[EM] Express any wish for democratic choice of poll-parameters. Do-able if desired.

Mon Apr 8 20:21:17 PDT 2024

On the other hand, nearly all of those computation-intensive methods elect
the CW when there is one…& there usually is one. So probably only the one
exhaustive pairwise/count would have to be done.

On Mon, Apr 8, 2024 at 20:06 Michael Ossipoff <email9648742 at gmail.com>
wrote:

> The hard thing about Voter’s. Choice is that we would  have to do a count
> for each method that someone specifies.  …which could be as many as 23.
> …all but 4 of which counts would have to be automated, due to the number of
> methods being voted among.
>
> On Mon, Apr 8, 2024 at 18:26 Michael Ossipoff <email9648742 at gmail.com>
> wrote:
>
>> Oh, another ballot-addition that I forgot to add:
>>
>> In. case anyone specifies STAR, my STAR-ballot:
>>
>> Alternative #1: 5 stars
>> Alternative #2: 0 stars
>> Alternative #3: 5 stars
>>
>> On Mon, Apr 8, 2024 at 17:45 Michael Ossipoff <email9648742 at gmail.com>
>> wrote:
>>
>>> Typo-correction:
>>>
>>> Here is what I meant to write in this paragraph.
>>>
>>> Whatever method you’ve specified for the count, that’s the method *whose
>>> winner* you’re counted as approving down to…in the base-method, which is
>>> Approval.
>>>
>>> On Mon, Apr 8, 2024 at 17:32 Michael Ossipoff <email9648742 at gmail.com>
>>> wrote:
>>>
>>>> I should add, to my hypothetical  Voter’s Choice ballot on those 3
>>>> alternatives, I additionally cast an Approval-ballot, in case some people
>>>> specify Approval:
>>>>
>>>> Approval set:
>>>>
>>>> {Alternative #1, Alternative #3}
>>>>
>>>> On Mon, Apr 8, 2024 at 16:58 Michael Ossipoff <email9648742 at gmail.com>
>>>> wrote:
>>>>
>>>>> When proposing this poll, I assumed that people would prefer the
>>>>> simplest & least elaborate poll-proposal.
>>>>>
>>>>> I wasn’t trying to be autocratic or favor my favorites when I
>>>>> suggested Schulze & Approval as the polling-methods. Schulze isn’t my
>>>>> favorite Condorcet anyway, but it’s popular.
>>>>>
>>>>> So I suggested what seem to have been the 2 favorite methods at EM.
>>>>>
>>>>> But it could be that I was wrong about that, because maybe the
>>>>> favorite has changed.
>>>>>
>>>>> Or maybe, contrary to what I assumed, SIMPLE & STREAMLINED aren’t as
>>>>> desired as much as DEMOCRATIC or THOROUGH or ALL-INCLUSIVE.
>>>>>
>>>>> A few people have suggested different methods for the polling.
>>>>> Democratically choosing that poll-parameter is easily achievable if people
>>>>> so desire.
>>>>>
>>>>> All that’s necessary is for a few people to express that wishes in
>>>>> this thread. If a few do, then there’d be justification for proposing a
>>>>> vote on the polling-method parameter. I hesitate to propose that vote now,
>>>>> with only two people having hinted about it—otherwise I’d propose such a
>>>>> vote now.
>>>>>
>>>>> So, if anyone would like for us all to democratically choose the
>>>>> poll’s voting method, then please either make that proposal, or express
>>>>> that wish, because it’s easily do-able.
>>>>>
>>>>> In the meantime, let me comment on 2 possibilities for that purpose,
>>>>>
>>>>> …& a possible (but not yet proposed) 3-alternative vote among those 2
>>>>> possibilities & the default stays quo (Schulze & Approval, with Schulze
>>>>> predominant).
>>>>>
>>>>> The Schulze & Approval status-quo, I’ll call “alternative #1.
>>>>>
>>>>> Alternative # 2:
>>>>>
>>>>> I don’t think you’ll like this one, because it takes twice as long,
>>>>> requiring 2 polls—the 1st one to choose the method for the 2nd one.
>>>>>
>>>>> Of course that vote could be conducted immediately… like right now if
>>>>> desired.
>>>>>
>>>>> To save some time, the existing nominations, + any additional ones,
>>>>> could be regarded as being for the 1st of those 2 votes.
>>>>>
>>>>> What method for the 1st vote? I suggest Schulze, due to its long
>>>>> popularity here.
>>>>>
>>>>> There isn’t much more to say about Alternative #2.
>>>>>
>>>>> Alternative #3:
>>>>>
>>>>> A long time ago I conducted an EM poll by a method that I call
>>>>> “Voter’s Choice”. It doesn’t require any agreement or choice on a method
>>>>> with which to start:
>>>>>
>>>>> Voter’s Choice:
>>>>>
>>>>> Each voter indicates 1) A method for the poll;
>>>>>
>>>>> & 2) a ballot for at least that method, & optionally also ballots for
>>>>> 1 or more other methods that others might specify, if those other methods
>>>>> require different balloting.
>>>>>
>>>>> e.g. Right now there’s interest in ranked-methods, Approval, &
>>>>> Smiih//Score (which requires both a ranking & a Score ballot). So then, if
>>>>> you yourself have specified a ranked-method, you post a ranking. …with the
>>>>> option of additionally adding an Approval-ballot & maybe a Score-ballot
>>>>> (for Smith//Score).  But you needn’t add those others unless you want to.
>>>>>
>>>>> What if the person specifying Smith//Score is the only person who
>>>>> posts a Score-ballot?
>>>>>
>>>>> No problem!! …as will be seen below, in the description of how that
>>>>> ballot + method-specification is counted:
>>>>>
>>>>> Whatever method you’ve specified for the count, that’s the method that
>>>>> you’re counted as approving down to. …in the base-method, which is Approval.
>>>>>
>>>>> Of course, optimally, on should approve (only) down to the CW.
>>>>>
>>>>> So, the optimality & effectiveness of automated Approval-vote depends
>>>>> on the Condorcet-efficiency of your specified-method.
>>>>>
>>>>> Optionally, you can, instead of specifying a count-method, indicate
>>>>> “Manual”, indicating that, instead of the automated Approval-vote, you want
>>>>> to have-counted for you the Approval-vote that you have posted.
>>>>>
>>>>> So, if only one person posts a Score-ballot, then, automatically, his
>>>>> top-scored candidate wins the Smith//Score count, & so he is counted as
>>>>> approving down to his top-scored candidate.
>>>>>
>>>>> If that sounds elaborate, it’s all reasonable & logical, with nothing
>>>>> arbitrary or unexpected.
>>>>>
>>>>> It eliminates Alternative #2’s need two elections, & it doesn’t
>>>>> require any agreement on a method for counting the poll.
>>>>>
>>>>> That’s Alternative #3.
>>>>>
>>>>> If anyone proposes, or indicates preference for, a 3-way vote,
>>>>> immediately, among alternatives #1, #2, & #3, then let’s take a quick vote
>>>>> on that immediately.
>>>>>
>>>>> …by what method?
>>>>>
>>>>> Either Schulze ( because of its long popularity) or by Voter’s Choice.
>>>>>
>>>>> Whichever one gets more thumbs up…either figuratively or 👍
>>>>>
>>>>> I’m not actually proposing that 3-way vote, but I’m inviting anyone
>>>>> who wants a democratic choice of polling-method to say so, & we can then
>>>>> proceed with it.
>>>>>
>>>>> I’m agreeable to that if it’s desired. If no one wants to go that
>>>>> route, that’s fine too, & that means that you want to just stay with
>>>>> Schulze & Approval, as initially proposed. The choice is yours—as of course
>>>>> it should be.
>>>>>
>>>>> In case people prefer a vote among those 3 alternatives, here’s my
>>>>> ballot for that vote:
>>>>>
>>>>> If it’s to be counted by Schulze:
>>>>>
>>>>> 1. Alternative #1
>>>>> 2. Alternative #3
>>>>> 3. Alternative #2
>>>>>
>>>>> (Of course, with 3 candidates, Schulze = MinMax(wv). )
>>>>>
>>>>> If it’s to be counted by Voter’s Choice:
>>>>>
>>>>> Specified method: MinMax(wv)
>>>>>
>>>>> Ranking:
>>>>>
>>>>> 1. Alternative #1
>>>>> 2. Alternative #3
>>>>> 3. Alternative #2
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240408/7c53823e/attachment.htm>