Do we all live on one massive social graph?
What if all human relationships were mapped, and if all your connections were always seamlessly available to you for some form of remote information sharing and communication? A good plan? This concept has been bouncing around more and more this year. Brad Fitzpatrick offers an enthusiastic overview of the idea in his hugely-viewed post “Thoughts on the Social Graph.”
I’m thinking about the patterns humans have made while socializing in other times and spaces. People have evolved with social and physical walls, borders, and segmented places where people can behave differently or be in different group contexts. We carry this desire to extend, block and retract our connections into our online interactions, so we create filtering, privacy controls and the ability to convene a new group and invite or exclude people.
Perhaps the most socially powerful online social context is the “Third Place” (as in the classic description of The WELL by Howard Rheingold.) We’ve all experienced some of that kind of scene at various times. It’s not home or work. It’s an interactive setting. It’s relatively open to newcomers. It’s informal. It’s not a confessional or a therapy session. It might not work for you if you dragged your boss, therapist, priest or parent in via an underlying grid of all your social relationships. It works because it is not universal or enduring, and parts of your graph are dark to people in other parts of it.
Historically, that great good pub in the village, while beloved and important, was not a place for blanket confidentiality. However, it was a place where there were certain understandings, and where the things you had talked about once you might not be talking about now because of who’s here or who you might have had a falling out with. It was easier to adjust with a smaller number of connections, face to face in real time. Those adjustments are still needed.
Groups of people have mixed degrees of cluefulness about what to gossip about, and when to be silent.
In fact, when individuals in an online site that functions as their third place decide such a spot is totally safe and secret, their choice carries a tiny but real chance of putting them and the people they talk about at risk by trusting the apparent context too much, socially and technically. A good privacy agreement may remind them, if they read it. Trust is risk-taking by definition, and gossip is a very old concern for human society. The third place and informal social information circulation also provides the chance of accidentally revealing things that make some corner of the world better. Those ideas or tidbits of social information want to be spread, so any confidentiality rule is usually unspoken, informal and situational.In a text world with archived interactions that presents a new set of social and design challenges. Modern culture is stumbling through the shards of broken learning curves, working out these dynamics.
So now let’s imagine that universal graphing of relationships fed into databases and extended out. It’s easy to grasp that each human could be seen to be at the hub of a number of lines representing relationships, and those connect to others in a classic simple sociogram. The hard thing to “graph” or design in an application is that not all lines are reciprocal and not all the people involved have the same information-sharing understandings. Also, as we all know, the trust and information-sharing understandings shift over time. People do or say things some people must not know about, and even if you are willing to be utterly transparent, that doesn’t mean you want to force the others in your invisible web of connections to be betrayed to third parties. You also may have no reason for secrets now, but in the future the laws, the government, your activities, your work and other factors may change radically in ways you can’t imagine. Your friend may flee an abusive relationship while you keep both in your address book. You may not stay in touch with someone you worked long ago, and may know they are now an agressively evangelical cult member who wants to recruit all your contacts. Contexts change, only partly because we want them too.
So network migration tools have to allow extremely thoughtful opting in, or people will have to abandon them at intervals, zeroing out their participation in that corner of the universal graph.
Maybe more modest goals are valuable. I have a few short term wishes:
– Make it easier to import information and to minimize spamming of your friends if one moves elsewhere and wants to replicate the network, without exposing their existence to one another in a new place or context. I think this is the problem people are experiencing, but I am not sure a universal mapping is needed to address it. This would mean designing for more elegant overlay of affiliation from multiple sites, with an intentional partially overlapping venn diagram outcome, rather than one grid to rule them all.
– Make it possible to preview new third-party disclosures that arise as a side effect of expanding a network. In other words, click on “see who will have new information about people other than you because of this action.”
– Raise awareness that the person you trust today may not be the person you trust (or even remember!) tomorrow, so perhaps we can start to both design and tailor our own user behaviors for shifts in affiliation and connections that age out unless renewed. (Some popular social network platforms are currently very bad at this!) Better to understand that than to see people discard one platform for another when too many contacts get stale or tainted by context changes.
I found a very informative post while I was musing about this issue, from microsoft research sociologist Marc Smith on Bill Johnston’s blog.