There has been a lot of chat in the blogosphere recently about Facebook overtaking Myspace, then on the value of international traffic compared with US domestic traffic – both on Techcrunch and GigaOM, and finally asking if the implication of Facebook over-taking Myspace is that network effects for social networks have been overstated (this is a great post from Andrew Chen).

All of this has got me thinking that social networks are better understood as media or entertainment rather than a communications tool or network.

People talk more about ‘wasting time’ on Facebook than staying in touch with people or any other constructive activity. They profile hop, check out people’s pictures, and that sort of thing. Thought about in this way socnets can be seen as a game. The main activities are collecting friends and building out a profile. The brilliance of the Facebook API was that it extended the game by adding vampire bites and other similar activities.

Then like any game it started to get boring. The overall stats kept looking good because so many new people were playing the game that the growth kept going for quite a while, but as is well documented growth is plateauing now.

On this analysis socnets are vulnerable to the same here today, gone tomorrow, phenomena as any film or TV show. Their sheer size and presence in the lives of so many people gives them some protection – but if they don’t stay interesting they will start to fade away.

Om Malik put it this way in the GigaOM post I also linked to above:

In other words, Social Networks need to find new purposes for people to come back every day and be loyal. I had argued in my previous post that the world of social networks is going to be divided into two – the big players (MySpace, Facebook) and niche players (Dogster, Dopplr etc.) [I include the reference to niche, because niche drives interest]

The two main initiatives at the major socnets recognise this risk – data portability and open APIs are both attempts to become platforms that are an important part of some other interesting activity – that either happens inside the socnet via widgets of Facebook apps, or outside of it using the data from the socnet. Either way the socnet gains a relevance and importance that is lost once the user has finished building their profile, got bored of surfing other people’s photos and of sending vampire bites.

For a while now I have been saying that virtual worlds are best understood as media properties, perhaps analogous to TV channels. We all have favourites, but we have several that use regularly, and they all change over time. The good news is that this means there is space for lots of successful virtual worlds, the bad news is that none of them will be worth as much as Google, or probably even Facebook’s famed $15bn.

I am now thinking that socnets could be similar. Whether they are or not will depend on whether they successfully transform themselves into platforms.