On 2011-05-18 08:01, Kerry Raymond wrote: > I am somewhat curious about your question. You seem to be trying to come > up with some kind of "quality" or "genuine-ness" metric based on the > characteristics of a single tree. > > I would have thought most genuine researcher's databases would contain a > majority of people who are all connected by some sequence of > parent-child, spouse-spouse relationships (the "family tree", but > mathematically probably a connected forest rather than a tree) ...... > > Kerry > I often don't bother to distinguish between a tree and a forest. it might be viewed as either. I agree with your comments. I started out looking for some sort of "goodness" metric, but this quest has now become one of defining a set of metrics describing the characteristics of a file, for instance how densely populated is an ancestral tree. That is an example of how it can be quite tricky because there could be several such trees contained in a forest. In addition the population density might be different at different points in the tree. The overall intention is to provide an indication of how much further work there might be to do on that file. I am still looking at some of the example files suggested by others, but I expect to come back with a list of suggested metrics in a few days perhaps. Peter