Re: Glossary from NIST/EAC

From: David Mertz <voting-project_at_gnosis_dot_cx>
Date: Mon Apr 04 2005 - 16:57:43 CDT

On Apr 4, 2005, at 4:34 PM, Alan Dechert wrote:
>> I agree that SPB should be a separate entry from VVPB. I don't agree
>> that SPB is the most important thing.

> VVPB is not very interesting to OVC. I vote on a VVPB -- an opt scan
> ballot. Mail-in ballots could be called VVPB.

I guess it's good I'll be making suggestions. In particular, the
definitions I will propose to NIST will be that VVPB is EXACTLY what
the central goal of OVC is.

I'm not sure exactly how Alan votes (mechanistically that is; I also
don't know the content of his votes). If it's one of those SAT-style
fill-in-bubbles thing, it's not a VVPB. It *is* a paper ballot, so I
don't really mind that system. But it would not be a VVPB.

The main thing is to distinguish VVPAT (or VVAR, or VVPT, depending on
who writes it) from VVPB. I'm not sure if I should suggest Mercuri's
paper as a citation, or Keller, Mertz, Hall & Urkin. Part of it is the
authority game, and I think there's an ACM cite for Mercuri.

Basically, the definition is something like:

Voter Verified Paper Ballot (VVPB): A human-readable Voted Ballot
produced with the aid of an Electronic Ballot Printer. In contrast to
a Voter Verified Audit Record which is generally treated as a secondary
safeguard against failures in electronic records, a VVPB is the
considered primary in identifying a voter's intention.
See also Voted Ballot, Voter Verified Audit Record, Electronic Ballot
Association: Voting
Source: ??

And then:

Summary Paper Ballot (SPB): A type of VVPB in which only affirmative
voting preferences are contained on a human-readable ballot. For
example, a SPB might contain the name of a voter's preferred candidate,
but omit the names of non-preferred candidates for typographic and
handling convenience.

Basically, a SPB might be a bit more convenient if a VVPB would run to
multiple pages for a particular election. That's the whole of it: in
*some cases*, a SPB *might be* an administrative convenience.

Actually, human factors considerations *could* come in here too. My
intuition is that voters are more likely TO accurately verify their
votes against a SPB than against a non-summary paper ballot. If so,
that's a pro for SPBs. But my intuition could be wrong; it is not
based on actual studies of test subjects examining different formats.

Most likely, different categories of common verification errors would
pertain to different layouts of VVPBs. For example, a voter is
probably more likely to overlook an omitted contest on a SPB than a
non-selection on a non-summary ballot. It's true that OVC's ideal SPBs
would contain a phrase like "No Preference Indicated" to make the
matter clear... unless, of course, the preference just wasn't printed
because of a software/hardware error! It's important to consider what
happens not just in case everything operates smoothly, but also the
routes things take when they break down.

> SPB is very important to the OVC project, specifically. Consider this
> language from SB1438:
> sb_1438_bill_20040927_chaptered.html
> (e) "Paper record copy" means an auditable document printed
> by a voter verified paper audit trail component that
> corresponds
> to the voter's electronic vote and lists the contests on the
> ballot
> and the voter's selections for those contests. A paper record
> copy is not a ballot.

This is pretty much irrelevant. SB1438 contains a discussion of a
VVPAT, not a VVPB. In other words, the thing discussed, being "not a
ballot" is logically therefore not a VVPB.

OVC discuss mailing lists
Send requests to subscribe or unsubscribe to
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Sat Apr 30 23:17:01 2005

This archive was generated by hypermail 2.1.8 : Sat Apr 30 2005 - 23:17:22 CDT