h a l f b a k e r y
A hive of inactivity
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
Electronic voting can eliminate human counting error, reduce the
number of individuals that have an opportunity to cheat, and
provide very fast results. Most systems proposed to achieve these
benefits are not fully observable from start to finish as a white-
Proposed is an electronic
election system with a keyed vote ID
assigned to every voter at every election. (This is not a voter ID
but rather gets assigned uniquely every year.) When the voter
votes, a ticket will print at the voting machine with their vote ID
(basically a random GUID). Later, they can go online and download
the list of EVERY vote, and they'll be able to match their vote ID
with the results that were actually recorded.
This vote ID will also display on the screen for the user to write
down if the printer is out of paper (because you know they'll
overwhelmingly run out of printer in the precincts that lean in
favor of the underdog, by coincidence of course).
The media will be able to download these lists to ensure the
reported totals match the individual votes.
Individual voters who wish to have a secret ballot will still get
Individual voters concerned with whether their vote actually
counted will be ensured of such.
Fake votes will be curbed because the aggregate count of the
votes can be tracked.
||I quite like the idea of blockchain administered voting
systems - both for the public at large, but also as a means
of recording the actions of the legislature. Which seems to
be sort of what you're proposing here.
||This idea might ensure that votes get counted, but it
doesn't ensure that extra votes weren't generated.
||This needs one more step. When voting, a photo of the
voter is taken. It must be verified as a clear photo by the
person manning the voting station and retaken as
necessary and certified as a good photo before voting
could take place. The ticket printed would have two
GUIDs: one associated with the photo and one associated
with the ballot. So in the end we have a two databases:
one with photos and the other with votes. You can go
online and verify your photo and your vote. The number
of photos and votes from each polling location must
match which would be enforced by the software. Votes
and photos would be stored locally then reported in
batches with the order of both lists randomized to ensure
||People looking for voter fraud could use facial recognition
software to find matches. If the number of twins is
higher than the statistical number of twins in the
population, more serious investigations can be
undertaken. Police can scan the photos for wanted
criminals, but it will be delayed, and they would only
know the precinct. If a person was caught by looking at
photos, if they destroyed their receipts there would be
no way to uncount the votes, but they could be put in jail
and fined. Alternately if a single precinct had more than
x% fraud, the entire precinct could be disqualified.
||This wouldn't absolutely prevent voter fraud, but would
make it more difficult to do on a large scale.
||Not a new idea, or even remotely unique. This is what
everyone uses as the introductory example to secure