Dear Tom,
That's really revolutionary. A real change to all election model.
The only point is how can it affect the vote response. I mean, the "election day" works as a call to action. People know there is only one day to vote and they take it in mind. If you give a wider period to vote, the call to action is less strong and it could reduce the respond rate. Additionally, nowadays there is a constant information about voting statistics: not based on real data, but in "declared vote"; a longer period would allow people to speculate with their vote. Anyway, it must be explored as an interesting option.

I had similar ideas.

In our swedish electoral system, we are allowed to send in postal votes 4-5 weeks in advance (from home or abroad). And some of the larger public libraries have got polling stations set up in the last week before the election day. So we have 3 different options to cast our vote, in case you are on a vacation trip, or working on election day.

But there is a neat little thing in this; if we change our mind during the last weeks of campaigning, we could go to polling station on the election day and vote again, and only this last vote is counted!!! (and our earlier votes strucken away from the record)!!!

It can be true that having one official election day is required to get high attendance. But I would think that it could actually increase the turnout if you were allowed to vote (by mail or online) as often as you want to. (less people that "stay on the couch" during election if they can vote whenever they can)

If you only voted once, and never changed your party allegiance or political ideology, you are automatically voting on them in all elections. You would never need to renew your vote!!!
Until you change your mind, of course, when you want to vote for someone else. (or choose to erase your vote in an active act of protest...)
I believe this could raise the actual voters to nearly 100%.

Several counting systems are of course possible: there could be only one day, when the official record is counted of all the votes during the last period of voting (similar to these election days, once every fourth year)

Or it could be a vote-count on regular intervals; every year, or four times per year if you wanted. or even more often; monthly/weekly/daily... that would be very interesting! voters could become extremely interested in checking the news constantly for any corruption or new propositons and legislation that we don't agree with ;)

The public would be influenced by daily news and any scandals, and politicians would be able to check the numbers directly, to see the effect that their work and efforts has on the support.

It could cause higher competitiveness amongst the politicians to do proper work and stop messing around with stuff that we don't want. Consumer Power? (i dont think it would become a risk of ending up with silly laws; like getting cigarettes and alcohol for free or anything like that?)

Plus on the practical side, you could rearrange the polling stations to suit certain groups of voters, according to a schedule. To offer special support and assistance for one type of disability per day of the month.

I think this concept really stands out because of two variables this solution effectively tackles.

1) Time: which clearly constrains accessibility for voters (people who are abroad, sick, have other commitments...).

2 and most importantly) Trust: giving people the chance vote during a longer period of time and to even change their vote could prove to be a very powerful tool that would foster transparency and reduce questionable campaign-focused tacticism. This would hopefully help politicians looking beyond the election date and start building strong, long-lasting relationships with engaged voters (an anti-disaffection).
As I see it, the current model is like hunting (one shot, short term, high risk), and this change would hopefully get us closer to something more like 'growing and harvesting' (long term, reliable).
The idea of doing this through trusted institutions like banks, hospitals, etc. has also great potential for trust building and presents many advantages (economies of scope) as there's already a secure and spread network that would need minimal set up.

With reference to Vicenc Alvaro's concerns below - harvesting votes early but with reviewing them for results would provide a means to count participation and could easily generate data that could spur voters and parties to greater participation and shame people living in low voter response zip codes. A key feature of the he system would be to allow a means to change/update ones vote - perhaps by invalidating barcodes of previous votes registered to a voter?

I wonder if this could impact the way people think about changing their minds on political issues; would they vote as soon as candidates were announced, on clear campaign bullet points, or would they wait and look to absorb more campaign information/misinformation down the line?

I would also say that a No Vote should only be allowed as a half vote towards a party - to provide a means to criticize the party who should have served your interests. To vote but not vote would be sitting on the fence & the message would be lost between the two extremes and not empower people in a party to fight for change. Feedback is an important part of all systems.

Great input. This concept touches on some of the data points we gathered during the initial phase of our Voting Systems Assessment Project (VSAP) through voter surveys and focus groups. It would be interesting to delve further into how the length of the voting period would be defined and how voters would "finalize" their ballot and then be enabled to track the receipt and processing.

Provocative thoughts, Tom! We're looking forward to collaborative discussion here. If there are comments that you would like to use to build upon your concept – you can add them using the Update Entry button on the right. Bring on the builds!