Brynn Marie Evans

I just signed off Tummelvision, a weekly podcast/videocast (in some cases) led by Deb Schultz, Heather Gold, and Kevin Marks, where we had a fascinating conversation about designing for social interactions. I was a lucky guest along with Julie Hamwood from Adaptive Path.

The conversation was really stimulating, and wound in and out of perspectives about social interaction design (sxd). At first we were discussing Google Wave’s demise and Google’s inherent lack of social understanding, and then moved onto Facebook missing the mark with Facebook Questions. My take on this is that Facebook already has users asking and answering each other’s questions every day — and effectively at that — but in their status messages. What Questions appears to accomplish with its wide open format is merely to generate data for some presumed search platform later. I’m not convinced the data will be any better than Yahoo! Answers, and, as users, we’re certainly not compelled to answer (or read) questions by people who aren’t in our friend group. It’s the chicken and egg problem if Facebook wants to grow their database to search on later; so a better strategy, in my mind, would be to target questions to the right friend groups. Context matters.

This brought us to a discussion about fundamental differences in the way Twitter and Facebook are designed. Facebook is a network where relations are the social object; while Twitter is a network where information is the social object (although emotion and play are also social objects at times). Yet, with all of Twitter’s openness, it brings with it a sense of presence of who you’re talking to. It doesn’t feel like you’re tweeting into a void, whereas using Facebook Questions does. We also discussed, here, that relationships that form around content sharing (e.g., Twitter) develop into the sort of network where this openness can be supported. But privacy still reigns in the Facebook networks where the only glue is, in many cases, a former relationship.

Of course, this led to asking whether there are rules for designing effective social interactions. The short of it is no. Twitter created a versatile platform by using only simple rules for engagement. But these rules cannot be carbon copied, replicated elsewhere, and expected to work in another system in quite the same way. Things like context, culture, and personalities (among other sxd considerations) change the way a given feature will be used and adopted by the community.

Instead of developing rules or principles of social interaction, a better approach is to think of the questions we can ask when designing for social. Who are our users? How do they think of themselves? Who do they want to connect to? Who do they want to connect to tomorrow? Why? What’s the point of the network in the first place? Where are they when they’re using the network? What is the outcome of an interaction? What’s the role of strong versus weak ties in the network?

We didn’t have any glorious resolution to our discussion but instead emphasized how we need to keep trying to understand how social interactions play out — which includes doing research into social and developing case studies which illustrate principles that we can take to our clients and product teams.

A few references related to social interaction design include:

• Adrian Chan’s original introduction to Social Interaction Design. And related presentations.
• Paul Adam’s research into Real Life Social Networks.
• A great list of references by Paul for thinking about sxd.
• David Casali’s Social Usability checklist.
• A case study by us fine folks at Bolt | Peters about iPad usability and how social and context play a part.
• Our group blog on social interaction design.
• My article on using remote research to inform social interaction design. And a related talk.