Thus far, Rich has been trying his hardest, but hasn’t managed to do much for me. He doesn’t have access to the documentation for myReach, so he can’t give me any advice. He doesn’t seem to be capable of performing batch operations on my data according to parameters given in natural language. All he seems to be able to do so far is tell me about what I have.
What I think would be a useful function for Rich is for him to look through my data for nodes with high similarity, words and phrases in common, that have no current relationship, and then present me with a list of proposed relationships that don’t currently exist, sorted by his degree of confidence that there should be a relationship, allowing me to cruise through the list and accept or reject the proposals. And proposed relationships that are rejected should be noted somewhere, so that he won’t propose them again the next time around.