Yesterday I gave a talk at an AAAI Symposium on privacy, and wanted to share a few of the thoughts I presented there.
A lot of attention has been paid in the past several years to the reputation angle of the social web, and especially the way in which this interacts with privacy. I talked today about one challenge in particular that existed even before the web became social, for which fixes may appear more readily as the web becomes more social. It is the challenge of giving users a modicum of influence over their self-representation.
My choice of words here is very deliberate. At the core of the reputation challenge as it relates to privacy lies an inherent tension between free expression and control. Historically, this tension has been managed by libel and defamation law, areas I don’t claim to be expert in. As the web has expanded access to publishing methods, as well as increased discoverability, and enabled the wisdom of the crowds to take hold, there’s an argument to be made that these current mechanism for handling attacks on one’s reputation just aren’t enough. I’m interested in helping people influence the representations that appear about them online.
Now, I’m not talking about situations in which the commentary being made about a person is untrue, situations that would typically be handled defamation or libel law. Instead, I’m interested in the commentary that is entirely factual, and which has been transformed into an indelible and inescapable mark on the Web. There are cases in which it seems appropriate to have such character statements be made permanent: elected officials, for example, go on the record throughout their careers in favor of or against particular policy positions for the purposes of allowing the electorate to make informed votes. But there are countless other examples of events that would 20 years ago have been forgiven and/or forgotten with the passage of time but have today become indelible marks on a person’s record. I call these look-back-and-laugh events, the things you did as a teenager or even as an adult that everyone looks back on and laughs about a couple years later.
The Wikipedia entry on Internet vigilantism lists several of the most well known examples of these look-back-and-laugh events that were exploited by the crowds and, in some cases, had detrimental implications for the reputations of those involved. I took a look at two of these and drew some conclusions about how engineers might push the envelope in the next five years and produce some really valuable solutions in the self-representation space.
The first example I looked at was that of Stephen Fowler. Go type that into your favorite search engine. You don’t even need quotes. What did you find? The first hit says stephenfowler.sucks.com, doesn’t it? And in the top 5 somewhere there is maybe a headline (now this varies by search engine) that reads, “The worst husband in the world,” right? Poor guy. I never watched the show, but I can’t imagine anyone deserves that. Most of us are kind people by nature, but this crowd behavior doesn’t appear to be very kind.
The second example worth mentioning is the Star Wars kid. This was an Internet meme centered on a teenage boy who had videotaped himself using a golf club as if it were a light saber. The video was uploaded without his permission and became viral, embarrassing him and leading to mass media attention. If you read the Wikipedia article, you won’t find the boy’s name, but click on the first Reference link and you will be taken to a BBC article that cites his name in the first paragraph.
There are a few conclusions one might draw from these examples. First, don’t do something potentially embarrassing if there is a digital recording device anywhere nearby. (That is, I think, wholly unrealistic advice today). Second, if you do something embarrassing and it draws attention … well, tough luck.
I’m not satisfied with those conclusions. There are a couple ways we might think about the problem. In the case of Stephen Fowler, how can we make it easier for him to participate in his own self-representation? I don’t mean, how can we make it easier for Stephen Fowler to get that content removed from the Internet – no, it’s factual, and free expression is important, so short of valid legal process it ought to stay. What I mean is, can we give Stephen Fowler a way to act as his own SEO, within reason? At the moment, he is at the mercy of the crowds. People publish – via a blog, or Twitter, or Facebook for example – commentary about Stephen Fowler and what a horrible husband he is, and all that content begins to link to each other. Some subset of it draws a lot of links, and thus becomes reputable. How can one man be expected to duplicate a network of millions of independent publishers, to write enough that his views would be noticed in that crowd? He can’t. We need to find a way to enable the subject of these look-back-and-laugh events to participate in the crowd in a vocal and reputable way, to stand up in front of everyone and tell his story (either of what happened, or how he’s changed, or what other things about him you might like to know), without being muffled by the raucous hooligans down below. In examples like this one, we need to give him a megaphone just to give him a voice. No one really knows how to do this on the Web today without distorting free expression and the actual wisdom of the crowds, but sorting through the tensions here would make a great research topic.
The additional challenge I see in all of this, and I credit the idea somewhat to Jonathan who planted pieces of it in my mind vis-a-vis his TED talk on kindness, is to build social signals into the Web that would more naturally enable us to behave like the community of Wikipedians who chose not to name Star Wars kid in their article. I’ve written about social signaling before, and it seems like every week I see a new application for its eventual implementation. In the context of self-representation, we need to give Star Wars kid or Stephen Fowler the ability to blush online, perhaps even cry online, to show the raucous hooligans we might become in crowds that we aren’t being particularly kind anymore. Let’s be honest: the crowd sometimes needs little reminders that we are all human and deserve a degree of compassion.
Ryan Calo is doing some work that looks at the intersection of legal notice and neuroscience (I hope that’s an accurate characterization…that’s my impression of his work). The basic idea is that we are hard-wired to respond to particular anthropomorphic cues, and he’s thinking we can use that hard-wiring to rethink the law around notice. I’d like to take that thought a step further. We’re clearly hard-wired to respond to particular anthropomorphic emotional cues in particular ways; most of us see a child crying and want to comfort it, for example. How can we take that basic reality of human neuroscience and transform it into a human computer interaction that will facilitate a smoother social web? I’m absolutely fascinated by this problem, and would love to hear from (or about) anyone doing work that touches on it.
My talk touched on a couple other ideas as well, but unfortunately in 10 minutes I really couldn’t do them much justice. Perhaps in another venue I’ll be able to deliver a more polished version with adequate time. If the paper that formed the basis of this presentation, which I co-authored with my colleague Alma Whitten, appears online in the near future I’ll post a link to it here. Until then, curious to hear your thoughts!