[ponderings]

I often think a full rewrite is needed.

I woke up this morning with an odd thought. User interfaces are hard, they are not many but we have people who are really good at designing those, we also know how to user test them. Election programs are non binding, you can say one thing then do the exact opposite after winning. (Referenda are also complicated.) It doesn't seem to make sense in the modern age to be able to change your vote every x years. It's a great formula for [say] the 17th century.

What if we designed a really neat configuration page for the government and allow people to change whatever settings exposed to them. You check a candidate you like and with each option their choice is the default. If you don't agree with something you simply change it.

Then, one by one we take the topics away from the politicians so that they can focus on the rest of the work.

For choosing the party you get 3 up and 3 down votes that you may spend however you like.

A separate election is held to chose the team to populate and work on the interface so that gradually more and more topics get exposed. Their job would also be to research and estimate how familiar the population is with a topic along with a budget to educate the voter.

It would be enlightening for the candidates as well.

Eventually we can shut down the legacy system and have one or more nudgeable robot overlords.

Don't worry, I'm sure the dream will fade in a few hours.

[rrrrrrrrrrrryan]

Seems like you've independently invented a form of liquid democracy:

https://en.wikipedia.org/wiki/Liquid_democracy#:~:text=The%2....

It basically lets each citizen choose between direct or representative democracy, per issue, and does away with arbitrary things like election dates and even candidates (you can "elect" any other citizen to vote on your behalf, as can they). The only reason we don't have it is because it's basically impossible without software. Even very complex ranked choice voting can be tallied manually.

[ponderings]

The interesting thing in the moment was not the formula necessarily but the idea to start with the interface.

If you are dealing with such monstrosity of an application all battles are lost at the interface level. It has to work for everyone, there is no room for excuses about dumb users, they are the target audience.

While in the US the number seems infinite, in the Netherlands we have roughly 140 000 laws that each could have a series of check boxes and sliders. Say we all do 4 per day, that would be only 1460 annually. It would take 100 years which seems to long. At 40 per day it would require to much effort.

If the interface is to work as desired a large amount of law needs to go.

We should burn the books most worthy first. Experts can compete finding the most nonsensical laws. Short videos can be made to explain why the law exists.

We assign a good number of test subjects to pick the least likeable ones until we have a good list of candidates unlikely to survive.

There must be a good feedback report of the terrible implications after a law is deleted.

I can see it already, naked people around camp fires, drinking booze in public, selling food after sundown, making music without a license, singing songs insulting the monarch.

[RandomLensman]

We also don't know if the outcomes would be preferrable. Do we need to move democracy into a high-frequency trading kind of world to do the things we want to?

[_factor]

How about you make a database of questions related to voting history. The algorithm matches the candidate whose history aligns closest with your selections.