Re: will making opinion polls faster and more common improve life? [cr-95/11/14]


Sender: •••@••.••• (Marilyn Davis)


In response to your message:

Yes.  History has shown us nothing good.  But that doesn't mean that
something good can't come of the future.

And, I may be all that you project, but that doesn't make me wrong.
Facts are facts.  1 divided by 0 is infinity, even when a lunatic says

eVote turns the structure of polling inside-out.  So far in history,
only an elite and powerful few (mostly) men have had the privilege
to concoct polls.  Soon all online people will.

But, I doubt if eVoted polls will be fast.  I expect some polls will
never close.  eVote was designed to support a deliberative process.
My goal was to emulate a Quaker meeting for business -- online.
Because we can *change* our votes, we can discuss online and change
each others' minds.  Because we can see who votes how, the group can
hold the individuals accountable.  eVote is support for a process;
decisions are the by-product.

This revolution will have no leader.  Or, rather, this revolution will
have 10,000 leaders to start and billions before we're through.  This
revolution is about individual empowerment and group cooperation, not
about following another prophetic charismatic character.

I do find it interesting and even comforting that history's prophetic
charismatic characters are looking so right.  But, it calls into
question the concepts of free-will and predestination.  Kind of eerie.

> Sender: "David E. Anderson" <•••@••.•••>
> Marilyn,
> Your note (extracted below) scares me.  It comes across as condescending,
> naive, and anti-rational and completely ignores the real issues.  You sound


> We have a duty to make sure the electronic polling idea does not also backfire
> and wound us.
> Dave

Marilyn                               *
Marilyn Davis, Ph.D.-------------- * ---- eVote - online voting software
|                                 *       demo at (415) 493-8683       |
3790 El Camino Real, #147  *     *        Weekdays, 10-5 PST           |
Palo Alto, CA 94306 USA     *   *         log in as "eVote" - no quotes.
(415) 493-3631 ------------- * * -------- •••@••.••• ------------


Email version of a talk given by Marilyn Davis at the first evening
meeting of the Palo Alto Community Network, October 10, 1995.


What is God-given and true to our nature is the expectation of
self-determination.  We should be free.  Free to do as we please, free
to think as we please, free to *be* as we please.

This quirk of human nature is responsible for the terrible twos,
the troubled teens, and the willingness to fight to the death against
the world's despots.

Also, what is God-given, true to our nature, and a necessity to
survival is our need for each other.  We *want* to be together.
We want to belong to a family, to form groups and gangs and teams
clubs and communities.  We *love* each other.

One of the great challenges of ourkind is to reconcile these two
powerful drives: the instinct to be free, and the need to support each
other in groups.

How do we organize ourselves into communities and still meet our
expectation of personal freedom?  How can we get along together?

Human beings have experimented with many different forms of civic
organization; some, like facism, have sought to obliterate
self-determination, and some, like the Quakers, the Cuna Indians, many
peace activist groups, and co-housing groups, seek to encourage and
protect it.

These individual-supporting groups make their community decisions by
consensus.  Consensus means that no one disagrees.  Each individual
has veto power so no one is required to forfeit their free will.

Consensus, where no one disagrees, is much easier to achieve than
unanimity, where everyone agrees.  However, the consensus process is a
tedious and tiresome process and it requires that the group be
strongly connected in purpose and that they each be committed to the
process itself.

Here's how consensus works.  After an issue comes forward and is
discussed, there will be a call for a show of hands.  If all agree, or
abstain, except a few, these few are required to state their
objections.  I once saw a whole meeting turn around at this point --
when a quiet person whom I had never heard speak before was forced to
state his objection and it was an excellent objection that no one else
had thought of.

These meetings can go on and on, all afternoon and night.  But,
whenever a group really needs a decision, they always manage to pull
together on one.  Like I said, it's a terribly arduous process but
one that almost always produces the best answer.


Now the first point I want to make is that, although consensus-run
meetings are the most difficult to experience in the non-virtual
world, they are the easiest and most natural style of meeting
for cyberspace at this time.

For an online consensus-run meeting, you only need an email list.

An "email list" has an email address where I can send a message and
from there it will be sent to all my chums on the list.  They too can
a send message to the whole group by sending it to the list's address.
This is an online emulation of a meeting in a dark room.  We can hear
each other but we can't see each other.

To acheive consensus, I can say, "Are there any objections?".  If
there are none, the proposal becomes law for the group.  If there are
objections, we talk on and on, just like in face to face meetings.

I want to interject here that, at this time, the Palo Alto Community
Network is without legal status or bylaws.  But, when we do make our
by-laws, I hope we'll consider this consensus-driven online process
for at least part of our decision-making mechanism, and at least until
we have more tools available for facilitating decisions online together.


The purpose of this talk is to start a project to give us the best
tools imaginable and I'll get to that soon.

But first, I want to give a little background about the concept of
"Electronic Democracy" and critic a few experiments that have been
prominent in the news so that, when I present our dream system, you
will understand how different it is, how much better it is, and how
badly we need it.


Perot, most visibly, and many before and after him, espouse an event
where he gives us a question of the month on TV and we get to vote by
phone or cable box.

This type of *TV/phone voting event* was criticized in the Nation
Magazine in 1982 by Jean Betheke Elshtain.  Quoting from that article:

"The interactive shell game cons us into believing we are
participating when we are really simply performing as the responding
'end' of a prefabricated system of external stimuli."

Also, "... the advocates of interactive television display a
misapprehension of the nature of real democracy, which they confuse
with the plebiscite system.  The distinction is not an idle one....
Plebiscitism is compatible with authoritarian politics carried out
under the guise of, or with the connivance of, majority opinion.  That
opinion can be registered by easily manipulated, ritualistic

Brian Fay, a political theorist, says this about democracy, what "is
most significant is the involvement of the citizens in the process of
determining their own collective identity".

A real democracy involves deliberating together, changing each other's
minds, and bringing up our own concerns, not just pointing at a
multiple choice option that someone else cooked up.  In a real
democracy, the vote is the icing on the cake; discussion is the meat
and potatoes.

So when Perot calls his plan an "electronic town hall meeting", it is
a gross misrepresentation of what he is proposing.  In a real town
hall meeting, everyone has a voice.  In a real town hall meeting,
anyone can bring up a new issue.  A real town hall meeting is run by a
facilitator whose job is to ensure equal voice.  A good facilitator
tries to keep her own agenda out of the meeting, not to *lead* the
group, but to facilitate the will of the group and ensure equal
opportunity for all the participants.  Perot does not have this in


Another experiment under the Electronic Democracy banner is the
"national electronic open meeting" run by the federal government
earlier this year.  This experiment, and others like it, give us week
or so to participate in an online internet-wide discussion on a
particular topic -- the "Information Superhighway" was the topic last

This is an electronic emulation of a big shouting match.  There is no
way to process what is being said so that a conclusion can be reached.
There is absolutely no theoretical basis on which to stage such an
event.  Sillier still, such an event is sure to clog up the
communications systems, and it did.  It was a total frustration for
anyone who tried to participate and was reported to be so in the
papers and around the net.  To be in harmony with the physics of
cyberspace, we can't hold big, fast, events.  It puts too much strain
on the pipes.  By its nature, cyberspace is for supporting a steady
stream of communication, not big intense bursts.


My final example is Newt Gingrich's "Thomas" system, much like our
city's Web page.  Thomas and our city's system provide information
about congressional legislation, or about city services. This is a
lovely thing, but we should not confuse it with "town hall meetings",
as Newt does.  These government Web-pages are emulations of glassed-in
bulletin boards.  We cannot post, we can only read.  Although it is a
good thing to provide information, we must recognize that when the
government uses the medium this way, they are seeing the medium as
a tool for propaganda propagation.


For our system, we want to be able to truly emulate a town hall
meeting online, to truly emulate a consensus-run meeting online, and
any other style of meeting that the participants can think of.

We certainly want our polling facility to be embedded into the
conversation facility, so that discussion is the meat of our system;
and we want the lights on so we can see each other's hands when we
take a vote.  In cyberspace this means that I can query the system for
how Ann voted, or, I can ask for a list of those who voted "no".  Then
the group can demand that dissenters explain themselves and, using
these tools, some groups will be able to develop consensus.

Also, we want to be able to develop collaborative outlines.  This
organizational tool will help us to identify our issues.  As we
develop our outline, outline points that are added and not
contested become a decision for the group.  Contested contributions
can be discussed and then, if agreement doesn't emerge from
the discussion, we can take a vote.

Finally, we want everyone to have the facility to start up a private
group without asking anyone else's permission, or even without anyone
else knowing about it.  This is the meaning of the words, "right to
free assembly."

In summary, this is what we need:

*  User-generated discussion groups -- public and private.

*  Outline support -- branching topics.

*  User-generating polling, embedded into the discussion.

*  Consensus development support:

   *  Ability to change our own votes.

   *  Lights on!  We can see how others voted.

*  Privacy support.


Such a system does not exist today, but we are close to having it.  In
order to explain how close we are, I need to explain one technical
concept about software architecture.

In all software applications, the source code, the stuff the
programmer makes, and the stuff that makes the computer *do* whatever
it is that it does, has a natural division in it.  It divides
naturally into two pieces: the user-interface and the engine.

The user-interface is the part of the program that talks and listens
to the user.  The engine does the work that the user asks to have

The WWW is driven by this natural separation.  The Web browser handles
the user-interface; the Web-server at the Web-site is the engine.

The trick of the Web is that the user-interface is available from the
engine side so that it can be attached to all kinds of engines, not
just to the information servers the Web surfers are used to seeing.

The first great example of an engine, more powerful than an
information server, by no coincidence, is the voting sites on the Web.
If you poke around the Web, you can find some sites that invite you to
vote on things.  You can vote for your favorite music somewhere.
There's VOTELINK that invites you to vote on a different subject each
week.  So these sites have a fancier engine than an information
server.  They will also process your votes.

These systems, however nice they are, are still plebiscites because
some central authority invents the voting topics.

Our dream system, of course, needs another dimension in interactivity
so we, the users, can set up our own polls.

Currently, we have two engines, Participate and eVote, that, when
working together, will give us everything we need.  However, they
aren't connected together yet, and they both have clumsy and obscure
text interfaces that would discourage all but the most ardent
democracy supporters.

However, these engines, when attached to the Web user-interface,
will make participation intuitive, productive and fun.

Let's turn down the lights and have a look.


Here we see my opening screen in our future system.  See, "Welcome
Marilyn", it knows me.  And my email is arranged for me.  I designate
which of my email correspondents get into my "first priority" list.
And, I like to have my private email listed separately from my public

At this point, I can read my email, I can write email, I can ask to
see an overview of the topics under discussion in any of my online
groups, or I can start a new online group.


Let's ignore all my messages at first and suppose I came online to
create a new group for people who are interested in gardening.  Here's
the screen that allows me to do this.

The process works like this.  I make an invitation list and whomever I
list in this box gets an email invitation.

There's an editing screen behind this button where I compose the

For each invitee, I decide if that person can read, write, vote, and
poll.  For a gardening group, I think it's appropriate for everyone to
have the same powers and for everyone to be able to do everything.
Also, I allow people who are not joined to the group to lurk.  I don't
expect we'll have secrets to keep.  But because 'others' cannot read,
write, vote or poll, lurkers must join before they can participate.
Anonymous contributions are not allowed here, although in another
group they might be.  When people are allowed to participate
anonymously, they are more likely to say destructive things -- flame,
or just waste everyone's time with senseless nonsense.

Finally, I say 'yes' to the question, "List Publicly", so that
people who are looking for a gardening group can find us and join.

This rare facility, the facility to start a group without asking
permission and without having to set up any special software and
hardware is essential.  My son wants it so he can have a private
meeting with his best friends.  I want it so I can have a meeting with
the people who are interested in the system I'm describing.


Ok.  Let's go back to the opening screen, and from there, let's look
at the overview of the PA-COMNET discussion.

This Web page, and the two others I'll be showing have no
functionality behind them.  However, Participate can provide the
engine for all these features.  The connections have to be made
between the Web interface and the Participate engine.

Here we see the topics under discussion in the PA-COMNET.  When
the number '2' is listed beside the topic, it means that I have
two messages to read under this topic.

>>From this page, once again I can read messages and add messages,
or start a new topic, or move a topic to a new spot in the
organization if I have permission.

The topics are arranged like branches.  A new topic can be added
anywhere among the branches, which allows us to develop outlines
collaboratively.  I imagine that this process, developing outlines,
will be an important step in group decision-making online.

For example, in the ToDo topic, people can list all the things they
think the PA-COMNET should do.  If someone disagrees with a
suggestion, there can be an discussion online, and if the discussion
does not resolve the disagreement, the group can take a vote right
there in the midst of the discussion and finalize their decision.

If the group lists more things in the ToDo topic than the group has
the resources to accomplish, the ToDo list can be prioritized by a
vote.  This asterisk on the "Development" topic indicates that a poll
is under way there and it is, in fact, a poll designed to set our
priorities democratically.  We'll look at it next.

So far, all the functionality these Web pages promise is provided
by Participate.

Now we're going to move away from the lovely Web interface and look at
eVote's contribution to the system by looking at eVote itself.


And here, as promised is a vote underway about which of the
development tasks that PA-COMNET considers worth doing, should be done

Don't trust these data, by the way, I invented them and they seem to
match my feelings exactly about the things we should do and about what
is most important to do first.

People have been invited to vote 1 to 10 on each of four projects.
You might recognize #4, "Web/Parti/eVote Links" as the project I am
describing tonight.


Let's vote 10 on it.  Ooops, I guess I missed the zero and only voted
1.  Luckily, I can change my vote as long as this particular poll is

There's a real 10.  That raised it's score a bit.


Let's see who gave it a low score.  Hmmm, look at this, Al gave it a
1.  I could write him and ask him if he made a mistake, and if not,
ask him to state his objection to the group.  Maybe he has thought of
something that rest of us have not.

We can only see how everyone voted because, in this case, the poll was
set up as PUBLIC.  Another poll can be PRIVATE where we don't see how
each other voted.  Also, a poll can be designated as IF-VOTED where we
can see *who* has voted, but now *how* they voted.

The one thing I'm not demonstrating is the reading and writing of
messages.  That's the part we are used to seeing and, although I'm not
showing it, it's in our new system and it is still the most important

This column of 0's in eVote's demo should be the number of messages
that have been written to each of these topics.  eVote does not handle
messages on its own, that's why they are all zeros.  If Participate
and eVote were linked together, not only could we see the number of
messages and read them and write more, we could also see which of the
messages Al, and anyone else, has read.  The lights are truly on
in the software we are about to build.


Here's a poll where we are asked to decide which entity we think
should fund this project.  There are two polls here really, one for
funding of the development work; and one for funding the maintenance
of the system once it's going.

These polls demonstrate another type of voting.  For these polls, we
are each asked to distribute 100 vote points over the various choices,
what percentage of the money do we feel should come from each entity.

(This same technique of distributing 100 vote points over a list of
options can be used to spend a budget democratically.  Only then we
have to set reasonable limits on the maximum and minimum percentage
that each item can have.  The actual budget will be spent according to
the averages that result.)

You can see by my votes that I feel it's important that community
networks be funded by local taxes so that funding is stable.  I can't
think of a better use of our taxes than to provide a cheap
democracy-enhancing tool for our community.

By the way, the City of Santa Monica provides a community network for
its citizens.  $140,000 is the yearly operating budget.  So we're
talking peanuts here -- in terms of city budgets.

The development work, I think, should mainly be funded by a commercial
outfit, Netscape or EIT, for example.  The browser-makers are going to
make money from our system, so they should develop it for us.


Now, we have one thing left to do, and that is to invent a new vote
question.  This particular feature, that anyone is allowed to poll the
rest of us, flattens our decision-making system so that everyone
has the same power.

I find poll-inventing to be tremendously empowering so I want to share
this experience with you.

Does anyone have a suggestion?  -- a poll you would like to take?

[Suggestion and Group Participation]

0.  Concept - frame the question.

1.  Title?

2.  Tallied? - Will there be a poll generated with this new topic?

3.  Public/Private/If-voted - Will people be able to see each others'
votes, or not.  If not, can people see *who* has voted?

4.  Show tally?  - Will people be able to query the system for the
current tally while the vote is still underway?

5.  Grouped? -  Will this be a single question or a "distribute your
votes over the following topics" question?  If grouped, go to 7.

6.  Is this a yes/no question or a rating question?  If it's a
rating question, what is the minimum and maximum that someone
can vote?

7.  Grouped:  How many in the group?

8.  Sum-limit: What is the number of votes that each person gets to
distribute over the group?

   The sum-limit is 1 if the question is "Vote for one of the following".

   The sum-limit is 100 if we are distributing a budget.

9.  Do all the choices have the same minimum and maximum vote?  If so,
go to 11.

10.  If not:  what is the minimum and maximum for the first choice?
What is the title for the second choice?  It's minimum and maximum?
The third? fourth? etc.

11.  All the choices have the same maximum and minimum vote.  What is
that maximum and minimum?

12.  What is the title for the second choice?  The third?  fourth? etc.


Before I take your questions, I want to impress upon you the importance
of what we are about to do.  I know that this is only a software tool,
but it is a tool that suggests a new way of being together, a new
way to solve our problems, a new way to take care of each other, and

And, it is a tool that requires a new discipline, a new engagement,
a new thoughtfulness and experimentation, trust and courage.

But, make no mistake, *this* is our future, and it is ours to create.


Lori Aratani, "All booted up, nowhere to go", San Jose Mercury News,
May 5, 1995.

Jean Bethke Elshtain, "Interactive TV - Democracy and The QUBE Tube",
The Nation, Aug. 7-14, 1982, p. 108.


Marilyn                               *
Marilyn Davis, Ph.D.-------------- * ---- eVote - online voting software
|                                 *       demo at (415) 493-8683       |
3790 El Camino Real, #147  *     *        Weekdays, 10-5 PST           |
Palo Alto, CA 94306 USA     *   *         log in as "eVote" - no quotes.
(415) 493-3631 ------------- * * -------- •••@••.••• ------------

 Posted by Andrew Oram  - •••@••.••• - Moderator: CYBER-RIGHTS (CPSR)
You are encouraged to forward and cross-post messages for non-commercial use,
pursuant to any redistribution restrictions included in individual messages.