Identity
Filter
Visitors
(0) (0) (0) (0) (0) (1) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (61) (0) (0) (0)Subscribers
(0) (0)
Highest posts
Latest posts
Links
Blogs
Democracy
Search
[top-politics] Dance-Monkey algorithm
-
At 03:41 PM 9/5/2006, Mark wrote: >-M: Unlimited depth – like PageRank. >But I notice that it is only one delegate per voter.
Mark assumed this, it was not stated. In fact, what was stated implied one delegate per issue. I’m not sure that was meant.
I have assumed that in an FA/DP society, that is, a society which was extensively making use of FA/DP concepts, an individual might efficiently and productively belong to many of organizations, perhaps hundreds of them. Not only might they have a single proxy in each one of these, but they might have more than one in a given organizations.
I’m proposing that organizations set up data structures that can be analyzed flexibly. The basic data structures, the simplest ones, would be a proxy file plus vote files. Then, to be sure, there may be tools to analyze these files. If people in an organization find it useful, there can be more than one proxy file.
It is also possible to build systems that will assign proxies based on some set of criteria. However, my concern is that such assigned proxies, with present technology, would not be particularly likely to create the personal, bidirectional connections that DP, as we foresee it, should create.
That is, you might build information-filtering systems that would function better than what we have currently, but
information-filtering is only one aspect of DP. The relationships of trust are really the core. If you are inclined to trust a machine to tell you whom to trust, fine. I’m not….>Again, because there is only one, there is no ability to fork around >dead ends and loops. This creates instabilities and is overly >centralized compared with having multiple proxies. Well, what is your >response?
Again, the “only one” is for the simplest system, in an FA. In an FA, there is no need to formally “fork around” dead ends and loops. At most, what is needed is to inform loop members, for example, that they are in a loop. (If everyone assigns a proxy, there are no dead ends. But there can be, indeed there must be, loops.)
The problem with loops and dead ends is the same problem: under certain conditions, they can result in a member not being represented. In my view, the simplest and safest way to deal with this is to inform the member who is not being represented. As to alternate proxies, anybody is free to propose that another field be added to the proxy list. If proxy A does not vote, then proxy B.
My opinion is that smaller organizations won’t need this. DP creates, however, in theory, organizations that can change rapidly. I don’t see DP organizations developing the kind of inertia that I’ve described so many times that it has come to be called the Lomax effect by some.
So if there is a need for alternate proxies, not a problem, it can be easily done. Unless everything is written into the basic code for the main communications structure of the organization. Which is why I want analysis to be distributed. In FAs, we can get away with this, particularly because the use of the information is distributed. In governments, the usage is centralized, so the exact mechanism of analysis becomes crucial.
> >ec: One can change anytime his actions. A direct vote will, of > course, override an indirect delegation. > >-M: ‘of course’ – again, why the override?
Because the proxy is two things. The proxy is, for himself or herself, a free actor. But in casting proxy votes, the proxy is only a servant of the voter. The voter clearly has the authority to override any decision of the proxy. Or what we have is not a proxy, it is an assigned representative, or a representative who was at one time chosen, but who now has assumed authority superior to that of the client.
Mark may want this, may consider that in some way it will institutionalize government by the elite, but I won’t buy it, and, indeed, I wouldn’t buy any system that did not allow such an override. Consider that my free choice, a personal one. And I might very well be buying a system.
I would not buy stock in a corporation that did not allow me to personally vote that stock, that required me to choose a proxy, in advance, and that then barred me from voting because I had chosen a proxy. It’s not common law, it is not, indeed, that anyone with resources and choices would accept.
> Why not let the voter decide >if there is going to be an override or not? Maybe the voter wants both >a direct vote and more than one delegate. SD2-S allows for this. How is >your way better?
Huh? “Let the voter decide if there is going to be an override or not?” Doesn’t the voter decide this by voting? If I want my proxy’s vote to stand, why do I bother voting directly? It’s work, you know. I’ll only vote if I consider myself competent. And who is to say who is more competent, me or my proxy?
I’ll tell you. I am. Legally. That is, I have the right to choose to act directly or to assign my rights to a proxy. This is common law, it’s been that way for centuries. And now Mark comes and asks us why. No, he should justify why not, since he is proposing a serious change to the status quo regarding proxy voting.
Of course, he does not use the term proxy…. which is correct. He is not apparently talking about proxies, but about something else that merely resembles a proxy in some way.
>-M: SD2-Smartocracy offers: >1. general trustees and/or >2. issue position and/or >3. specific delegates and/or >4. decision thresholds and/or >5. deliberation thresholds > >The only two that are mutually exclusive are #1 and #3. >This yields about 20 different combination possibilities for the voter. >How is your way better?
That’s easy. Basic DP offers two choices to the voter: participate or delegate. If you delegate, you choose one person to represent you (to an organization, meeting, special topic or issue, if you want, but what is fundamental is: pick the person in this organization whom you trust most. If that person is too busy, that person will make a recommendation to you as to who would be a good proxy. Follow that unless you’ve got a good reason not to. After all, if you trust the proxy, in a delegable proxy system, you are effectively trusting whomever the proxy trusts….)
Simple. This simplicity is what would make it practical to participate in hundreds, maybe even in thousands of organizations.
But, once again, if the members of an organization find that they need the additional complexity, it can be easily done. The matter of decision threshholds, I commend Mark for mentioning it, but time does not allow full discussion of it.
My view is that the majority in any meeting has the right of decision, but that a wise majority will seek broader consensus, if it cares about the unity of the organization. In an FA, it becomes moot, because the various factions that may exist may form and act through caucuses without any consent from anyone else. Seeking broader consensus is voluntary. The only coercion is the natural one: if you act before you gain broad consensus, you may end up wasting your efforts. A two-thirds majority with an equally determined one-third in opposition ends up with a vector sum of one-third, compared to near-unity for substantial consensus. Consensus is powerful.
But institutionalized rules requiring supermajorities, and I’ve certainly worked in plenty of organizations that had them, generally lead to forms of minority rule, where the status quo favors a minority. It would take elaborate systems to get around this, and it is much simpler to recognize the basic right of a majority to take action when it sees the time ripe for it. And democratic societies generally recognize this.
Decision threshholds are a matter of meeting rules, and the common Robert’s Rules are excellent as a default set…..
> >ec: Some differences: > > – no rank or other such silly concepts > >-M: Delegates all have differing voting power relative to one another. >This is rank whether it is called ‘rank’ or not. (I have said this >before, and this is an uncontested point of mine.)
I’ve used the term “proxy rank” to refer to the relative voting strength of proxies. Because, in the systems I’m considering, the actual vote of a proxy is not an independent variable, it depends on who else votes, the vote count on which proxy rank is based, as I’ve used the term, would refer to the number of votes which would be counted for a proxy if only the proxy voted. Because of loops, there is a little bit of manipulation needed to get a reasonable ranking (otherwise all members of loops would have the same rank) — but I won’t go into that here.
Proxy rank is one means of determining the right of participation in meetings. However, it is not the only one. Still, a system which automatically calculates proxy rank could be used to provide automatic provision of posting rights in a meeting-list. I’d guess that this is what Mark is talking about.
I do consider that communications restrictions are essential to the scalability of any democratic system. Otherwise the noise becomes overwhelming. However, I’d leave to each meeting (physical, mailing list, forum) what restrictions are appropriate. Small meetings need none.
FA/DP systems would be direct democracies in the sense that the right to vote is not alienated from the members (there is a double protection because votes don’t bind power; that is, an FA cannot command a member to do anything. Nor does it collect significant assets and spend them by majority vote, so it can’t spend the funds contributed by a member without that member’s consent.) However, direct democracies break down when the scale gets large — this is the problem we are all trying to solve, I think — and this makes the restriction of communication essential. To say it simply, if everyone can speak to everyone, almost nobody will hear anybody. There has to be a means of limiting the traffic. Filtering, in a word. In a DP system, the proxy is the filter.
And access to high-level meetings is through proxies. But voting is not so restricted, so, quite obviously, whatever constraints exist on communication to that meeting must be acceptable to, at least, the majority. And in an FA, if members don’t like that, they are totally free to form another meeting without such restrictions. If there are very many of them, I think they will soon realize their own foolishness….
>Does anyone have a better way?
Sure. Convince one person here that your proposals are best, and then let that person, shall we say, second your motion…. Simply proclaiming that your way is best isn’t going to cut it. Two people arguing for it is far more persuasive.
It might be best. I’m not arguing against it. But I’m far short of convinced that it is what we want now!
+1
