Best posts made by tec

@Jack-Waugh, I think that would be welcome.

@jack-waugh on the original question regarding languages, I don't really know FORTH (or its variations) and I have a strong attraction to LISP. But for the simple and practical reasons that simulations have to be run on some system, having JavaScript client code is really the only way to go if you want more than a handful of users running simultaneous simulations.

Hi All,

I haven't posted here since sometime in 2021, and I have a lot of reading to catch up on. What I have been doing in the meantime is working on the issue of consent and how it can be implemented at scale.

The idea is we would have a series of votes as part of the process of building our collective consent. What we would be voting on would be policies or policy bundles over a number of issues. The goal is to reach a consensus on what we collectively want our government to do. Also, this is done as an iterative process, with periodic votes until a consensus is reached with no explicit time limit.

My last post here was on Serial Approval Vote Elections (SAVE), and while there was some interest in it then, I realized I needed to do something to make it possible for folks to actually play with the idea. I now have a web site, SerialApprovalVoteElection.org that I invite you all to take a look at. It is not quite ready for the general public, but I think the folks on this site might well be interested, and could provide feedback to me on what to do next with the site.

The more general issue is whether a system like SAVE would be useful to us in providing information to our representatives about what we really want our government to do for us. To actually consent to the way we are governed. And to indicate when our government representatives are not acting with our consent.

The way I currently envision how the full process would work is that some non-government organization starts the process with an official master website of all the initial motions, with each of these motions a topic in a Collaboratorium or Deliberatorium. (I have not contacted Mark Klein at MIT about this, but I think it might well be consistent with his goals.) The overall structure of these motions is the basic text along with commentary both supporting and objecting arguments. (Duplicate "me, too" type comments are collapsed into a number for each argument regarding the motion. (The numbers are not votes, but rather a way to allow people to show they support a given argument without swamping the page with extra verbiage.) The reason to use documents of this type is they can be structured so as to be comparable to other motions.

The other thing the NGO needs to do is have a way to come up with the voter list, or voter eligibility criteria. (I am personally in favor of a vote going to every person subject to the rule of the government regardless of citizenship, on the grounds that anyone physically present and able to act independently can take action anyway, and it is better that the action takes the form of a vote than of an act of terrorism.) (I'm also in favor of some type of online voting, and some work on that has been done by Ben Adida, Helios: Web-based Open-Audit Voting.)

Once the initial set of motions is in place, the SAVE process begins. The first vote is a simple approval vote (AV), in which voters simply vote for which of the motions they like. The winner of the first round becomes the focus of the next round which is the first focused approval vote (FAV) round. There is time between the votes for voters to absorb the results and to discuss what happened. In the FAV round, voters are no longer voting for just the motions they approve of, but instead are asked to vote for all the motions they consider better than the focus. They can also vote for the focus itself, which is interpreted as a vote to end the process with the current focus being the consensus result.

After the vote tally for an FAV is counted, the system either finds the focus motion received more than half the vote and move votes that any other motion, in which case the consensus motion has been found and declared as the final winner. This round would be followed by a mandate approval vote, which is a normal AV vote with the sole purpose of measuring the mandate or support for every motion on the ballot.

What happens more often, particularly in the early rounds, is the focus does not get more than half the vote or, rarely, gets more than half the vote but less than some other motion. Then the system determines the focus for the next FAV round. If no motion received more than half the vote, the focus is a (strong or weak) Condorcet winner and remains as focus for the next round. If some other motion gets more than half the vote, the focus from the current round will be replaced by another motion as the focus for the next round. The algorithm for the next focus decision is strictly deterministic and public, as is the data driving the decision.

Since there are only a finite number of motions, at some point a motion will be chosen to be the focus that has already been the focus motion in a previous round. This, obviously, happens when there is a Condorcet winner, but it also happens when there is a top cycle. Whenever the next focus is a repeat, the voters are allowed to propose new motions. There are no restrictions on the motions other than their having a format that allows comparisons with existing motions. If a new motion introduces an entirely new issue, previous motions are presumed to be unspecified regarding the new issue. New motions (with or without new issues) can be thought of as being in one of two types: diversity motions, or compromise motions, although the division may not always be clear. If a voter does not like any of the initial motions, that voter might propose their own ideal motion as a diversity motion, an the motion will get votes from like-minded voters. If voters already have one or more motions they like in the mix, a voter might propose a compromise between the focus and a subjectively better motion that lost to the focus, with the thought that the compromise might get more votes against that focus.

The introduction of new motions means it is possible to defeat Condorcet winners and break top-cycles, and the electorate is thus freed from any limits from the initial set of motions, and more importantly it is quite likely that the final winner of a SAVE process will be better than any of the original motions.

As a political theory process, this vision of a future process is distinct from votes for individual offices or roles. It is a way for the electorate to provide guidance and feedback our legislative, judicial, and executive branches regarding what, exactly, we consent to, and perhaps more importantly what we do not consent to. Moreover, as a system that can be implemented over the internet, it allows voters to express their views in a nuanced way without having to write to a particular person or take part in a demonstration. With a large enough electorate and a secret ballot, it is safe to express one's true feelings on political issues.

So after all this, I have questions: Does this seem like a reasonable system? Is it worth pursuing?

I'll be posting a bit more about my website on SAVE a little later tonight. In the meantime, I hope you are all doing well, or at least as good as can be expected.

Best Wishes,

    --tec

Arrow's impossibility theorem is frequently cited in the discussion between ordinal and cardinal voting systems, with the idea that AIT only applies to ordinal voting systems. It is true that Arrow was only looking at ordinal systems, with the goal of a social welfare function (SWF) that took as input a collection of individual preference orders and returned a collective order. But what Arrow actually proved was his conditions were inconsistent. Specifically, his condition allowing voters to have unrestricted preference orders (citizen sovereignty or CS) is inconsistent with the idea that a majority preference of B over A would somehow be reversed by the introduction of some alternative C.

The Independence of Irrelevant Alternatives (IIA) condition seems reasonable at first glance. Yet there is a real question about what is meant by irrelevant. Ignoring ties, there are four possible states for an electorate preferring B over A created by the introduction of C, and at least two of them are highly relevant to the collective choice.

• When both B and A are preferred over C, there is no problem, and B remains the collective choice.
• When B is preferred over C and C is preferred over A, there is still no problem because B remains the collective choice.
• When C is preferred over both A and B, it can be argued that C is anything but irrelevant, and the collective choice should now be C.
• When A is preferred over C and C is preferred over B, the introduction of C creates a majority cycle. This is also extremely relevant because the cycle creates something like an equivalence class. This is a problem because the selection of any one of the three alternatives requires the SWF to reverse a known majority preference.

It is easy to come up with preference profiles that result in majority cycles, such as this one:

|-----------------+-------+-------+---------------+-------+-------|
| voter region    | order | count | voter region  | order | count |
|-----------------+-------+-------+---------------+-------+-------|
| left top        | A>B>C |    15 | right top (B) | B>A>C |    19 |
| left middle (A) | A>C>B |    18 | right middle  | B>C>A |    15 |
| left bottom (C) | C>A>B |    15 | right bottom  | C>B>A |    18 |
|-----------------+-------+-------+---------------+-------+-------|

In this preference profile A is preferred over C by 52 to 48, C is preferred over B by 51 to 49, and B is preferred over A by 52 to 48. But just coming up with a preference profile doesn't really help us understand why the voters would have these preferences, or how common they might be in a real election.

To get a better sense of this type of cycle, it helps to put the election in some type of context. In this case, consider a group of 100 voters in a small community trying to determine a location for a common resource such as a cell tower. Each voter has an ideal location (gray dot) and their utility of the resource is inversely proportional to the distance the single tower will be from their ideal. The locations are indicated by labeled locations (black dots) A, B, and C. In the two images, the voters are positioned identically, and the indifference lines (perpendicular bisectors between each pair of alternatives) are also identical. The only difference is the alternatives in the right image have been moved closer to the center.

The first point I want to drive home here is that majority cycles are real and are the result of real voter utilities for the specific alternatives in the cycle. The voting method is irrelevant to the existence of cycles. When there is a cycle, the use of a cardinal voting method does not magically make that cycle go away.

The second point I'd like to make is that no voting system that tries to do everything in one or two rounds can pick a good collective alternative when all the alternatives are collectively bad. The differences between the alternatives in the left image and those in the right image are what drove me to work on iterative voting systems.

Hi All,

I've just finished enough of a new website SerialApprovalVoteElection.org that I would like some direction on what should go in next. It highlights SAVE (obviously) but also is meant to show how one might use a framework to look at different voting systems and how that could resolve various debates about voting systems.

At the moment the site has eight simulations: tournaments, preference profiles, spatial models, an IRV exploration using Voronoi diagrams, a simple introduction to SAVE, a model for exploring different voting metrics used by a voter to rank motions, a fairly complicated look at aggregating voter preferences to find the best possible outcome for an electorate, and a more complete SAVE explorable (Explorable Explanations) showing how voter actions drive the result.

This all grew out of the difficulty of explaining SAVE without visuals. I think the folks here were interested in SAVE in the abstract, but couldn't really get a sense of it. I'm now trying again, only this time I'm giving you all something to play with.

The site is far from complete, but I'm feeling more and more pressure to get something out where it can be explored by others.

If you can, please take a look. And let me (and everyone else) know what you think in this thread. At the very least, it points to a way of comparing different voting systems in an automated fashion, with a real "best outcome" for comparison.

As an aside, I originally intended to put a number of other voting systems up for comparison, but I wanted to focus on SAVE and comparing results from an iterative system like SAVE to any single-round system just isn't fair. The architecture is still capable of doing the comparisons, but that isn't my highest priority.

So please have a look, maybe enjoy it a bit, and let me know what you think. Thanks!

Best Wishes,

    --tec

@cfrank Thanks for the comments. I certainly agree we need dialog, and I also think there is a hunger for something different. Whenever I see or hear about protests with the chant "Tell me what democracy looks like --- This is what democracy looks like." I feel like they have it wrong. What is going on isn't democracy, but rather an expression of a desire, a need, a hunger for democracy.

Since the 60's, there have been calls for structural change in our government, something more than just a change in leadership or representation. I developed SAVE with that in mind. SAVE is (or is supposed to be) a powerful method for making collective choices. It is designed to take input from everyone and come up with something we all can live with. It is, or could be, a way for we, the governed to tell our government what we want it to accomplish.

And our message isn't just directed at our government; it is directed at all of us. It becomes, or could become our own internal polylogue (monologue = 1, dialog = 2, polylogue = many) where we determine what it is we collectively want, and what we will consent to.

The way I see it, SAVE, if it really does what I designed it to do, is a tool to fill an unmet need. The start of the public phase of this project starts here, with other knowledgeable folks looking at SAVE and confirming (or refuting) that it addresses the issues that Arrow and Gibbard / Satterthwaite raised about voting systems, that this system will end with an outcome that if not completely ideal, is actually pretty close to it. And once the process is understood, it can spread to wherever groups want to make collective decisions.

From a voter's perspective, in the first round the question is "Which, if any, of these motions do I approve?" As this is AV, you can vote for as many items as you wish, and there is no penalty for honesty. With the second round, a focused approval vote, the question is different: Which, if any, of these motions are better than the focus motion? And, should we stop with the current focus motion as the final winner? Then, at the end when we have our final winner, we are back to a variant of the first question. Which of these motions do I support?. Added to this set of questions, we have the situation that whenever the focus repeats we can propose new motions that can provide a different take on possible solutions, introduce new problems to consider, or try to find a more acceptable motion to resolve our differences.

I'm at a conference this week so will be delayed in any responses, and I also just updated my site with some fixes and new features. Please be patient with my lack of immediate response. I am paying attention, and will respond. Thank you very much for providing feedback and opening this dialog.

Warren D. Smith's Range Voting paper (2000) and his C code simulation goes over a number of different voting systems and simulates both honest and "rational" ballots for many of the voting systems he covers. The 30 voting system and strategy permutations in the paper are: (In the paper, WDS starts his list with zero and the "as 25" strategy reference is "as 26")

1. Honest range voting (scaled utility vote)
2. Honest Borda
3. Honest Condorcet Least-Reversal (CLR)
4. Honest Coombs STV (most least-liked candid eliminated each round)
5. Honest Hare Single Transferable Vote STV (least most-liked canddt eliminated)
6. Honest Copeland (win most pairwise elections)
7. Honest Dabagh point-and-a-half
8. Honest Black (if no Condorcet winner use Borda)
9. Honest Bucklin
10. Honest plurality+runoff for 2 top finishers
11. Honest plurality (1 vote for max-util canddt)
12. Honest bullet (1 vote against min-util cand)
13. Majority vote on random candidate pair
14. Random "dictator" voter dictates winner
15. Random winner
16. Worst-summed-utility winner
17. Honest approval (threshhold=avg canddt utility)
18. Strategic range/approval (average of 2 frontrunner utils as thresh)
19. Rational range/approval (threshhold=moving avg)
20. Rational plurality (vote for 1 of 2 frontrnnrs)
21. Strategic Borda I (1 frontrunner top, 1 bottom, rest recursively)
22. Rational bullet (vote against 1 of 2 frontrnnrs)
23. Strategic CLR (strat same as 25)
24. Strategic Hare STV (strat same as 25)
25. Rational Borda (1 frontrunner max, 1 min, rest using moving avg to decide if max or min vote)
26. Strategic Coombs STV (strat same as 25)
27. Strategic Borda II (1 frontrunner max, 1 min vote, rest honest)
28. Rational Dabagh point-and-a-half (moving avg)
29. Strategic Copeland (strat same as 25)
30. Strategic Black (same strat as 25)

(A Google search for rangevote.pdf will get you to the paper. His c code is at: https://www.rangevoting.org/WarrenSmithPages/homepage/votetest2.c)

I hope this provides some good options for your strategy simulations.

--tec

@BlainCellars the code I use is what I cobbled together since I started looking at voting systems and trying to figure out what worked best. Much of my work was done without consulting the (somewhat vast) literature on voting theory, and the tools and processes I used to help me think about things are all either free/open source or my own coding. This specific graphic was originally developed using Gnu Emacs with the svg package. Since the discussion code on this forum does not accept svg files, I used inkscape to convert the svg image into a png file.

More generally, when I started looking into the problem of voting I started with NetLogo because it provides an easy graphic display for agent-based models. Using NL, I could create and display 100 voter-agents randomly distributed in a 2D field, then create some alternatives for them to vote for under whatever system I was looking at. It worked great for a while, but as my questions got more involved, and my code-base became more extensive I realized I needed something a little more robust. I switched to Python because it allowed me to write tests for my code and make it more reliable.

At the moment my code is not well documented and is in need of refactoring. So while I can get it do to a number of interesting things I'm not sure how useful it would be to anyone else. On the other hand, I think it might be very useful for voting theorists to have easy access to spatial models. There is a data hierarchy to the various electorate models I've seen in recent voting theory papers. Tournaments have the least amount of data and have essentially eliminated the electorate altogether. Preference profiles have an electorate but lump voters together based on preference orders without any indication of the relative strength of individual voter preferences. Lists of individual voters with randomized utilities for each alternative allow some comparisons, but are still limited in that they really can't handle the addition of alternatives. Spatial models allow just about everything and are known to cover all possible tournaments, preference profiles, and utility models.

I'm assuming you have something in mind with your inquiry about programs. What are you looking for? Are you looking for images of electorates? Or do you want an election simulation environment for running tests and evaluations? I am thinking of putting my code on GitHub, but it will take a lot of work before it is ready for someone else to use. Are you looking for an answer to a specific need? Or are you willing to be an alpha-tester for a generic voting system evaluation framework? And more generally, would people such as the folks on this forum be interested in such an evaluation framework?

(This was a bigger answer than I suspect you wanted, but perhaps it will get us further in our collective quest for better collective decisions.)

@cfrank You have brought up several valid issues, and I think I have answers for some of them.

I agree that representatives can serve to make our government more efficient. I am not convinced that they always do so.

One real-world example of representatives making government more efficient is the citizen referendum system in Switzerland. Citizens can initiate petitions on a referendum and with enough votes the referendum is included in a vote. (I don't know a lot about the details, but there is a lot of material on the web.) Historically, most referenda do not pass, but I think in some of those case it has been because a referendum has raised the visibility and priority of an issue to representatives of the government, and as a result the government proposed a more politically feasible option that was accepted by the voters. (There are up to 4 elections each year.)

The key point of the Swiss government example is that the representatives can get feedback from their constituents in fairly short order if they stray too far from the will of the people, and that feedback gives them the opportunity to correct course.

Structural issues in the USA severely limit such feedback. We, as the general population, have to write letters, make phone calls, attend meetings. While in the meantime potential and active office holders contact lobbyists to set up fund-raising events with donors in order to raise campaign funds. Money does not always decide elections, but it can certainly influence them.

One of the things SAVE can do (at least how I see it) is raise issues and provide real, solid numbers about what is important to the electorate. While using SAVE for a simple yes/no petition is severe over-engineering, that is not its purpose. SAVE allows multiple issues to be processed in parallel. Single issues can be combined with other issues either to build coalitions, or because the issues are related and part of a larger whole.

I'm beginning to think that my choice of the word 'consent' is not quite right. SAVE is a way to implement multi-party, multi-issue negotiation at scale, to determine what we want our representatives to do. It is not passively consenting to something someone else has proposed, but actively and collectively constructing something acceptable to a super-majority of the participants.

Part of SAVE will definitely involve consent as part of a review of our current collective state, as we express what we think is correct or appropriate. And part of SAVE will definitely involve denial of consent as we express what we think is incorrect or inappropriate. Yet the most important aspect of SAVE is that it can involve everyone concerned with an issue.

US politicians, whether they won or lost, have often said 'the people have spoken' when the election results are in. I strongly disagree. At best, under our current system, the people have grunted. SAVE allows us to articulate our collective will, and speak it as truth to ourselves, and to make clear to those of us who have chosen to represent us exactly what we want done.

I'm not sure I've responded to all your issues, but we can certainly iterate for a few more rounds.

@rob Your question was answered by Ken Arrow (of Arrow's impossibility theorem fame). AIT is more formally called Arrow's general possibility theorem, and is the second theorem in his Social Choice and Individual Values. His theorem 1 (Possibility Theorem for Two Alternatives) covers exactly this situation, and unlike his General Possibility theorem, the answer is affirmative. It is possible for a society to choose between two alternatives without violating any of his desirable conditions. The method to use is simple majority rule.

After Arrow published his book, Kenneth May published an article in Econometrica (1952) titled, A Set of Independent Necessary and Sufficient Conditions for Simple Majority Decision. That paper augmented Arrow's result and states that not only does simple majority rules satisfies Arrow's conditions, but every voting system that satisfies those conditions is equivalent to simple majority rule.

In essence, choosing between two alternatives is easy. The problem comes when you try to expand beyond two alternatives. That's what AIT shows and what Gibbard and Satterthwaite expanded upon.

So when limited to two alternatives simple majority decision works, and any other reasonable voting system devolves into simple majority decision when restricted to two alternatives.

@jack-waugh, the main argument against instant runoff voting (IRV) in my mind is that it seems very easy to eliminate a Condorcet winner. The story line starts with an existing plurality system with two major parties (Duverger's law). Over time, the two parties will tend to converge to the center along any issue that is in play in an election. This leads to party members not being very happy with the results even when they win. This leads to challenges to both party establishments from their extreme wings. Under plurality, this can lead to parties losing due to the spoiler effect. Some of the arguments in favor of IRV claim it solves this problem because voters can vote for their extreme wing as their first choice and the party establishment alternative as their second choice, and the implicit assumption is the extreme alternative is eliminated in the first round and support flows to the "more electable" of the parties' offerings.

The problem with this argument is that IRV is being proposed in order to allow multiple alternatives, and the more choices their are, the fewer first place votes there will be on average for any individual alternative, and the greater the likelihood that alternatives in a crowded middle will be eliminated. Once that happens, we start seeing IRV electing fringe alternatives instead of moderates, which is almost exactly what we have now with plurality.

Note that this argument is against the standard IRV in which if there is no winner, the alternative with the fewest first-place votes is eliminated. The dynamic is much better in a variation of IRV called the Coombs method, which successively eliminates the alternative with the most last-place votes. (The fact that a single alternative can have both the most first-place votes and the most last-place votes seems to have made it a hard sell, but that's a different issue.)

A different argument against IRV [that was actually presented as an argument in favor of IRV, (Bartholdi, 1991)] is that strategic voting under IRV is so complicated that it is nearly impossible to determine the best strategic vote. The author's argument is that faced with the computational difficulties of determining the best strategic vote, voters will instead choose to vote honestly. To me, this line of thinking seems unsound, and I thus take it as just one more reason to avoid IRV.

I hope this helps you in your discussions.