[EM] Re: Later-no-harm, Mono-add-top, etc.

[Kevin Venzke]

Chris,

 --- Chris Benham <chrisbenham at bigpond.com> a écrit : 
> Yes, I may have goofed. I (it seems wrongly) imagined that Mono-add-top should be so easy to
> meet that only a method saddled with some extraordinary burden would fail it.

It might seem so, except that pairwise methods don't consider who is at the
top of new ballots.

I think if we're set on Condorcet (especially Smith), then we shouldn't worry
too much about those monotonicity criteria which involve adding or substituting
randomized ballots.

> I assume that Margins does better than WV at meeting your Earlier-no-harm/help criteria.
> Am I wrong?

I don't know.  If it could be shown that Margins "does better" than WV at
meeting later-no-harm and later-no-help, then I would assume it would also
be better at ENHarm and ENHelp, since both methods operate solely on the
pairwise matrix.

But I'm now pessimistic that ENHarm is useful, since it doesn't imply any
kind of guarantee about not having to order-reverse.


Kevin Venzke
stepjak at yahoo.fr


_________________________________________________________________
Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français !
Yahoo! Mail : http://fr.mail.yahoo.com