[EM] I meant Tideman(wv) !

[MIKE OSSIPOFF]

When I spoke of Tideman as a possible alternative to SSD or CSSD
in public proposals, I was referring to Tideman(wv) only.

Since "Ranked Pairs" is the term more widely used here, I suppose
I should use it too. I abbreviate it "RP".

I neglected to add in my previous letter that RP(wv), like SSD & CSSD,
meets all 4 majority defensive strategy criteria, SFC, GSFC, WDSC, &
SDSC. Since anything that meets GSFC also meets SFC, and anything
that meets SDSC also meets WDSC, it would be sufficient to just
tay that SSD, CSSD, & RP(wv) all meet GSFC & SDSC.

Also, RP(wv) meets the Independence from Clones Criterion (ICC),
as does CSSD. SSD is efectively clone-independent in public
elections, because pair-ties are rare there.

As I said, I don't suggest RP for small committees because it's
more awkward to program for those applications.

What I was emphasizing in my previous letter was the differences
between CSSD, SSD, & RP(wv). In small committees RP can choose outside
the initial Schwartz set, and ordinary SSD isn't clone-independent. CSSD
has neither of those problems even in small committees. In
public elections RP & SSD don't have those problems.

The other difference, as I was saying, is that when RP chooses a
different winner from that of SSD or CSSD, RP's winner more often
than not pair-beats the other method's winner.

I suggest that the merits of SSD, CSSD, & RP(wv) are so nearly equal
for public elections that the choice of which of those to propose
should be based entirely on which one is more likely to be accepted.

Also, I should add that my girlfriend listened to my explanation of
the Schwartz set because I'd asked her to give it a fair hearing,
to test the definition. If someone had come up to her on the street
with an initiative, that might have been different. The test suggests
that the Schwartz set won't be a problem to people who've had no
contact with voting systems, and that it will be accepted if the
person listens to the explanation. Of course RP(wv) has the advantage
that it's definition just starts right out with the procedure, without
a preliminary definition of the Schwartz set. If there's a question
about how much people will listen to or read, then maybe RP(wv) has
the advantage there. But, as I said, finding out which is more
acceptable will depend on polling.

Mike Ossipoff




_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.