Re: Comments on Voting Project

From: Alan Dechert <adechert_at_earthlink_dot_net>
Date: Sun Aug 17 2003 - 09:31:30 CDT

Arthur wrote:
> I think it is important for each of us to understand the nature of
> intellectual property on this project, the future direction we
> contemplate, and whether any of us intends to make money off of this
> effort. Those of us who volunteer need to know if their efforts will
> go to line someone else's pocket.
>
I would like to try to clarify a few things along these lines. I fear I may
have contributed to the confusion by the use of the word "free" in the
announcement about the project. "Free" voting system, in this context, was
meant to mean more like "open" or "freely available." It did not mean "no
cost."

I am proposing to swap the voting system in the U.S. for a new one. The old
one costs money. The new one will cost money too, although we're expecting
it to be much cheaper, more efficient, more reliable, etc.

It costs roughly one billion dollars per year to administer elections in the
U.S. (this includes all costs -- the budgets of all the organizations
involved -- this figure is from a phone conversation with Ansolabehere of
MIT). Of the one billion, something like $200 million goes to election
vendors -- companies that provide equipment and services to counties.

The Help American Vote Act (HAVA) is slated to provide $3.9 billion to
states for purchasing new voting equipment. Since a lot of the equipment
that would be purchased is very expensive and has a high depreciation cost,
the $200 million figure could escalate quite a bit. Let's say the useful
life of the new equipment is 10 years (I think that's generous). We would
be looking at about $400 million per year in terms of depreciation of fancy
new machines.

Quoting Matt from an earlier post:

> The scenario I put to Alan was this:
> Say Essex County, MA, had its own IT infrastructure available
> to the point where all that was needed was the source code[2].
> Would they be able to get the source code for free? Yes.
>
> More needs to be explored: for example, how would Essex Co.
> certify the results (which begs the question of certification).
> I'm sure you all could come up with other scenarios that better
> explore the issues here.
>
We are not going to make election administration literally free. This is
not possible. I suppose we could make the software literally free for
download and use, but this is a trivial point.

Suppose we suddenly finished the production software (never mind for the
moment how we got it all done and raised the $100,000 needed to get the
software certified for use in elections) and post it on the Internet where
Essex County could download it for free and use it.

Now Essex County can choose vendor A or vendor B or vendor C or they could
Download the software and do it all themselves. What will they choose?
They will choose A, B, or C. Probably they already have a contract with A,
B, or C. The vendor takes care of what they need to have done. The
Download option is completely unknown to them. Where would they get the
PCs? How do they know the software is good? How does the contest data get
loaded into the machines? They will have a million questions and no time to
figure it all out. Do they trust their own IT people to figure it all out
for them? The fact that the software is free for download is almost
irrelevant.

In order for Essex County to choose the download option, a whole
infrastructure will have to exist to support the elections people. The
Essex County elections people will need to have someone come to them and
say, "We know how many machines you need and we can provide them. We
guarantee that the machines will run the voting software correctly. The
machines are all tested and ready to go. Our PCs meet all these published
standards regarding using commodity PCs for the voting application. We know
how to situate them in the voting booths, and we'll do all this for you."
They need other people to come to them and say, "We know the software inside
out. We will train your people to use it. We will give you a tech support
number you can call and get help anytime you need it." and so on. These
support people might also say, "We're members of the Open Voting
Consortium."

The Open Voting Consortium (OVC) will be the vehicle by which the system is
delivered. The OVC itself will be a non-profit (really a trade
organization -- 501c(6) organization). The OVC members will include people
that make money doing this work (voting system delivery). I will also
propose that the OVC include government membership. Members pay to be
members like any other consortium. The consortium engages in activities
like other consortia. The success of the OVC will not depend on its ability
to deliver a system that costs nothing. The success of the OVC will depend
on its ability to deliver a voting system that is better and cheaper than
what we have now (and what is projected to exist in the near future).

EVM project members need to realize there is a lot of money involved here.
We're all volunteers right now. Some of us may find a money-paying role in
the OVC. Some of us may find a money-paying role in the development of the
production software. There are no guarantees. It's, like, up to you.

--Alan Dechert
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Sun Aug 31 23:17:11 2003

This archive was generated by hypermail 2.1.8 : Sun Aug 31 2003 - 23:17:18 CDT