Using Social Network Sites the Wrong Way

This post was written in response to danah boyd’s post, “Facebook and Techcrunch: the costs of technological determinism and configuring users.” danah focused on recent (and not so recent) attempts by social network sites like Facebook to regulate how individuals relate to others when using their service. I noticed that danah’s argument—expressing a consistent point of view, whose development you can trace in her writing—reinforces the criticisms I made of the Spock service last December.

At that time, I pointed out that Spock was actively discouraging individuals from initiating and maintaining multiple online identities, but that this attempt to regulate user behaviours was incompatible with much social life as it is actually lived on and offline. And it is becoming evident that if valid user concerns and social needs are not met—in relation to issues like privacy, contacts management, and the overall balance of “open” and “closed”—then people will simply look elsewhere. The abundance of web-based services contributes to the way we think about these kinds of profiles—as social artefacts, they are pretty evanescent; especially, as danah has noted previously, from the perspective of teenage users.

The reference in the title of danah’s post to Woolgar’s (1991) influential concept of “configuring the user” wasn’t lost on me. And I’m familiar with one particular social scientific/research response to the problem of companies trying to configure their users (via bureaucracy, restrictive ToS, or whatever). That response is generally to say, OK, well, we just need more context about the ways people are actually using these systems, and then we’ll be able to design them better. To caricature, “thick description cures all ills”. I’m not suggesting danah was necessarily advocating this perspective, certainly not in a naïve way, although I suspect that some aspects of it are compatible with her preferred research approaches and overall perspective. And I’ve experienced and been influenced by elements of the “contextual” approach myself, both in my academic training and in my subsequent research experience. However, here be dragons.

The notion that “designing for diversity” means enriching design formalisations with social data is pretty prominent right now. Conferences like EPIC are promoting a healthy interest in ethnographic approaches in HCI and user-centred design in industry. But—and repeating that I write as someone deeply invested in social research—I think that ethnographic and social scientific approaches to design don’t necessarily hold all of the keys to meeting users’ needs. I’ve been thinking back to the arguments James Stewart and Robin Williams proposed in their 2005 paper, “The Wrong Trousers?”. In that piece, Stewart and Williams critiqued what they termed the “design fallacy”: “the presumption that the primary solution to meeting user needs is to build ever more extensive knowledge about the specific context and purposes of various users into technology design” (4).

I think Stewart and Williams are basically right, that we can’t necessarily overcome companies’ design flaws by building the “right” values into their systems… Values are constantly being politicized; as a consequence, they are also very much subject to trend cycles. Currently, the research buzz is around openness and sociality, and the bias is towards sharing. Yet these values are as “political”, as “constructed”, as any others; researchers are not neutral participants here. Neither are educators; who, where social media is concerned, are often responsible for promoting learner understanding of what constitutes “appropriate” or “correct” use (see Tracy et. al.'s paper, "Using the wiki the wrong way").

Finally, and perhaps more importantly, user needs and behaviours are constantly shifting; they are inconsistent; they are difficult to predict. “Designing out” context may be more important than designing it in. Like danah, I don’t see a point in chasing the techno-determinist tail of cause and effect. I believe that where social research can have most impact is in design iteration and studying overlapping patterns of design and use; realistically, we can improve, far more than we can "optimize". As researchers, we are almost always playing catch-up, and we capture or describe the social “what” long before we start to understand the “why”.

References

Stewart, J., and Williams, R. (2005) "The Wrong Trousers? Beyond the Design Fallacy: Social Learning and the User ". In: User involvement in innovation processes. Strategies and limitations from a socio technical perspective. Ed. by Harald Rohracher. Profil-Verlag, Munich.

Tracy, F., Jordan, K., and Johnstone, K. (2007) "Using the wiki the wrong way: A case study in plant sciences." Solstice 2007 Conference, Edge Hill University.