http://digitaldemocracy.freeforums.org/

• People’s bills : The top 6 petitioned legislative proposals would get to a second reading in Parliament.
• People’s veto : If more than 1 million people petition against a law in 60 days, there would be a referendum on that law.

Lib Dem Motions (pdf), page 11 (discussion here).

Whilst this is a great step forward, my big issue it does not protect issues that are voted on in Parliament, but aren’t laws. The whole Iraq war is a case in point, no matter what stance you took on it, there was a lot of opposition to the war which wasn’t really reflected in the division in Parliament.

One issue with vetoed taking place after the law comes into effect, is what happens to people charged under those laws if they are then vetoed. Making the public vote take place before the law comes into effect is a far more sensible mechanism for all but emergency legislation.

There is also no mention of the other pillar of direct democracy, Recall Elections.

To achieve this we need two items, at the minimum

1. A Unique Vote identifier, that can cross voting system boundaries (think Welsh and Scottish Parliament)
2. Their vote recommendation

The unique vote identifier would best be implemented as a permalink. E.G. “http://www.directdemocracy.org/resolutions/425″

Representing the vote is slightly made more complex by the need for multiple options. Personally the Schwartz method would be used to tally these votes and has seen adoption in many open source bodies. Wikimedia, Debian, Gentoo.

Currently my line of thinking is:
<?xml version="1.0" encoding="utf-8" ?>
<rss version="2.0" xmlns:dd="http://www.directdemocracy.org">
<channel>
<title>An Organisation</title>
<description>Vote Recommendations by the Organisation</description>
<link>http://www.theorganisation.org</link>
<item>
<title>Bill 256 will destroy the universe</title>
<link>http://www.theorganisation.org/opinions/256</link>
<description>Bill 256 is bad for the Environment and bad for business</description>
<dd:vote id="http://www.directdemocracy.org/resolution/256">
<dd:recommendation id="6" sequence="2" />
<dd:recommendation id="3" sequence="1" />
</dd:vote>
</item>
</channel>
</rss>

(Sorry about the nasty formatting)

This would indicate that they wanted to vote for option 3 first, then option 6, and not vote for any of the others.

A similar system could also be used by the main voting software to enumerate the currently open resolutions including additional tags. The validity of the document can be ensured by using XML-Signatures to sign the document.

One nice addition, is that by using RSS, any feed reading software could display it, great for when your on the move.

How can you mix being able to make sure your proxy is someone you can trust to vote with your ideals in mind, while at the same time ensure the secret ballot to all?

On a large scale you probably can, by placing legal safeguards to separate you as a person from you as a voter as the current UK system does. But this doesn’t really stop small scale problems such as inside a family “Your under my roof, you’ll proxy your vote my way” kind of attitude. Even if legal safeguards were in place I doubt that it wouldn’t be rife.

I doubt any system would be acceptable to the whole of society without the secret ballot intact, the fight for the secret ballot took 34 years since proposal by the Chartists to actual implementation. So is there a system that retains the proxy element of liquid democracy and still allow the secret ballot to exist?

One method talked about in the kuro5hin article is separating the proxy element from the direct democracy element. Users would subscribe to feeds for each ‘party’ and be able to manually or automatically choose their vote based on those feeds. The vote weighting system could be more tuned to the users way of thinking than a simple proxy, they could take the majority view from those ‘parties’ they are subscribed to. You could also still delegate your vote to your local MP (if you so desired) by subscribing to their feed. Since nobody has named this varient of liquid democracy, I’ve nicknamed it Recommended Direct Democracy. But it really needs a snappier name.

As well as just a yah, or nay vote the feed could have a fuller analysis of why they feel that their position is the right one. There wouldn’t be any limits on who could setup a feed, all the user would have to do would enter in a URL for the feed and it’d be subscribed. There would be no systematic way of finding if a feed was used to decide the outcome of an individuals vote. But it might be useful to track the statistics if the user voted manually or automatically.

From an implementation point of view it is also easier and more scalable as the system no longer has to hunt through data structures following proxies. If it was implemented as a distributed system it would be much simpler as the system only has to be presented with a block of data instead of having to constantly ask for the proxies results. To actually implement their proxy vote, the users software (or the web server) would do that proxying for them if they decided not to vote.

From the point of view of the secret ballot, the current system separating the user id and the user identity could be used. If there was a case of voting fraud the user id and real identity can still be matched up, but the requirements to access the real identity would have to be regulated carefully. From my point of view the only time when a persons voting record should become public is when they are elected or up for election to a political body, as the voters have a right to know where they stand.

When you aren’t interested in a topic you can delegate it to someone you feel will best represent your ideals, but when you are interested in a topic you can vote your own mind. If a person never votes and their proxy is their local MP, this is nearly equivalent to representative democracy. This mechanism would allows any person to take as much or as little part in politics as they so desire.

In terms of delegating to other voters, you would rank your proxies in list of preference. If everyone had 3 or 4 delegates then the likelihood of those delegates not voting is substantially reduced.

For example, Alice decides not to vote, but her proxies include Bob, Charlie and Dave (in that order). As Bob has the priority, his vote is used. Additional Examples will be discussed in subsequent posts.

There are substantial issues to resolve before any Liquid Democracy system could become widespread system.

• How do you ensure that the person voting is the person they say they are. Obviously public keys will take a part in this, but making it effective and usable by the end user is a challenge.
• How do you ensure the availability and scalbility of the system that could potentially be counting 60 million (UK), 305 million (USA), 1 billion (India) or even a whole 9 billion (World) votes, with multiple votes taking place at any one time (obviously a distributed system)
• How do you ensure that the system is secure against the almost certain hacks that will be attempted on it.
• How do you manage the secret vote alongside with the right to proxy. Should the status of proxying be delegated, and who the person delegated to be visible? What legal protections need to be in place to protect the individuals voting record. Should a realtime display of the current vote result be available?
• How do you allow individuals without an Internet connection to take part in the system.
• How to resolve some of the nuances of the proxy algorithm (Re-entrant voting, what the priority is in terms of depth where multiple proxies are involved)
• How to protect minorities and how to counteract NIMBYism.
• How to get it from a theoretical idea into a deployed system. How could it fit with systems already in place?
• Would voter fatigue cripple the system?
• What about Preferential Voting systems?
• How can you delegate different individual for different areas of voting?

These topics and additional ones that crop up are the main subject to be explored with this blog.

With this system, democracy could scale above the possibilities of direct democracy, a person representing the people of a certain area could then travel to the parliament without any real time delay in voting.

One side effect of representative democracy is political parties. The number of political parties depending very much on the voting system used for elections, First Past the Post tend to have two parties (or with a third minor party), and of course the party in office does not want to alter this status quo.

In the past political parties have stood for something, but after the big arguments have been settled (e.g. Welfare State, Public/Private ownership of companies), parties tend to move towards the centre ground to appeal to all voters. But having little to separate the parties realistically means that the parties are only out there to get into office and leads to voters being disenfranchised in effect.

Only being able to vote for a representative of a political party causes every voter to choose the lesser of n evils. Nobody agrees with the whole policy document and votes for the closest party to their beliefs. But then the elected representatives use these votes as a mechanism to get unpopular ideas into law by claiming they are just doing the will of the people.

And so we come to the point of this blog, what alternatives are there? From my point of view the best way forward is to take the best of direct democracy, the best of representative democracy add a dash of technology and create a compromise system. Liquid Democracy.

However Direct Democracy does not scale well. With a state the size of Athens, it was feasible to hold debates and votes with all members of the citizenry. However past this size, given the lack of communications technology and computers, it was infeasible for mass real-time discussions on laws
and not practical way to count vote on a regular basis in a short span of time.

To bypass these problems, the form that all democratic countries currently use is Representative Democracy, in one form or another.