ICANN Voting Deemed Confusing

Brian Reid warns us about a potential confusion while casting an ICANN vote. His complaint in a nutshell: "I was tricked into voting for people that I did not want to vote for." I haven't cast my vote yet (two days left!) so I'm glad someone else found this out first. If accurate, outrageous; I'm surprised nobody else has complained.

michael: Reid doesn't understand the voting system. A brief explanation:

Your first-place vote counts as long as that candidate is in the running. Then your second-place vote counts, if that candidate is in the running, etc. Thus your seventh-place vote is exactly the same as a blank vote - it can never, ever, count (since the system will be resolved in no more than six rounds). Your sixth-place vote says, "In the event the election comes down to only these two candidates (the one I'm about to rank sixth and one other, who isn't any of my first five choices), I prefer the one I'm ranking sixth over the one I'm ranking seventh or leaving blank." Only if your first five candidates have been ELIMINATED from the voting pool does your sixth-place vote count for anything at all.

Once more: as long as your first-place candidate is in the running, none of your other six rankings count for anything at all. If number 1 gets eliminated, your number 2 vote counts and none of the other five count for anything at all. Etc.

There is only an advantage to you for ranking all the candidates, no disadvantage. Reid was annoyed that he ranked the "bad" candidates at all. But all he was saying was that "In the event that ALL of my "good" candidates have been eliminated and only these "bad" candidates remain as possible winners, I prefer this one over that one." Nothing wrong with that.

Let's go through a brief example. We have seven candidates for the North American seat. We compare all of the 1 votes. No one candidate has a majority (though one may have a plurality...). The candidate with the lowest number of 1 votes (let's say it's Miller) gets eliminated. Now, all of the people who voted for Miller as their number one choice (say, 2%) don't have a candidate anymore. So their number 2 votes get promoted, and that's their new candidate. Most likely, this doesn't create a majority, because it's only 2% of the votes being redistributed among the other six candidates. So we eliminate another candidate. (Maybe this candidate had 5%.) So now anyone who had those two candidates as their top two picks (in either order) is on their third place choice, and we have only five candidates remaining. If one of those five was your top choice, it still is; no one has even looked at any of your other choices, nor will they until your top choice is eliminated.

This sort of run-off voting is much, much fairer than single-choice, and much, much more likely to get "good" candidates chosen. Reid just doesn't understand it.

Re:This seems rather disturbing

[pbryan]

I don't know what you mean by corporate influences on ICANN's the voting system.

This voting system is, by far, superior to the standard popular vote. It is used in democratic countries such as Australia, and once you understand the fairly simple process, is easy appreciate just how good a system it is.

It allows you to vote your conscience (the best candidate, in your opinion), without fear that your vote will be "wasted" on a candidate who has no chance of winning.

If your first choice places last in the first round of tallying, that person is eliminated from the next round, and everyones' votes are moved up to replace him/her on their ballots.

This continues until it gets down to the final two, and the one with the majority of votes wins.

So, it's a recursive process, but it yields a far better result in my opinion compared to the kindergarden voting system in the United States and Canada.

"Of course, that's just my opinion. I could be wrong."

This seems rather disturbing

[SaxMaster]

I'm also an ICANN member who hasnt voted yet, but I'd have to think that the corporate influences in ICANN have something to do with this "confusing" voting system. However, I may not vote at all because my PIN is nowhere to be found :(

Re:STV

[jachim69]

Say two democrats and one republican stood, chances are both democrats would gain about 30% while the republican gets 40%. Thus the republican guy wins despite 60% rather having one of the two democrats.

You mean like the way Bill Clinton has won two elections without ever getting a majority of the popular vote? IIRC, he never got more than 48% of the vote. We have the funny-looking hand grenade with a bad haircut to thank for that!

addition to post

[ponxx]

oops sorry, just realised that ICANN chooses more than one director from each area, in which case one has to include into the description the redistribution of surplus votes of successful candidates as described in the link mentioned by another poster above (post #4 [slashdot.org])

read it twice though if you don't think it is fair... it is in fact one of the most logical systems i have ever heard of!

This whole ICANN thing needs to be redone

[Masem]

First, we had the problem where they were expecting only a 'handful' of people to respond to register to vote, but were swamped with thousands of requests. The server crashed and many of those that tried to get registered weren't able to. ICANN went on with it anyway.

Then, early in the voting process, election.com failed to recognize some of those voters, and had problems. They have appeared to be fixed the problem, but still, there was about a week in the narrow voting period where people that could legitimaly vote couldn't. And maybe those people had to leave on trips or vacations without access...good planning here with no way to absentee.

(As an aside, I also question this election.com. The internet as it is today has already seen one presidental election in 1996, so you cannot tell me that 1) election.com wasn't registered back then and 2) the ICANN election is their first use. It would seem to me that they might have bought into election.com therefore violating everything ICANN was *meant* to do in first come first served, but I don't know how to check on this.)

Finally, this screwy voting system, which appears to be inadequetely explained. Is this how ICANN wanted? Maybe not, as the replies the fellow here got were from election.com. Could they have 'tampered' with how the election was going to be run to put their own goals forward? Even if ICANN sanctioned, the lack of details is unnerving. In most US ballots, they tell you exactly how many (or "vote for no than X" for some places), and it's plainly obvious on the winning conditions. Here, he had to weave through too many details to get his answer, and only until after he voted.

Mind you, this could work in the favor of those that read /. (and therefore against the panel-sponsered ones), as I would figure most people would read the instructions as demonstrated here, and fill in all 1 thru 7 (or whatever number in other areas), which is suggested to be a vote for each one (e.g a wasted vote). Those that know what the vote represents will only vote for the non-panel sponsered ones and Lessig (IMO), and therefore, they will get votes that the others dont. Of course, there could also be the continginent that want the panel members in there, and might know these facts too. We'll have to see.

However, I believe there is sufficient evidence that demands for a re-election based off the flubbering of the situation, starting from reregistering. Two of the incidents above are well-documented, and if this was a standard ballot election, such incidents would force a reelection.

STV

[ponxx]

Single Transferable Vote (STV)is a system that is basically equivalent to having several rounds of voting in which the weakest candidate gets eliminated. It has the advantage that "all votes count" as the guy from election.com said. What this means is NOT that all candidates you rank get a vote, but that all voters have a say in the final outcome between the two most popular voters. The advantage over having lots of election rounds of eliminating the weakest is simply that you do not need so many rounds which would be unpractical and very time-consuming.

"Most votes win systems" such as the US presidency election (not really, cause of some bizarre rules from two centuries ago which makes it one of the strangest systems to vote a single person i have ever heard of) have one HUGE disadvantage:
If two people share a similar platform they hurt their cause by both standing! Say two democrats and one republican stood, chances are both democrats would gain about 30% while the republican gets 40%. Thus the republican guy wins despite 60% rather having one of the two democrats. In STV systems they would have rated the respective other democrat second and one of the two would have won.

anyway, what was i trying to say? yeah, good system, lacking some explanation, but as one other poster commented a simple google search would have brought it to light. Don't tell me you can construct a situation where this system also produces weird results, cause i know that, and I am sure you will find a friendly mathematician somewhere to prove to you that there is no "fair" voting system, and this is the best one i have seen for choosing individual people for a post. And anyone who followed the description will have achieved the results he desired...

Suggested disclaimer for ICANN:

The voting system is STV (some link). By giving lower preferences you do not affect the chances of your first preference candidate.