[EM] Time-independent MMV definitions (conditional & unconditional)

[Michael Ossipoff]

Ernest and EM:

First the unconditional definition. Then I'll state the conditions
that are added to it in the conditional definition.

Conditional MMV:

1. A defeat is a discarded defeat if it contradicts a set of
not-contradicted defeats that are stronger than it is.

2. Additionally, a defeat, D, is a discarded defeat if it contradicts
a set of defeats that 1) aren't designated in paragraph 1 as discarded
defeats; and 2) consist of one or more defeats that are not designated
as discarded, and are stronger than D, and one or more defeats that
are equal to D.

3. Additionally, a defeat, D, is a discarded defeat if contradicts a
set of defeats that are 1) not designated in paragraph1 as discarded,
and not designated in paragraph 2 as discarded; and 2) are equal to D.

[end of unconditional MMV definition]

Conditional MMV:

Same as Unconditional MMV, except that:

1. Each paragraph after paragraph 1 doesn't apply if there's a
alternative that doesn't have a defeat that isn't designated as
discarded in a lower-numbered paragraph.

2. Paragraph 3 has the following clause added to the end of it:

...But D this paragraph doesn't designate  D as a discarded defeat if
there is a weaker defeat that is designated in this paragraph as
discarded, and the alternative that has that weaker defeats doesn't
have any defeats that aren't designated as discarded.

[end of conditions added for Conditional MMV]

I don't know which of those 2 versions is better.

Conditional MMV can't be clone-proof. But these methods are for ideal
majoritarian conditions, in which people won't do chicken-dilemma
defection. . So shouldn't we likewise assume that they won't do
clone-strategy?

So maybe clone-indepence doesn't matter for these methods.

Conditional MMV is more decisive. Isn't that always a good thing?

Another comparison:

If A = (B>C>D>B) the voters are rather ambivalent regarding B, C and D.
But they're clear and definite in voting that no one is better than A.

So isn't a a more rightful winner?

But though the voters don't have a decision _between_ B, C, & D, they
do say that {B,C,D} is as good as A.

That suggests that we should solve the cycle and return a 2-alternative tie.

So, which of the above arguments is more compelling?

And which is more important in ideal majoritarian
conditions--clone-independence, or desisiveness?

If voters won't do chicken-dilemma defection, doesn't that mean that
they won't do clone strategy either?

Michael Ossipoff