[conlang_learners] Now or in September?

[Jim Henry]

On Tue, Jul 14, 2009 at 11:41 AM, Philip Newton<philip.newton at gmail.com> wrote:
> On Tue, Jul 14, 2009 at 17:33, Jim Henry<jimhenry1973 at gmail.com> wrote:
>> With range
>> voting, or the variation of it that Larry proposed, you have to choose
>> between helping your favorite option as much as possible, or hedging
>> your bets by spreading your votes out among several acceptable
>> options, to minimize the probability of something you don't want being
>> selected.

> That sounds like a reason to use straight range voting -- rank each
> language on a scale of 0-5 (or 0-10 or 0-100 or whatever), without the
> limitation on total number of votes.

That would be better, but if I'm not mistaken is still has a similar
disadvantage relative to IRV, if in a lesser degree than Larry's
variant.

> Then you could rank seven languages as (for example)
> 100-90-87-85-13-2-1 if you wanted, or 100-100-100-100-100-0-0, or
> 100-80-60-40-20-0-0, or anything else.

Suppose that, without your vote, it came down to a  near-tie between
two conlangs, both acceptable to you, but of which you strongly prefer
one to the other, and of which your favorite would lose by a small
margin without your ballot being included.   Then your best strategy
would be to rank your favorite as high as possible and all others as
low as possible.   That is, with range voting there are still some
situations where you can hurt your favorite by giving a nonzero
ranking to your less preferred acceptable options; that would never
happen with IRV.

But the fact that range voting lets you finely gradate your
preferences, which IRV doesn't, may be a good enough reason to prefer
it.  And range voting is easier on the vote-counters.


> It would also make it easier for the vote counter since they wouldn't
> have to add up the numbers to make sure the voter hadn't spoiled their
> ballot by using more than their number of votes.

That's important; also easier for the individual voter, I expect
(relative to Larry's proposal; I think it's equally easy on the
individual voter as IRV).

-- 
Jim Henry
http://www.pobox.com/~jimhenry/