Tuesday, April 27, 2010

Social context, Facebook Likes, activity and action streams

This post began as a comment on the following post by Adina Levin, but quickly became too long, so I am posting it here instead. Read Adina's post on social context first (excerpted here).

Where is social context?
In yesterday's post on the problem with Facebook Like, I wrote that Facebook is trying to be the sole provider of social context. This got me thinking about the various places that social context may be represented in a networked system:
  1. in the object or message (which ActivityStreams helps enable)
  2. in the context where it is created
  3. in the contexts where it is seen and used
  4. in each node of the social graph
  5. in sets of social graph elements
  6. in decentralized elements of the social graph (e.g. aggregated/syndicated profile elements)
  7. shared understanding in participants minds
  8. unshared understandings in participants minds
Facebook's model is seeking consolidation in two places. By replacing a metadata-rich, standardized, ActivityStream based representation of the message with a proprietary API call, Facebook is foreclosing opportunities for the adding of context in creation and in viewing and utilization (items 1-3 in the list).

By acting as the sole provider of social graph and profile services, Facebook is seeking to own those aspects of context (item 4, 5, and 6 in the list). Is Facebook doing anything to enable the exchange of subsets? (item 5 in the list)


What is context? Is it a matter of where or what or something else?
Interesting post — it raises for me the question: What is context? or perhaps, what is the value of context. I am guessing that context means original context, but that begs the question: What's the value of preserving original context? And in the question of context is the presupposition that shared context is valuable (shared context or shared understanding) — but that is a normative claim and the post also argues for diversity and difference.

Several kinds of original context then spring to mind: context tied to to original intent; context tied to original audience addressed; context tied to object references; context tied to linguistic references; context tied to original activity or practice; and context tied to social or public in which the content is produced.

Any one of these may arguably supply context, if context is meant to include:
  • what does the author/contributor mean (to be doing, saying?)
  • what does the content mean to communicate (internal references, external references)
  • what is the content's social status (what social or audience does it tacitly address, and what contribution does it make to what practice within that context)
  • what routine practice does the content refer to or belong to, that might help in understanding its meaning
  • how might one respond (convention, activity, situation, and other kinds of interpretive context external to the content)
  • who is involved (audience context so important today because intended audiences are always involved or presupposed)
  • there are certainly others

On the loss of context
As I'm not a huge fan of the value of content of original production, being rather more interested in creation, production, interpretation (re-contextualization), I don't mind loss of context as I believe:
  • that there's no particular normative privilege involved in original intent — the contribution was made online and w some understanding of what this results in!
  • there's no normative claim in consensus or agreement, or in other words, the original context doesn't preserve truth or rightness of interpretation; the contribution is made online and thus w an expectation of multiple uses and interpretations
  • neither the intentions of the contributor nor the interpretations of the reader supply "truth" — the online world is a communication space in which contest and commentary are assumed — and therefore context as a supplier of original meanings vs context as a referential system that informs interpretive schema are each valid forms of context

Which leads me to believe that context needs further critical reflection. What about context is so important?

An alternative to context: frames, and communication and action
Systems theory provides one way around this — communication. A difference is a difference that makes a difference. The question, then, for social interaction design, would be: what action is possible, what communication can be made more probable?

I would then (no surprise here) nominate different types of action, activity, and social practices as contextual frames of reference, from local and onscreen user interface selections and actions on up to routinized social practices meaningful only over time and within the shared practice of a number of actors. Both action and communication can be pretty clearly articulated, and neither requires a regression to original context, be that of intent, reference, linguistic claims, or what have you.

I know that this contradicts some of the common assumptions made in system design about context. But I don't think we developed these paradigms with social action in mind — I think they were conceived to facilitate effective and efficient user interaction with systems of information (applications). Thus the very notion that original context ought to be preserved is a problematic one — it assumes that meanings ought not go astray of originally intended activity.

We assume, often, that this original context belongs to the object — that, too, is problematic, for much of what is going on is not object centric but is embedded in ongoing communication and social practices (actions).

An example: games and rules
Game rules, for example, better supply context to action and interpretation than do objects — and as Wittgenstein showed long ago, such rules are tacit. Frames can refer to other frames — a move may be understood within its application context or by means of its reference to another frame of activity — in this case, the game.

In social gaming, the game itself may be understood by participants as a social pastime in which several members are contesting supremacy, and this in turn may be a social interaction whose consequences are known only to a small group of individuals in which long-standing social contest for status is re-enacted repeatedly by gaming (game within the game).

Thus the entire question of context may be reframed in terms of action and communication, each of which can involve application-specific meanings on up to social and cultural references. Context might then be better understood within the practices that reframe and recontextualize online contributions, thus permitting ongoing action and communication. Social theory doesn't have a special place for original context, for action supplies its own context.

Practical reflections: activity streams, action streams
Finally, and to get practical for a moment, some thoughts on the matter of activitystrea.ms and Facebook's monopoly of distributed social web activities are warranted. Activity streams supports a broader range of activities than does the anti-social Like. The Like is a one-size-fits-all solution to Facebook's need to venture into social search and socially-contextual advertising and marketing.

Likes eliminate differences of kind and of degree: liking is simple affirmation and association with an item at a minimum, passionate and loyal commitment and dedication at a maximum. But the Like itself neither captures nor represents the degree. And Liking fails to capture nature or kind: does a person Like because s/he identifies with the item, brand, cause, person, etc; own or want to possess it; feel social affinity with the scene or culture it is associated with; mean to gesture or signal activity or engagement (in a game, an offline practice, etc); or what have you.

Activity stream meta data would permit a greater number of updates and qualify them by attributes that supploy more context around the update — which in turn would enable richer and more differentiated interpretations and responses. But these updates, too, are unilateral and monological. Social web updates are a monological system of self-referential declarations. Updates are posted into the open and held open because there is no action possible on them that transforms the update into a move of some kind — a social action.

A transactional system would offer coupling of action updates and closure of activities in which the simple yes/no response essential to social action and communication would be represented within stream updates. A dialogical system would not only solve some of these context problems (not just the where but the what of social context) but would facilitate forms of social networking around messages themselves: distributed or federated, dis- and re-aggregated.

Activity and action streams might not solve the audience context problem, but would permit greater linguistic differentiation of statement types and corresponding responses (invites: accept/decline; offer: accept/decline; news: like/share; purchase: buy/do not buy; and so on). Language itself supplies context, in its grammar and in its role within communication practices.

The fact that so much social web content is treated as information, not as communication, is re-inforced by the loss of context. But could be addressed if we were to standardize the handling of linguistic types and enable reciprocation — or response. The fact that present-day streams now dominate social web activity just seems to beg for this solution of transactionality around coupled messages.

Labels: , ,