Re: Fwd: [Votingtech] AVANTE on VVPR Implementation - A WP

From: Alan Dechert <alan_at_openvotingconsortium_dot_org>
Date: Sat Nov 29 2003 - 23:09:00 CST

Charlie,

> two comments:
> one is that it might be wise to not specify exactly how votes are
> to be counted as part of the OVC spec. ....
>
Of course we will. It is normal for vendors to play a part in setting the
rules for how elections are to be administered with their systems.

We will be introducting new technology. Our voting machine is NOT a DRE.
This is a subtle point but very important. Rules about how to count votes
from DREs are not directly applicable.

> States will come up with different prescriptions for what constitutes
> the official vote and OVC should be adaptable to this. ....
>
The OVC will write the rules for how the system is to be implemented.
Again, it is normal.

Look at when punch cards were introduced. (BTW, invented by a Berkeley
Prof.... I co-authored the first incarnation of the current proposal with
another Berkeley Prof., Henry Brady). The rules for administering elections
adapted to the new technology. When other technologies came along, the
rules adapted to that too.

> Even HAVA and HR2239 do not say that votes cannot
> be tabulated electronically, it says the paper will serve as a
> recountable manual audit trail. .......
>
You need to think again. This is a different technology. These are not
electronic votes.

> From what I've been able to find I expect the best compromise
> model that stands a chance of winning acceptance is that votes
> will be tabulated electronically; the paper ballots will serve only
> in a recount or in any mandated partial vote count to spot check
> accuracy. ....
>
You are juxtaposing current rules into the future -- a future that will be
different for lots of reasons. And these rules vary from jurisdiction to
jurisdiction. If the OVC grows as I think it will, we will eventually play
a significant role in remaking rules about election administration.

> By using an all electronic official
> primary ballot count states will save a lot of money and effort in the
> vast majority of races in which the election is not contested. Of
> course I would prefer to see votes counted only by optical scan of the
> paper ballots, but mandating this might be outside of OVCs bussniess?
>
We can't mandate anything but we can certainly influence things. It is our
business, absolutely.

> two, some states specify, in law, how electronic vote
> storage is to be accomplished. ....
>
Again, this is new technology we are introducing. Rules will change as they
have for other new technologies that have been introduced over the years.

> For example, New Mexico law says that ballots not only will
> be printed in random order when dumped from memory, but they
> will be stored in random order in memory ....
>
For DREs? Remember, our system is not a DRE. But it has been the plan all
along that when the votes get written to disk when the polls close, they
will be sorted by ballot ID (thus random order). The purpose of this is so
that no one looking at the list of ballots can figure out who it belongs to
by knowing the the order they were cast.

> ... and that each vote will be recorded three times in
> memory with each replica in a physically separate memory
> chip. pretty specific eh?
>
Again, our system is different. Most of the code on the books applies to
punch cards. A lot of the rest of it is a very mixed bag which varies from
one jurisdiction to the next. There will be legal code that is developed as
a result of our project -- lots of it.

Alan D.

==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Sun Nov 30 23:17:12 2003

This archive was generated by hypermail 2.1.8 : Sun Nov 30 2003 - 23:17:13 CST