clear, I'm only partially on your side. I support NRAV in principle, I oppose this Democrat version of the bill because the e-mail/fax portion shifts the balance in mind. Hence I see Amash's argument and the principles he used to arrive at his vote, disagree with his conclusion, but strongly disagree with what I think is a careless attack on his personal integrity even though I suspect the attack itself was well-motivated itself.