A Social Software reading digest: What goes in?

Question

Although I'm a programmer looking to explore social software patterns, I'm also interested in their psychological and sociological origins.

My proposed format is to limit each reply to ONE reading with a short summary of why it's important.

The idea is that we will then be able to see which particular readings are most important. Since community wiki posts don't count towards one's reputation, hopefully there won't be too many complaints.

I've read a few papers and a few books myself, so I'll get the conversation started...

Answer 1

The Delicious Lesson

The Del.icio.us Lesson, from Bokardo, is that personal value precedes network value in social applications. In other words, you shouldn't expect your users to contribute meaningfully to the health of a group function. What you can and should do, is to assume that users will act in their own self-interest, and then try to learn from what they do. That derived information is what's useful to the group.

Anyway, that's what I take from it. There's also a bunch of talk about the merits of tagging in general, but the reason this reading is important is for the lesson outlined above.

Answer 2

Shirky: Community, Audiences, and Scale

A 2002 essay arguing that communities don't scale, since humans have a finite capacity for connections with other people, and that those connections are what make communities useful. I agree.

Answer 3

The Lessons of Lucasfilm's Habitat

The principle developers of Habitat, a "multi-user online environment" write about their experiences and lessons learned. Habitat was a virtual community that can be thought of as a very early precursor to Second Life. In the paper, the authors declare some guidelines for social software, some of which are widely accepted today. Some of the guidelines, such as the assumption that bandwidth will always be a scarce resource, were prescient. The authors spend a lot of time trying to say, I think, that complexity in social software should come from the users, not the programmers.

In another feat of foresight, the authors warn administrators of social software to try to "work within the system". I've seen a good number of social apps get a lot of negative attention when they censored a user's action for reasons that seemed unjust. Flickr and Digg come immediately to mind. The authors might have recommended that the moderators of flickr and digg attempt to discourage unwanted behavior without pulling the adminstrator card and wiping out user content.

Like the LambdaMOO paper, there's an implicit warning against assuming any sort of higher-mindedness from the users of your software. Assume that online identities will display every baser behavior that offline humans do. The authors remind us that real people are behind those avatars.

There's also an extensive bibliography at the end of this paper, although the listed material is by now quite dated.

Answer 4

designing for the social web

Author Joshua Porter is a web designer so the writing has advice in how to design social web applications; however, there is plenty of psychological and sociological analysis here that leads up to the advice. You may also be interested in this NY Times article on ambient intimacy.

Answer 5

Shirky: A Group is Its Own Worst Enemy

In 2003, Clay Shirky (shirky.com) gave a speech at The O'Reilly Emerging Technology Conference about social software, and this article is essentially a transcript. Borrowing from Experiences in Groups by W.R. Bion, he begins by talking about ways groups, online and offline tend to self-destruct.

This is a wide-ranging article in which Shirky explains many of the typical problems in social software design, lists assumptions that designers of such software should make, and outlines solutions to common pitfalls.

Overall, it's a fantastic read. If you're thinking about writing some new social software offering, I suggest you read it periodically to stay on track.