PEP 308: Alternative conditional operator forms

Carel Fellinger carel.fellinger at chello.nl
Wed Feb 12 04:27:53 CET 2003


On Tue, Feb 11, 2003 at 06:12:30PM -0800, Erik Max Francis wrote:
> Carel Fellinger wrote:
> 
> > When 50 of those are pro and 50 are contra, then Guido has a problem:)
> > Whatever he decides will upset half the population.
> > 
> > On the otherhand, when 1 is pro and 1 is contra and all the others
> > don't
> > care, then whatever he decides will be fine with the masses.
> 
> The issue with this analysis is that you'll never know how large the
> "don't care" camp is, almost by definition -- because if they don't care
> then they won't vote.

Not so, as I tried to point out.  There are those who *do* care, but
are indifferent to the outcome but for the fact that to warant any
addition the need should be substantial.

There are also those who are indifferent to the outcome but willing
to vote anyhow if they could only indicate their indifference.

Likely there are those in favour (or against) who don't care (enough)
to vote or who think the outcome is fixed so their vote won't help or
whatever.

Bottom line is that in any open voting, especially when the population
size is unknown, you don't know much :(


> Even if you hold a vote where you have a "yes," "no," and "abstain," and
> the numbers come in as (say) 45%, 45%, and 10%, respectively, that
> doesn't mean that 10% of the Python community at large doesn't care.  It
> could be that 99% of the Python community at large doesn't care, in
> which case you're in the second scenario you depict above.  The problem
> is, just by including a "don't care" on the ballot, you'll never be able
> to know that.

same holds for pro's/contra's.  Just think about all those who don't read
meandering threads:)

> By the way, Andrew Keonig's suggestion of a voting process where all
> alternatives are listed, including "no change," the "don't cares" can
> still indicate their position by submitting an empty ballot which
> selects none of the options.

great!


-- 
groetjes, carel





More information about the Python-list mailing list