On Sat, Nov 01, 2003 at 10:16:39PM -0600, Manoj Srivastava wrote:
> > As I said to Raul:
>
> > [...] if the technique is consistently used, and people don't
> > adapt their voting practices to compensate for it, that it
> > could result in zero progress in an infinite number of steps,
> > regardless of the actual collective desired pace of progress.
>
> Since you have yet to see this technique work even once, I
> think that you have yet to make your case that it can work in the
> first place.
Interesting.
Your premise: I have yet to see this technique work even once.
Your conclusion: I have yet to make a case that it can work.
You are specifying your premise as a necessary condition for the
realization of the conclusion. Thus, by your reasoning, the only way
*I* can make a case that it can work is for someone to try it
successfully. Moreover, you are declaring that it is flatly impossible
to convince you (or anyone?) by any means short of an empirical
demonstration.
This seems a somewhat prejudicial approach, and inconsistent with the
grounds on which we adopted Condorcet Cloneproof/SSD as a modification
to our voting system. As I recall, the election-methods mailing list
folks warned us that our original ("Concorde") method was prone to
undesirable behavior about once every twenty votes or so. We felt that
was unacceptable, so we embarked upon a long process of revising our
voting system prospectively, before such an outcome came to pass (along
the way, as we know, the voting method resolution accreted a few other
changes, and became more of an omnibus resolution to overhaul many
aspects of our voting system).
Am I to understand that your position is that people who would use this
technique should get N free bites at the apple, until it works once?
Even if we can convince ourselves that it is possible, we do not want to
erect procedural preventative measures?
This does not seem to be an approach that promotes a robust resolution
procedure. Is it an approach you are reserving for this occasion?
--
G. Branden Robinson | I have a truly elegant proof of the
Debian GNU/Linux | above, but it is too long to fit
branden@debian.org | into this .signature file.
http://people.debian.org/~branden/ |
Attachment:
signature.asc
Description: Digital signature