Re: Money, democracy, and transparency

From: Alan Dechert <adechert_at_earthlink_dot_net>
Date: Sat Aug 16 2003 - 18:16:16 CDT

----- Original Message -----
From: "Anand Pillai" <abpillai_at_nospammail_dot_net>
To: "voting project" <voting-project@lists.sonic.net>
Sent: Saturday, August 16, 2003 12:37 PM
Subject: Re: Money, democracy, and transparency

> I am certainly not involved in this project for any money, because at
> this stage I would like more experience and interaction with some of the
> top
> experts in programming concepts and security like Dr. David Mertz.
> I am solely involved in this project for that purpose. Any name it brings
> to me if I do a good work, or any money if there is some at the end, is
> of a very distant concern to me.
>
It's just as well that these concerns get aired now.

> I also joined this project because ....
>
We're glad you did, BTW.

> .... it was announced in c.l.py group
> as an opensource project. ....
>
Yes, that is correct.

> Now according to my belief, an opensource project is
> one which uses a license approved by the OSI and nothing else. ...
>
Then you have a mistaken belief. The OSI definition is one of many
definitions of "Open Source." I'm no expert on all the possible variations,
but it's really an emerging concept as far as I can tell.

The most significant aspect of "open source" in relation to voting software
is that the code is open for inspection and testing by anyone that wants to
look at it and try finding problems. Also, having it all out in the open
makes it possible to discuss best practices and standards for voting system
software.

> It seemed natural that this project will adopt some kind of opensource
> license along the way when I initially agreed to take up lead developer
role
> and contribute to this project, both in my time as well as my efforts.
>
Yes. And thanks again, Anand, for taking on this role.

> After reading through all the 30 or so mails on the discussions about
> this topic, I have some strong concerns. I spend upto 3-4 hours a day
> on this project, apart from managing my regular career in the office
> and my personal affairs.
>
I appreciate your dedication. I think you are spending as much or more time
on it than anyone -- except me. I think David is spending quite a bit of
time on it too.

I hope no one spends so much time on it that it negatively impacts their
ability to earn a living. I can't say that about me, however. The project
began as a sidelight but has taken over my life. I've gone from being
fairly healthy financially in 2000, to being about one step away from
homelessness.

> If at the end of this day all of my work is (apparently) going
> towards making money for person or a group of persons I have
> not even met, I have strong reservations about continuing in this
> project.
>
I can understand that. And if you have to quit, there will be no hard
feelings from me. Maybe I didn't spell this all out adequately. However,
maybe you didn't read enough. I think the project has been accurately
characterized. The whole project -- EVM, and later, the UCVS and the Open
Voting Consortium (OVC) is enormous in scope. You are new to the project
and are not expected to have all the relevant documents read and memorized.
There is a lot of material here:

http://home.earthlink.net/~adechert

The original project proposal (UCB study for California) is here:

http://home.earthlink.net/~adechert/uc_voting_study.html

I was scheduled to be paid, yes. The project was recommended for funding by
some pretty prominent people in the State Senate. It did not happen for two
basic reasons: 1) the budget crisis (at that time it was mainly the energy
crisis drain on state money), and 2) the fact that some prominent
politicians were sold on the idea of paperless touchsceen systems for all
the voting booths.

You have taken on the role of Lead Developer for EVM.

> Who knows whether the University of California or whoever,
> if they take over will even keep my name in the source code?
>
I don't know either. But, again, at this point your role only has to do
with EVM. Nobody knows who will do what after that. It's unlikely that EVM
source code will wind up in the finished production system. I expect EVM to
roll into the requirements and specifications for the production system
(UCVS). The demo could be a flop in which case it's all a moot issue. I am
dedicated to making the demo a success. If the demo is a success and we
wind up getting some serious funding to develop the production system, it
will open up all sorts of possibilities.

> I may not even get the narrowest visibility an opensource
> developer expects, that is having his name in the source
> code files (in the comments/headers) because perhaps the
> people who take over the code might decide to remove it. (who
> knows?)
>
You have taken on the role of Lead Developer for EVM. If you see the
project through, you will have quite a bit of control over what's in the
source code. Your contributions will be acknowledged far and wide. You are
recognized as Lead Developer in the annoucement I sent out. You will be
acknowledged in other publicity about the project. What more do you want?
If you want your picture in the New York Times, I'll try to get it in there.
I am happy to keep a low profile and let others take all the credit.
However, I do want to get paid eventually because this is taking ALL of my
time.

> The people who commercialize the code might decide to "throw away"
> the original code, but still borrow its concepts. Since an opensource
> kind of license (OSL, LGPL, GLP, BSD etc) is not enforced, this will
> be a pretty easy task since AFAIK only the OSI approved licenses
> enforce any kind of control over derived source code or derived licenses
> based on the original work. A commercial license does not do that
> since the licensing terms based on the actual source code is not really
> relevant in a commercial setup, as it is based on the binaries or the
> bytecodes.
>
The people you are working with on EVM take this sort of thing very
seriously. I don't think you should worry about not receiving proper
attribution for your efforts. Arthur Keller has served as an expert witness
in a number of cases involving copyright infringement etc. I believe he is
involved in two such cases right now.

> It is surprising that this project apparently looked on the way to
> adopting a proper opensource development methodology, but now is
> talking in terms of intellectual property to a person or a group of
persons. ...
>
David pointed out (and I tend to believe him) that my claim of "intellectual
property" may not have a firm legal basis. I don't really know and I am not
really concerned either. EVM-UCVS-OVC is my idea and I've fought long and
hard to gain acceptance for the idea. I note others have said that the idea
is "obviously correct" and that voting technology experts will see the light
if they haven't already. However, I recently got an email from MIT
Professor Ted Selker where he argues against several key features of our
proposed system.

BTW, I should try to use "our idea" as much as possible as opposed to "my
idea." Even though anyone looking at this project from the outside
wondering "who's behind it?" would conclude that Alan Dechert's the one
spearheading the project, the fact is that many others have contributed to
the idea. I need to be careful about acknowledging this fact.

The list of contributors is very long and someday I will try to write it all
down. Some of the early contributors that come to mind include Larry Sokol
of Speaker Hertzberg's office (CA State Assembly). Chris Reynolds of the CA
Secretary of State's office also spent a lot of time with me in the early
months (Dec 00 - Feb 01). I had a long detailed and sometimes tedious
discussion with Peter Neumann that began Dec 12 00 and went on for some
weeks (I met him in person 17 Jan 01). David Jefferson also gave some good
feedback in Jan 01.

Jill Levine of the Sacramento County Registrar of Voters office arranged the
first meeting I had with elections staff people (see bottom of my web page
for handout given at that meeting). Carnegie President Vartan Gregorian
wrote a nice letter after I shared the materials with him.

I started talking with Roy Saltman in March of 01 and this dialog has
continued to the present. I had lunch with him in San Franciso on June 2nd
of this year. Elections Chief for San Mateo county, Warren Slocum, became
an advocate for me after our meeting in March. I met with the Elections
Chief (ROV) for Alameda county, Bradley Clark, on the same day. Also in
March, I got a boost when Mike Antonovich, Mayor of Los Angeles County,
wrote a letter asking the ROV to investigate the feasibility of my scheme.

Chris Lehman of Senator Perata's office was my first real advocate in the
state government. He introduced me to Sandi Polka of Senator Burton's
office. Senator Dede Alpert was the first legislator to do anything to
advance the idea (she handed my materials to the consultant for the Senate
Elections Committee and told him to consider the idea). Don Perata wrote a
couple of letters on our behalf.

Sandi Polka introduced me to Bruce Cain of UC Berkeley's Institute of
Governmental Studies (IGS http://www.igs.berkeley.edu/ ). Karin McDonald of
IGS warned me that UCB administration would have a hard time with the idea
of free software. Bruce introduced me to Henry Brady (UCB's resident expert
on voting technology) in April 01.

Henry E. Brady, Professor of Political Science and Public Policy Director,
Survey Research Center and UC DATA, worked with me greatly in 2001. He was
my co-author for the California proposal and we attended countless
meetings -- defending and promoting the proposal -- with University and
government officials. Henry introduced me to former congressman Leon
Panetta (and former Clinton Chief of Staff) who help behind the scenes
arranging a couple of meetings. Recently, one of Henry's former students
(when Henry taught at Harvard), MIT Professor Stephen Ansolabehere has given
valuable input.

Chancellor Berdahl of UC Berkeley pushed the project along with
encouragement and a letter or two here and there.

Mark Hayes, University Relations, Microsoft Research, pushed the idea around
Microsoft (first half of 2002 and a little bit in 2003).

The project was reborn in 2003 with the help of David Dill. David
introduced me to quite a few people, including the Australians involved in
the open source (GNU GPL) PC based voting system they developed there. Ed
Cherlin (Dill referral) was the first to suggest Python for the demo.
Dennis Paull (Dill referral) introduced me to several more people and
invited me to participate in the recent panel discussion on Electronic
Voting (where I met a bunch of people). Arthur Keller (Dill referral)
introduced me to Adrianne Yu Wang.

I met with Mohan Paturi, chair of the UC San Diego CSE department, on April
25 of this year. I started a long dialog and partnership with Doug Jones
shortly after meeting Paturi.

After we got the kernel of the demo project formed (thanks to Arthur
Keller!), lots of others, including David Mertz and Anand, joined in.

> Let me ask then what the poor developer who did the *throw away*
> demo gets? A word of mention maybe? 100 or so mails in this list?
> Or some kind of "thanks a lot for your work and your motivation"
> mail?
>
I don't know what you will get out of it. If you get nothing of material
benefit, then you will get more material benefit than I have received so
far. What do you want out of it?

If the project is successful, there will be plenty of recognition and
benefits to go around, I believe. You are in a key position and so the
recognition and benefits would probably be greater than people playing more
peripheral roles. I can see all sorts of possibilities for paying work to
come out of a successful project. Do I need to spell them all out? Besides
the possibility of getting paid directly for working on the production
system (I don't think anyone will get any pay for working on the demo),
there is the possibility of future work having to do with technical support,
maintenance, training, etc. with the finished product.

If the project is successful, you will join a very long list of
contributors. None of these contributors so far has made an issue out of
what they will get out of it -- except maybe me. The only reason I have
made some issue out of it is that the project is taking all of my time and
energy and I have a wife and two children to support. As Doug Jones and
Arthur Keller have acknowledged, it's reasonable for someone playing my role
to receive compensation. As the project grows and develops, I will likely
be the first person to get some compensation (as project developer with
2,000 hrs invested, seems reasonable, no?). Others will likely get some
compensation as funding sources develop. If it gets funding as a university
project, the vast majority of hours and money will likely go to student
researchers. There will be some money available for consultants and project
administrators.

> If I am working in an opensource project for making someone else
> rich, then I might as well devote that time doing a closed source
> project of myself, which might one day benefit me financially. I dont
> care about my mails being archived in this list for the future, since
> I am involved in many other opensource community projects apart
> from this. They are actually truly opensource by the way. I might not
> have said this before, but in the light of the mails exchanged here,
> I would like to stress that point.
>
I will also stress the point that the eventual finished product will entail
commercial possibilities. I think it is unreasonable to demand that nobody
can make money off of voting technology. My goal is to make this voting
technology as cheap as possible. The current crop of electronic voting
machines are rip-offs. They are unaffordable even for jurisdictions in the
USA.

The whole EVM-UCVS-OVC project carries the possibility of a very high
quality, reliable, fully auditable, easy-to-use election administration --
CHEAPLY! We can't make it totally free, but we can make it very cheap.
There are costs associated with election administration and there is no way
to totally eliminate them. BTW, MIT's Ansolabehere told me that voting
equipment cost is running about 15% of the total cost of election
administration in the USA. I think we can reduce that as a percentage but
also reduce the entire cost of election administration -- all while making
it much better.

> I am at present involved in many community (especially python)
> and personal projects most of which are truly opensource. Heck,
> my python page albeit a small effort is listed in google under the
> heading "Opensource software written in python". All my
> freshmeat projects are under OSI approved licenses and their
> sourcecode is available for download. I have listed EVM 2003
> also in my homepage thinking that it was a true opensource
> approach at promoting ransparency and security of election
> processes, whose code would be written in python, a language
> I love to program in. Exactly the same reasons David has mentioned.
>
Okay.

> It would have been a gentleman's decision to let us developers know
> beforehand that the project was aimed at developing a commercial
> solution managed by a select coterie one day right during the
> announcement of this project in c.l.py. That would have given at least
> a person like me who belives firmly in the opensource philosophy
> sometime to ponder over before actually joining this project.
>
Perhaps you joined without reading enough material about the project.
Nothing was hidden from you or anyone else. The project is evolving. I
might add that the Open Voting Consortium will be designed to be the vehicle
by which the voting system can be delivered to election officials. There
are many issues to consider regarding how the PCs get to the voting booths.
It will be nice if we produce great software for all aspects of election
administration, but it also has to be established how the software will be
used.

> I am not sure of continuing my contribution to this project in any
> capacity before this issue is not sorted out. And in the way I see
> it there might be only one way to sort it out, that is by announcing
> an OSI approved license for this project in its sourceforge page.
>
I don't know. I thought David was recommending Public Domain.

> I want to inform everybody else in this project that I am suspending
> my time and effort devoted for this project till then. ....
>
Okay. Please let me know soon if you want to continue.

Don't worry that the project might die if you leave. It has died 20 times
already. If it dies again, I will bring it to life again -- perhaps with
more wisdom than before.

I'm sorry for any misunderstanding about your role. To clarify, you were
chosen for the Development Lead for EVM. This is designed to develop demo
software. You may or may not have a role in the development of the
production system. That would probably be your choice and would depend on
the success of EVM. If the production system is developed through UC, you
cannot be the Lead Developer. This have nothing to do with your skills.
The only way we can get grants through the University is to have academic
engineers listed in the proposal.

> I am also a busy person in my own little way and has my
> own ideas and projects, where I can devote a better time in terms of
> trusting trust, adherence to ideals and the spirit of opensource.
>
Right. But you can't expect no one to make money on the finished product.
People make money on Linux. People find ways to make money on all sorts of
"free" products.

Alan
==================================================================
= 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 23:17:10 2003

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