PEP 308: Alternative conditional operator forms

[Carel Fellinger]

On Wed, Feb 12, 2003 at 10:53:22AM -0500, Terry Reedy wrote:
...
> 'Indifferent' literally means "I do not care whether Python remains as
> is (modulo changes in other areas of the language) or whether this
> change is made".

It must be my lacking English then, but "don't care" was not what I
had in mind when I wrote "indifferent".  Some "don't cares" care
enough to vote so calling them "don't cares" seems strange.

 
> > Sure you'll agree that *if* pro/against groups both were real
> minorities
> > that then changing the language was out of the question.
> 
> But here you suggest that 'indifferent' be counted as 'against'.  I

No I don't!  I'm conservative to any changes made to Python, it's
a wonderfull language as is, so any change should bear it's own weight.
And one way of weighting is to measure pro and contra against the
total population.  Laura hints at why it's difficult to get a correct
overview of peoples likings, you'll never reach the whole population
for one, but if you only count pro and contra you'll have no way to
put those numbers into perspective.  We have *no way* to establish a
relation between yes/no voters and yes/no `believers' in the total
Python community.  Counting indifferent voters too would be a way to
put the yes/no counts into perspective.  True again, we have no way to
relate the number of indifferent voters to the number of indifferent
users in the whole population, so whether using the indifferent votes
to weight yes/no votes might be silly, but I feel left out otherwise.


> don't agree.  It would be just as sensible to me to either split the
> 'indifferent' vote or distribute it in proportion to the against/for
> votes.  Concretely: if 2 people vote no, 20 vote yes, and 30 vote
> don't care, I disagree that the 30 don't cares should all be treated
> as nos.  The issue is binary; either a patch is made or not.

Well, 2 no, 20 yes, 30 don't care's: that's quit a lot saying yes
isn't?  The no-people, now they form a real minority.  So i would
think that in this case they outcome could well be yes.

On the other hand, 2 no, 2 yes and 48 yes: that's an overwelming
majority not particular keen on getting it in.  Or keeping it out:)
So in this case I would deem it wise to refrain from adding it,
and luckely to me that seems to match Guido's point of view.


-- 
groetjes, carel