The short answer is everything. There’s so much left to be explored in this field and I feel like we’ve only dipped a tiny toe into the water. That being said, here’s a random grabbag of stuff I’ve been pondering over the last few months:
- The behavior of gifting and the social narrative that implies, in particular the difference in gifting behavior of created content vs curatorial content (think, I compose a song for you vs I make a mixtape for you). How do you leverage gifting of curatorial content in interesting ways (Quora has an interesting example of this where you can suggest topics during your invite, thereby “gifting” them a personalized experience).
- Using online social interaction to trigger, augment or discover real-life interaction.
- Social affordances: Xianhang Zhang’s answer to How do you build a community of users that give high quality comments on a website?
- The “evaporative cooling” problem in meeting place social dynamics. aka: The people who most want to meet people are the people who the least number of people want to meet. In its natural state, over time, the best people in a community stop gaining value and so they leave, the worst people glom on and it devolves into a steady state of mediocrity. What are some ways to try and prevent that?
- Micro-social currencies (eg: http://en.wikipedia.org/wiki/Cap…) and a full listing of all the social pathologies inherent in the system so that they can be designed around.
- The “meta-judgment problem” where a community does not possess the skill to determine the quality of a piece of content (see: Dating & Relationships). Such communities exist in a reflexive paradox. Even if good answers are provided, they get downvoted and so the good contributors leave. Even when you try and tell them how to recognize good contribution, they can’t make use of it because they can’t distinguish good advice from bad advice. Is it possible to bootstrap a community like that into a healthy state or is it always doomed to fail (this also applies to Google & social & a host of other problems)
- How do you do good social design when you’re working in the valley and people’s social relationships here are so outside of the mainstream?
- What are the methodologies around social software design in the first place? How do you gather data? How do you evaluate success? How do you tease out causative factors?
- The “kickstarter” model of actions contingent on a threshold for participation and where are the interesting social spaces this can apply to.
- How do we design to allow users to better understand the social consequences of their actions?
- What are the natural cleaving points on group identity and how do we build social ontologies that reflect better reflect people’s social realities?
- What are possible social gatings that allow people to self select into high quality participants?
- The concept of circumspection as a term distinct from privacy and what implications that has for the privacy debate.
- The cultural difference between social game designers and social software designers and how that brings different perspectives to the problem (the legacy of social software design is from interaction design, the legacy of social game is from marketing).
- Quality metrics that move beyond upvote/downvote, thinking about how do you even define quality and also how to meld curatorial control with crowd sourcing in a scalable way.
- How do you design a system that allows for multiple presentations of self and how do you build the UI to make this usable for both the contributor and the reader?
- What are the design primitives around real world, asynchronous, shared device interaction? The iPad is potentially the major driver in supporting this use. The “family iPad” always sits on the coffee table and it’s used to co-ordinate the social activities among multiple members of a family.
- Humor on the web and how to stop it. Internet humor can be a particularly corrosive force that relies more on recognition & the demonstration of cleverness than any genuine desire to please (cf: reddit comment threads). I’ve had some interesting discussions with Quora engineers about their particularly dour approach to humor and, while I disagree with their implementation of policy, I appreciate the degree of thought that has gone into them being humorless pricks.
- The concept of social software as being an inherently sociotechnical system, that policy and code are two sides of the same coin and they both serve as constraints against which a community is shaped. Especially how this view gives perspective on how to keep the community stable as the social constraints alter by buffeting them with technical constraints. (eg: in a tiny community, you probably shouldn’t implement a reputation system as it’s better for people to keep & manage reputation in their heads. As the community grows larger and that social constraint weakens, you could consider allowing people to leave reviews for other people in order to let technology add back in the social constraints that kept that community successful).
- An ethical debate on whether you should deliver to people what they intrinsically want or whether you should use your privileged position to impose your values onto your audience. This comes up whenever I meet someone who wants to design a tool that allows people to see both sides of a political debate and have warring sides come to a consensus. Our empirical evidence shows that “open mindedness” is only a true value for a very tiny fraction of the population. For everyone else, the freedom to seek information means the freedom to craft their world such that no contrary thoughts intrude. If this horrifies you, do you design for these people, do you remain agnostic or do you actively design against your customer base?
- The “Mom, keep out” technique of controlling privacy by controlling context rather than using access control. Teenagers control the privacy of their room by sticking a sign on the door that says “Mom, keep out”. This sign has no security force, it makes it 0 times harder for Mom to gain access to the sensitive information. But what it does do is re contextualize the narrative around the privacy violation. If mom discovers a pack of cigarettes in your room, it switches from “What were you doing with these cigarettes?” to “Why were you snooping in my room in the first place?” Can this be used in clever ways to provide “good enough” social privacy?
- Some interesting research from a colleague at the University of Washington on using content analysis to provide a richer analytical framework and metrics that actually considers the content of posts rather than just implicit behavioral metrics like time-on-page or clickstream data. How can something like this be integrated into the heavily metrics driven design of facebook or google to make them at least slightly less socially autistic.
- Using Theory of Mind analysis to systematize and dissect narratives that are performed on social software and then using this to understand how changing the information flows affects what behaviors manifest.
- Building a solid case for why social interaction design is as different from interaction design as interaction design is from visual design and arguing that it needs to be treated as it’s own separate discipline and given the respect it deserves.
- Should it be called Social Interaction Design, Social Design or Social Experience Design. I am personally pushing for the latter and trying to figure out how to get impetus behind this.
- Consensus & Conflict. What are the areas in which communities must be in consensus to remain healthy, what areas can people be brought into consensus and what areas will people remain eternally in conflict? For example, I’m not really aware of any online community that half the members believe in evolution and half in creationism. They engage in a state of perpetual warfare until the majority of the other side quits in disgust. Quora is currently experiencing a series of low intensity brush fires on topics like Paleo Diet vs Veganism andQuestions That Contribute to the Sexism/Misogyny on Quora vsYou’re Being Too Sensitive, Get Back in the Kitchen (this is a somewhat reflexive post, I created this last topic precisely to explore how conflict like this gets handled by Quora :)).
- The phenomena of “side channel revolt” where when the desires of the community admins come in conflict with the desires of the users, the users will use their intimate knowledge of the system to subvert the system through side channels that work around the normal political organization. Digg had to deal with this when it decided to censore the HD-DVD DRM key in 2007 and the user base exploded into popular revolt.
- How to combat existential risk for a winner-takes-all social platform. In 2005, everyone switched en masse from Friendster to MySpace. In 2007, everyone switched en masse from MySpace to Facebook. I remember people in 2007 being evenly divided between those who believed there would be a Next Big Thing and those who thought Facebook would dominate in perpetuity. The Next Big Thing brigade seems to have quietly disappeared over the last few years but if anything will kill Facebook, it will be the same existential risk factors that brought down the previous two incumbents. What will that be? How will it work? Will it ever happen?
- I claim there are at least 2 different social networks: “people that are useful to me” & “people I enjoy hanging out with”. How do you design software that is aware of this distinction?
- How come the web chooses not to replicate the real world structures of class and progression? The web is ideologically tuned to flat structures and an “everyone is equal” ethos. Is that the natural ordering or a cultural quirk?
- Plazas & Warrens. How to decompose every social software design down into these two basic primitives and what the use of doing so is.
- The concept of sunlight as a disinfectant for social spaces. It’s often the neglected corners of a social space which are the most abused.
- “Literate”, “Illiterate” & “anti-literate” cultures. Entering into a literate cultures, the social norm is that you’re expected to be read up on some “canon” of the culture. Conversation proceeds under the assumption that all participants are familiar with the literature and it is considered the fault of the ignorant for not learning the requisite knowledge themselves. Illiterate cultures don’t expect anybody to have read the literature and it’s the responsibility of those who have read whatever piece of the literature to be able to educate the rest. “anti-literate” cultures are those which are actively hostile to appeals to the literature. Wikipedia is a great example of a anti-literate culture, by design. It actively disdains “expertise”, to the extent that danah boyd actually lost the battle over the correct spelling of her own name. I’m starting to formulate some basic arguments about how Illiterate cultures cannot retain Subject Matter Experts over the long term due to the fundamental way that discourse is structured. The problem is, Subject Matter Experts fundamentally don’t really know anymore how they know something. That adding programmers to a late project makes it later is something most programmers know. That it comes from Fred Brooks is something most programmers forgot. When asked for [citation needed], they’ll just leave for more productive venues of conversation rather than have to talk down to the masses.
- Mapping the “Bespoke”, “Made to measure” and “off the rack” distinction in fashion design to online community design as a metaphor for the various tradeoffs and also exploring how the business models also mirror each other.
- Exploring the notion of a “social stroke” and the difference between social applications that use objects to enhance the relationship between users and those that use users to enhance the relationship between objects.
