[Sketchy, semi-random notes from an afternoon of 20-min presentations, with much much much backchannel chat]

Warren Sack gives his twenty minute presentation. Social computing addresses two questions: 1. How can the insights of social science be applied to design better software? 2. How can software be designed to address social problems? He talks about a software prototype that does a “translation map“: “The Translation Map is a prototype system designed to facilitate collaborative translations and geographically-based messaging” (from the site).

Now Warren asks if software should be evaluated in terms of social capital? Are there private, public and social capital? Nah, it’d be better to think about this in terms of space. When you introduce a new technology, it redistributes the private, public and social space.

Q: (Clay) Danny O’Brien says that on the Net we have public and secret speech, but not private speech…

Paul Resnick talks about reputation systems: A system that aggregates and distributes info about what people have done in the past so people can make decisions about what to do in the future. He’s done a study that has some preliminary results: A sense of uniqueness leads to more ratings, and people respond to challenges to create more ratings.

He talks about eBay. About 1 in 100 transactions gets a negative feedback. But we don’t know if that reflects the actual rate of satisfaction. His lab studies show that reputation leads to more trust and trustworthiness, but long-term partners is even better. We know from eBay that having a positive reputation brings in about 8% more money, in his study. Reputations are useful when interacting with strangers, but aren’t so important if you already know the person because your experience will trump what others say. Short histories create the best incentives but long term tell you the most about the person. (Paul’s research confirms what we suspected.)

Susan Herring talks about “Weblog as Genre.” Her group randomly sampled blogs from blo.gs They looked at the producers, purpose and structure of the blogs. They coded 44 features and quantified the results: Adult males produce blogs that are filters, while women and young people do more personal journals. 50% of blogs didn’t have links to anyone else. The average blog had 6.5 links out. The blogosphere is densely interconnected: The average degrees of separation of the blogs in the sample was 3.8.

Jonathan Grudin talks on “IM and Blogs in Work Environments.” IM he says will be the predominant form of information exchange in business. He says that IM is playing much of the role that email did in 1984. Business is hot on IM, he says, which is different from email 20 yrs ago. In one project, he interviewed 20 people in the Puget Sound. He found they’re technically adept but don’t now much about blogs. In another study, they looked at 400 early adopters of a new IM client at Microsoft. Managers and older users use it differently. Technophobia and switching costs are dropping. Socially, you can IM down but not up, which is maybe why the managers like it.

Steve Whittaker writes about “Designing for Informal Communication and Social Organization.” How do we manage complex social orders? Animals like fixed roles. Apes like dominance orders. The social view says that there are two aspects to being human: Social representation and informal communication. But what is informal communication? His research, within one domain, shows that it takes place between two people, impromptu, and lasts about 2 mins. They looked at ContactMap and tried to build a complex social representation of who communicates with whom by analyzing email. People liked its graphical view of the social net. ContactMap worked better than email for some particular social tasks. But there are issues around scaling.

Wade Cunningham talks about wikis, which he pretty much invented. He says that if wikis and blogs had been invented first, maybe we wouldn’t have had email. He says that wikis generally aren’t trashed, so maybe people are good. Clay quotes Wattenberg and someone who said that the cost of trashing a wiki is higher than the cost of repairing it, so that’s why they generally aren’t trashed.

Elizabeth Churchill talks on “Social Computing and Lightweight Collaboration.” She shows a video of Sticky Chats, chats that attach to portions of a document. Looks useful. Another project, the Plasma Poster, uses a large screen as a public board by which distributed groups can post shared info, leave msgs, etc.

Categories: Uncategorized dw