When Files become Friends


Domus Magazin. April 2009

Someone to whom we are linked may be called a friend, and these links as friendships tend to spread. Friends of friends become friends. Not like in the 19th century, when rather the enemy of our enemy was to be considered a friend. Today’s neutrality of the link does not allow for enemies. Links are contagious, but they do not spread like viruses. Networks have experienced some jumps in their growth. At first they consisted in little more than a cable linking one mainframe computer to another. Instead of voices one would hear the beeps and blips of non-human technical beings talking to each other over the phone. One can still imagine those songs of electric cicadas on the backstage of the network. Their music only became social on the surface. The connections have reached out from the basic beginnings to ever-higher levels, each constituted by a protocol whose name still sounds familiar when it accidentally appears, hidden like an archaeological relic under layers and layers of communication. The Internet protocol IP still connects computers. The transfer control protocol TCP still serves to transmit files. And it was HTTP, the Hypertext Transfer Protocol, that pushed the link to the surface of readable texts. Thus it marked the human’s entry into the realm of links. At the fundaments of the so-called social network now reigns the FOAF, short for the format named Friend Of A Friend.

Was that development inevitable? Or did it just happen by chance? What will be the next steps? And where might they take us?

A key philosophical question consists in asking what something “is”. But this essential and ontological method of defining things is under threat, as Google’s core algorithm shows. A file’s page rank derives not from how its significance is defined, but from the sheer amount of links connecting to it. Being has been replaced by being linked. The definition of an object is no longer given in terms of existence. It appears as a vector of relations.

However, there seems to be a difference between objects or subjects. Some entities are linked passively, while others establish connections by themselves. But does it really make a difference? What would the word “themselves” refer to? We seem to enter a cycle here. Human beings are able to define their links, but at the same time they are defined by their links. A familiar effect shows up. Every time we accept a new friend, we necessarily appear as a friend on his or her side as well. We cannot avoid being simultaneously object and subject of that bidirectional relation named FOAF.

Let’s take a step back in order to look at how we have arrived at this point. All the networks we are talking about are overloading an underlying structure. The overload occurs at defined entry points. HTML – the hypertext mark-up language – worked as such an entry point. When Tim Berners-Lee came up with the idea to turn a word into a link with the command he created an entry point. He overloaded readable letters with a pointer to a file, expanding text to a network of data.

So where was the entry point of the social? Links exist only among files. They do not extend to the outer world. The world must take the form of data in order to be technically linked. So the entry point of the social came with the existence of a file that would represent our self. The basic attribute of that file, apart from being easily accessible, consisted in its links to other files of the same kind. Server-sided languages like php helped to solve the first task, allowing us to transfer ourselves into databases. The virtual self did not need to take the visible form of the human, as the adherents to immersive worlds imagined in the 1990s. It needs no shape at all. It is simply a file linked to other files that we may call friends.

What happened next? Three main facts quickly became apparent: the takeover of public communication, the reappearance of the time dimension, and the risk of enhanced control.

Networks of friends generate separated spheres. Earlier forms of communication migrate inside these semi-public spheres, as the examples of mail and event notifications demonstrate. Thus social networks can establish a new meta-format, taking over the dataflow as HTML once did.
As the friends are active users, time becomes an issue. User participation creates a constant flow of events. While the early Net had lost any sense of time, as pages were uploaded and left unaltered, now the dimension of time reappears. Facebook became a time machine, and with twitter that movement gained momentum. Soon the flow of events will be accompanied by an increasing amount of visual information, creating a hitherto unknown sense of timely presence.

But the future is not all bright and shining. As never before details of our lives are entering the web. Twenty years from now, someone looking back will see our times as obscure. We are about to leave the dark ages to enter the bright light of data exposure. This bears some dangers. Individuals will become their own profiles, traced in their slightest movements, intimate talks and private interests. Most network companies claim the data entered as their property, hoping to generate revenue by channelling consumption. Here another cycle comes to a close. The file called “friend” has been appropriated by someone other than it’s own self. And so it may turn into an instrument of social control. In the future we may need to reclaim the networks for ourselves. Not in order to abolish it, but to turn it into a kind of file-sharing framework of free friend-sharing.