<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
</head>
<body class='hmmessage'><div dir='ltr'>
> but i thought they were reporting results in the near future. how are they reporting their results? <BR>
> will they be telling us who the STV winner is, the Schulze winner, the Ranked-pairs winner, the Borda winner, <BR>
> the Bucklin winner, the Coombs winner?<BR>
.<BR>
...<BR>
.<BR>
> what good is the site if it only reports the data and does not digest it in some way?<BR>.<BR>
Kristofer is right--It's ok if the the polling website or the poll-conductor doesn't do the count. It's enough if<BR>
the rankings are easily available to anyone who wants to count them as they choose to. I was a bit unfair<BR>
when I implied that the poll-conductor should do the count.<BR>
.<BR>
But yes, how to count them is a problem. That problem is avoided in the Voter's-Choice count that I'd do if <BR>
it were my poll. Maybe I'll do a poll myself, by Voter's Choic.<BR>
.<BR>
I may have mis-stated Voter's Choice yesterday. Let me reword it here:<BR>
.<BR>
Each voter, in addition to voting a ranking, an Approval ballot, and a Score ballot, also designates<BR>
the method that he wants to decide the winner. <BR>
.<BR>
For a given alternative, and for each method choosing that alternative, that alternative gets a point for each<BR>
person who designated that method.<BR>
.<BR>
The winner is the alternative receiving the most points.<BR>
.<BR>
This amounts to weighting the methods according to how many people have designated them.<BR>
.<BR>
It's the fair way to aggregate the results of several different methods.<BR>
.<BR>
> if they assign the highest ranked choice to the "single affirmative vote"<BR>
> that the traditionalists want, will they be telling us who the FPTP winner is?<BR>
<BR>
They can't assign the win to a method, without, first, established what method to use. That's why<BR>
I will use Voter's-Choice in my poll<BR>
<BR>
In the poll that I will conduct, if anyone wants to designate FPTP, they may. Likewise, of course,<BR>
anyone may nominate any method they want to.<BR>
<BR>
The method that you, as a voter, designate, needn't be one of the nominated methods that we're<BR>
voting between--though it of course could be. And your designated method needn't be simple enough<BR>
for a public proposal, though I think it would be much better if it is--so that our election can demonstrate<BR>
publicly-proposable methods.<BR>
<BR>
I'm talking a lot about my poll, though I haven't yet proposed it, or set it up at a website (if that's how I'm<BR>
going to do it).<BR>
<BR>
That brings up another point:<BR>
<BR>
Website polls, of course, have nearly no security. Yes, you can require people to register their e-mail address.<BR>
I have two e-mail accounts and addresses. If I wanted to cheat in the election, I could register and vote<BR>
with both accounts. The current poll's registration requirement helps avoid the most blatant ballot-<BR>
stuffing, but it doesn't prevent ballot-stuffing.<BR>
<BR>
I know of one, and only one, secure way to conduct a poll on the Internet: Do it at this mailing list. <BR>
That's how we, at EM, did polls for years. I understand that, now, it's more popular to do polling<BR>
at other websites, because the voting is easier.<BR>
<BR>
But is that worth abandoning any chance of having a secure, un-ballot-stuffed poll? I certainly don't<BR>
think so.<BR>
<BR>
In a poll conducted in the EM mailing list, each ballot will be a mailing-list posting (like this one), whose<BR>
subject line is "Ballot" (without the quotes).<BR>
<BR>
Either the voters coudl copy and paste a pre-written ballot into their ballot postings, or else they<BR>
could simply rank the alternatives (and approve and rate them too), writing the ranking, ratings and <BR>
approvals themselves. I'd suggest that the choice should be optional, between those ways of voting.<BR>
<BR>
I'm going to conduct a poll. Is it ok if I do it at the EM mailing list, instead of at a polling website,<BR>
so that the poll can have security against ballot-stuffind?<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
i do not know how to > <BR>
> translate a marked ranked ballot into an approval ballot or a score ballot without making assumptions that ma<BR>
> not be valid.<BR>. <BR><BR> </div></body>
</html>