ICANN/DNSO
DNSO Mailling lists archives

[ga-full]


<<< Chronological Index >>>    <<< Thread Index >>>

Re[2]: [ga] Voting rules, take 4


Hello Kent,

Wednesday, June 21, 2000, 12:24:14 PM, you wrote:

> On Wed, Jun 21, 2000 at 10:56:27AM -0700, William X. Walsh wrote:
>> 
>> No, this method is very flawed.

> It is not any more flawed than any other voting system.

>>  People can vote yes for something
>> that they think is...a bare minimum acceptable, and it would win if
>> the most people thought that, even if the option that the most people
>> would PREFER has over 50% yes as well.  You end up with the less evil
>> rather than the most good.

> For example:

>     Option A gets  65 yes, 35 no
>     Option B gets  51 yes, 49 no
>     Option C gets  20 yes, 80 no

> You are claiming that there are some situations where Option B would be 
> the "best" one, because it would be the "most good" instead of the 
> "least evil".  I'm having trouble making any sense out of that.


Things almost never end up that simple, Kent.

Most likely a better example of what I am saying is:

Option A gets 75 yes, 25 no
Option B gets 65 yes, 35 no
Option C gets 20 yes, 80 no

And where most of the votes for option A were by people who supported
Option B above all, but found Option C so reprehensible that they
voted for Option A as well as an alternative just in case, to block
Option C from being selected.

In that case, you have a situation where the consensus of hte group is
that Option B is the best choice and the consensus of the group, but
Option A was adopted because those who voted for Option B found it
palatable compared to Option C and there were a small number who
actually prefered Option A.

-- 
Best regards,
 William                            mailto:william@userfriendly.com


--
This message was passed to you via the ga-full@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga-full" in the body of the message).
Archives at http://www.dnso.org/archives.html



<<< Chronological Index >>>    <<< Thread Index >>>