RE: Michele Friend's Institutional Compass.
The idea of a single compass is interesting. Practically, I believe that we will need to sort an aggregate compass into different compasses or sub compasses so different policy makers with different situations, models, and variety have compasses that are responsive to policy changes they influence. A compass needs to fit the terrain which they need to navigate their particular territories.
I am attracted to your Vedic threesome: Excitement, Suppression, and Harmony.
My threesome is Freedom, Solidarity, and Participation. In the extreme, mine may map well enough to excitement, suppression, harmony, but they have different overtones that may be important.
One thing that I do want to have a conversation with you about is the role of Conversation in change or stasis. I find myself more interested in differences of personal opinions among people who know one another or could reasonably be expected to get acquainted. Unfortunately, I have seldom seen raw facts (scientifically provable) change social behavior. Beliefs do affect behavior, facts, not so much. I think that the most efficient way to change behavior and even discover affective facts (science that affects behavior) is in conversations between people who are co-located and necessarily interactive.
So I want to use compasses to show differences that lead to conversation about differences that matter and that may lead to self-adjustments in beliefs or assessments of relevance (priorities).
I do believe that Vester's interaction matrix and its implications are critical to be added to the mix. I also think that Statistical Process Control (SPC) methods are useful to remove noise from signals. Stafford Beer used a Bayesian form of SPC taht I think we all need to be using. Bayes theorem should be used any time it is appropriate. It is the equivalent of contextualizing statistics in time, place--based upon what has recently actually happened. Lots of variation is eliminated when historic particulars are taken into account. See:
A Bayesian Approach to Short-Term Forecasting Author(s): P. J. Harrison and C. F. Stevens Source: Operational Research Quarterly (1970-1977), Vol. 22, No. 4 (Dec., 1971), pp. 341-362 Published by: Palgrave Macmillan Journals on behalf of the Operational Research Society Stable URL: http://www.jstor.org/stable/3008187