Election Integrity
A count is real because you checked it, or because someone you do not trust checked it and you could reproduce their work. Not because the people who ran the election said so.
That is the entire standard here, and it is deliberately indifferent to who won.
Prevention beats trust
Every "trust us" election has the same defect: the assurance comes from the party with the most to gain from it. Prevention moves the guarantee out of the promise and into the technology:
- Open-source voting technology — the code that counts votes is readable by anyone, including the side that loses.
- Records designed to be reproduced, not merely stored.
- No component whose correctness rests on the operator's good character.
An honest operator loses nothing under these rules. Only a dishonest one does.
The guarantee test
The test this area is built around, stated so it can fail:
An outside team, using only the public record, reproduces the official result.
Not "observes the process". Not "is briefed". Reproduces — independently, from published data, with no special access and no permission. An election that cannot pass that test has not been shown to be wrong; it has been shown to be unverifiable, which is a separate and serious finding.
Routine audits, and gaps recorded in public
- Audits are routine, not a concession. An audit that only happens when someone complains is a political event; an audit that always happens is a control.
- Gaps are recorded openly rather than argued away. Where a jurisdiction cannot produce what reproduction requires, that is written down as an open gap and stays written down until it is closed.
- Both feed the same catalog everything else does — Problems — and the same public data at Open Data.
What a citizen can verify without permission
The point of publishing the method is that a citizen is not dependent on us either. With public data and open code, a person can check their own jurisdiction's numbers, compare published totals against published detail, and see for themselves where the record runs out.
What this movement will never do
Certify who won any particular election. Not now, not later. The moment a citizen platform starts declaring winners it becomes one more party asking to be trusted, and it forfeits the only thing that made it useful. We publish the method, the data, and the gaps. The conclusion is the citizen's.
Pages in this area
- Open-source voting technology — move the guarantee out of the promise and into the technology.
- Reproduce the result — the guarantee test, and why unverifiable is not the same as stolen.
- Audits and open gaps — routine audits are a control; on-demand audits are a political event.
Where to go next
- Influencers — the area before this one.
- No Forever Wars, No Draft — the area after this one.