[EL] (Not Really) ranked-choice voting in SF

[Joseph Lorenzo Hall]

On Mon, Nov 7, 2011 at 2:40 PM, David A. Holtzman <David at holtzmanlaw.com> wrote:
>
> San Francisco’s hardware limitation of three rankings is a function of the availability of equipment and the unwillingness of elections officials to administer ballot counts conducted by hand.  Those problems, in turn, are related to political unwillingness to invest public funds in upgrading the actual machinery of democracy, compounded by regulatory barriers to private sector entry into the election equipment market.

Hmmm, this doesn't seem exactly right.  I may be misinterpreting your
words. If so, please correct me!

First, to run an IRV/RCV election using optical scanning technology,
as it currently exists in elections, each "choice" in an IRV contest
gets it's own space on the ballot (you essentially have an election
for first ranking, an election for second ranking, etc.).  This is
because current optical scanners can only reliably detect one kind of
mark (which varies by vendor but is usallay an oval, square, rectangle
or the wacky "broken arrow" scheme).  Current equipment cannot detect
hand-written ranking numbers and no vendor has certified a system that
uses a "matrix" style IRV/RCV ballot design (where each candidate is
listed as a "row" and the rankings are "columns" and there can only be
one choice per column, once per row or no choice).

I'm not sure what the unwillingness to count by hand refers to.  Of
course, each county in CA must perform the 1% manual tally per CA
Elec. Code 15360, which is a count conducted by hand (and there is
currently a statutory pilot program, on which advisory board I sit, to
develop methods much more accurate and efficient than the 1% tally).
Counting hundreds of thousands of IRV/RCV ballots with arbitrary
numbers of rankings sounds like a total nightmare.

best, Joe

--
Joseph Lorenzo Hall
Postdoctoral Research Fellow
Media, Culture and Communication
New York University
https://josephhall.org/