Category: Peer Networks

Facebook has been under constant fire for more than a year now and seems unable to answer its critics. Under such criticism the company’s executive team has promised to make user privacy its primary concern, until the next revelation exposes its duplicity. Now it seems every other week another article is written demanding that Facebook be broken up or regulated by government oversight.

We might wonder what exactly is wrong with Facebook and why can’t they fix it?

The answers are in the faulty logic of Facebook as a social network that connects the world and the financial business model required to fund that mission. Both efforts are fighting a natural contradiction when it comes to real reasons people use Facebook.

Let’s address the social aspect first. Facebook started as a on-campus online gossip network at Harvard University. This is the secret of its appeal – people like to gossip about others within their network of peers. The behavior went viral and expanded from Harvard to Yale and Princeton and other Ivys. Then it spread to universities across the country. Nobody really is as concerned with social status as young people between the ages of 13 and 21.

But then Facebook decided its gossip model should go public and proudly marked its rapid growth of the social network across the globe – to the tune of more than 2 billion users. We even got a movie out of it. But let’s consider the logic of such a global gossip network because, frankly, it makes no sense.

Gossip serves a very useful social and evolutionary purpose, despite it being popularly dismissed as “small talk” or “idle talk,” or even malicious or “nosy.” Robin Dunbar (he of Dunbar’s Number = 150) explains how gossip helps us maintain social relationships in groups and also helps community members sanction free riders or those who break established social norms (“Gossip in Evolutionary Perspective”). In this way, gossip provides a means of gaining information about individuals, cementing social bonds, and engaging in indirect aggression; helping people learn about how to live in their cultural society. Gossip anecdotes communicate rules in narrative form, such as by describing how someone else came to grief by violating social norms.

Certainly there appears to be something about gossip that is innate: our entertainment world is pretty much driven commercially by celebrity gossip. But we don’t know these people!

Dunbar actually extended his research to online social networks, specifically using Facebook as a test case of whether network technology relaxes the constraints that limit the size of offline social relationships (link). What he found was that the 150 number still holds for any meaningful social networks. In other words, the human brain is developed enough to maintain 150 social connections, after which the connections fall to the level of casual acquaintances. According to surveys, this is the experience of most Facebook users. Facebook “friends” are not really friends in the vernacular meaning of the word.

So, a network that connects us to roughly 2 billion users across the globe doesn’t make a whole lot of sense for the benefits of gossip. Rather, a gossip network that extends to people we have no personal relationship with tends to reinforce the negative aspects of gossip, i.e., meanness and rudeness. We can observe that celebrity gossip tends to focus on caricatures that emphasize the extremes of hero worship and cruel pettiness. In similar fashion, Facebook is very useful for small friendship networks that cohere around common interests or personal relationships, and the limits on that tend to approximate around 150 people.

The second reason Facebook is failing as a social network relates to its ad-driven revenue model. If I am using Facebook as a way to connect to my friends, I certainly resent a third party advertiser trying to insert itself into the middle of that communication channel (just imagine advertisers interrupting in the middle of your phone call!). How many of us were turned away from Facebook about 2+ years ago when our feeds were suddenly flooded with advertisements for things we had no interest in? The network data Facebook is selling to advertisers is weak, not robust. We know what our friends like, if they are friends, and Facebook algorithms do a poor job of approximating that. “Like” clicks are not really likes and digital advertisers know it.

The problem here is that Facebook ad rates are a function of the number of users FB claims to reach and the flow of network information across those user nodes, even if it’s Candy Crush games or humorous cat tricks. Facebook cannot really evaluate the subjective value of the information flow, so it merely sells it all in targeted user bundles. This does not serve end users (or advertisers) very well and the attrition rate is evidence of general user dissatisfaction. I would guess that most users stick with Facebook for the positive value they receive from far-flung friend networks and the lack of a viable alternative. But then we end up ignoring most of the white noise on our feeds, threatening the financial viability of FB’s revenue model.

So where does this lead?

Frankly, I would argue Facebook’s longevity under its current business model is challenging. Gossip makes sense and can be tolerated in small community groups, while wider social networks make sense if they are somewhat limited to common interests. Facebook “Groups” seem to exhibit some of these qualities, so perhaps that is a direction FB can move towards. But the problem then is that it is a much less valuable Facebook under its ad revenue model. Market competitions and alternative OSNs may eat into FB’s global network, forcing FB to adapt to a smaller footprint. That is likely to be a difficult financial adjustment for a company of FB’s size and reach. But technology cuts both ways and today’s Facebook may just be tomorrow’s obsolescence. Personally, I would prefer a social network that delivers more meaningful connections to other people and allows me to filter out a lot of the white noise. That can’t happen as long as the network servers make money off white noise.