Menu

Sanspoint.

Essays on Technology and Culture

What The Next Great Social Network Needs

If anyone is going to break the hegemony of Twitter and Facebook for being the focus of our online social lives, they’re going to have to do what Twitter and Facebook—and Ello—are not doing. They’re going to need a understanding of how human social interactions work offline, and find a way to reflect that in interactions. The closest implementation of this is Google+’s Circles, though Google+ failed by requiring a huge amount of cognitive overhead to build and maintain them. Google at least understood that we don’t share everything about us with everyone we know. The binary nature of Facebook friends is not even close to representative of human friendship.

Getting this right goes a long way in solving most of the problems in social. It’s a huge step in preventing both violations of privacy, and preventing abuse. If Facebook understood this, gay students wouldn’t be outed to their homophobic families by joining a Facebook Group. Our relationships with our families, our friends, and our coworkers are all different. We even have different relationships with certain groups of friends. There are things I would tell my partner that I wouldn’t tell my parents. There are things I would tell my parents that I wouldn’t tell my friends—and vice-versa. There are things I would tell my real life friends, but not my online friends. There are things I would tell all those groups, but not my coworkers.

If you feel that you can be free and open with all the people you know, you have a luxury that most people don’t have. There’s always the risk that something you say or do online will become public. It shouldn’t be easy for this to happen. When a secret told to a friend leaks out, it’s a violation of trust, and that person should not be your friend. The structure of most social apps makes it all the more likely someone will make a private statement public. It’s caused by a mix of apathy in implementation, failure to understand the nuances of real social structures, and the needs of advertisers to see data before they give you money.

This is why it’s so bothersome that Ello is considered the vanguard of “private” social networking when its idea of privacy is just to not sell your data for ads. For real privacy, you’ll have to pay for it. If the choice came down to an ad-supported social network with fine-grained personal privacy controls, versus an ad-free social service that forced users to live in public (even with a pseudonym), I’d take the former any day. If half of the effort that went into Ello’s artistic site design and manifesto went into trying to find a smart way to incorporate the same nuances, filters, and limits in our online social lives that we have in real life, this is a debate we wouldn’t be having. “Friends” and “Noise” are not enough.

Understanding is the start. Implementing it the next step. With all the machine learning algorithms we have applied to our social graphs, it strikes me that it would be possible to algorithmically determine the relationship between two users—to a certain extent. If you’re friends with someone who shares your last name, for example, but is a few decades older than you, chances are they’re a parent or other older relative. So, the theoretical network can say “this person may not be someone you want to share everything with. Can I put them in your ‘relatives’ group?” The analytical tools used to target ads can be used to help users target the audiences of their posts. It would take some machine learning and trial and error, but at least they’d be doing something new in a moribund space.

Modeling real life, pre-Internet relationship models combined with strong and usable privacy controls that put real people in control of who sees what, when, and how. This is the real future of social networking. Getting it right, so it doesn’t require much more thought than just clicking “Post” is the hard part, but we have all those smart programmers and designers out there wasting their time on slicing and dicing data for marketers. There has to be a group of them somewhere willing to turn the tables. Once someone rolls out a service like that, I’ll be first in line to sign up—and I’ll bring as many friends as I can with me.