I got a twitter (thanks @lindastone) about danah boyd’s SXSW talk on Privacy and Publicity, and read it online. A couple of days later it was summarized by Jason Kincaid on TechCrunch: http://techcrunch.com/2010/03/13/privacy-publicity-sxsw/. It’s a great talk, worth the read.
I’ve been doing very related research recently so I have been thinking quite a bit about the topic, one thought being I don’t really like the word “privacy” and I’ll explain why.
First, some background. This problem has grown due to several recent trends in social media. Social networking systems tend to provide tools for only a unified identity, reflecting the assumption that one identity fits all situations. In reality, people live very faceted lives, and a deeper understanding of how people manage identity across areas of their lives would greatly benefit the design of social media.
Second, within new forms of social media, sharing is often defined using a network-based model, where sharing is managed by a social connection, where two people explicitly articulate that they know each other and have access to each others’ content. These network-based sharing models present new challenges to the average user as she struggles to retain boundaries between areas of her life. Although Facebook is primarily a context for personal sharing, because of its increasing penetration, and the difficulty denying friend requests from people they know, many individuals find they have “friends” from multiple social contexts, including friends, family, and co-workers.
Finally, social media systems are increasingly pushing and pulling user generated content and activity streams outside the context in which they were originally shared. Explicit posts and implicit activity updates are scurrying about madly across Facebook, Twitter, social aggregators like Gist, and so forth. This final issue is at the heart of danah’s talk, where she describes it as a privacy fail on both Facebook and Google’s part.
I certainly agree privacy is an important topic of great concern. I just read a great academic article  that indicates that about 36% of people (admittedly, a biased sample of college students) have some aspect of their identity they make some effort to conceal. Similarly in my own research, I am finding about half of our users indicate they have “faceted identity” — where they have different sides of their identity that are expressed differently depending on the situation – and a third are saying their facets tend to be incompatible.
That said, I also have some reservations about concluding that users don’t realize how much their privacy is being threatened, and that’s why they continue to share content that inappropriately crosses life boundaries online (e.g. pictures of excessive partying with friends leaping inadvertedly across the stage, so to speak, from personal life to work). What I am also finding in my research, is that even though folks are worried about sharing in social networks, they still do it. Many people will continue to choose sharing.
The real problem for social technology designers, I believe, is NOT so much how to improve privacy controls or educate users in how to use them. Users for the most part already have privacy controls, and even when they are aware of these controls will chose to overshare anyways. Here’s an example from my own life. A friend from college posted somewhat embarrassing old pictures of me in Facebook and tagged me as in them. I left the tag up even though some co-workers can see the pictures. Why? Because I want some of my other old college buddies to find me. It reminds me of grocery store discount cards: i’ll gladly trade my privacy for five cents off my milk.
The point is, users really want to share, and many will choose messy, socially awkward sharing over not sharing at all.
The real problem is that users are forced to choose between two suboptimal alternatives: excessive privacy and inappropriate sharing. What users really want, and we should be focusing on in this age of excessive web 2.0 dissemination of social media across applications, is helping users focus their sharing to the right people, at the right time, in the right area of their life.
For social media designers, this is a very important distinction (even though you might argue it’s purely semantic). There’s just a lot more to be excited about when it comes to helping users optimize appropriate, targeted sharing, vs optimizing privacy. “Privacy” is a much dryer, yet fearful word. Who wants to say “I’m working on privacy” vs. “I’m working on optimizing sharing”?
 Quinn, D, Chaudoir, S. (2009). Living with a concealable stigmatized identity: The impact of anticipated stigma, centrality, salience, and cultural stigma on psychological distress and health. In Journal of Personality and Social Psychology 2009, 97, 4, 634-641.