Friday, October 28, 2005

Reading Notes on Social Capital


  • Social capital seems to be a slippery term. I'm tempted to think of the term "civic society," which would combine not only the health of social relations, expressed in a population's reservoir of good neighborliness, with its sense of a common good. In other words, not only the acts of generosity, but a shared set of values—and healthy ones.


  • Then again, it seems that some definitions focus more on one's social network. Here, the definitions tend to emphasize the "asset value" of social capital.


  • Which is it?


    • Something spent, like money? Fungible, like currency? And bankable? Capitalist's capital, in other words?


    • Or something proffered, as a gift, out of generosity and a commonly shared sensibility that what I do for you will come back to me. And that what I do for you is simply the right thing to do.



  • Either way, it does seem that social capital has little to do with software... "The second is to ensure that the software also supports sociability, that is, effective social interaction online (Preece, 2000)." Preece mixes the two: social interaction and social capital.


  • My sense is that social capital has to do with the degree of concern we feel about others, and the degree to which it can be mobilized if called upon. If we count a social network of 200 friends, none of whom would come to a birthday party, are we rich or poor in social capital? I think poor...


  • The confusion stems from terminology: capital, when accumulated, counts as wealth. Well that's not how interpersonal relations work. In order to "spend" social capital, one would have to be able to move people (to do something on one's behalf).


  • I see this as an attribute of communication then, not of assets. A rich man may be alone; a poor man, among friends. The rich man's social capital may in fact be poor if they desert him at the witness stand.


  • Because it takes communication to bind people, and to move people to act on matters beyond their own immediate personal concern, it's in the health of communication that real (authentic) social capital is found.


  • For that, even the most transparent software only plays a role in facilitating interactions. Two people can hate each other and communicate this with utter clarity...


  • The relationship between what binds people, their communication (in terms of content), and a society's well-being are all much more complex than is suggested by the simple equations:



    • Social capital = Number of friends

    • Social capital = Sociability of software

    • Social capital = Trust in a network

    • Social capital = Degree of Collaboration



  • It is worth considering the relationship between trust and social capital separately. For trust in an individual is not the same as "trust" within a group, or in a social network. I often find that these attributes of "sociability" are drawn out of interpersonal interactions and pasted onto a groups, companies, crowds, networks, and other collectivities AS IF there were no difference between the emotional commitments people make one-to-one and those they make to groups, etc. We all know that the psychology of interactions changes from, say, a couple to a crowd. Why then do we assume that if a number of people who trust one another as friends, and who set themselves up as a group online and count themselves as a network, share that same trust as a group that animates their paired relationships? I bring this up here only because I sense that it's trust that's behind the term "social capital."

Tuesday, October 25, 2005

Reading Notes on Niklas Luhmann's Social Systems

Reading Notes: Social Systems and Niklas Luhmann
The complete pdf is available here


  • Double contingency is a fact of meaning, and the domain in which social software must be situated is a meaning-based domain: each individual must come to understand what others' communication and action means. This fundamental point seems to have eluded a lot of what I find in the literature on social software. Authors, designers, and critics alike tend to view meaning-based events such as communication, transactions, exchanges, and interactions as straight-ahead and straight-up phenomena. What happens on one side (with one user) happens equally on the other (Other user). As Luhmann points out—and he's only drawing on the hermeneutic foundation of contemporary sociology; this is not his own invention—each actor's interpretation is implicated in the Other's actions and vice versa. This double contingency throws a wrench the works of any simplistic views of social systems.

  • It is easy to confuse terms here: social system as Luhmann uses it is an application of systems theory to societal systems. It is not a “technical system.” As we're many of us designers and engineers, we have to keep this in mind, if we're to use Luhmann's approach.

  • Luhmann's modification of communication theory is brilliant. He notes that we need to ask whether society comes out of communication or action, and ultimately, he combines both by integrating action theory into communication theory. His steps are this: that communication first involves understanding (each actor must understand what is said); this creates the possibility of a yes/no response. It is the response that integrates action into communication, and which furthers the interaction, thus creating and limiting further communication/action.

  • Luhmann's distinction between communication and action leads to a distinction between communication and interaction. The meaning, and the process of understanding meaning, is distinct from the production of a linguistic utterance. This is critical in my approach because I view most of these social software systems as talk systems. The production of talk (e.g. speech in the form of text) should be viewed as communication; the handling of talk, and speaking (IM'ing, emailing) actors should be viewed as interaction. Each sets up its own needs. The distinction is powerful in that it provides ground on which to separate communication tools and interaction tools, along with the necessary design and use constraints belonging to each.

  • Ambiguity, which naturally accompanies any conversation or speech situation, is exacerbated in talk systems or social software systems by the intervention of technology itself. Not only the technical design and interface but also the user practices built around it substantially impact users' means of going forward with communication and interaction. In talk systems, software facilitates (even while it puts its own stamp on) these linguistic and interactional proceedings. Now we can proceed to describe the user competencies required of a social software system. In other words, standard HCI and human factors stuff, as well as sociological, linguistic, and interactional stuff.

  • I've been wanting to ground social interaction design in practices, rather than in design, cognitive science-based user analysis and modeling, or engineering. My claim that practices are first and foremost social and sociological works only if I can show that all social software is embedded in social practice. These practices assimilate technical solutions—they are not determined by them. We never leave the social domain.

From User Practices to Social Practices

Designers like to establish who their users are before commencing with a system's engineering and architecture, and for obvious reasons. Knowing one's user is tantamount to knowing how one's product or service will be used. And yet it's impossible to know one's individual users, so generalizations have to be made in their stead. Which is to say that stereotypes must stand in for actual users, providing guidance during development for what is hoped actual users will do with technology or application.

A corrective to the generalization of user types, termed by its inventor Alan Cooper as "personas," has become quite popular as a heuristic by which designers can get closer to actual users and usage, and thus design better. It's not possible to simply extend the personas model to social software, and for one simple reason. We're dealing here neither with individual users nor with individual needs, goals and objectives. We're dealing instead with social phenomena, and that means that we need to understand the manner in which social practices satisfy individuals and their pursuits. What we wish to model our systems on is not a single user's (persona's) agenda, but rather a social system and its (more complicated) organization.

The type of social practices a system is designed to support, and which it counts on for its own ongoing sustenance will be manifest in a wide range of behaviors and modes of participation. Shortly we will cover the manner in which we might model some of these social practices. For now, simply consider some of the ways in which a view towards social practices exceeds the language and concerns we know from an orientation to user practices.

An emphasis on social practices, and designing towards social interaction, would shift our point of focus:

  • From users to groups

  • From individual use to community participation

  • From direct interaction (HCI) to social interaction

  • Our usability concerns would shift also, from first order interaction design to second order, social interaction design:

  • Interaction tools designed for the transparency of the interaction between users, rather than transparency between the user and the tool.

  • Attention to the way in which users might confuse interpersonal ambiguity for technical shortcomings or errors.

  • In organizing, archiving, and indexing discussions and other kinds of member contributions, concern for the contributor as well as for the text of his or her contribution.

  • An appreciation of the way in which users present themselves to others in their profiles and other contributions—that is, how they might second guess their audience, attempt to "game the system," and so on. Usability concerns here amount to the difference between choosing a photo of oneself and uploading it with a form page.

  • Constraints built into the system should involve not only technical and interface choices (or better, limiting those choices), but should involve the cultivation of social behaviors and normative restraints.

  • Error management and handling should focus on risk management as a social undertaking, rather than a technical one. The kinds of mistakes that happen among users of communication technologies are more often than not fraught with social repercussions.

  • Similarly, help systems and documentation should involve not only standard online help and troubleshooting. Expert users should be encouraged to help others, in the spirit of neighborliness and good will.

  • User competence will involve social considerations as much as technical ones, and designers should be aware of the difference. Many novice users will lurk before becoming actively engaged participants, seeking to understand how the community works and what it values before making their own contributions.

  • We could cite other examples of the shift from user-centric interaction design to social interaction design, but points noted above should suffice.

  • As always, our systems should satisfy tried and tested success criteria. They should win points for their:




    • Effectiveness. In short, they should do what they were built to do, as well as possible.

    • Efficiency. They should require as little overhead as possible. Steps ought to be kept to their minimum, and users should not be required to absorb application shortcomings.

    • Safety. Users should be spared unnecessary embarrassment or other unwanted surprises. To the usual list of annoyances (crashes, mistakes, disappointments, misundersanding and confusion, lack of confirmation/acknowledgment, etc), we should also list:


      • Our systems should not add to the ambiguity that can already exist among social transactions and encounters!

      • Our systems should avoid further obscuring a member's identity or intentions, as well as how we are to interact with him/her.

      • Our systems should give us feedback, and quickly if possible, as to whether or not we have behaved properly, done the right thing, etc.



    • Utility. Systems should be useful, or at lsat users should find them useful!

    • Learnability. It should not require a PhD to navigate and learn a community's or system's architecture.

    • Memorability. Labeling systems, functions, features and so on should be transparent (no, not see-through), sensible, and familiar.



Monday, October 24, 2005

A Case for Social Interaction Design

My Social Interaction Design project is in progress. This is an excerpt from an introductory section.

Social software services that fail to develop into social practices among their users will fail in the marketplace. Social interaction designers, by designing with social outcomes in mind, can help shape the feature mix and design the user experience such that these sites produce results benefiting individuals and organizations.

Social software applications work on several levels simultaneously: they enable communication, interaction, and social relations. We need to think of these three separately because they are different kinds of systems. The recording and distribution of communication as text is different from getting a member's attention, or, say, making connections based on who knows whom.

Here, in brief, are the key differences between communication, interaction, and society:

  • Our primary concern in communication is reaching understanding about what we're saying: making ourselves understood.
  • Our primary concern in interaction is handling the dramatic character of social performances and our performances as participants, including such psychological features as personal comfort levels, insecurities, dispositions, attention-sharing, and more.
  • Social systems are built on relations among members, and they are maintained only as long as those relations are reproduced. Any online community, in other words, needs to succeed at the very basic task of connecting members and compelling them to stay in communication. Only participation will do that; no software can do that for them.

Social software engages each of these, certainly, but it does so differently. And that's one of the reasons it's so interesting. Its manner of facilitating communication results in searchable archives. Its manner of mediating interaction protects us from embarrassment. Its manner of connecting people permits relationships between some unlikely bedfellows. So in communication, we can focus on how these technologies enable the capture, storage, and distribution of information. In interaction we can look at how non face-to-face encounters are shaped by their removal from physical immediacy and co-temporality. In relations, we can examine the conditions that permit or block connections, with an eye on the groups and communities they support or empower.

Technology designers are a feature-driven bunch, as are user interface designers. We tend to think that failed user experiences can be repaired with better-designed interfaces. Social interaction design would embrace and extend this approach, but with the added premise that any time two or more people use a technology for communication, issues pertaining to social interaction become relevant. Be they matters of interpersonal misunderstanding, or of social performance and public behavior, the successes and failures of social software involve a social interface. Now of course none of us can legislate how people should behave or what they should say. So how then do we design social interactions? We don't. Rather, we design the architecture that enables it.

In this paper I suggest that social interaction designers will need to leverage interfaces, back-end architecture, navigation schemas, features and functions to obtain second order effects. If a first order design feature is a common web feature like a popup window or a form, then second order effects are the social phenomena that unfold when a whole number of users engage with that form—when members all say that they're single, for example, though they're not. Second order effects are more complex than their first order ingredients because they're social. And because they're social, all we can do as designers is apply leverage with our balance of first order widgets.

Communication technologies force us to explicate the implicit

"Note that insofar as participants in an encounter morally commit themselves to keeping conversational channels open and in good working order, whatever binds by virtue of system constraints will bind also by virtue of ritual ones. The satisfaction of ritual constraints safeguards not only feelings but communication, too." Erving Goffman, Forms of Talk, p. 18

"Sincerity is incommunicable because it becomes insincere by being communicated."
Niklas Luhmann, Social Systems, p150

Cultural processes require implicitness
Culture is traditionally described as the sedimentation of sanctioned behavior in the form of spatial order and temporal organization. Whether defined restrictively as code, or loosely as habit, these recurring patterns of behavior become embedded in culture as a shared stock of knowledge in which members participate consensually and for mutual benefit (disregarding any systemic bias or inequality). Social practices instantiate these embedded claims, but without forcing them to the surface. Cultural context thus enables interaction while constraining it to the reproduction of a shared set of familiar routines. It is this stage upon which performance takes its liberties.

Technical processes require explicitness
The introduction of technology into social interaction always makes some of the implicitness of human communication explicit. Even the most user-friendly and "transparent" technologies require users to attend to the technical "means of production" of communication, by committing some degree their attention to the successful execution of pre-determined processes, operations, functions, etc. Similarly, the undoing of face to face interaction rituals can sometimes require that users explicitly state their interests, intentions, affections, etc. with one another. What can be conveyed implicitly face to face, is forced to the surface by technology. This foregrounding of process can interfere with the spontaneity of the exchange.

These are excerpts from a set of "principles of media" I'm creating as a way of organizing my thoughts on the interventions of technologies of communication.

Sunday, October 16, 2005

Blogger in the Bathroom, or Blogging as Graffiti

Graffiti or conversation? Messaging or interaction? Which is blogging? Is it a form of writing or a form of talk?

We're tempted and perhaps even required, by insufficiency of alternatives, to understand our new technologies and their forms of use by analogy to familiar activities. Technologies of communication, of course, are hence often compared to various kinds of realtime and co-present interactions, be it private conversation, (chit)chat, an announcement, what have you. The point is not what form of interaction do we think its technically mediated form takes; it's that we lose something in translation if we refer only to familiar face-to-face comparisons.
Technology transforms, and we need to tease out the manner in which it transforms. The transformative operation that a technology performs (I'm not a determinist; in fact I think that culture anticipates its technologies before bringing them to discovery/market, to wit, the stirrup and the gun, neither of which were instruments of war in many cultures that had access to them) may be simple or complex. Most of the time that transformation is along a perceptual axis.
The magnifying glass amplifies along the visual axis (Marshall McLuhan wrote that all technologies amplify along one axis while bracketing along others, thus tunneling or focusing our experience by extending awareness just along the visual, the acoustic, etc.), the phone amplifies along the auditory, and so on. Those are the first order operations; second order operations are where things get more interesting. And to keep things to blogging, where I think we need to address a major oversight in our use of analogies...
McLuhan was famous for his (somewhat unorganized) perception of the social dimensions of technologies. Technologies, for him, conducted a two basic operations: each new technology referred to a previous technology as its content (tv referred to radio; film referred to theater), and extended our perceptual faculty beyond its normal physical power and reach. Note that the "content" of a new technology is not simply an earlier version. Viewed by McLuhan, or viewed sociologically, cinema is not a advancement of photography (which it is, of course, technically and historically speaking); it's an advancement of theater. It was either Laurie Anderson or Brenda Laurel who remarked that "we go to the movies, not to the projectors." Theater, as a form of expression that involved scripts, actors, roles, a stage, a non-participatory audience was extended as film. It's no accident that the two use the same building. (Necessary aside Foucault would add here that it's no accident the prison and the school use the same building!)
So, is blogging conversation? Though it looks more like graffiti in the bathroom (a mess of faceless messages half of which have been edited by somebody else, all written to be read, but happily free of consequence and accountability and thus tending to be heavily self promotional—in other words, names of bands)? No.
Conversation is bounded in space and time and to its participants, and blogging, clearly is not. Blogging's one remarkable feature, in fact, is in its dislocation of speaker/hearer, or speaker/audience (true of graffiti also, so, point graffiti). Conversations are a focused and directed interacton/participation. Utterances are addressed to the company present, and a great deal of the interaction, both in content of what is said and in its meta-language, as well as in physical participation, is directed towards sustaining the engagement of all participants and thus caring for the group dynamic... Conversation involves facework about the face, in other words (the first words of conversation say it all: How are you?).
I would venture to say that blogging is not a form of facework. Also, that it's not bounded in time and space in the way that conversation is. I think it's more accurate to call it a form of talk, and then go from there in our efforts to be more specific. First, though, and picking up on the McLuhan trail again, to the modes in which the blog is a transformation. And I say this about all communication technology:

technologies of communication and interaction perform a temporal operation

This sneaks up on us because we tend to be visually inclined (as a culture) and spatially and object oriented. We like to see things and their relations in space (McLuhan claimed that Americans were visual culture; Russians, an oral/acoustic culture, and that our mode of spying on them was thus to conduct U2 spying overflights, their mode was to plant bugs and listen in...). Time is much harder to plot, but it's there, and it's the primary mode in which interaction technologies tear the interaction from its ground and foundation.
Technologies of communication, like blogging, IM (which is near-synchrony), email, message boards, what have you, "connect" people in spite of physical distance, and separate them in time. In two times, in fact: first, absolute (real, objective) time; second, subjective time (duration, lived time).
The first dislocation puts us out of step with one another, and is the reason that asynchronous communication technologies tend to have problems distributing or focusing participants' attention to one another. We need to be face to face and thus in real time to pay attention to each other as people. Asynchronous technologies enable us to "talk" to each other, but not to attend to each other (in meta-linguistic terms; whether or not "attention" is a currency of blogging, of posting and commenting, is a psychological matter and one we can address separately, but I think the answer is smal attention but not ontological, or big attention, if you get where I'm going with that). This is one reason that chat rooms are filled with so many capital letters and punctuation marks: those are attention getting devices, facial or physical gestures, if you will, transposed to writing.
The second dislocation removes us from common time, and is, philosophically speaking, more profound. Sociologically speaking, it's more frightening. We might be adapting to asynchronous interactions, to the practice of messaging, emailing, posting, etc, as a form of talk, and having little problem with it as a first order technology. But on the second order, where asynchrony and physical separation still the rhythms of shared, inter-personal time, I'd venture to say that we're having a tougher time.
In fact our entire culture may be out of time, may be losing "good time" and "quality time" to the multi-tracked, discontinuous, and interruptive time that is the time of asynchronous communication. We pride ourselves on being connnected, and we get connected, presumably, to maintain our connections. We get in touch to be in touch. But connections facilitated by technologies don't produce connection. Technology's transformation is lost in the terminology. Analogy, which is a comparitive operation, doesn't capture a dislocation-amplification-extension (which is what technology does).

My point—and I apologize for the length of this one, but it took a while to cover the technical and the social separately here— was that in understanding social practices in which technologies like blogging have become embedded we need to better grasp the manner in which a technology-in-practice amplifies and extends, how its operation intinsic to new second order social and cultural practices, and most importantly (because we see it so poorly), how it works with time.

Tuesday, October 11, 2005

Blogging as public and private

The sociologist's main concern is understanding the manner in which the individual stretches transcends his/her physical presence with relationships that stretch across time and space. "Big" sociology goes after the structures and systems that comprise "society," the institutions in which common values are sedimented and through which they are codified and maintained, the various forms and positions of power by which certain interests and groups articulate their preferences, and so on. I'm captivated by the smaller sociological concerns, those that describe the interface of technology and individual, where the rubber of cultural practices meets the road by which we stretch out and navigate our interpersonal relationships.
Big sociology used to distinguish between the public and private spheres—the former being the zone of social integration, the latter, a space in which the individual had freedom of movement (and thought). But that distinction belied a spatial and physical bias of thinking that we can no longer maintain. Communication technologies today permit us to be present to one another without physical co-presence; they place demands on our time and attention that exceed the physical boundaries of a situated individual presence. Presence negotiation now requires that we finesse our handling of interactions and relations with and through mediation by technologies. Put in terms of primary sociological interests, technologies have become deeply embedded in the very means by which we stretch our relations across time and space.
Take blogging, for example. Among the numerous things already said about blogging, mostly by bloggers, a few misconceptions stand out.
Misconceptions:
--Blogging is writing
--Blogging is publishing
--Blogging is public

Corrections:
--Blogging is talk, what Erving Goffman called "an open state of talk," and more closely resembles forms of speech than writing (though it takes the form, of course, of text). (Note, for Goffman, a monologue is a kind of talk; talk does not have to be conversation.)
--Blogging is a form of messaging, and though blogging technologies are online publishing technologies, we shouldn't confuse the product with its means of production. As a form of messaging, blogging involves a loose and ambiguous mode of addressing (as in, a speaker addresses his statements to a listener). Audiences in other words are not as anonymous as they are for the published text, for the medium does permit commenting, quoting, and other kinds of "interaction" with the text or author.
--Blogging instantiates a hybrid public/private "space" (as much as I dislike the term) in that the blogger may compose his or her piece as if it were a private journal, or alternatively, have a specific audience (of one or many) in mind. The blog form does not presuppose an anonymous readership simply because it involves a communication/interaction tool/technology.

The sociologist could derive any number of conclusions from this, but the ones that captivate me again pertain to our relationship negotiations. To what extent do we hide "private" messages within our "public" blogs? To what extent do we feel involved in conversation when blogging (looking for our piece to be picked up, or for its comments, direct or not, on other bloggers, to be acknowledged)? What does it mean that we engage in these modes of indirect interaction? What temporal markers might characterize the blog form—from how long we expect to wait for comments or continuation, to the effect of the blog's persistence in time on the bracketing of conversation (a blogger's utterances long outlive the blogger's act of uttering them). And so on....

Wednesday, October 05, 2005

Speech Act theory applied to Social Software and other talk technnologies

Speech acts, which were first noted by Austin and which involve utterances that "do something," are often cited by interaction designers as a way of structuring a software system's conversational interactions. There's a glaring problem with this approach, however. Speech act theory is a way of analyzing the particular kinds of statements and exchanges in which people do things with words. The most oft-cited example being the "I do" pronounced during the marriage ceremony. Speech acts are types of utterances that involve persons, and persons in the act of making commitments to one another, in particular. The speech act is a type of action that transforms (wherein the action occurs in) each interlocutor. That's what makes it so interesting. The event, action, is of a psychic and relational nature.

To transfer the analytic of speech act theory onto sofware design, as a means of building software wizards, surveys, dialog boxes, action sequences and steps, etc, is to miss the point of speech act theory. Software events are verifiable. They occur within a "virtual" (though nonetheless real) system of organized information whose relations and events involve transactions that may or may not involve "real world" activity (banking, messaging, retailing, what have you). But the software with which we interact is not a person. The system does not "promise" to do what is says, or what we ask of it. The "OK," "Submit" and many other buttons we "utter" as we click our way through an application are only analogically conversational. The software itself makes no personal commitment to us and establishes no special personal relationship to us.

The proper use of talk or conversational analysis in software/application design is with standardized speech, not with speech acts in particular. Designers fail to see that it's not about bringing natural speech to software. What's happening is the reverse: software brings standardized, sequenced, stepped interactions involving clear binaries (yes/no commitments) to its users. We get standardized, not the other way round!

Monday, October 03, 2005

Social Networks: Color by numbers or connect the dots?

I'm knee deep in research about online community, interaction tools, wireless communication and technology, social networks, you name it. Through all the theories, methodologies, and other design approaches I've found, one thing stands out. And I don't know if it points out a fundamental misconception of the transactional character of human relations.

There is in the various design theories, whether they're influenced by motivation and intentionality theories, agency theory, activity theory, cognitive science, rational actor modeling, social network mapping, even instructional design and learning theory, a tendency to *VIEW* relations, that is describe them structurally. That's to drop time out of the picture (you cannot draw time). Even many of those methodologies most interested in relations (social networking theories) map relations between people without indicating directionality. Lines are drawn between actors without any indication of tendency.

Not only do we need to bring temporality (and not incremental time, but lived time, or duration) back into our analysis; we need to understand and model the fact that in human relations, we are always in motion: towards or away from; identifying with or differentiating from; becoming close or losing one another.

This is in part because all human interaction involves a double contingency -- we account for the Other. That other, be it a real "object" or "virtual object" (your lover or your mother) has been left out of the vast majority of the literature I'm reading on communication technologies and their design.

Whether or not we agree with rational actor modeling, that is, regardless of where we stand on agency being rationally motivated, the social context is far more dynamic than rational actor models recognize. If we're going to successfully grasp the mechanics of interaction technologies, we must do more than map object-oriented design thinking (user, goals, needs, transparency, feedback, yes/no, etc.) to social relations. The double contingency (A's action anticipates effect on B) undoes the clarity and structural causality of the standard interaction design approach.

More on this later...