. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nap Time!!!

Wednesday, April 22, 2009
More drop-shaped fun

Oscar Mariena has filed suit against the Elections Council in the Judicial Council in order to prevent them from retabulating votes after Student Action drops Arya Shirazi. As you might suspect, this would take the seat away from CalSERVE's Viola Tang and give it to Matt Samuels.

There's an advisory opinion (PDF) exactly on point that rejects his claims, and CalSERVE did nothing about the By-Laws in the year since that advisory opinion was issued. Now, however, it's suddenly very important that the By-Laws be interpreted in a different way that just happens to give them another seat in the Senate, even though they couldn't be bothered to actually change the By-Laws to reflect the "correct" interpretation. Good show, CalSERVE. Next time, you might have more luck getting Senate seats by getting more votes, rather than pulling arguments out of your ass for why the procedure should be retroactively changed. Student Action beat CalSERVE by well over 1000 votes in Senate races, and deserves that extra seat.

(This is, by the way, an excellent benefit of Alex Kozak filing that advisory opinion before the election results. He got a decision separate from a particular instance of political conflict so folks couldn't figure out which side they should be on)

The first argument is that only disqualification triggers retabulation.
ASUC By-Laws, Title 4, Article 15, articulate a detailed procedure for the tallying of votes. They clearly outline the disqualification of a candidate as the only reason for which a retabulation of votes may occur.

"If any candidates should be disqualified following the preliminary tabulation, a second and final ballot tabulation will commence following all decisions and settlements of lawsuits regarding elections by the Judicial Council." (ยง4.15.2).

This clause establishes the disqualification of a candidate following preliminary tabulation as the necessary condition for triggering subsequent tabulations. In other words, votes may be retabulated if and only if a candidate is disqualified from his/her position.
Besides telling me that Oscar doesn't know the difference between implication and equivalence, this statement has no informational content. The most casual reading of that By-Law shows that it provides a sufficient condition, not a necessary one. While he goes on to say:
No reading of the By-Laws or Constitution provides for a retabulation of votes in any other case.
one could just as easily note that no reading of the By-Laws or Constitution prohibits a retabulation of votes in any other case, either. (This is the difference between something being necessary and being sufficient) In any case, at least one reading allows retabulation for another case, and that's the reading the Judicial Council used last year in their Advisory Opinion. The AO is nonbinding, but all parties acted under the understanding that it existed, which means that Student Action had good reason to believe that they could wait until after tabulation to do their drops. To change the rules retroactively would be far more unfair.

(Interestingly, though Mairena quotes this advisory opinion on the topic of when candidates can drop, he doesn't mention that it directly contradicts his argument about retabulations. This means he doesn't give the Judicial Council a reason to reverse its opinion on that topic)

Mairena also takes the position that drops in the middle of executing the algorithm are allowed, despite the complete absence of any indication that such a process is allowed in the By-Law description of the algorithm. His refusal to read the By-Laws as allowing things not mentioned in the retabulation section apparently doesn't preclude him from reading the By-Laws as allowing things not mentioned in the algorithm.
When a losing candidate withdraws after results have already been tabulated, it can only be done with the intention of manipulating votes in order to alter the results of the election. Council should not allow the will of the voters to be subverted by a withdrawal tendered in bad faith.
What makes this bad faith? CalSERVE dropped candidates, too. Are they manipulating votes? Doesn't the fact that they run candidates that they have no intention of seeing in office mean they are manipulating the votes in order to alter the results of the election, too? How much difference does the fact that they drop while ignorant make?

I see no reason why dropping candidates to change the result is somehow wrong. The votes are there, and no one's vote is going to someone they don't vote for. In fact, I expect voters would prefer that candidates drop in order to allow their transferred votes to affect the election.

In fact, this whole case is post-hoc political manipulation. Oscar was a Senator for a whole year while the advisory opinion explained the drop mechanism. He made no effort to change the By-Laws to fix any problems he perceived. He now is acting in bad faith to a much greater extent than Student Action is.

Even if he wins this case, won't Shirazi just break some campaign rules, get sued by Student Action, settle for disqualification, and trigger the retabulation anyway?

posted by Beetle Aurora Drake 4/22/2009 10:24:00 AM #
Comments (29)
. . .
beetle, are you completely in bed with student action this year?
or are you just blind, because surely you can see how corrupt this move is?
Only if Student Action's your mom.

It's not my fault CalSERVE keeps doing stupid things, but I do recall calling Student Action utterly classless and tools in recent history.

Explain to me how this move is corrupt. There's nothing being hidden. There are no rules being violated. They are using the structure of ASUC elections to get as many Senators as possible, a property shared by every party in any republican government in existence. So please, Anonymous, enlighten us as to how this is corrupt.
Calserve is engaging in a vitriolic discourse to try and scare student action from acting in a reasonable matter and pick up that last senate seat. I don't think it will work.
You are right that the By-Laws neither explicitly prohibit nor endorse retabulations for reasons besides the one mentioned in Article XV, but the fact that this is the only case that is mentioned in the By-Laws, to deal with a very conceivable scenario, I think, emphasizes the exclusive character of that provision. I really think the simplest and clearest way to interpret that section is to read it as being the only case when a retabulation should occur.

That aside, I agree that the by-laws should have been clarified last year, but unclear by-laws are not a new phenomenon in the ASUC. The fact that the senate did not revise them this year does not really deprive anyone of their right to challenge the way the ASUC decides to interpret its rules.

As for the your claim that Student Action "deserves" the seat, I think that's pretty much bankrupt. In our voting system you don't vote for parties. You vote for candidates. You can't really take the aggregate number of votes for a party to mean that the votes were for the party, rather than a bunch of very popular candidates. CalSERVE could have earned more votes to avoid this, but so too could have Student Action. This is really not the issue. The practice of dropping candidates to manipulate election results, after they're already announced, complicates the process beyond any reasonable measure, for little reason except to allow big parties to play games. Both parties engage in this when they can, but that doesn't make it any less questionable.
It does, however, make complaining about it when it doesn't go your way opportunistic hypocrisy. Oscar is, after all, accusing SA of manipulation of votes while engaging in manipulation of the process.

The fact that dropping can get SA another Senator means that their votes weren't just for a few very popular candidates. Student Action has the votes they need to get 9 people into office. The fact that they didn't know that in advance does not, to me, justify prohibiting them from getting 9 people into office. No one's vote is being moved to someone they didn't want to vote for. What we're seeing is certain candidates are bowing out of the race so that their voters can have their remaining vote impact the result. There's no evil to this kind of "game-playing," and the voting system we have encourages it. The only ideological complaint here is that now it's being done in an informed manner.
I'll tell you what would be wrong: not giving party chairs and/or candidates all the info they need in order to make informed drop decisions.

This is an important part of the voting system.

People are deferring to the pointless/meaningless public tabulation as if it is sacrosanct. It means nothing.

This is just one more example of unbelievable CalSERVE bias. Obviously if CalSERVE had the chance to get another senator they would do it in a second. And Student Action wouldn't attack CalSERVE the way CalSERVE is attacking SA now.

For CalSERVE if something doesnt go their way, then the greatest injustice on the planet must have occurred.

Why doesn't the algorithm simply account for this and do it automatically? This is all quite retarded.
Do what automatically?
To clarify before people miss the picture,

I am not suggesting that the algorithm play party favorites, or even suggesting that the algorithm mutually try to help each party to its maximum capability. The algorithm should be party blind.

I am simply suggesting that the algorithm assume that out of the losers x, that [0, x-1] may be dropped, and then calculate from there.

I'm sure someone could figure it out.
Thanks for clarifying Anonymous, your reputation was really at stake.
What you fail to realize was that I was clarifying information. Information is far different from reputation.
Around here, information apparently isn't worth much.

I suppose you could do it automatically, but I'm not sure there's a clear idea of what "it" is. Dropping everyone but the winners and one more screws with alternates and such. I don't see any need to automate it if folks are willing to do it themselves, anyway.
The problem with that is that there are too many permutations for that to be reasonable. While 90%(yes, i am taking this number out of my ass) of the time a human could spot a possible move-up, a computer could calculate thousands if not millions of permutations of scenarios.

While it is a long shot, it is quite possible that Viola does in fact deserve that senate seat with certain others dropping out. The computer should go through all the permutations, and see which 20 senate candidates end up with the most votes in any given permutation.

I think.
...or the humans could write a program, if the humans were smart, that would the humans who to drop, and then they could go drop those people.
As Beetle pointed out, these humans do not understand the difference between a necessary condition, a sufficient condition, and a condition that is both necessary and sufficient. Or anything else for that matter.

God that's funny.

In any case, I think the underlying problem here is Arrow's theorem.
What would the computer do with multiple drop permutations with different results? Who decides which one works? To do so would require some kind of favoritism. In the current system, dropping is in the hands of each individual candidate (or their party), who can do whatever work they feel appropriate to find the correct drops. There's an entire department of CS nerds who could write programs to do the search, and the game theoretic implications work themselves out. outside of the program. The current system has the advantage of being well-defined, with individual candidates in charge of their own drops.
I don't think it would be a difficult program to write or require any CS nerdage. I could do it in matlab in maybe 90 minutes, and I'm no CS nerd. Figuring out how to read the data in would be the only hard part.
it's different when you drop candidates before preliminary tabulations than after. that's the key difference in what student action is doing here and what other parties (CalServe, Squelch, etc) do
Yes, they describe slightly different times. That doesn't explain why you think it's different in any way that matters.
no, actually if you drop before preliminary tabulations, you do so without knowing the tabulation results or what impact your drop will have on those results.

dropping after you know the preliminary results allows you to calculate what candidates should be dropped in order to influence and change the tabulation results.

and to reiterate my previous point, that is the key difference between what student action is doing in this case and what other parties (calserve, squelch, etc) do.
I'm familiar with the linear nature of time and its impact on the acquisition of knowledge. What I don't understand is why you think the difference is of moral significance. Why does being ignorant make the manipulation process different in a sense that matters?
It's of moral significance because what is the point of the STV system if elections can be rigged in this fashion?

If this technicality were exploited every year, we would have a situation where parties could easily dictate the outcome of election in their favor time and time again. It's not so much about being ignorant as it is about knowingly rigging elections in your favor. That undermines the entire concept of free and fair elections.
The point of the STV system is that if your candidate wins by a lot, some of your vote goes to other candidates, and if your candidate loses, all of your vote does. Dropping with knowledge doesn't affect either of these goals.

You work under a very different definition of "rigged" than I do. No one's vote goes anywhere the voter didn't want it to go. There's no obstacle to "free and fair elections" here. Student Action is openly using the process that has existed for a year, that all voters had the ability to learn about and that all candidates ran under in order to achieve their political goals. I don't perceive the moral horror that you do, and those who did perceive this moral horror should have done something about it. Suddenly having an enlightenment about how wrong it is now just looks like sour grapes.
When the tabulator program ran two Friday, the system is designed to drop by order from least votes to greatest votes until you get to the top 20. During that process, you are not sure who is going to end up where... its all in the hands of who voted ppl number 1,2,3,etc. Moreover, you do not know how your votes will transfer each round. I think that is part of this fair voting system.

By being able to mess with the tabulator program, and figure out the exact values the vote transfers would work... that completely manipulates the program.

If you were not going to drop before they ran the program, you should not be able to drop now just because you lost the election.

Technically, you already did during that process within the confines of how the system was supposed to be run (on good faith that you had no idea how the votes would turn out - since thats how everyone else got in)
Beetle: You have defied the CalSERVE Collective. Similar to John Moghtader (whom we kicked out), and Attorney General Sinanian (whom we tried), you are now persona non grata. We have applied to the ASUC Commission on Diverse, Representative, and Nonconfrontational Websites to shut this page down.

Resistance is Futile. Long live the Borg!
The problem is Arrow's theorem. It really is.
it's called a violation of independence of irrelevant alternatives and it's blatant manipulation.
You say "blatant manipulation," Arrow says "mathematically necessary aspect of voting systems."
Post a Comment

. . .