Paul Ford is talking about editors and editing the web, and I think he’s getting at another twisty strand of this thing I’m thinking about as I keep coming back to the problem of how best for humans to make sense of data. There’s no way around it—computers are very good at finding patterns and humans are very good at finding meaning within those patterns. Computers are not so good at finding meaning in patterns. We keep trying to skip ahead, asking machines to find meaning for us, but the most momentous way to use the computation power available to us right now is to have human beings edit the patterns machines find in data, rather than to have human beings edit the data itself to make it more palatable to the machines in hopes that the machines will extract something important for us. Machines think humans always want single answers when really humans just as often want to be able to wonder about things, to ask half-baked questions on a hunch and select or reject patterns in the same way. We want to be able to guide the computations toward the meaning we’re looking for.
And to step sideways for a sec and come at the idea of meaning from a different direction; I have a fuzzy theory that the meaning we are looking for has a visceral component. We need to make a connection that maps the external world of patterns to our bodies somehow, and we have to do it for ourselves. Mitsu often talks with frustration about how even when we (both the big we of Society and the small we of me) have precisely the information we’re looking for, we are often unable to bring ourselves to use it, or maybe even to really and truly believe it. Financial information from a time before we were born is a fine example of this. No matter how relevant, we can’t quite trust it. For some reason, once things fall outside of a certain scale—too far before or after our lifetimes, too big or small in comparison to our bodies—we seem unable to internalize even information we rationally find sound.
Part of the reason I am so fascinated with genomic data and personal tracking data is that I see it as a link between the scale of our bodies and historical, geographic, and molecular events that we might not otherwise find meaningful. (Ferchrissake, I can’t seem to stop saying meaning.) But imagine (and this is a totally scifi example, but still), if I were able to relate a present physical sensation to a set of epigenomic markers which in turn were associated with an event that took place in a specific geographic location some forty years before I were born. That historical event, the descendants of the people involved, the location, everything surrounding that event and the cascade of consequences take on a new kind of urgency for me. They are literally part of me.