[EM] ERBuckliln(whole) with AERLO meets Strong FBC

[James Green-Armytage]

	A couple of quick questions, Mike. One, what is the difference between
strong FBC and regular FBC? I couldn't find the answer where FBC is
defined on your web site. Two, how would AERLO apply to Bucklin? The
definition of AERLO that you sent to me earlier was based on the existence
of a majority rule cycle, which wouldn't be detected by Bucklin. Would it
come into effect when more than one candidate passes the majority
threshold at the same stage?

James


"MIKE OSSIPOFF" <nkklrp at hotmail.com> writes:
>For rank methods,  the automatic equal ranking line option, AERLO,  makes 
>FBC compliance into Strong FBC compliance.
>
>ERBucklin(whole) is the only rank method that I know of that meets FBC.
>With 
>AERLO, it meets Strong FBC.
>
>Should I say all that more tentatively? In case there's something that
>I've 
>missed, maybe I should just say that it seems to me that ERBucklin(whole) 
>meets FBC and that, with AERLO, it meets Strong FBC.
>
>If someone objects that AERLO merely dodges the issue of Strong FBC, I
>reply 
>that when you rank using AERLO, and you rank the above-line candidate X
>over 
>the above-line candidate Y, you're voting X over Y by my definition of 
>voting X over Y. In fact you're voting X over Y by any meaningful 
>interpretation of that term, since Condorcet counts and uses that
>pairwise 
>vote unless you need the automatic equal ranking to get an above-line 
>winner.
>
>Mike Ossipoff
>