[gnso-osc-ops] meaning of 'in the lead' was Re: ACTIONS/SUMMARY: ...

[Avri Doria <avri@xxxxxxx>]

On 2 Sep 2009, at 22:33, Julie Hedlund wrote:

> 1. GNSO Council Rules of Procedure: Review Sections beginning with  
> 5.2 to the end of the document. See attachment.  Julie will get  
> clarification on suggested language to define the term “leading” in  
> the context of Section 4.1 Selection of the Council Chair.

On this one, I know, I would be much more comfortable with a dialogue  
on the list as to what 'in the lead' means then a dialogue among the  
staff that we are not party to.  I totally accept Julie coming up with  
a suggestion as was asked in the meeting, but would prefer that any  
dialog, including the participation of other Policy Staff, be  
transparent if at all possible.  Obviously people, including Julie,  
will talk to whomever they wish, but formalizing a request for staff  
consultation is what strikes me as non-transparent.
One thing I did fully understand during the meeting was why 'in the  
lead'  had to mean in both houses.  i guess i do not see the danger of  
one house having a clear advantage for the following reasons:
- by dealing with percentages in the houses, we effectively equalize  
the influence of the house's votes.  using percentages is a  
normalization.
- by requiring a positive outcome of 60% in each house for election of  
the chair, having an advantage in just one house in the previous  
ballot does not translate into an automatic victory in the runoff
- as long as the 'non of the above' option remains active in all  
ballots, it becomes impossible for one house to force its will on  
another.  just because there is only one candidate left, it does not  
mean that candidate will take the election.  people have to actually  
vote for her.
Assuming this reasoning is acceptable to others and we do move to a  
notion of 'is in the lead' meaning an overall percentage lead, i think  
the simplest result would be to do it mathematically.  The set of  
possible results is not very large and i  guess that either a formula  
or a table could be generated that clearly shows the leads.
since i tend to think in terms of formulae and models one that i think  
works for me takes its basis from the notion of proof in whiskey (i  
like whiskey and hence use it as opposed to some other alcohol)
200 proof = 100% alcohol

if we take 100% of both houses, we get to 200 proof

assuming that using percentage equalizes the influence of the houses  
then
assume house A and B with two candidates x and y

 A                    B
x    y             x   y

73 30          30 70   = 100/100 hence a tie and we go into election  
timeout
60 40          40 40   = 100/80   x is in the lead - x versus 'non of  
the above'
50 50          40 60   =  90/110  y is in the lead - y versus 'none of  
the above'
etc...

(the same formula works for candidates > 2 )


i.e. whoever has the highest Proof is in the lead.
does not _ win_ unless both houses meet the threshold, but the lead is  
clear.
thanks for reading

a.