Thursday, March 26, 2009

Children Should Vote, Too, in Ranked-Choice Elections !

I originally wrote this entry on September 8, 2004, and published it on

Here, I'd like to discuss a proposal for a possible forum project on a voting system that could be made available to school children to experiment with voting which includes a certain degree of formality and complexity. It will not only introduce them to voting on the web but also to some of the less-than-apparent issues in our current voting system.

We're approaching a national election in November, and a number of people have been writing about voting systems on and elsewhere, including at

For example, Professor Lawrence Lessig has pointed to ranked-choice voting in San Francisco this November.

SeongBae ("Bits and pieces") has written about the use of SunRay machines as voting tools. Azeem Jiva has echoed SeongBae's idea.

On my part, I've been writing about ranked-choice voting and the mathematics of elections and like SeongBae and Azeem Jiva have been thinking about possible ways to host a project for bringing Sun technology into the voting foray.

My suggestion is slightly different from Azeema Jiva's and SeongBae's but I think it can complement theirs.

Basically, what I'd like to do is to help some people to start a forum project that would focus on bringing Condercet or Ranked-Choice voting to schools across the nation, as a Web Service which can host both pre-configured and configurable elections.

[The pre-configured one would, for example, be a national election, say a presidential election. The configurable ones are minor elections held for various purposes. More on this below.]

In other words, this is an almost-real voting system that allows school children to participate in all national votes and educates them regarding ranked-choice and Condercet voting.

What is interesting about this project is that some server-side computation will be required to resolve voting result ambiguities.

On the client side, we could use Java Card, we could distribute voting numbers to each school principle who requests it, we could use SunRays or whatever else to make it practical. The most practical approach may be the distribution to school administrators of unique voting numbers and "ballots" which could be entered by teachers or students who own them through the web.

This could be a very exciting project because it brings together a great deal of ready-to-go technology, allows for a twist on the voting system by its inclusion of ranked-choice / Condercet and has a great deal of social value which could motivate contributors. Finally, multiple people from various companies can collobrate, to implement and host the web service system on the server-side, possibly in different geos if it involves competing companies.

I think the first version of the application could be decently implemented by two to three engineers in less than three months.

Furthermore, the Web Service which will be hosting this voting system, will be available for people to configure or create votes of their own upon registration. It could even host a voting service for mobile devices. (Even silly votes, for example on who should lead a soccer team or be the captain, etc., could be hosted and carried out through any web-ready device.)

Finally, I'd be surprised if such a system doesn't already exist! If it doesn't, it should be started as a forum project.

No comments: