indicators of statistical properties

[Leonard Soicher]

Dear All,

We have had a healthy discussion about where partial balance properties
should go.  Now it is time to *decide*.

Peter C's summary of the argument *for* putting partial_balance_properties
at the top level was:

> 
>  - it is of interest (in different ways) to both combinatorialists and
>    statisticians, and so doesn't really belong within either of those
>    two categories;
>  - it has several analogies to automorphism group which suggest a similar
>    treatment.

On consideration of Peter C's email, and a good look at the rnc, I would
add that there are also analogies to (the top level) resolutions element.
Both partial_balance_properties and resolutions are of interest to both
combinatorialists and statisticians. Both enclose (lists of) what are
structures in their own rights (resolutions and their automorphism
groups, resp. association schemes and their automorphism groups and
other properties).

I now believe that it *is* consistent with what we have done so far to
put partial_balance_properties at the top level, and we should now *do it*.

Peter C goes on to discuss RAB's suggestions about what to put in to this
section. Clearly, this will have to wait until after version 1.0 of the
rnc is published. 

Regards,  Leonard