A Final Report on the Election

[Jon Seattle]

There is an issue with our electoral system which can allow an organized group to use strategic voting to gain more Borda points (and potentially more representation) by decreasing the value of their neighbors points. Thus happened in the January 2008 RA election and very fortunately did not significantly change the outcome of that election, but it has very significant consequences for our community and electoral system.

An Example

To explain what happened I will give an example using faction names A, B, C, and D, and two voters: Jane and Paula. Paula approaches Jane before the election, knowing that Jane supports faction B, and tells her that she favors faction A and asks Jane if she would please list A second on the ballot. Jane expects something like this:

Jane: B, A, C, D
Paula: A, B, C, D

The resulting scores would be: A = 7 (35%), B = 7 (35%), C = 4 (20%), D = 2 (10%). But instead the actual votes are:

Jane: B, A, C, D
Paula: A x x x

The resulting scores are: A = 7 (41%), B = 5 (29%), C = 3 (18%), D = 2 (12%). Notice two things about this example:

By using this tactic instead of a mutual ranking, Paula has won an extra 12% of the Borda points for her preferred faction over the Jane’s choice. The tactic also decreased the percentage score of Jane’s third place faction and increased the score of the her last place faction. In other words this tactic decreased the entire value of Jane’s vote except for the faction favored by Paula.

This Election

The reporting features of the new election system are designed not to provide names with election data, and I have never seen those. What I have seen is aggregate frequency counts (basically what I published), and I also have a data set with unidentified votes in random order.

In the January 2008 election, very early on the voting I noticed an anomaly with aggregates was pushing one party ahead in the Borda counts without a larger number of voters putting it first in the ranking. When I looked at the anonymous vote data set I noticed that a number of voters from that faction were eliminating all but their preferred faction, and that all the votes in question arrived within a short period of time.

I spent several sleepless nights worrying that election results would be substantially changed by these votes. CSDF had enough of a lead that it never was at risk, but this action clearly disadvantaged the other two parties that depend heavily on second and third rank votes for their points. I appealed to the SC so as to keep those smaller parties from being damaged through this strategy.

I am very very relieved to report that the number of voters who cast ballots later in the week corrected this situation, so that by the end of the election there was no difference in the outcome, either in the number of representatives nor in the ranking of the factions. We were very lucky this time.

Asking some questions, I was told that the faction in question held a meeting where this voting strategy was discussed and recommended (I assume the votes in question occurred immediately after the meeting.) I was also told that the faction did also intentionally canvass second place votes.

In spite of this, only about one third of the faction eliminated all other factions from their ballot, with those votes arriving in a block at the beginning of the election. It was by no means all of the voters in that faction. If end this strategy failed in that it did not improve the electoral outcome of the faction that used it, and may well provide a disadvantage in the next election.

A Quote from Lincoln

Several people have told me that the issue is a problem with the RA as it passed a bill that allowed this form of electoral manipulation. While, I do think the system should be changed (more on that below), just because something is legal is not necessarily a great reason to do it. Those who take an action are responsible for it’s consequences. To quote Lincoln’s Cooper Union address (via Krugman):

That is cool. A highwayman holds a pistol to my ear, and mutters through his teeth, “Stand and deliver, or I shall kill you, and then you will be a murderer!”

Long Term Prospects

If the electoral system is not changed, there will be serious consequences for future elections. The question is, knowing that someone may well use this approach in the next election against you, can you afford not to use it also? It becomes a use-it-or-loose-it situation.

In game theoretic terms, this approach has the structure of a non-iterated multi-player prisoner’s dilemma. (see http://en.wikipedia.org/wiki/The_Evolut ... ooperation ) There are cases of the Prisoner’s dilemma where cooperation can arise, but those are generally in two player games where the game is played again and again an indeterminate number of times. Whenever you know that the game is about to end, the only rational strategy is non-cooperation, or taking advantage of the other player.

If we do not change the law, we will most likely revert to a simple majority electoral system. It will be the end of preferential voting in the CDS. Two of our factions that have, in my opinion, an important voice in our politics get most of their Borda points from second and third rank votes, and their participation will be all but eliminated. This is something we should fix before the next election.

[Cindy Ecksol]

If we do not change the law, we will most likely revert to a simple majority electoral system. It will be the end of preferential voting in the CDS. Two of our factions that have, in my opinion, an important voice in our politics get most of their Borda points from second and third rank votes, and their participation will be all but eliminated. This is something we should fix before the next election.

There's some great analysis here, Jon, but also some places where it seems that you jump to conclusions that cause alarm bells to go off when in fact no such alarm is warranted.

1. When I first read the description of how the ranking of parties and individuals worked, it was clear to me that the tactics that you describe were not only available, but that a well-organized party would certainly recognize the availability of those tactics and could easily execute them. Since this system was obviously discussed extensively in previous sessions of the RA, I would be surprised if those who approved it didn't understand the possibilities. Certainly you, as the one closest to the programming of the voting machines, must have understood the possibilities since you spotted the "anomaly" in the data as it came in. You would not have been able to do that had you not understood what the data would look like and noticed the pattern when it appeared. But for you to complain about these tactics as "unfair" (no, you didn't use that word, but you certainly implied it) strikes me as...well, amusing at best, and disingenuous at worst.

2. In your example you assign a certain expectation to Jane, saying that she would expect Paula to vote in a particular way. But there's no indication that Paula made any such promises: she merely asked for Jane's second place vote. Jane was free to grant that request or not, and Paula was free to vote as she pleased. To imply that deceit was involved in Paula's solicitation of Jane is farfetched at the least and perhaps even libelous. In any case it's probably not the best way to make your point.

3. For one individual to have access to the results while the voting was going on and to actually appeal to the SC on the basis of that access strikes me as a violation of the integrity of the vote of the worst kind. In fact, if you mentioned this analysis to any voting citizen of the CDS during the week, it was YOU who changed the outcome of a perfectly legal, by-the-rules vote, not those who understood the system and created strategies to maximize their return under it. I don't know how the constitution can be amended to remedy this, but it certainly needs to be discussed in the current session of the RA to ensure that no such interference occurs in the future. I can only imagine the constitutional crisis that would result in the US if voting booth techs had the ability to check the counts on the machines at noon and then mentioned to other voting citizens that an anomaly had been spotted that indicated that one particular party was "getting the vote out" better than the others!

4. In stating that "In spite of this, only about one third of the faction eliminated all other factions from their ballot, with those votes arriving in a block at the beginning of the election" you have completely violated the privacy of an entire faction, potentially causing dissension in the ranks and perhaps a complete disintegration of the faction. As in #3 above, I'd call this a mis-use of your access to the data of the worst kind, and action needs to be taken to ensure that it does not happen again.

5. As for the "prisoner's dilemma," I'll just point out that it's fairly difficult to structure a non-trivial game where the prisoner's dilemma does not arise. Our voting system is not an exception to this rule, so for you to raise such an alarm about it is....well, as I said before, "amusing and possibly disingenuous."

Now that I've got all that out of the way, let's talk about the future. I agree with Jon: the system is flawed and needs to be changed. One set of changes involves access to the data and can possibly be handled by the SC. But the voting system itself needs improvement. I do not believe that the "doomsday scenario" that Jon envisions will ever be a reality, but there is one simple way of encouraging people to rank more than one faction. The flaw that I see involves an assumption built into the system that the faction ranked first is the faction to which the voter "belongs" and limiting their ranking of candidates to that one faction. Unfortunately there's no way to ensure that this is a correct assumption. I can envision scenarios in which a voter might want to support another faction than theirs as #1, and under the current system they would then be prohibited from ranking their faction's candidates. That's not particularly democratic, now is it? What I would propose is that each voter be allowed to rank not only the factions, but also the candidates within the factions. True, the winning candidates might not be the ones who the faction members alone would choose, but on the other hand I can envision scenarios in which an out-of-control faction leader is beloved only by faction members and should not sit in the RA. If everyone could rank individuals in every faction, not only would they have more incentive to rank more than one faction, the system overall would be safer from potential self-perpetuating demagogues.

So here's to an RA that can manage not only to identify problems, but to quickly and efficiently propose legislation to remedy those problems and civilly work through the compromises necessary to implement that legislation.

Cindy Ecksol

[Beathan]

Jon --

Very interesting analysis. However, I think that the system is not broken, and is self-correcting.

I have previously posted my position:

http://forums.slcds.info/viewtopic.php? ... a&start=15
(The post at the bottom of the link is more instructive than the post at the top.)

That said, I also think that such manipulation of the vote is self-correcting in a multi-party system so long as the party guilty of the manipulation does not have a super-majority. I certainly plan to penalize any faction that uses such tactics by actively voting against them in the next election. This might become that issue I never thought I would have -- the issue that causes me to cast outright rejection votes rather than nuancing my vote by ranking. Thus, we will get to the "tit-for-tat" correction in the Prisoner's dilemma.

For this reason -- as a tactic, I think it might work once and must work the first time tried or else it backfires. Hopefully, the next election will prove me right on this. Of course, if any faction gains a super-majority, it will be able to use such tactics with impunity. However, it will also not need to use such tactics to maintain total electoral control. We will have that worst-of-all-systems -- the one-party democracy.

Beathan Vale

[Jon Seattle]

Since this system was obviously discussed extensively in previous sessions of the RA, I would be surprised if those who approved it didn't understand the possibilities. Certainly you, as the one closest to the programming of the voting machines, must have understood the possibilities since you spotted the "anomaly" in the data as it came in. You would not have been able to do that had you not understood what the data would look like and noticed the pattern when it appeared. But for you to complain about these tactics as "unfair" (no, you didn't use that word, but you certainly implied it) strikes me as...well, amusing at best, and disingenuous at worst.

This is incorrect. I was asked by the CDS government to program voting machines implementing the new law just prior to the election. I had never been involved in the vote counting before this. I did not realize this strategy was available until it started happening, and given the reaction from the SC, neither did they. In fact, I still have a very hard time explaining it to people who confuse the idea of ordering a set of objects with rating them.

2. In your example you assign a certain expectation to Jane, saying that she would expect Paula to vote in a particular way. But there's no indication that Paula made any such promises: she merely asked for Jane's second place vote. Jane was free to grant that request or not, and Paula was free to vote as she pleased. To imply that deceit was involved in Paula's solicitation of Jane is farfetched at the least and perhaps even libelous. In any case it's probably not the best way to make your point.

This assumes that Jane has perfect knowledge of the possibilities. I will say that almost everyone who voted, myself included, did not know this trick existed or assumed other parties would vote their honest preference. Paula would not have bothered with soliciting Jane if she had reason to believe that Jane would use the same strategy. The trick works only though deception. You may want to call me and others who ranked all factions idiots (perhaps you are correct on that), but we would hardly had voted in that way had be known what was going on.

3. For one individual to have access to the results while the voting was going on and to actually appeal to the SC on the basis of that access strikes me as a violation of the integrity of the vote of the worst kind. In fact, if you mentioned this analysis to any voting citizen of the CDS during the week, it was YOU who changed the outcome of a perfectly legal, by-the-rules vote, not those who understood the system and created strategies to maximize their return under it.

No, there was a disagreement about the meaning of the law and still remains some ambiguity about it. It was not clear to me, nor should it be to anyone, that it allows elimination or more than one faction, or gives a complete account of how additional eliminated factions should be counted. It was only when the SC reviewed the history of the act that it found the interpretation I used in counting to be the correct one.

In addition, it seems to me, there is an open question if a system that gives an advantage to one group on the basis of special knowledge and in fact penalizes expression of one's political preferences is consistent with the UNHCR that is our bill of rights. That document guarantees equal suffrage and freedom of expression.

I don't know how the constitution can be amended to remedy this, but it certainly needs to be discussed in the current session of the RA to ensure that no such interference occurs in the future.

To the contrary, elections and voting conditions are challenged all the time in the US by people who monitor and run elections and those who represent political parties etc. Take, for a glaring example, the 2000 US presidential election.

In this case the appeal was done in private messages to the SC and it was up to them to decide if the problem was serious enough to warrant remedy. I am very amused to think that you would limit the SC from hearing about an election it was supervising if there were serious anomalies. And never, ever, should someone who is running election software be limited from reporting on and appealing to the courts. Such appeals are necessary to insure that votes are not tampered with.

4. In stating that "In spite of this, only about one third of the faction eliminated all other factions from their ballot, with those votes arriving in a block at the beginning of the election" you have completely violated the privacy of an entire faction, potentially causing dissension in the ranks and perhaps a complete disintegration of the faction. As in #3 above, I'd call this a mis-use of your access to the data of the worst kind, and action needs to be taken to ensure that it does not happen again.

This fact is already completely available in the aggregate frequency counts (see Pat's message to that effect). Such statistics have been released and discussed in past elections. It violates no person's privacy because it releases no specific names nor the information needed to reconstruct any particular individual's votes. Furthermore, aggregate vote counts are always released for every election in every locality in the US, usually in very specific geographic detail. I know this for sure, because I have done spatial and statistical analysis for US political scientists who are studying voting patterns.

Cindy, sorry, you manage to void the remainder of your argument. One moment you argue that the system is just fine and fair and there is no deception involved, and in the next breath tell us we should rush change it. It gives the appearance of wanting to avoid the consequences.

[Jon Seattle]

http://forums.slcds.info/viewtopic.php? ... a&start=15
(The post at the bottom of the link is more instructive than the post at the top.)

That said, I also think that such manipulation of the vote is self-correcting in a multi-party system so long as the party guilty of the manipulation does not have a super-majority. I certainly plan to penalize any faction that uses such tactics by actively voting against them in the next election. This might become that issue I never thought I would have -- the issue that causes me to cast outright rejection votes rather than nuancing my vote by ranking. Thus, we will get to the "tit-for-tat" correction in the Prisoner's dilemma.

For this reason -- as a tactic, I think it might work once and must work the first time tried or else it backfires. Hopefully, the next election will prove me right on this. Of course, if any faction gains a super-majority, it will be able to use such tactics with impunity. However, it will also not need to use such tactics to maintain total electoral control. We will have that worst-of-all-systems -- the one-party democracy.

I happen to agree with you that the system is self-correcting, and that this is at most a one-time trick. (That is assuming voters have complete and accurate information.) I just fear to see the consequences of the correction because I am not at all sure they will be the best for our community. Our system leans over backwards in the direction of being more inclusive, not less inclusive. This is something I like about it.

As for the super-majority party, you are right, but I would be very surprised to see it happen. Because seat allocation is proportional and Borda counts are distributed between parties, it would be very very difficult to achieve. However, It becomes more conceivable if people stop ranking and vote for a single faction.

[Jon Seattle]

A few more comments..

I think that this attempt at vote manipulation was a mistake. A new party who, afraid that they did not have the votes, took some bad advice. I very much doubt they would do it a second time. It was not a majority of their members, and their party does represent a substantial group that deserves a voice in our legislature. So I argue that if we can close the door on future abuse, lets be generous.

Without a correction in the law, the next election, or any election where people are allowed to refrain from voting in second and third rank may be bad for smaller parties. Beathan has one solution to that, declare that a smaller party will never use that strategy and so people will feel confident in ranking that party second. (Given what I know about the SP, I absolutely believe them!) I wonder though, if this may make our system more volatile than it would be otherwise.

I understand and sympathize with Pat and Beathan’s goal of making the system more expressive. We are reaching for a system that will allow us to express, not only our preference ordering, but the strength of our support for each faction. The problem is that when you try to graft that on to a preference ordering system it can have serious unintended mathematical consequences. We ran into those consequences.

Right now our system gives an advantage to minorities, and this does mean that we have to listen and negotiate with them. My fear is that in moving towards a system that is more expressive, we may be moving away from that consensus building aspect of our system. I hope we can preserve that aspect.

[Jon Seattle]

This is for those who are hell bent on shooting themselves in the foot.

The facts of the matter are on the table now and nothing you can do will put them back in the dark. I felt it was absolutely my duty to all of the electorate, without revealing individual votes, to make those facts known by publishing aggregate statistics and an explanation for the pattern in those statistics. People have a right to make decisions based on complete and accurate information.

Right now, if you want to be effective, you need to build confidence among those who are not in your inner circle. Your faction has a good level of representation in the RA and executive branch, so I feel you have a real opportunity to do that. There may be no retribution vote the next time if you prove that it is unnecessary. I truly hope you will do that.

As to attacking me personally, you are really wasting your time. I have a moderate position on this, in favor of reform. I have no official office, and currently am not a member of a faction, just an ordinary citizen.

I spent four weeks of full time work, my complete Christmas and New Years holiday developing the software and making sure all elements were in place for a fair and accurate election with an audit trail that also allowed anyone to get an instant receipt to make sure their vote was registered correctly. I designed the software so that with a bit more work it could be run by someone non-technical, hopefully someone other than myself (because I would rather not be in the business of running elections, thank you!)

Alas, I will probably never complete the project. Once again, you are helping the CDS to eat one of its most dedicated volunteers, and perhaps putting your party in a worse spot.

[Cindy Ecksol]

Since this system was obviously discussed extensively in previous sessions of the RA, I would be surprised if those who approved it didn't understand the possibilities. Certainly you, as the one closest to the programming of the voting machines, must have understood the possibilities since you spotted the "anomaly" in the data as it came in. You would not have been able to do that had you not understood what the data would look like and noticed the pattern when it appeared. But for you to complain about these tactics as "unfair" (no, you didn't use that word, but you certainly implied it) strikes me as...well, amusing at best, and disingenuous at worst.

This is incorrect. I was asked by the CDS government to program voting machines implementing the new law just prior to the election. I had never been involved in the vote counting before this. I did not realize this strategy was available until it started happening, and given the reaction from the SC, neither did they. In fact, I still have a very hard time explaining it to people who confuse the idea of ordering a set of objects with rating them.

Jon, you're either insulting your own intelligence here by claiming that you didn't understand the system that you programmed or insulting the community's intelligence by claiming that the average person is incapable of understanding something that was obvious to me after one reading. There was no "privileged information" here that was restricted to one group. I know that once the subject came into discussion, I myself brought it to the attention of several others and not all of those were in the faction I support.

2. In your example you assign a certain expectation to Jane, saying that she would expect Paula to vote in a particular way. But there's no indication that Paula made any such promises: she merely asked for Jane's second place vote. Jane was free to grant that request or not, and Paula was free to vote as she pleased. To imply that deceit was involved in Paula's solicitation of Jane is farfetched at the least and perhaps even libelous. In any case it's probably not the best way to make your point.

This assumes that Jane has perfect knowledge of the possibilities. I will say that almost everyone who voted, myself included, did not know this trick existed or assumed other parties would vote their honest preference. Paula would not have bothered with soliciting Jane if she had reason to believe that Jane would use the same strategy. The trick works only though deception. You may want to call me and others who ranked all factions idiots (perhaps you are correct on that), but we would hardly had voted in that way had be known what was going on.

That's complete nonsense! Let's dispense with loaded semantic terms like "trick" and "idiots" and "deception" and talk about reality. There was no "trick" and there were no assumptions made that others would not decide to refuse the suggestion that they rank a particular faction second to their own and eliminate that faction altogether once they understood how the system worked. So the faction that you are accusing of devious behavior and of exploiting a trick unknown to the rest of the citizens was in fact the original source of information on the possibilities of this feature of the voting system and a key force behind dissemination of that information. You yourself permitted and even encouraged at least one member of the faction to test the voting machine multiple times, and it was that experience that helped the faction fully understand how the voting process worked and pass that on to others. There was actually a very broad effort made to reach out to everyone who might have been a source of those second-place votes. To encourage their participation in the campaign, it was necessary to explain how the system worked. Once again, you seem bent on insulting the intelligence of the average citizen by claiming that they voted like robots and were incapable of using independent judgment in casting their votes.

3. For one individual to have access to the results while the voting was going on and to actually appeal to the SC on the basis of that access strikes me as a violation of the integrity of the vote of the worst kind. In fact, if you mentioned this analysis to any voting citizen of the CDS during the week, it was YOU who changed the outcome of a perfectly legal, by-the-rules vote, not those who understood the system and created strategies to maximize their return under it.

No, there was a disagreement about the meaning of the law and still remains some ambiguity about it. It was not clear to me, nor should it be to anyone, that it allows elimination or more than one faction, or gives a complete account of how additional eliminated factions should be counted. It was only when the SC reviewed the history of the act that it found the interpretation I used in counting to be the correct one.

In addition, it seems to me, there is an open question if a system that gives an advantage to one group on the basis of special knowledge and in fact penalizes expression of one's political preferences is consistent with the UNHCR that is our bill of rights. That document guarantees equal suffrage and freedom of expression.

Again, there was no "special knowledge" involved here. The faction that understood the system the best actually passed that information along to as many people as they could contact. Those individuals then had the ability to act as they chose, by supporting the strategy, by ignoring it, or by countering it. For you to claim otherwise is still amusing, disingenuous, and possibly insulting to the intelligence of the CDS citizenry.

I don't know how the constitution can be amended to remedy this, but it certainly needs to be discussed in the current session of the RA to ensure that no such interference occurs in the future.

To the contrary, elections and voting conditions are challenged all the time in the US by people who monitor and run elections and those who represent political parties etc. Take, for a glaring example, the 2000 US presidential election.

In this case the appeal was done in private messages to the SC and it was up to them to decide if the problem was serious enough to warrant remedy. I am very amused to think that you would limit the SC from hearing about an election it was supervising if there were serious anomalies. And never, ever, should someone who is running election software be limited from reporting on and appealing to the courts. Such appeals are necessary to insure that votes are not tampered with.

Jon, you're insulting our intelligence again. No election is ever challenged until AFTER the election is completed. Your ethical violation was not in taking the issue to the SC, but in even looking at the incoming information, not to mention analyzing it and presenting it to the SC, before the election process was completed.

4. In stating that "In spite of this, only about one third of the faction eliminated all other factions from their ballot, with those votes arriving in a block at the beginning of the election" you have completely violated the privacy of an entire faction, potentially causing dissension in the ranks and perhaps a complete disintegration of the faction. As in #3 above, I'd call this a mis-use of your access to the data of the worst kind, and action needs to be taken to ensure that it does not happen again.

This fact is already completely available in the aggregate frequency counts (see Pat's message to that effect). Such statistics have been released and discussed in past elections. It violates no person's privacy because it releases no specific names nor the information needed to reconstruct any particular individual's votes. Furthermore, aggregate vote counts are always released for every election in every locality in the US, usually in very specific geographic detail. I know this for sure, because I have done spatial and statistical analysis for US political scientists who are studying voting patterns.

Jon, you too have clearly misunderstood the system. You made statements about the behavior of a particular faction that you have little support for. YOU DO NOT KNOW whether those who ranked a particular faction first were "faction members" or not because we have no real definition of "faction member" beyond a personal declaration that does not follow one into the voting booth. For you to say "only about one third of the faction..." is for you to speculate entirely about the behavior of a particular group when you can't even define the group. And you're making these claims with support from a classical fallacy, the "fallacy of authority," claiming that your experience is evidence that you must be correct. Again, you're insulting our collective intelligence by expecting us to support your breach of good manners and ethics.

Cindy, sorry, you manage to void the remainder of your argument. One moment you argue that the system is just fine and fair and there is no deception involved, and in the next breath tell us we should rush change it. It gives the appearance of wanting to avoid the consequences.

Yikes, Jon, what is it with you and the loaded semantics? I stand by my statements: there was no "deception" and the system is certainly fair in that it does not permit tactics on the part of one faction or individual that are not also permitted (and known, thanks to the faction who explored them first) to others. That does not make it perfectly democratic, nor does it make it a "perfect" system. If the experiment in CDS is about creating a system that maximizes participation and cooperation from all citizens while permitting distinctive voices to emerge on a factional or individual level, we've clearly got a ways to go. My suggestion that we consider allowing ranking of candidates in all parties is my first idea on how one might move in that direction, and I hope that I'm not the only one who agrees that it deserves discussion. But there are other possibilities too. I've already begun noodling about some of those, and I certainly hope that others among the CDS citizenry who are as interested in the structure and function of voting systems as I am will chime in with other ideas. That's the nature of political discourse, and I expect that we'll be having quite a lot of it over the next six months.

Cindy

[Beathan]

::sigh:: Here we go again. It is a shame to see an allegedly "reformed" party disappoint us again by repeating old behaviors. CARE was known for its attempt to manipulate elections -- and then for crying foul after the election when their underhanded tricks, while partially successful, were checked sufficiently by the operation of the system to prevent them from undermining the democratic process. I had great hopes for the "Nuness" of NuCARE. Sadly, it seems that the leopard has not changed its spots. (Hopefully the leaders of NuCARE will now realize that the opposition to CARE was not directed at Michel personally, but was directed at and motivated by his project and methods, which struck many of us as subversive to democracy.)

Jon did a great job on this election. As one of the main RA-member proponents of the election rules in question, I think that his handling of the election was entirely consistent with the intent of the RA. Petty sniping at him seems like nothing more than the sour-grapes of a minority Party that unsuccessfully seeks to use democracy to gain autocratic control in a working democracy. The failure of this attempt -- and what I expect to be the future ramifications to the detriment of the offending Faction -- prove nothing more than that our system works and works well.

Beathan Vale

[Justice Soothsayer]

Sigh, indeed, Beathan, but in my case the sigh isn't over the behavior of a faction, but how quickly the discussion has degenerated into recriminations.

Jon has done us all a service by carefully analyzing the implications of the Borda count system, and last term's decision to allow us to rank fewer than the totl number of factions competing in the election. While Jon sees that decision as allowing for the potential of mischief ("strategic voting"), those who don't want to award points to our least-favoured factions should be able to do so.

For me, one of the interesting things about CDS elections has been that voting generally has been more about the factions than the avatars. Indeed, one of the factions didn't have as many candidates as seats earned in the election. While some may see this as a flaw in our systems, I see it as a strength. One could easily envision a system where we vote for, or rank, 7 or fewer candidates for the RA, each of whom are identified by faction label. But such a system also allows for strategic voting through "plunking" of votes for less than the total number of seats available, and could become simply a beauty contest.

[Rose Springvale]

I agree Justice. While Jon and I have certainly had our differences, I think to criticize him for monitoring and pointing out potential issues with the system he created is not only not illegal, but responsible. And as a rather well informed citizen usually, i have to say i didn't know we could "no vote" more than one party... not that i would have. As a person who's run campaigns, monitored voting precincts and had my share of party politics in RL..I see nothing at all wrong with this process.

My concern lies in the necessity of a citizen to choose a party with more than one candidate running in order to be able to have a say in the outcome of the leadership of the RA... and even in that i was wrong.. we could have had more "empty" seats to fill! So let's blame the PIO! for not clearly understanding the system, nor communicating it. And for assuming that 'someone else had it covered."

I was asked by many many many new people "who" to vote for, and in every case, i refrained from giving them advice. I think as an INDEPENDENT public official it is wrong to even appear to endorse someone. Or the opposite. But i'm not a public official anymore, and there were so many things wrong with this election that it makes my head spin. I've asked that there be a special commission appointed to review election process and campaign reform. I'd like there to be a place where people can come and actually talk about what happened, how they felt about it and why.. .where misconceptions and outright untruths can be uncovered and corrected. I've made this request to the SC, but to the extent it falls under the jurisdiction of the RA,I'm making it here too.

The saving grace, as far as i'm concerned, is that the RA seems to have a healthy balance of qualified individuals.

Let's just not shoot the messenger for the message.

[Cindy Ecksol]

...there were so many things wrong with this election that it makes my head spin. I've asked that there be a special commission appointed to review election process and campaign reform. I'd like there to be a place where people can come and actually talk about what happened, how they felt about it and why.. .where misconceptions and outright untruths can be uncovered and corrected. I've made this request to the SC, but to the extent it falls under the jurisdiction of the RA,I'm making it here too.

The saving grace, as far as i'm concerned, is that the RA seems to have a healthy balance of qualified individuals.

Let's just not shoot the messenger for the message.

I'm 100% in agreement with Rose on this. And I agree with Justice that Jon has done a service by analyzing the results and presenting them. And also with Justice regarding the degeneration of the tone of this conversation. I was doing my best to stay on the issues and not get personal, but obviously I failed in that my last posting inspired a very emotional outburst from Beathan. I apologize for that, and I have already suggested directly to Beathan that he and I meet to discuss the real issues one on one.

Where I don't agree with Rose or Justice or Jon is on the propriety of "peeking" at election results before the election is complete. As Jon points out, everyone who voted received a receipt, so they can be sure that their vote was properly recorded. The time for reviewing the statistics and analyzing them is after the election is completed, not halfway through. If anyone knows of a democratic system where it is the norm for the vote counters to look at the votes cast before the polls have closed, I'd be interested to hear about it, but to my knowledge there is no such beast. The Soviets used to do that I understand....but I don't think that's the kind of system that we're attempting to develop here.

Once an election is complete, if analysis shows that some faction or individual violated the rules then there is plenty of time before the inauguration for a protest to be filed with the SC. Even in the 2000 election in the US no protests were filed until after the initial recount revealed problems with the voting process in Florida. With all the rumor and innuendo and now outright accusations alleging subversion of the election process flying around here I would have thought that someone would surely have filed a protest before the inauguration, so I'm really wondering why that didn't happen and hope someone will explain. I think perhaps I just don't understand the fine nuances of the SC process and need to be educated.

Cindy

[Jon Seattle]

Jon has done us all a service by carefully analyzing the implications of the Borda count system, and last term's decision to allow us to rank fewer than the totl number of factions competing in the election. While Jon sees that decision as allowing for the potential of mischief ("strategic voting"), those who don't want to award points to our least-favoured factions should be able to do so.

Its useful in this context to think about utility ratings, lets think of this as a rating of each faction by an individual according to their preference for that faction gaining seats. We really have three rating systems going on here, the voter’s utility ratings, the ratings that the voter would like to reflect in her ballot, and the Borda count ratings. The difference between the first and second kind of ratings arise when a system supports people gaining a higher level of utility with a vote that is not an expression of their actual ratings. In other words where there are opportunities for a group to get better outcomes by not voting their actual preferences.

Three things need to be considered when evaluating the change in the law. First, Borda with elimination must be better at reflecting a person’s utility ratings, I think this is what Justice is saying. Second, the system must not give extra incentive for indicating ratings that do not reflect her true preferences. Third, the system must be free of paradoxical outcomes. Our system fails all three.

I am going to address each issue in a post, but because I have limited time, I will post each separately.

1/ Expressiveness

A lot of the frustration with simple Borda counts is because they are an ordering system and not a rating system. The weighting forms a nice even gradual line, if you get mad you can’t really punish several factions because some one of them will end up with counts from, say, third rank points. Here is the typical Borda count weights, normalized:

1.00, 0.75, 0.50, 0.25

With elimination you get a sudden drop between the last ranked faction and the others. Since the system is differential, the correct way to compute the drop is relative to the average of the counts that would have been assigned to the eliminated factions had they not been eliminated.

1.00, 0.75, 0.50, 0.25 no difference
1.00, 0.75, 0.25, 0.25 drop: mean 0.125 for each of two factions
1.00, 0.25, 0.25, 0.25 drop: mean 0.250 for each of three factions

In other words the more factions you drop the more you disadvantage each faction on average. I you were mad enough to drop faction A, and also faction B, then you must be madder at faction A than if you dropped faction A alone! B made you madder at A and visa versa!This is really weird and I suspect resembles no person’s ratings.

We see how it helps the person who gets really mad at some of the factions, but what about a person who has one favorite but is indifferent to the remainder of the factions? Lets say the person’s ratings look like this:

1.00, 0.50, 0.50, 0.50

The voter has two options, either eliminate, or rank:

1.00, 0.25, 0.25, 0.25 difference: 0.25 each
1.00, 0.75, 0.50, 0.25 mean difference: 0.16 each

The system forces this voter to give points to one and take points from another faction arbitrarily, and the result does not match the voter’s intension. The same problem exists with normal Borda, but you see the addition of elimination unfairly aids the expression of one particular type of voter (the really mad-at-factions voter) while not providing the same advantage to all others.

A ranking system is not a rating system, and if you try to make it a more like a rating system, you are likely to mess it up. And so we did, myself included.

[Bromo Ivory]

Wow --- what a back and forth.

For what its worth, the Borda system was more or less "mutated into something else" the second the RA granted the voters the right to not rank some factions. I believe as a response to the analysis, the CDS can do one of 3 things:

1. Accept the new system, warts and all
2. Change the system (further reform) to remove things people think are unfair
3. Revert to the old system

I myself did not like the new reform for precisely the reasons that Jon stated. And that Beathan inadvertently re-iterated by threatening to "punish" (presumably) nuCARE in the next election by not ranking them. Now it is feasible that "everyone" is gearing up to "punish" others - thereby moving further away from the system that Jon (and I) prefer. And this is not unethical, it is the new system at work in its awful new form. The incentives are there. If we do not like a particular behavior, we have no business rewarding it, and that is exactly what we are doing here.

I would say that we can use loaded words and accuse each other of "abuse" - all I see is "use" of the new rules on a limited scale which obviously has a lot of folks very uncomfortable (me included).

It made me nervous when Jon reported the findings the way he did (with obvious access to insider knowledge during the election) - and I understand Cindy's concern about it - though I know Jon well, and I do believe he is trying to duly report the results and has done nothing actually wrong. I do think it would be prudent in the future for anyone privy to results have some sort of "no comment" period to avoid even the appearance.

Nevertheless, back to topic - it may serve little purpose to accuse people of being unethical for using a new right granted to them by the RA. Let's figure out if there is anything we can do about it, eh?

Oh, and if anyone asks, I did not eliminate any parties in my ballot.

[Beathan]

Bromo and Cindy --

For the record, I don't plan to punish NuCARE by "not ranking" it in the next election. Rather, based on a personal appeal and an assurance that NuCARE had changed in a real, meaningful way, I did not rank "NuCARE" fourth (as I would have ranked old CARE). Because I now believe that NuCARE engaged in a lawful manipulation of the system, a manipulation which I oppose on principle, I am now inclined to rank CARE last in the next election. I will "not rank" a party only if I "not rank" more than one party -- and that will be only if multiple parties engage in this tactic. The DPU came somewhat close -- but not close enough for me to conclude (as I have with NuCARE) that the DPU engaged in a partisan misuse of a tool designed to enhance the individual voices in CDS elections.

In other words, NuCARE has lost my goodwill, after having tenuously and momentarily gained it. I am reverting to my now-accustomed skepticism. NuCARE can win me back -- but only by showing that it will not seek to again exploit the seams and cracks of our electoral system. This is not to say that NuCARE did anything illegal. If I were on the SC and were deciding the lawfulness and Constitutionality of the behavior of all participants in this election, I would uphold each and every action of each of every party of which I am aware.

However, as a matter of electoral policy, I categorically oppose factional direction of individual votes other than the first vote. I think that the remaining preference votes should be left as matters of pure individual, rather than factional, preference. That is part of the principles of the Simplicity Party -- and one reason why the SP will never engage in "machine politics" of this kind.

Further, I don't think the "punish" response is an emotional response. Under game theory, and the tit-for-tat corrective to the prisoner's dilemma, it is not only rational, but optimally rational. Thus, I don't think my post was "emotional" so much as "indignant" -- which is a rational reaction (at least on Aristotlean ethical theory). I was personally misled by pre-election vote mongering by a Faction which, it turns out, was engaged in electoral manipulations specifically calculated to damage the standing of my Party. That is politics, yes -- but so is the rational response to penalize such tactics next time around.

Beathan Vale