[specialist]

Forgive me, I don't know what you're advocating for here. I greatly appreciate that you're chewing on the policy, problem, tech.

For the USA, I no longer think polls are useful predictors, nor are exit polls useful audits or verification. For polling to work (be useful), like is done in Germany, requires the whole system to be designed as such.

Our gold standard is the Australian Ballot. Private voting, public counting. We weaken this system to extend the franchise, eg absentee ballots.

But the real kicker is our FPTP (winner takes all) election system. It's so brittle. The inevitable error rate intrinsic in any form of voting (casting ballots) coupled with durvergers law virtually ensures drama.

Said another way, my militant defense of the Australian Ballot, this recurring national spaz attack, would be mooted by switching to a more robust form of elections. Ranked choice and proportional representation have the most interest and support, though I prefer Approval Voting for a better balance of fairness and simplicity.

Back to your point about disclosure, if I follow you: I very much would like to see tech, POCs, research into time boxed privacy. Like maybe all election materials, including ballots, are released when an election is certified. Versus hidden and then destroyed a few weeks later. My motivation is to find other balances, equilibria, between people's privacy (and protection) against society's need for confidence in the results.

[mjevans]

I agree with proportional representation applied within a bounding area of authority and letting representation by __area__ and/or __area of interest__ be by decided by the vote rather than distracting (which is subject to jerry mandering).

I prefer the "Path Voting" method (which I have to google for with wikipedia path voting every time, since I can't remember how to spell the inventor's name). https://en.wikipedia.org/wiki/Schulze_method

"The Schulze method (/ˈʃʊltsə/) is an electoral system developed in 1997 by Markus Schulze that selects a single winner using votes that express preferences. The method can also be used to create a sorted list of winners. The Schulze method is also known as Schwartz Sequential dropping (SSD), cloneproof Schwartz sequential dropping (CSSD), the beatpath method, beatpath winner, path voting, and path winner."

[ClayShentrup]

http://scorevoting.net/rangeVcond.html

[Ill_ban_myself]

From one side of your mouth you're demanding perfect privacy and from the other side I'm sure you'll tell me its necessary because of the laws in the US for voting procedure and policy to be bottom up starting with localities, then states, then federal elections.

I think we both know perfect privacy is cost prohibitive at the local and state levels and largely a talking point that ends with the listener assuming it is impossible to achieve online voting with perfect privacy in our lifetime.

If we can't afford to change the way people vote then why not change the way we measure it so its not so cost prohibitive?

Between exit polls, aggregate voter data held by the national parties, private and public data sources it is already possible to accurately predict the outcome of most elections.

Its already being used by national parties as the foundation for when and where they choose to challenge election results and allocate resources.

There is no such thing as a, "Secret Ballot" with modern analytics. There are just those who can afford to find out how you vote and those who can't.

Those who can't in today's world includes regulators and the government themselves.

[specialist]

Well then.

I am explicitly not shutting you down. I shared my views and why. I encouraged you to do the same.

For instance, one cheap and obvious way to improve uitilityof exit polling would be to implement compulsory voting.

Also, please work on a campaign. To the best of my knowledge, modern campaigns don't rely much on polling. Most effort is put into GOTV (voter identification and balloting chasing).

Edit: Oops. I'm not familiar with Republican campaigns. In my area, their GOTV is less potent and they rely far more on advertising. So they might still be more reliant on polling. It's a good question for me to follow up.

Opinion polls are still useful in other ways. Depending on who's paying. Sanity checks. Push polling. Message crafting. Talking points, a la Voted San Fran's Favorite Pho Restaurant. Feeding corporate media's horserace narrative. Policy groups trying to gather intel on both opponents and allies. Consultants fleecing noob candidates. Arm waving because old school operators expect it.

YMMV.

[Ill_ban_myself]

KOBE! https://www.youtube.com/watch?v=tcMe2H3mIKk

[Ill_ban_myself]

At the danger of DOX'ing myself I'll just say I was a hair's breadth from closing the loop on full online voter registration in 2008 but apparently local offices are not REQUIRED to accept faxed registrations and might FORGET to change the toner or turn them on.

This is not my first brush with radical voting ideas and I'm not afraid to put them into practice when legal signs off :-p

These days I settle for election judging. Isn't that a scary thought?

You should google project Houdini, narwhal, and orca