Deindividuation and Networks
Still thinking about that Guardian piece.
Obviously there’s no question that deindividuation is a phenomenon online. And yes, the way you get round that - unless you actively embrace it, a la 4chan - is by raising the social costs of bad behaviour. The way in which sites have done this in the past is by allowing community standards to emerge and then letting the community broadly police itself.
The standards that emerge in places like Metafilter, Reddit, Slashdot et al are very different: a lot of them exclude as much as they include, and in ways that don’t necessarily make those sites very nice places to hang out. But what they have in common is that identity is authenticated by the site and community, and rests on activity within the community rather than activity outside it.
Social networks, broadly speaking, work differently. Identity in social networks rests on offline identity and so its ideal guarantor is proof of that offline identity. One of the really fascinating things about Facebook - and a secret underpinning of its success I think - is that it’s one of the first online phenomena to have abolished the idea of the newbie. There are no experience counters on Facebook - no equivalent of “joining date” or “number of tweets”. There are network metrics, of course - number of friends - but they don’t necessarily correlate with experience. To be a noob on Facebook would be like being a noob in real life - which we do have words and ideas for, but not quite with the same flavour.
So social networks guarantee identity differently from online communities - but are those versions of identity actually in conflict? I think they might be. The relationship of the network to the community gets interesting at the content level. Communities online are built around communal content - whether published (by a blogger, newspaper, site owner) or self-generated. Networks are built around shared content, which works rather differently - it’s pulled from a variety of places and exists as part of a stream, not a destination.
This seems like a finicky distinction, but think about how it affects one particular piece of content - a newspaper article, for instance. The article is a piece of communal content - readers of the newspaper are likely to encounter it, and if there’s a community formed around the paper it will weigh in on the article. But it’s also a piece of shared content - someone who reads it can rebroadcast the link to a network and attract an audience who have no real connection with the paper or any community formed around it.
These transient audiences have no investment in the community, they simply follow interesting content and move on. So they also have no investment in any system of identity or standards that have formed within that community.
It’s also the case, though, that these transient audiences often represent a higher and higher proportion of a community’s readership. While average traffic still determines a site’s advertising rates, it’s increasingly misleading for thinking about how audiences actually behave. There are still loyal readers who visit a front page and then decide what they want to look at. But there are plenty of others who parachute in and out when sent to a particular article. And since it’s a numbers game, every site now offers plenty of “likes”, “+1”s etc - incentives to encourage more drive-by readers in.
Do they leave comments? Are those comments more or less likely to be constructive? I don’t know. (I’d love to know!) My hunch is they leave fewer comments and their comments are less useful. But it’s only a hunch. What I am convinced of is that this spiky, fluctuating audience makes creating a sense of community in a site a lot harder. Comments on communal content are an investment in the community. Comments on shared content are an investment in - what exactly?
Well, the personalised network. The Guardian piece that kicked these thoughts off opens with a series of excerpts from social media collected by Stewart Lee, all of which are threatening violence to him or are otherwise deeply abusive of him. Few of which are, strictly speaking, anonymous, only pseudonymous. The first compares Lee to a child murderer, and is posted from a Twitter account. That Twitter account presumably has followers, who are the audience “Wharto15” is addressing - trying to impress, provoke, make laugh. This isn’t some anonymous troll, it’s a person talking to other people who know exactly who he is. It’s just Stewart Lee (and the rest of us) who don’t.
Networks are formed of millions of sub-networks - the imagined audience (and therefore the imagined community) their users are reaching out to. Each has their own (privately negotiated) sets of standards - there are surely lines Wharto15 wouldn’t cross, where he’s happy to assume a comedian is fair game. These networks are essentially destabilising to online communities and their communal FAQs, cultures, hierarchies and codes of behaviour. If I was more economically literate I’d throw out a comparison to globalisation and government regulation here.
This destabilisation occasionally becomes opposition - the debate over online “real names” is part of this wider issue, I think. The community and the network definition of online identity are at loggerheads. The former says that identity derives from your activity in a community. The latter says that identity is both fixed and transferable across communities. The latter is winning - see this fine piece by Virginia Heffernan - but the triumph of network over community comes with real costs.
