Meeting notes 7/31

From: Alan Dechert <adechert_at_earthlink_dot_net>
Date: Mon Aug 04 2003 - 13:03:28 CDT

In attendance at Arthur Keller's house in Palo Alto, CA: Arthur Keller,
Adrianne Yu Wang, Alan Dechert, and, by phone, David Mertz.

I. Introductions/Roles

We introduced ourselves briefly. Adrianne mentioned that she is a recent
UCSC computer science graduate. Arthur talked about his Stanford background
(CS Ph.D. from Stanford). He mentioned his association with David Dill and
Terry Winograd. Arthur has started several companies - including a couple
of successful ones and a couple of less successful ones. Arthur recently
completed a complicated NSF grant proposal, which got him involved with top
administration at UCSC. He likes this type thing and knows how to do it.

Alan mentioned his background as a computer consultant - including
assignments with Borland and Intel as a QA engineer. Alan also discussed
the history of the voting project, which began as a proposal to develop a
pilot project for Sacramento County but expanded over a period of 2.7 years
as more and more people gave input.

During these presentations, many people asked Alan if a demo was available.
Alan realized the importance but did not have the time or resources to put
together a demo. Finally, Arthur came along and offered to help with a
demo. Arthur recruited Adrianne who was a student of his at UCSC.

David talked about his career as a professor and writer. He mentioned that
his areas of interest with respect to the voting project involve

1) Human Factors
2) Security
3) Legal Compliance

Alan mentioned that a problem with the "Legal Compliance" issue is that the
body of laws having to do with voting issues is a mess. Furthermore, our
system is a hybrid that does not fall easily under any of the voting machine
categories covered in legal code. Our project should get involved in the
process of regulatory change - our Election Rules Database is designed with
this in mind.

David adds, "It might be more accurate to say that I felt/feel these are the
three areas that must be addressed in the requirements document, and by the
project. While recognizing its importance, for example, I'm less interested
participating in the details of human factors personally."

We decided to include David as an "advisor." [I tend to see this Board of
Advisors, namely, Arthur, Alan, Doug, and David, along with Team Lead
Adrianne, as the governing body for this demo project].

We are looking for someone to be Lead Developer.

We nominated Matt as QA Lead [Matt accepted the role yesterday]. As we go
along, we will look for others to take the lead in other areas. Arthur
mentioned that we will want someone to specialize in the UI, for example.

After the meeting, Adrianne mentioned that she will be going to NY soon for
about a week and will have only sporadic access to email. I think we need
to also find an assistant for Adrianne so that gaps like this can be
covered.

It is understood that the people and roles are not static - we expect the
team to grow and change.

II. Requirements and Specification

David argued for a non-MSFT format for these documents. Arthur suggested
LaTeX [in subsequent list email, David has suggested APT, but we need some
input from Van and Ed since they are the ones that will be producing the
documents].

David added, "While running by Van and Ed is certainly a good idea, much of
the reason I argue for a non-proprietary format (and an easy-to-learn one,
e.g. APT rather than LaTeX) is so that *anyone* who becomes interested can
contribute to the documentation process (and revise/enhance prior
documents). This will not only include the technical documents as such, but
might also include things like press releases, mission statements, summaries
of legal issues, comparisons with other voting systems, or all sorts of
background documentation maintained over time. I assume not only Van and Ed
will produce such things."

Arthur made the point (we agreed) that the Requirements should address the
whole system: we will only demo some of the attributes spelled out in the
Requirements but we should have it all there since questions will come up
about how it all fits together.

III. Sourceforge.net Status Report

Adrianne reports that the project is now on Sourceforge.net. The name is
evm2003. Here's the direct URL:

http://sourceforge.net/projects/evm2003/

There is nothing there yet. There are some questions about which features
we will use [Anand has since made some comments about this we should
consider].

Adrianne mentioned that project members need to register with Sourceforge
and give her the Username in order for her to add them to the list. She
said she would send a message to the list about this [she has sent the
message but we still need to have members follow through with this].

We want to make the list archives available. David offered to deal with
that. Arthur needs to send the digests to David so he will have all the
messages. We have an issue with whether or not the archives should be
publicly available.

Arthur expressed concern about how some members' comments could be construed
as politically partisan (presumably inspiring political opposition). Alan
mentioned that while we're probably mostly Democrats, we had Republicans on
the list too ... as well as several non-U.S.citizens. Alan also mentioned
that the discussion on this list has all been on-topic so far and that I saw
no problem with political comments; he is in favor of open discussion.

Alan suggested that the public/private question should be posed to the group
[this has since been done, no responses yet].

IV. Mailing List

We agreed that the advisors (members of the Advisory Board) and Adrianne
could add people to our list without debate. To remove someone
involuntarily, we will need to have a majority of the management team
(advisors and Adrianne) in favor. Project participants are welcome to make
suggestions for additions.

We considered whether or not Arthur might need a back-up administrator, but
we decided it was not necessary at this time. Arthur will take care the
list by himself unless it becomes obvious we need to change this policy.

Anyone on the list can get a list of current members by sending a message to

majordomo@lists.sonic.net

with the command

         who voting-project

in the body of the message.

We agreed on the need for list archives. David agreed to handle that.
Arthur mentioned that he gets a digest every day that includes all the
voting-project messages. He said he would forward the digests to David.

V. Timeline (PERT chart)

We decided (Arthur suggested and we agreed) that mid-October would be a good
target for presentation of the demo. Arthur argued that we probably would
not be ready before the Oct 7 election (here in CA, recall etc.) After the
Nov. election might be too late since people would be tired of election
stuff.

Also, we want to get this done sooner than later. It's still possible to
have the certified software done before the 2004 elections if we can get the
demo done soon. It's debatable how many jurisdictions in which we could
have our system deployed, but if we had some at least by then it would be a
major accomplishment.

Adrianne agreed to be responsible for the development of the chart. We
talked about the possibility of using MS Project for this. Arthur said that
Adrianne could get a copy through UCSC. We are not concerned about the
proprietary nature of MS Project so long as the chart will be exported to a
non-proprietary format (i.e., PDF).

We identified a few things that should happen very soon: Namely, we need to
get some early drafts of Requirements and Specifications posted.

Arthur mentioned that we should have an architecture document and that the
Lead Developer should be responsible for it. Arthur agreed to help with
that.

VI. Bug Tracking

We agreed that the bug tracking system is important. Instead of "bug
tracking" or "problem reports" we agree that "issues database" would be
better. All of the issues we need to resolve should go into the database so
we can all view and participate in the resolutions. We will probably start
with the system included with Sourceforge unless someone can make a good
case for using some other tracking system.

VII. Demo date etc.

As mentioned, we agreed on mid-October for a target date.

VIII. Demo hardware set up

We nominated Jay and Ben to come up with a design for this.

IX. Announcement about the demo project kickoff

Alan said he would be sending an announcement soon. He will draft a copy of
the announcement and run it by others before sending. The announcement will
go to Alan's announcement only list that now includes about a thousand
addresses, including most everyone that has gone public with their concerns
over voting modernization issues.

X. Funding ideas

Arthur has agreed to take the lead working on a NSF grant proposal. He says
he knows how to do it. He gave us a handout regarding a specific NSF
program solicitation (NSF-02-156 re: "Digital Government"). This program
could provide up to $500,000 per year for three years.

Due Second Wednesday in October [October 8, 2003 is our target]. I think
that aligns well with our target of having a demo a week or so later, since
we must have the demo software completed first.

Arthur will need a lot of help from Alan and others with the contents of the
proposal. We'd need a PI at each participating institution. Arthur would
either be PI or co-PI at UCSC (with a more senior person as PI). We'd also
need to decide the institution where Alan will affiliate. Alan could be an
employee, which would be facilitated by having him named in the proposal.
We'd also need to decide which of the institutions serves as lead
institution.

Arthur mentioned that with more than three or so campuses involved, it can
get unwieldy. Alan mentioned that some money might be funneled to the
project via HAVA money to the states. These might be discrete projects
associated with the larger project that is expected to have its core at one
or more of the UC campuses. Besides UCSC (where Arthur is affiliated), we
have faculty interested in participating at UC San Diego and UC Berkeley.
Other campuses around the country where we have interested people include U
Mass (Amherst), NYU, Stevens Tech (NJ), Stanford, and U of Iowa to name a
few.

Alan received a letter from the Ford Foundation indicating some interest.
There are other foundations to approach and there is also the possibility of
other private and government sources.

Doug Jones has also heard from some people in Europe that might be
interested in funding part of our project.

Arthur has expressed some concern about the politics of getting foreign
money unless it's for development and demonstrations elsewhere.

David added, "Btw. I mentioned some contacts among humanities/social science
academics at the end of the conversation. Those I had in mind were chiefly
the Radical Philosophy Association (RPA) and the Association for Economic
and Social Analyisis (AESA). But I should probably forward announcements to
those lists myself, since I'm known there. When I see Alan's announcement,
I'll chat with him about any wording changes that might be relevant to
target those groups (maybe none needed, I dunno).

I've been thinking about whether there might be other funding sources that
have more to do with the legal/political aspects than the technical ones. I
know we talked about having to pitch the science aspects to NSF. But I
wonder if an organization like the ABA would be interested from a different
perspective.

Part of the reason I have thought of announcing to lists like the
abovementioned is that we might find folks with expertise and ideas in
relation to such other grant sources."

Adrianne, did I miss anything?

Alan

- -- Alan Dechert
Received on Mon, 4 Aug 2003 11:03:28 -0700

This archive was generated by hypermail 2.1.8 : Wed Aug 06 2003 - 12:50:26 CDT