An Instant Message

As editors, marketers, and writers look to the web for revised models of social interaction between readers and writers, how might this change the book market? At the MIT Media Lab, Judith Donath directs the Sociable Media research group, which addresses the communication of identity in mediated and face-to-face interactions. Using the highly mediated exchange of IM, we discussed the social dynamics of online interaction.

Colleen Asper (Rail): What is your interest in signaling theory?

Donath: I use signaling theory from biology as a starting point for looking at a lot of issues about how we make sense of other people, and in particular, how we establish the reliability of people's claims about who they are.

Rail: What drew you to look at mediated communications in this way?

Donath: Most of what we want to know about other people are hidden qualities: are you a nice person? Did you really like the cake I baked? If we got married would you be a good parent to our children? So we rely instead on perceivable signals that indicate these qualities. Signaling theory is about understanding the relationship between signals and the qualities they represent--and whether they do so reliably. Using signaling theory as a basis for examining online interaction first makes you analyze what are the costs and benefits of a signal, it makes you articulate what qualities you think the signal is meant to indicate, and it helps you figure out if it is a reliable way of indicating them. As a designer, you can then decide whether reliability is important in that area, and it provides a way of understanding what designs might be well suited for achieving that.

Rail: How has the Internet changed the way reading is designed?

Donath: Several ways. People reading something in print are generally not immediately able to participate in a debate about it, but much online reading invites public comments, thus making the original material more of a jumping off point for further interaction. Paper books are also linear. Many online articles are too, but some incorporate links and other forms of more complex navigation. This has been done with books (e.g. The Dictionary of the Khazars), but is much more flexible and easily integrated with computer based media.

Rail: Some writers are attracted to their medium because of a natural confidence they feel with all verbal communication, but others turn to writing because of difficulty with speech. The written word is a place for clarity they can't achieve in speaking. People who use the Internet as a social tool may often be more comfortable in a mediated space then they are in face-to-face communication. How is that changing as social network sites not only become more ubiquitous, but also more closely linked to other forms of socialization?

Donath: As generations grow up who take online social interaction for granted, certainly skillfulness in that area becomes more important. I think it also becomes more nuanced - its not just the words you use to express yourself, but your choice of medium - phone, IM, twitter, status updates, etc. The structure of social networking profiles allows a lot of social and subjective detail - do you always have the most recent cool apps (and does that make you seem fantastically knowledgeable or hopelessly trendy), is the sparseness of your profile a sign of your ironical edginess or social obtuseness.

Rail: As social network sites become more complex and subtle, they have the possibility to give users more and more information about the impression they are giving others and to allow users to consciously design that impression. How might this augment the sophistication with which people construct their everyday identity when dealing with others in person?

Donath: That's an interesting question. One thought is that it will make people much more self-consciously aware of the extent to which their identity is constructed. And it is likely to make people more acutely aware of both the richness and limitations of meeting face-to-face. Thus far, mediated meeting has been a shadow of meeting in person - people feel that they don't really know someone till they've met them in person. This certainly is changing and there are many, many anecdotes about people developing relationships with others online. But an interesting point will come when people who meet face-to-face, say at a party, feel that they haven't really met the other, can't really say they know them, till they've checked them out online, seen their social network, etc.

Rail: That has already happened with the use of search engines to evaluate our peers; how googleable one is becomes a status symbol that changes how we think about the people we meet. If our lives are increasingly archived and publicly viewable, how will that change the stake of social interaction? Will the amount of information people disclose online only increase, or do you think that at some point repercussions will be widely felt in a way that will make that openness seem naive?

Donath: That will depend on what the repercussions are. There will certainly be many social situations, employment issues, divorce situations, etc. where people will feel that they have naively over-shared and choose to become more guarded. My guess is that if that is the extent of the repercussions, people will continue to disclose more and more, finding that the benefits of participating very fully in an extended social world outweigh the disadvantages. This will be even more true if privacy safeguards become part of our legal system. For instance, today when someone interviews you, they can see your race, but they are not supposed to use that information in their hiring decision; similarly, one can imagine legislation that said you cannot use information you found out about a person through online investigation in your hiring decisions.

That said, there may be much more serious repercussions - if our government becomes very repressive and uses information revealed online (by you or your connections) to detain you, or keep you from flying, or working, etc.

Rail: Fashion is not relegated to the material realm of clothing and personal possessions, but is also evident in how one presents oneself on a social network site or signs off an email. Such immaterial fashions have the possibility to change at a faster rate than material ones. How will this effect how we think about personal presentation?

Donath: This is one of the topics I'm writing about [in Signals, Truth and Design, forthcoming from MIT Press] and it's a fascinating one. The rate of change in fashion is determined by the speed of communication and the mobility of a society. Today, both are increasing rapidly - and fashion both online and off is changing very quickly. At the same time, we're facing clear signs that we cannot continue rampant material consumption at the current pace - and fashion is part of what drives our consume and dispose society. In an ideal world, we would shift our fashion based competitive consumption to the world of environmentally friendly bits, and make a fashion of conserving atoms.

Rail: If shifting fashion to the dematerialized and therefore environmentally friendly realm of cyberspace is one positive scenario, a way to use the Internet to change something as basic as consumption, what other possibilities are you excited about for virtual societies that are unavailable for physical ones?

Donath: It's certainly not a new idea, but the expansion of our social reach - of being able to truly engage with a bigger and more diverse group of people. The simple existence of the net was one step here, the existence of global newsgroups another, now social networks add a new dimension to this.

Also, I am excited about finding new ways to communicate with people. Thus far, most of our computer based media are very similar to our paper based ones - we write text that could be in a letter - we post photos that could exist on paper. But what if we could easily compose simple simulations as a way of communicating? One of my students is developing a site now that is a bit like Twitter in that you use it to publish everyday goings on - but it is all in the form of statistical graphs. It's not something that has been used in social communication before, but it's fascinating to use, and to follow what others record of themselves. Sites like scratch.mit.ed that let kids communicate by making games and interactive art are, I think, a very interesting step in the direction of communicating computationally