The idea of a memory base is simple. We define a memory as a matrix, keeping counts in the matrix cells between the names of things on the rows and columns. Let’s say we had a memory of me, called “Person: Manny”. If I query the row called “City: London” and ask for associated columns for “Carrier: ?”, I’d see “AA” and “BA” for American and British Airlines. Moreover, AA would be returned with a count of 6 and BA with a count of 1. More than just the existence of my travel relationships, we can also see the strength of my travel habits – in the context of going to London at least.
The idea is simple but fundamental. When we started Saffron and began working with one of the big intelligent agencies, one true believer in what we were doing would provoke others by saying, “It’s all just counts. What else is there?” What did he mean? When dealing with the analysis of massive data, so much of what is computed needs to be computed over counts. More deeply, information and knowledge is based on the frequencies of what we see in the world. Counts are fundamental to knowing what we know.
