[EM] Richard reply, 14/2/04, 1142 GMT

MIKE OSSIPOFF nkklrp at hotmail.com
Sat Feb 14 03:42:02 PST 2004


Richard--

You said:

It occurred to me that the phrase "The method allows full ranking of
all candidates" might be sufficient, instead of the more difficult
wording of my last message, depending on the correct interpretation of
the following:

"votes sincerely"

and

"falsely voting two candidates equal"


I reply:

That's your business entirely, and I don't have an opinion on it. For me to 
judge your criteria with those wording changes, you'd have to send to me the 
full wording of the criteria.

When you speak of the correct interpretation of those two phrases above, I 
can answer about "votes sincerely", but not about "falsefly voting two 
candidates equal", because no criterion of mine contains the phrase "falsely 
voting 2 candidates equal". Maybe one of yours does. If so, then you must 
specify what you mean by it.

You continued:

To help interpret the first phrase, we have the following statement:

"A sincere vote is one with no falsified preferences or preferences
left unspecified when the election method allows them to be specified
(in addition to the preferences already specified)."

The ambiguity here lies in the phrase, "already specified". I don't
think that refers to the chronological order in which the ballot is
marked. So what does it refer to? Top-to-bottom ranking order?

I reply:

That isn't my definition of sincere voting. I didn't write that definition. 
As I was saying before, our website is a committee effort, and, as is 
usually the case with committees, no one committee-member gets his way on 
everything. So definitions are often written differently from how I write 
them.

Are you aware that I've frequently posted my definition of sincere voting 
here? But you're instead concentrating on the website's definition, because 
that's where you can find ambiguity. If you want to ask about my 
definitions, I suggest that you find them in the EM archives. Of course the 
most recent take precedence over the less recent.

Though I've  told you that many or most definitions at the website were 
modified from my definitions, and though I've  posted my definitions to this 
mailing list, you prefer to concentrate on the website's altered 
definitions, because that's where you can find ambiguities.

Is it that you didn't know that I've defined sincere voting here, or are we 
displaying a little dishonesty?

But I'm going to mention to  Russ that someone is complaining about the 
altered definitions, and I'll ask him to use my definitions, to avoid 
criticism of the website.

The change back to my wordings for criteria and methods might take a while, 
so be patient.
But, in the meantime, if you want to know what my definitions are, check the 
recent list archives.

My definition doesn't say "already specified". Here's my definition:

A voter votes sincerely if s/he doesn't falsify a preference or fail to vote 
a sincere preference that the balloting system in use would have allowed hir 
to vote in addition to the preferences that s/he actually did vote.

A voter falsifies a preference if s/he votes X over Y, but doesn't prefer X 
to Y.

A voter votes a preference for X over Y if s/he votes X over Y.

You already know how I defined voting X over Y, and how you define it. Take 
your pick.

[end of definition of sincere voting]

You continued:

As for the second phrase, we now know that "voting two candidates
equal" has the following special meaning, provided by Mike: "A voter
votes X equal to Y if s/he doesn't vote X over Y, and doesn't vote Y
over X, and votes X over someone, and votes Y over someone."

I reply:

We also knew, before I posted that definition, that leaving 2 candidates out 
of one's ranking is not voting them equal, in fact is not voting them any 
way at all, unless you believe that "unranked" is a rank position, an 
assumption of yours that I've never encouraged. Not mentioning 2 candidates 
says nothing about them. Any meaning that you give to it is your own.

You continued:

So if a ballot is truncated (but fully ranked above the truncation
point), it cannot be "falsely voting two candidates equal"

I reply:

You have "falsely voting two candidates equal" in quotation marks. Where are 
you quoting it from? Did you get it in one of the modified definitions at 
the website? That phrase doesn't occur in any criterion definition of mine.

You continued:

...since it
isn't "voting two candidates equal". However, if a ballot has two
candidates voted equal at the top, or somewhere in the middle, and
this equal voting is forced by constraints of the method, is this
considered *falsely* voting the candidates equal?

I reply:

I have no idea. I don't have a definition of "falsely voting two candidates 
equal". As I said, that phrase doesn't occur in any criterion definition of 
mine.

Yoiu continued:

To be sure I have the correct understanding of both phrases, I ask
Mike the following questions:

Suppose a method allows candidates to be ranked on a maximum of four
levels. Also suppose there is a restriction in this method that the
top and bottom levels may each have at most one candidate on any
ballot, but the other candidates may be distributed over the middle
levels in any way the voter chooses.

Now, if my preferences are A>B>C>D>E>F, which of the following are
sincere votes, and which ones have two candidates falsely voted equal,
according to the electionmethods.org usage of these phrases?

I reply:

As I said, I don't have a definition of "falsely voting two candidates 
equal". You might want to ask Russ? Are you saying that our website uses 
that phrase in a criterion definition?

However, I have posted my definition of sincere voting above, in this 
message. And so I'd be glad to apply that definition for you, _a few times_.

You continued:

Ballot 1: A>B>C=D=E>F
Ballot 2: A>B=C>D=E>F
Ballot 3: A>B=C=D>E>F

I reply:

I'm going to apply my definition of sincere voting for you now. And then, if 
you have any questions based on the my definition that I've posted in this 
message, I'll answer those questions too. But that's it, for answering your 
questions about the application of my definition of sincere voting. So if 
you have more questions about  my sincere voting definition posted here, or 
the application of it, ask them in one e-mail. That's a reasonable request.

Of course, even after that, it would be reasonable to ask questions about my 
answer. But only if it contains a genuine ambiguity.

Right now, I'm giving you the benefit of the doubt, and assuming that you 
haven't seen the sincere voting definition that I've repeatedly posted to 
EM.

Ballot 1:

The ballot fails to vote the 3 sincere preferences between C, D, & E. In 
order to vote some of those, it would be necessary to move one of those 
candidates to another rank position. But doing that would rank one of those 
candidates with someone with whom s/he wasn't previously ranked, and 
therefore would prevent you from voting another sincere preference of yours.

So, when you fail to vote your sincere preferences between C, D, & E, are 
you failing to vote a sincere preference that the balloting system in use 
would have allowed you to vote in addition to the preferences that you 
actually did vote? I'd say no. What would you say? It's pretty obvious. 
Voting one of those unvoted preferences would require you to not vote one of 
your voted preferences.

Ballot 2: It's the same as ballot 1.

Ballot 3:

Different ballot, same thing: If you move one of your equal-ranked 
candidates to another rank position, in order to vote one of your unvoted 
preferences, doing so will unvote some of your voted preferences.

So the answer to your question is: By my definition of sincere voting, those 
3 ballots are sincere.

Mike Ossipoff

_________________________________________________________________
Check out the great features of the new MSN 9 Dial-up, with the MSN Dial-up 
Accelerator. http://click.atdmt.com/AVE/go/onm00200361ave/direct/01/




More information about the Election-Methods mailing list