Re: Hand Recounts of votes recorded on DREs

[Paul E Condon <pecondon@xxxxxxxxxxxxxxxx>]

On Tue, Dec 07, 2004 at 12:18:10AM -0800, Arthur Keller wrote:
> At 11:14 PM -0800 12/6/04, David Aragon wrote:
> >Barbara Simons wrote:
> >
> >> A "hand recount" of paperless DRE votes should come up with
> >> the same tally that the DRE originally produced, unless the
> >> DREs are really really broken.
> >
> >Clearly it can't detect errors in the recording of votes to
> >produce the DRE output.  But it can detect errors in the
> >tallying of those outputs downstream of the DRE's -- which is
> >certainly an area of concern and potential source of errors.
> 
> It can also detect errors in the tallying of votes WITHIN the DREs.
> 
> >So it's not pointless.  But neither is it the end-to-end
> >audit function that "hand recount", with its connotations
> >of painstaking meticulousness, has formerly meant.
> 
> Certainly it is better to have an AVVPAT for each DRE.  But let hand 
> recounts of DREs go forward with the painstaking auditing process if 
> there's a chance it's not pointless.  We don't want election 
> officials deliberately avoiding a great deal of effort by choosing 
> DREs without AVVPATs.  Be careful what you wish for, you just might 
> get it.

Please excuse a question whose answer is obvious to most everyone
on this list: What does AVVPAT stand for?

-- 
Paul E Condon           
pecondon@xxxxxxxxxxxxxxxx