ENnies discussion thread

[MavrickWeirdo]

Thing is, with IRV, the author of A does not help his cause by using this system of voting. It's not strategic voting for advantage, merely strategic voting from bitterness. Unless we are assuming the voters will vote from spite, I don't think this should be an issue. If are assuming this, then, IMO, the Ennies are a farce anyway.

One of the main points being made by the IRV advocates in this thread is that, once he has been eliminated, there is no reason for Author A not to give kudos where it is due.

1. Yes, I admit that in my example has strategic voters "vote from spite". I was demonstrating that such a vote is possible in IRV, where the "spite" doesn't kick in till the first choice has been eliminated. A "vote from spite" is less likely in AV because approving any other canidate can hurt your chances.

2. I find it interesting that some here feel that most people will vote "Strategically" rather than "Honorably" (by ranking everything). On the other hand they do not expect people to vote "Spitefully" because there is no "Logical" reason to do so.

Do we expect the voters to be Lawful Neutral, or Chaotic Neutral?

By the way, before I made the examples I actually prefered IRV method. I'm still on the fence because the following statement is also true.

In IRV

anyone can vote for those that they think are deserving, without damaging the prospects for those that they believe are best.

 

[Psion]

I didn't - at least I don't think I did - that's who I meant when I said PC (Pirate Cat). BTW - I did not see your hat in the ring, but I would assume you would be another heavy hitter!

It's a tough choice -- and not a final one -- but I have a bad habit of taking on obligations that I don't have time for. I am trying to look ahead this year. As some of you may know, I have a relatively new child and a new job (at a job site that is less friendly to non-work-related internet access.) I thought I would look at the player field and see if I felt it was in "good hands" before I decided one way or the other. So far, the contenders seem to be shaping up nicely.

I will be involved in the ENnies in another capacity, however.

 

[Nisarg]

Thing is, with IRV, the author of A does not help his cause by using this system of voting. It's not strategic voting for advantage, merely strategic voting from bitterness. Unless we are assuming the voters will vote from spite, I don't think this should be an issue. If are assuming this, then, IMO, the Ennies are a farce anyway.

One of the main points being made by the IRV advocates in this thread is that, once he has been eliminated, there is no reason for Author A not to give kudos where it is due.

On one hand, I think you are seriously underestimating the bitterness constituency.

On the other, it doesn't make the Ennies a farce, because what you are asking people to do is vote for what they feel is the best product. If they don't vote for it, its because they don't think its the best product. Period.
It doesn't matter if their reasoning is because "Company A's product has poor editing" or because "Company A's product is a steaming pile of written by a dirty self-righteous little dickweasel who I hate".

Either way, its a legit vote, the fact that you might not like the latter's motives has nothing to do with the validity of his right to vote out of sheer bitterness.

Hell, how many people voted in your country's last election out of sheer bitterness? I know I did...

Nisarg

 

[Morrus]

Either way, its a legit vote, the fact that you might not like the latter's motives has nothing to do with the validity of his right to vote out of sheer bitterness.

Not to quibble, but nobody has a right to vote in the ENnies. And any conditions imposed on their permission to vote are valid; for example, if there were some possible way we could eliminate certain voting habits, we would. I don't like strategic voting, and I really disagree with undervaluing a product below its worth deliberately just so your favourite will win. If you consider a product average, you should give it an average score, not a "1" so as to ruin its chances of winning.

Of course, no voting method will prevent this entirely. But last year, we did discard all "10/1/1/1/1" ballots. I refused to believe that anybody honestly considered 4 out of 5 nominated products, which had been selected for their quality by five judges, to all be worth only a 1, and the other - coincidentally - to be worth a 10. The odds on that being an honest vote are so remote that it's not even worth the effort of debating whether it might be.

So, ironically, those people who voted in that method actually deprived their fave products of a 10 vote. If they'd been honest and selected "I'm not familiar with this product" or given a range of scores which weren't all 1s then their chosen fave would have done better.

But still, that's not ideal either. There's no ideal, but I'd strive for enforcement of honest voting even if it's a goal that can't be reached.

 

[fusangite]

2. I find it interesting that some here feel that most people will vote "Strategically" rather than "Honorably" (by ranking everything). On the other hand they do not expect people to vote "Spitefully" because there is no "Logical" reason to do so.

I don't think there is any such thing as voting out of spite. In IRV, you vote for things you think should win and don't vote for things you think should not. Whether the reason you want something to lose is "spiteful" or not is neither here nor there.

The advantage of IRV over approval voting is that there is no conflict between voting in the way most likely to defeat the most hated candidate and voting in the way most likely to elect your favourite candidate. No voting system does or should inquire into the rationality of why you want certain candidates to win and others to lose. All a voting system can do is ensure that your dual agendas of wanting some things to win and others to lose don't conflict with eachother. IRV does this. Approval voting does not.

That stated, voting systems are premised on the idea that people will pursue their agendas rationally, however irrational those agendas might, themselves, be. No voting system can compensate for irrational or illogical ideas nor should they. Thus, while the question of what portion of the voters will behave rationally in their own interest is an interesting one, it does not meaningfully pertain to this voting systems debate. Because of this, what we should do is design our systems on the assumption that voters are rational actors -- not because all are but because rational actors are the only people we can meaningfully serve when we design voting systems so we might as well serve them well because no decision we can make will meaningfully serve people who don't rationally pursue their own agenda.

 

[fusangite]

Of course, no voting method will prevent this entirely. But last year, we did discard all "10/1/1/1/1" ballots. I refused to believe that anybody honestly considered 4 out of 5 nominated products, which had been selected for their quality by five judges, to all be worth only a 1, and the other - coincidentally - to be worth a 10. The odds on that being an honest vote are so remote that it's not even worth the effort of debating whether it might be.

You discarded a bunch of votes on my ballot because you suspected I voted strategically!?

You have now completely sold me on the need to reform the system. Any voting system where people following the voting procedure have their ballots invalidated because the people counting them don't like how the people followed the procedure has got to go.

So, ironically, those people who voted in that method actually deprived their fave products of a 10 vote.

Did you notify us of this beforehand or did you change the voting rules after people had already cast their votes?

Well, gee, if the people counting the ballots can change the rules of the election on the fly, how legitimate can the election be?

If they'd been honest and selected "I'm not familiar with this product" or given a range of scores which weren't all 1s then their chosen fave would have done better.

So, I should have voted 10/1/1/1/2 and then my vote would have counted? This is the problem with Borda and Borda-like systems; you can't really build a gentleman detector.

There's no ideal, but I'd strive for enforcement of honest voting even if it's a goal that can't be reached.

Or you could use a system that makes this an irrelevant impossiblity (ie. IRV).

 

[CRGreathouse]

But still, that's not ideal either. There's no ideal, but I'd strive for enforcement of honest voting even if it's a goal that can't be reached.

The whole point of these various voting methods is to make the best strategic voting the same as honest voting, so there's no incentive to dishonestly rank products.

 

[Nisarg]

Not to quibble, but nobody has a right to vote in the ENnies. And any conditions imposed on their permission to vote are valid; for example, if there were some possible way we could eliminate certain voting habits, we would. I don't like strategic voting, and I really disagree with undervaluing a product below its worth deliberately just so your favourite will win. If you consider a product average, you should give it an average score, not a "1" so as to ruin its chances of winning.

Of course, no voting method will prevent this entirely. But last year, we did discard all "10/1/1/1/1" ballots. I refused to believe that anybody honestly considered 4 out of 5 nominated products, which had been selected for their quality by five judges, to all be worth only a 1, and the other - coincidentally - to be worth a 10. The odds on that being an honest vote are so remote that it's not even worth the effort of debating whether it might be.

I think that's fairly ridiculous. Who are you to say that someone didn't actually read all the books in the category and happened to think all of them sucked but one, and that one was perfect?

If you don't like the way people vote then change the system so they can't vote in the first place, don't go around some florida backroom tossing out the hanging chads, because what you're doing just about amounts to that. Its also a very small step in the temptation range from judging whether or not a vote was in "good faith" to judging whether or not a vote was "informed" to just plain tossing out anyone who doesn't vote for the side you want to win.

I'm with Fusangite on this one: you don't want people to vote 10/1/1/1/1, then make a new system where they aren't tempted to do that, don't rewrite the rules in midstream because your accounting system sucks.

But still, that's not ideal either. There's no ideal, but I'd strive for enforcement of honest voting even if it's a goal that can't be reached.

The only "dishonest" vote is someone who votes on a product they have no knowledge of; and in that case its dishonest whether its a 1, a 10 or a 7.5.
There's nothing "dishonest" in choosing to give a product you know what you feel it deserves, regardless of WHY you feel it deserves that, unless the contest is only for a specific criteria (ie. don't factor in the written content of a book when the contest is over its art).

Nisarg

 

[Morrus]

If you don't like the way people vote then change the system so they can't vote in the first place, don't go around some florida backroom tossing out the hanging chads, because what you're doing just about amounts to that. Its also a very small step in the temptation range from judging whether or not a vote was in "good faith" to judging whether or not a vote was "informed" to just plain tossing out anyone who doesn't vote for the side you want to win.

I'm with Fusangite on this one: you don't want people to vote 10/1/1/1/1, then make a new system where they aren't tempted to do that, don't rewrite the rules in midstream because your accounting system sucks.

Err.... yeah. Isn't that what this thread is about doing? It's kinda the whole point of the discussion.

And the implication that I want to make this anything but as fair as possible ("tossing out anyone who doesn't vote for the side you want to win") is not appreciated.

 

[Wulf Ratbane]

I'm with Fusangite on this one: you don't want people to vote 10/1/1/1/1, then make a new system where they aren't tempted to do that, don't rewrite the rules in midstream because your accounting system sucks.

They didn't. I distinctly remember being warned of that in advance; in fact I told all my friends voting for me to make sure to vote honestly and don't down-grade the other products "stragecially."

I wanted every positive vote I could get to count. (Didn't help, but still...)


Wulf