Community Priority Evaluation

[Avri Doria <avri@xxxxxxx>]

This comment concerns the basis on which the calculations for Community 
priority ratings are done.

The Integer scale that the staff has picked does not allow for sufficient 
differentiation for those doing the calculations.  It is, I believe, a 
reflection of the one size fits all problem that ICANN often experiences in its 
implementations.  

In almost all of the criteria, the decision is on a 3 point scale, though in 
some cases it is only a binary scale.  The criteria, are not, however that 
Black, Grey and White (or Black and White in the case of the binary selections) 
so the discrete scale will end up evaluating various choices with the same 
metric even when they may be qualitatively different.  Since only 2 points can 
be lost in the evaluation, such a rough scale will exclude possible communities 
who have near misses on the criteria, but who must be marked down a whole 
point, instead of a more appropriate .5 or even .2 points.

I recommend that the evaluation procedure not use a gross integer measure that 
will miss the nuances in these difficult and crucial criteria.  I recommend 
that the scale be changed to a real scale that allows for discrimination in 
terms of tenths of a point instead of in terms of whole points.

I believe that making this change will make the decision procedure more 
sensitive, while still providing a bright line between those applications that 
merit community priority and those that don't.

Avri Doria
Individual Opinion