[EM] hmmm. Maybe I missed something before SF passed IRV then called it RCV?
robert bristow-johnson
rbj at audioimagination.com
Mon Jul 3 10:05:08 PDT 2017
---------------------------- Original Message ----------------------------
Subject: Re: [EM] hmmm. Maybe I missed something before SF passed IRV then called it RCV?
From: "Ken B" <kbearman at isd.net>
Date: Mon, July 3, 2017 12:22 pm
To: election-methods at lists.electorama.com
--------------------------------------------------------------------------
> [KB] This is a more complete description off our process in
> Minneapolis. After the poll (precinct) closes, we
> - transmit electronically the ballot counter's data to Elections HQ.
this transmission is not transparent. and its security might be questioned. what if someone had intercepted or "spoofed" your transmission and sent "creatively modified" data to
HQ?
> - print copies of the precinct totals of first choices for every race
that's not enough information third parties and watchdogs to recheck the outcomes. you need precinct subtotals of *every* possible way to mark the ballot for that race.
> and post one in the precinct entrance for anyone to inspect.
> - remove the thumb drive with its stored data from the machine and
> seal it in its own envelope.
> - remove all the ballots from the counter and seal them in boxes.
> - complete reports on various activities in the precinct during the
> voting day (7 am-8 pm).
>
> After we've completed all our paper reports and sealed all the other
> precinct records, two election judges -- including at least one of the
> Head Judge or the Asst. Head Judge -- drive the records (including the
> thumb drive) and ballots to one of two locations where HQ staff check in
> everything against a checklist of precinct requirements.
what if that thumb drive was fudged by someone on the way?
>
> HQ has the electronically-sent precinct ballot images and vote totals
> for each race; the ballot images and vote totals on the precinct ballot
> counter's thumb drive; and the paper ballots the counter processed.
>
> A recount would use the paper ballots.
as a recount *should*. that's the whole point behind paper ballots and optical-scan voting machines. so that an election has a natural "paper trail" to recount or for some other scrutiny. (the other important point is so
that, on each ballot, the candidate's name and the voter's mark is on the same physical instrument. this is *not* the case with those punch-card "butterfly ballots". if a ballot was misaligned in the jig in the voting booth, you might think you're punching for Gore and you end
up punching Bush and there is no way anyone will ever be able to tell, even in a recount.)
now what do they do with those paper ballots if there is a recount of IRV? how are the ballots transferred to different "piles" as candidates are eliminated according to IRV rules?
for a city the size of Minneapolis, that seems unwieldy.
Ken, i am not saying that your election judges are shady characters. but your election does not have the transparency it had before IRV. when Mpls was FPTP, you could, at each precinct location, count and
report locally each salient subtotal. with the IRV, the data goes into a virtual black box (or more like a "black tube"), is opaquely transported to the "HQ", and is not detangled anywhere except at the "HQ". but that is where some nefarious doings by one or
two persons can swing an entire election. it's much harder to do that when the counting is decentralized and diffused throughout all of the precincts. if the subtotals are made public at each precinct location and knowledge of all of the precinct subtotals is sufficient to compute the
outcome of the election, there is no way that just a couple of hackers can swing the election.
this is why precinct-summability is a salient feature of some voting systems.
--
r b-j rbj at audioimagination.com
"Imagination is more important than knowledge."
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20170703/0fbc218a/attachment.htm>
More information about the Election-Methods
mailing list