I have a 74-year old aunt who lives in a small town in the mountains in Greece. She has never used a computer. But when she asked me what e-mail was, I could explain it to her easily. It isn’t different from paper mail in any essential way. It moves faster and it’s cheaper. It, so to speak, smells different, but it’s pretty much the same thing.
But Facebook and Twitter are not things I could ever explain to my aunt. There is nowhere to begin. I can tell her what an individual does on Facebook, but this gives her no insight into the thing itself. The network cannot be explained without context, because there is no straightforward answer to the question: what is it for?
Likewise with Twitter. That people send short messages to each other on Twitter says nothing about its texture. The substance of Twitter, like that of Facebook, is its structure.
The structure of the network governs how information diffuses on it. If I post something on Facebook, or if I tweet it, who will see it, and how fast? The answer to that question dictates what it is I will post. Both networks are odd amalgams of the personal and the public. But the mix is distinctly different.
Few people send out baby pictures on Twitter. The interesting thing about this is that it is not by design. Had Twitter evolved into a baby-picture rich medium, as long as it had plenty of users, its creators would have been happy with their creation. But it didn’t.
The difference between Facebook and Twitter stems from one simple fact. To understand that fact, one must visualize them as graphs, which mathematicians define in the following way. A graph is a collection of points - called vertices by mathematicians - and of lines between the vertices, which are called edges. The vertices are individuals who are on Facebook, or Twitter, and the edges are the links between them. The edges on a graph can be arrows - pointing in a specific direction - or they can be “undirected” as just a line between two points. (See here for an illustration of this.)
Think of an undirected edge like a two-way street between vertices - there is no privileged direction. A directed edge is like a one-way street: it points from one vertex to another.
On Facebook, the edges are undirected. This is reflected in its semantics. Friendship is not directed. To say: “we are friends” is to imply a symmetry. I am your friend, and you are mine. Contrarily, on Twitter, the edges are directed. “She follows me on Twitter.” This is an arrow (the nomenclature is sort of backward here) from me to her—I am sending information her way, with no expectation that she send me information in return.
This difference between directed and undirected edges is what creates the different characters of Facebook and Twitter. We decide whether to follow someone in a fundamentally different way than we decide whether to accept a Friend request on Facebook. We make this decision differently because the reciprocity inherent in Facebook makes it, essentially, a gift exchange.
I give you the gift of my information (be this articles in Der Spiegel or pictures of my vacation) and you give me yours. This worked really, really well in the early days of Facebook, when the people who were on it were peers. They knew what sorts of gifts one another liked. But my parents’ friends don’t know my tastes well. Neither do my work colleagues. And I don’t know theirs. So the social texture of Facebook started breaking down when it grew too large.
The directed nature of Twitter, on the other hand, made it evolve in a totally different way. Because there was no expectation of reciprocity, people were more hesitant to share personal information. Few people are tempted to randomly send pictures of their kids to strangers. The brevity of Twitter is peripheral; it is evaded by routinely linking to longer content. The basic difference is that a tweet is an announcement; a Facebook post is a gift.
The individual decisions about how to use each network were shaped by its emergent structure, which was a result of the nature of its edges. And those individual decisions then created social norms.
The innovative thing in Google+ is the following: Some of the edges are directed, and some are undirected. This mixture of directed and undirected links will govern the evolving structure of Google+. Twitter allows directed links only in one direction. (If I follow you, the information flows from you to me.) But Google+ allows directed edges in either direction. By adding you on Google+, I can either push information to you, by sharing things with you, or I can pull information from you (like following on Twitter), by adding you to what Google+ calls “incoming”.
The “circles” that many have touted as the fundamental sociological innovation of Google+ (including my old Economist colleague Andreas Kluth, in an interesting, but I think ultimately incorrect essay) will not attain the significance they are currently given. Though Google has designed a fantastic user interface to the circles, they are still artificial privacy settings, and as such, they are bound to be cumbersome.
They will be useful for very specific purposes: I am selling my old sofa, and only want to inform people who are geographically close, and thus in a position to buy it. But most of the way we use the Web (a quip, an article about Yemen, a nostalgic photograph, a parodic video) we want to spread to more or less the right people, not to a rigidly defined group of "people I am comfortable sharing a nostalgic photograph with."
Apart from a subset of the population that worries ceaselessly about privacy, most of us care who the information is likely to spread to, not who could find it if they searched for it doggedly. And the likely spread of information is governed by the structure of the network, which in the case of Google+ will depend on what fraction of its edges end up being directed, and what fraction end up being undirected. This is impossible to tell in advance. But it will be the key to understanding Google+.
In part, we want to avoid sharing too much, or seeing too much: A joke in poor taste in the hands of a prospective employer, a wounding picture of an ex with a new squeeze, the vacation pictures of the high school acquaintance you never really liked. But we also, to varying extents, want to be part of the conversation. 'This article about the debt ceiling explains the debt ceiling better than other articles, friends, and you will want to read it. This picture I took of the harbor is persuasive, and you should see it.'
Figuring out the tension between these two desires is a matter of trial and error. Google Wave failed because it got this balance wrong. Google Buzz failed for the same reason. This balance between seeking attention and seeking to avoid it is dependent on the structure of the underlying network.
The mix of directed and undirected edges will give Google+ its character. So long as Google+ reaches escape velocity, achieving the scales necessary for the emergent behavior to take hold (which, at this point, looks likely, given Google’s commitment to it), it will be a new thing entirely. It will not be something halfway between Facebook and Twitter. It will be different from both of them. I couldn’t tell my aunt what it will be. In fact, until it succeeds in becoming itself, I can’t even tell you.