For me, presently the most impressive network in the world is not the network of email traffic, Facebook friends or cell phone calls. Both for sheer size and complexity, the personal network must surely take the cake. In many ways, it is not merely a superset of most communication networks. It is something different. It exists in minds. The rest is a proxy. This is the network of mutual acknowledgement, of individuals as particular beings, rather than as signifiers of something larger, such as a town, ethnic group or any classification. Like rebranding bank machines as “Personal Touch” machines, as Royal Bank did in Canada, rebranding this network of particular relationships as the personal network, we make it sweet and homely. It softly bumps up against community, and might mutter “sorry” for complicating terminology. It is anything but homely. It is not simply a friend when you’re feeling down or a neighbor to loan a cup of sugar; it is almost everyone who knows anyone.
In my work I show samples, like microscope slides, of this personal network. They are not perfect. But it is worth knowing that the people who try to access this network in a measurable way realize they are not perfect. But nor are they completely made up and hopelessly useless because of recall error. Firstly, it is not “error”, it is bias, and it is not noise, it is difference in how people understand the world.
Such samples, done through direct questioning are a snapshot of the world as is understood, rather than the world as made (via telephone calls, emails or twitter feeds). At a fundamental level, there is no need to be sneaky about this network, or seek to assert the primacy of the ‘communication network’ simply because it allows researchers to scale up their ventures more rapidly. The fact that humans presently cannot read each others minds, or directly access these minds should not be forgotten, or ignored. It is this limitation that prompts us to use representations like a call graph. But it is so obvious, or at least so taken for granted, that we often forget this network of minds is the real target. The fact that we cannot read minds means that we must infer them, not that data stands in for them, not that behavioral traces encapsulate them.
Of course, not every large scale network concerns the personal network. There are many designed systems that benefit from the understandings brought on by massive scale network analysis; we need efficient power grids, transportation hubs, waterways, commodity distribution systems and so forth. But that should not blind us to the fact that as humans, it is the relations to each other that matter most. The rest is just a proxy.