The largest difficulty in implementing adaptive democracy is enabling every voter to update any of their votes at any time. In private organizations like member cooperatives that can simply assume all members can vote through the internet, and where the stakes aren't quite as high, this isn't a serious difficulty. But for a government with citizens spread across a large geographic area, it becomes quite challenging.
It should be immediately obvious all the voting methods I've described in this book absolutely must be counted and stored digitally. The idea of election officials having to manually validate and track a adaptively incoming stream of paper Quadratic Range ballots and Democratic Weights udpates is just ridiculous. It's tempting then to simply perform all voting over the internet, and this would be more reasonable with a broadband universal service obligation. However there is one very good reason not to make this decision even if an internet voting system could be made available, accessible, and secure.
Quadratic voting systems of all kinds count votes differently depending on how many individuals they came from, and votes from more people can add up to the same total vote much more cheaply. The incentive to somehow gain access to extra fraudulent identities is therefore much stronger.
However it's also true voter identification laws have historically been used to discriminate. By unfairly disadvantaging those with fewer resources, less time, a longer distance to travel to licensing offices, missing documents, or documents made less carefully by previous institutions, a class of people can essentially be denied their right to vote. If we're going to universally require verified identity, we have to make it possible and tractable for everyone.
The obvious first choice is to make acquiring identification completely free. But to address the bulk of the remaining issues we need something more comprehensive.
The country is already dotted by a network of postal offices, a necessity since the United States Postal Service is required by law to serve every address in the nation. This network could be fairly easily adapted to also house small voting offices. These offices would be open all year at broad but reasonable hours depending on the location. They would have computer voting machines, and to ensure their software was the most secure and intuitively designed available, their software would be open source and selected by democratically elected officials. These voting machines could be used to view and update all a voter's democratic weight allocations and any accompanying ballots, print paper manifests of current weights and ballots to act as a receipt, and view all available elections.
Each office would have one or more voting clerks, democratically elected or appointed by someone who is. These clerks' main responsibility would be to act as a "proctor" or witness for the office. But they would also help voters who don't understand the voting machines by printing out their manifests, giving them educational materials about how the voting systems work, and providing scannable paper ballots if requested. These offices would also act as identification bureaus, making the tractability of voting and acquiring identification equivalent.
To properly serve those who are incapacitated or live very far away, we could provide a "mail lifeline" to interact with these voting/postal offices. A voter could send letter requests for paper manifests and ballots, or requests for a group of voting clerks to visit them at their residence to give them identification.
I've already advocated for residency granting citizenship, so the only requirements to gain identification would be to demonstrate residence, give basic information about yourself, and have a photo taken. To demonstrate residence these offices could mail verification codes to someone's stated address and have them bring the code into the office, or a group of voting clerks could pick up the code when they visit.
Verifying uniqueness of identity is a more difficult problem. This is an area of active research, and will likely take a much more complicated mechanism to robustly enforce. The best we can do without such a mechanism is to make it merely more annoying and costly to acquire and use more than one identity. If voting clerks give each voter a one-time code after checking their identification, the voter would have to come in on a separate occasion to actually use a non-unique identity. And if people can make accusations of duplicate identity to be resolved through a trial mechanism, then at least acquiring and using duplicate identities would carry some risk. These methods would likely be sufficient if most voting happens in person.
It's important to point out, any logistical hurdles someone might face while trying to vote are a much less serious problem in a Adaptive Democracy. Our existing elections present rigid date windows during which voters must clear any logistical hurdles or risk losing the chance to participate.
But in a Adaptive Democracy they can gradually clear these hurdles and thereafter fit voting trips into their schedule as their life unfolds. The elections they wish to participate in will still be there once they're able.
Voting offices would of course be tax-funded with the purpose of serving an essential public good, so they wouldn't be required to sustain themselves financially. However if we can make them more efficient without interfering with their purpose, we should be eager to do so.
The best ways to reduce these costs are with increased digital automation, and consolidation of more "in-person" governmental functions into these offices. Driver licensing, other permitting bureaus, courthouses, libraries, and community meeting spaces could all be examples.