At 07:26 AM 7/22/2008, Michael Allan wrote:
I'm grateful I was directed to this list. You're clearly experts. I
wish I could reply more completely right away (I should know better
than to start 2 separate threads). I'll just reply to Juho's
questions today, and tomorrow I'll look at Abd's work. (You've been
thinking about this longer than I have, Abd, and I need to catch up.)
Nevertheless, your thinking is very important, for I've encountered
nobody who has gone as deeply into this without having contact with
others working on it.
> 1) All voters are candidates and it is possible that all voters consider
> themselves to be the best candidate. Therefore the method may start from
> all candidates having one vote each (their own vote). Maybe only
after some
> candidates have numerous votes and the voter himself has only one vote
> still, then the voter gives up voting for himself and gives his vote to
> some of the frontrunners. How do you expect the method to behave from this
> point of view?
The basic rule of vote flow is: a vote stops *before* it encounters a
voter for a second time, and it remains held where it is. A vote is
always considered to have "encountered" its original caster
beforehand. So it is not possible to vote for oneself. It is
permitted, but the vote stops before it is even cast - there is no
effect.
Juho doesn't get it, yet. The system provides no special incentive to
vote for "frontrunners," and, indeed, a counterincentive, which
Michael describes below. "Voting for yourself" is actually the
default position. I.e., everyone is considered to represent
themselves, and to vote for themselves (on whatever issue is being
decided). But openly and deliberatively. I.e., the default is direct
democracy. Now, how can we make direct democracy efficient? Classical
answer: elected representation. But that sacrifices a fundamental
value of direct democracy. Proxy representation does not sacrifice
that value, particularly if individuals remain free to direct vote if
they so choose.
The problem of scale with direct democracy is *not* a voting problem,
it is a problem of how to conduct deliberation. Delegable proxy
breaks down deliberation into smaller units (what we call "natural
caucuses," a natural caucus consists of a proxy and all represented
by the proxy, directly and indirectly).
> 2) Let's say that the preferences of voter A are A>B>C>D>E. At some point
> he decides to vote for his second preference (B) instead of himself. B's
> preferences are B>D>etc. At some (later) point B decides to vote for his
> second preference D. A is however not happy with that the vote now goes
> directly to D (instead of C that was better). He changes his vote
and votes
> for C. The point here is that it may be that many voters will
vote directly
> the leading candidates instead of letting the voters in longer chains
> (according to their own preferences) determine where the vote ends at. The
> reason may be as above or maybe the voter simply prefers to vote directly
> for the leading/best candidates instead of being at the long branches of
> the tree (away from the main streams close to the root of the trees where
> the decision making appears to take place). Controlling one's own vote may
> also give the voter some additional negotiating power. The end result may
> be that the cascade chains may tend to be short rather than long. The same
> question here. Is this ok and how do you expect the method to behave?
The proportion of voters who preferred to vote for the "stars" would
act as a dead weight in the electoral system - a kind of irrational
ballast. To the extent they were fickle, they would act as a shifting
cargo on a rolling ship. Some factors that might reduce this:
* it can be detected and filtered from the results (as irrational
dross)
With, of course, loss of information regarding the free choices of electors.
* it will be boring, there's less scope to interact with a star
candidate, because a single vote has relatively little worth to
her, so:
- the voter's questions, and attempts to enter into dialogue are
likely to go unanswered
- the voter's freedom to shift the vote will confer no leverage,
no input to the candidate's behaviour
Yes. There is no advantage to voting for a star. Nor, in fact, is
there really an advantage from the star's point of view, *compared to
what stars will actually do,* unless the star is socially
dysfunctional. What a star will do is to identify clients (we call
those who name a proxy the clients of the proxy) who are trusted by
the star to act as filters between further clients and the star. This
filtering is actually a standard function of a proxy. The proxy
shields the client from excess traffic from above, and the client
shields the proxy from excess traffic from below (i.e., from clients
of the client).
So, you want to vote for Clint Eastwood, you email him. You get back
a mail that suggests you pick So-and-So. This gives Clint your vote,
effectively, but indirectly. Would you be offended? Not if So-and-so
is skilled at communication, not if Clint made a good recommendation.
And, in fact, should you have a legitimate need, you have *more*
access to Clint than you would have had if you were part of a
faceless mass of clients directly attached. All you have to do is
convince So-and-so that it's worth Clint's attention.
(Hint: things already work this way, but the structure is not clearly
known. DP documents it, and, in doing this, it makes it possible to
make it much more efficient, that is the big change.)
* the star voter will be open to criticism from better informed
peers, because the vote placements are public information
- "I see you're voting for a star. If you want to waste your
vote like that, why not waste it on me?"
Well, bad argument.... I wouldn't give my vote to someone who
solicits it. That is a radical change from present electoral structures.
> 3) In theory the method may also end up in a loop. There could be three
> voters (A, B, C) with opinions A: A>B>C, B: B>C>A and C: C>A>B. If A votes
> for A, B votes for B and C votes for A, then B has an incentive to change
> his vote to C in the hope that also C will vote for himself after this
> move. That would improve the result from B's (as well as C's)
point of view
> (from A to C). But as a result now A has a similar incentive to vote for B
> that is to him better than C. And the story might continue forever. This
> kind of loops would probably be rare. But do you think this is acceptable
> or should there be some limitations that would eliminate or slow down
> possible continuous changes in the votes? In this looped case is possible
> that when the voters note the loop they are capable of negotiating some
> compromise solution (e.g. A and C agree that C will get something
in return
> if he sticks to voting for A).
Maybe the rule of vote flow (1) will prevent that, since self-votes
are null? (I have to look at this one again in the morning.) There's
a little more detail on cycles here:
http://zelea.com/project/votorola/d/theory.xht#cascade-cyclic
The loop objection is a very common one. It's a sign that someone is
actually starting to consider the structure. However, it's a
non-issue. Juho's analysis is still coming from the electoral model.
DP can be used for elections, but that's not the best application, by
far. Rather, DP creates a deliberative body that can negotiate
consensus. It does so using traditional means (though it can
certainly utilize new ones and invent whatever it needs). If every
member names a proxy -- and, really, every member who cares should do
so, it is harmless at worst, since it only has effect if you don't
participate -- then there must be loops. The question is not, then,
whether there will be loops or not or how to avoid them, but what
their effect is. Whenever a loop includes someone who actually
participates in a decision, no harm has been done at all. When some
set of members isn't represented because no member of a loop
participates, and if this is considered a problem (it isn't,
necessarily), there is a simple solution: notification.
When notified, any member of the loop may connect the loop by
participating. Alternatively, any member of the loop may connect the
loop to a larger caucus by reassigning their proxy.
(This may be a general proxy assignment or a special one. Special
proxies are created for a ... special purpose! Special proxies
override general proxies within the special function. So if a
committee is formed to study something, and a member trusts some
particular member other than their general proxy with regard to this
special purpose, they assign a special proxy through a dedicated
proxy table for that committee. If they do not assign a special
proxy, the general proxy stands. Thus the system is simple (one proxy
assignment does serve for everything if that's what you want), but
flexible (you can be as particular as you like.)
(And here I must note that we have worked this out for application
within Free Associations -- this is FA/DP -- where nothing is really
binding on anyone, it is all pure, voluntary process, designed for
the negotiation of consensus, not for control. So anyone can create a
special proxy table for some purpose, but there is no guarantee that
it will be respected by others. Probably will be, though. If the goal
is consensus, why not allow people to participate as they choose?)
We have not attempted to design DP as a software tool. Our focus has
been on the personal connections, and analyzing the *significance* of
proxy assignments, which is what software tools would do, has been
considered secondary, and, in advance of actual applications,
probably a distraction. However, handling multiple proxy tables and
the like, as long as formats are compatible, is pretty simple. In the
FA environment, as well, security issues fade in importance. A few
sock puppet votes, or even many sock puppet votes, aren't much of a
problem, since (1) they don't bind and (2) it would be fairly easy to
smell a rat. If proxy assignments are public, and so are registration
dates and the like (as if, say, implementation is on MediaWiki), what
one would see with sock puppet proxy assignments would be very
different from real ones, and then we need to look at how proxy
assignments are used.... they are used by analysts. Who are the
analysts? Anyone who cares! If I'm a leader of a natural caucus, I
want to advise my clients how to vote *in a real election*, say. I
will, through the structure, attempt to negotiate a consensus,
because consensus is powerful. I see that we have found 40% of votes,
using DP, for a satisfactory outcome. Is that enough?
It is up to me to decide. For now, the point is that I might look at
what is in opposition, and I discover that a huge block of votes, say
30%, are from mysterious origins, smells like socks to me. And so I
conclude that, in reality, we have 40/70, a majority. Is that enough?
Depends. I can continue to negotiate to find higher consensus, or I
can decide it's enough and proceed. The higher the level of true
consensus I can find, the easier will be the implementation; on the
other hand, it may require postponing decision too long. It's a
deliberative process, and decisions are made by free agents acting
independently but in communication with each other through an
efficient structure. There is another name for this: intelligence.
This is what neurons do.
The FA/DP concept could revolutionize politics as we know it,
*without any changes in legal structure.*
There is are two major obstacle, each with two faces:
First, apathy and cynicism, which feed on each other. "It won't work,
nothing will work." "They won't let you do this." "It will be corrupted."
Second, there is a very natural filtering mechanism that we use to
prevent information and processing overload. New ideas are, pretty
much, automatically rejected simply because they are new, and this is
actually quite efficient, for most new ideas are not worth the effort
it would take to understand them. Naturally, we have filter bypasses,
or else we'd never be able to find worthwhile new ideas.
How to bypass the natural filters is the essential problem facing
FA/DP theory today. Some basic mechanisms are fairly well understood.
For example, it's been my experience that if I raise the issue of
FA/DP with someone a year later than their first contact, there is
far more receptivity. This points to a bypass: persistence of the
concept over time, which not only shows that the idea wasn't just a
wild imagination of a moment, but has some depth. Another bypass is
multiple sourcing. When we hear of an idea from more than one person
considering it worthy of attention, we likewise will be more inclined
to think about it.
The presence FA/DP plan (or at least my plan) is to simply discuss
the ideas, to implement them in projects where possible, and to allow
the ideas and the projects to percolate. It's possible that some
major implementation will appear that could accelerate the process.
FA/DP was proposed for Wikipedia and was (as I expected) roundly
rejected. First time. Based on the usual *total* misunderstanding
that it is about voting, and, hey, what about sock puppets and end of
discussion. But it's been mentioned, now, over the six months or so
since the first formal proposal, a few times here and there. I've
become much more widely known and trusted. In another six months or
so, there may be another formal proposal. While I would not go so far
as to predict success, it may start to happen that significant
numbers of proxies start to be assigned. *FA/DP does not require
central approval.* One thing proven by the initial proposals was that
(1) attempts would be made to crush it, and (2) those attempts would
likely fail. That is, the proposal was rejected, but the quite
remarkable (and anomalous) attempts to crush it, actually delete the
proposal, erase all trace, were likewise rejected. So what remains as
an obstacle are only the standard ones listed above.
Because FAs don't take, as an FA, any controversial position, they
are in opposition to nobody. To me, the theory indicates that FA/DP
could function even under relatively repressive regimes, particularly
in China, as an example. The FA would initially serve to develop and
coordinate public consensus to support the noble goals of the
Communist Party (and a particular example and organizational impulse
might come from environmental protection, which is an official Party
goal). But the medium is the message. I'd predict that the FA would
never actually oppose the Party, but it would, rather, become an
institution whereby the people advise the Party, coherently, and in
opposition only to those who, themselves, are the ones who corrupt
the Party and make it into an oppressive instrument instead of what
Marxist theory would have predicted and wanted. And it would do this,
not by actually opposing those ones (scapegoating) but by making them
less relevant, and those individuals would be continually invited to
participate and exercise proper influence. Ultimately, the Party
would be transformed and become what it should have been from the
beginning, if it had not been organized on what must now be seen as a
relatively traditional, oligarchical plan.
The hazards are many, but the theory is that FA/DP will make
decisions that are maximally intelligent, that are wiser than what
traditional structures (including purely anarchist ones) have been
able to make. If the students at Tienanmen Square had been organized
with FA/DP concepts, the history of China from that point on would
have been radically different. The government was actually
negotiating with them, there were sympathetic elements in the Party
that supported the students, that were excited by what they were
doing. But because there was no coherent organization, but only
firebrand leaders, in addition to some sober ones, there was nothing
to negotiate with and the firebrands, the loudest voices, were
seeking nothing but the total humiliation of the government. That
doesn't fly in China. The government took its only option
(considering their own limitations as well), they imported troops
from other regions that did not speak the local language, and crushed
the rebellion (which had involved massive support from the local
Beijing workers.) From the point of view of responsible government,
this was almost their only option, as tragic as it was. With FA/DP,
I'm pretty confident, the students would have negotiated changes,
would have peacefully dispersed, the crisis would have been averted,
the Communist Party would have become stronger, more popular, and who
knows where that would have gone.... but probably thirty years more
advanced toward a vibrant, healthy Chinese political life. And a
world political life, for the example would have been imitated elsewhere.
So, for now, it is quite enough of a task to disseminate these ideas.
If the right people in the right place understand this, the seed
crystal will form. My theory is that society is currently
supersaturated, looking for this, massive crystallization could occur
practically overnight. But first there must be a crystal of
sufficient size and persistence to not be broken apart by the random
fluctuations that afflict very small organizations. I do not know how
large this crystal must be. It could be as small as a hundred or so
individuals who understand the concept. Progress toward this, in my
view, has been ongoing, but I can't predict how long it will take;
except I'll say that I expect I'll witness it before I die. I'm 64,
which gives me roughly an estimated twenty years.
At this point, Delegable Proxy has possibly enough mention in what
Wikipedia considers "reliable source," to justify a return of the
article (which was deleted as part of the reaction to the proposal on
Wikipedia, and, at that point, the deletion was not outside of
notability policy). But I can't do it, I consider myself to have a
Conflict of Interest.... I will, however, have the article restored
to my user space, it may be time to start working on it.
----
Election-Methods mailing list - see http://electorama.com/em for list info
