Re: Consensus of the OVC

From: Arthur Keller <arthur_at_kellers_dot_org>
Date: Wed May 26 2004 - 17:34:51 CDT

At 1:01 PM -0700 5/26/04, Alan Dechert wrote:
>David,
>
>> When I refer to full-face voting--and I think I speak for Alan here
>> too--I don't really mean that every contest appears on one screen, no
>> matter what. Obviously, we're not going to fit 80 contests on one 17"
>> screen (nor even on a 19" screen). For purposes of this discussion,
>> there is a continuum between (A) Fitting as much as is legible on each
>> ballot screen, thereby minimizing page turns (often fitting on one
>> screen, but not always); (B) Restricting each screen to exactly one
>> contest, thereby focusing voter attention on the contest (perhaps with
>> some navigation/progress widgets on the same screen).
>>
>Good points.
>
>As I mentioned in another post (response to Jeff), we won't decide all of
>this in a vacuum. There are other bodies like NIST, EAC, and the IEEE, and
>maybe NASED that will have input in this area. The OVC will be (and to a
>certain extent, already is) involved in the process of developing standard
>ways of doing these things. I might add that if the UC proposal gets
>funded, we'll have far more talent applied to these issues than all these
>other bodies combined. NIST doesn't have any computer scientists currently
>working on this. EAC certainly has no scientists. IEEE has a small group
>(vendor dominated) in their standards committee. One of the main guys in
>the IEEE working on this is a ES&S employee. Others are mostly shills for
>existing vendors.
>
>One good question to ask--I have no idea of the answer, probably Doug Jones
>or David Jefferson may have some clues: To what extent is the on-screen
>ballot design automated with existing e-voting systems?
>
>It seems that no matter how complicated the rules are, if we can spell them
>out clearly we should be able to automate the process. Automating the on
>screen layout tends to favor simple rules (i.e., one page per contest), but
>may yield unacceptable rules (i.e., 76 pages for the 76 contest ballot--most
>of which are judicial retention contests).

I suggested putting multiple judicial retention contests on the same
screen. I think one can easily automate which contests were
compatible to be on the same screen. It could be done either with
generic rules (harder to design) or parameterization (easier to
design, but requires more work at the county level). I think
parameterization is the way to go.

How would you parameterize multi-column ballot displays? What rules
do you have, like avoid splitting contests over two columns if they
can fit in one? Can you rearrange contest to make them fit better,
or does that confuse voters?

I prefer to have a single-column vs. multi-column comparison rather
than a one-contest-per-screen vs. all-contents-on-the-same-screen
(i.e., full-face). It's a more realistic consideration that gets to
the crux of the matter.

All this aside, does it make sense to give the voter a single-column
view of the ballot, with the voter having the option to select a
multi-column view? I think the simpler-to-use interface should be
the default, and I would argue that single-column is easier to use,
albeit at the cost of being more tedious.

Best regards,
Arthur

-- 
-------------------------------------------------------------------------------
Arthur M. Keller, Ph.D., 3881 Corina Way, Palo Alto, CA  94303-4507
tel +1(650)424-0202, fax +1(650)424-0424
==================================================================
= The content of this message, with the exception of any external 
= quotations under fair use, are released to the Public Domain    
==================================================================
Received on Mon May 31 23:18:07 2004

This archive was generated by hypermail 2.1.8 : Mon May 31 2004 - 23:18:17 CDT