On Network Weaving, Valdis Krebs announces the "tipping point" for the term social graph. I followed some of the breadcrumbs to earlier blogs on the topic that Valdis referenced in this post, and am suitably complexified. Let's see how straight I can get it.
The term graph comes into the realm of social networks quite honestly: in mathematics and computer science, graphs are collections of objects that have relationships that can be represented as nodes and lines connecting them. (Thanks, Wikipedia.) So far, so good. But the question is, is the graph a representation or is it the real "thing?" The term social graph seems to be getting mixed up with the term "social network;" (Dave Winer) suggests that they are exactly the same thing. (So, apparently, does Wikipedia, so that if you search for "social graph" you land on the "social network" page.)
Valdis simplifies a long blog by Tim Berners-Lee by summarizing the insight as a set of layers, each of which can be described in 3-letter acronyms:
- III (International Information Infrastructure) - how computers are connected
- WWW (World Wide Web) - how documents are connected
- GGG (Giant Global Graph) - how people are connected
The GGG (in Berner-Lee's exposition) is the Semantic Web, which provides the enabling technology to reveal information about content and links: meaning. So, a social graph would be embedded with information about the connections among people.
How to explain all this in simple terms? I would like to say that the social network is the connections among people and the graph its representation, its map. Mapping is an attempt to shift something complex (relationships) into the domain of the complicated, where it can be revealed in mathematical terms. But the map is always (as I say when I give workshops on ONA/SNA) is just a snapshot in time. As soon as you capture the graph, the network changes.
Comments (1)
I find the word "graph" a little misleading. It's meaning may be too wide (anything "graphical"), or too narrow (mathematically, the graph is invariant to distortions and distances - it's only about connectivity).
And ontologicaly - what is it, then? It is noy the real thing - ceci n\'est pas une pipe, and moreover it does not have stable duration over time. Yes, it can be a snapshot from an observer's perspective. But even if we allow this (the shift from complex to complicated that you mention, plus from (multi)-local to global): why would someone want to do that?
The map is not the territory. Hillariously described in Umberto Eco's short essay "On the Impossibility of Drawing a Map of the Empire on a Scale of 1 to 1".
This is not to say that network is not a useful term. It does provide insight. But it's value does not increase if stretched too far.
Posted by christianhauck | November 29, 2007 1:04 PM
Posted on November 29, 2007 13:04