[EM] RE : Re: Range voting, zero-info strategy simulation (raphfrk)

[Abd ul-Rahman Lomax]

At 10:52 AM 11/1/2006, Kevin Venzke wrote:
>I have seen more concern that voters
>will vote the opposite way: Commit to a favorite candidate and cut
>off any chance of even electing the second favorite.

It should be kept in mind that Approval is a practically no-cost 
reform. It adds maybe half the value of Range Voting with *no* change 
in voting machines, extra ballot space. And it remedies what has been 
an injustice from the beginning, the tossing of overvoted ballots. 
(Which is a little more justifiable because voters are supposed to 
know that it is against the rules, though I think many don't really 
understand that. If overvoting is allowed, you have Approval. Just 
count all the votes. Nice slogan, eh?

Once you have Approval in place, people will want to express more. 
And then it is possible to go to Range, or to ranked ballots. Range 3 
would be an obvious step, the -1, 0, +1 scheme which has been suggested.

I think we really should be joining forces with the AV people, 
because we *should* agree on the next step. Later, we can duke it 
out, if we really want to fight for *our* way rather than theirs.

Given that Approval is generally admitted to be a Range method, i.e, 
Range 2, we might not even need to change the name of the umbrella 
organization, just make explicit that Range is a whole family of 
methods. Including Approval, the simplest of all.

If people bullet-vote Approval, then we have, for them, what we have 
now. Nothing has been damaged, rather the system has been approved 
at, as far as I can see, no cost.

The *only* legitimate issue, as far as I can see, is campaign 
finance. And splitting the money between approved candidates does 
make sense, if vote counts determine financing. (It shouldn't, except 
maybe to establish a threshold, and, if it does not determine the 
*amount* of the subsidy, then there is no problem with the overvotes. 
Overvotes, which really do represent public support of a candidate 
receiving them (since they risk electing him or her), do no harm. But 
I raise the issue because opponents will.)

There is *also* an apparent problem with CRV having fixed the Range 
definition to include bells and whistles. This allows opponents to 
argue against Range based on these rules, or on missing rules that 
may appear necessary, and if one notes that the system can be easily 
fixed by adding this or that rule, with respect to an anticipated 
problem, they will say -- and it is being said by Chris -- but that 
is not the definition, according to CRV. And is there any other authority?

"For our group purpose there is but one authority -- ... our group conscience."

Group conscience is deliberately not crisply defined in an FA. But it 
is generally understood that the higher the level of consensus, the 
more clear has been the expression of group conscience. Thus the RVFA 
*can* effectively set up consensus definitions. It publishes them, 
with the supporting votes and including minority reports. It remains 
utterly even-handed, allowing even a single dissent to be expressed 
in the full record. But if the approval of the standard is broad, it 
can effectively be considered as having been fixed, at least for the present.

CRV can certainly have its own definition, and if we have not come to 
agreement, I expect CRV to barge ahead. But the definition, if 
possible, should come out of an emerging consensus. You really have 
that when both supporters and opponents of Range Voting have 
generally agreed upon a definition.

(When they don't, they are clearly not seeking genuine agreement, 
when agreement is possible. There are exceptions to this, where a 
word is firmly established in usage and people are heavily attached 
to it. What is "marriage?")

I personally would keep Range to mean any method which sums the 
votes, votes being chosen from a set of possible votes (linear 
spacing?there is another possibility which would provide for fine 
distinctions at the top and bottom, though possibly only the top 
needs that) and which allows each candidate to receive ratings 
independently of all other candidates. That's my first thought, I'm 
sure Warren could nail it down.