While my blog was down, i was stoked to see Clay write about visualizing social networks. He references a great, albeit older (2000), overview article about visualizing social networks. In comments, folks also bring up Tamara Munzner’s thesis.
I’m in awe of all of the efforts to visualize social networks, but not surprisingly, i have my biases on this topic. The big question for me about visualizations is WHY? Most answers to this question fall into two camps: to efficiently understand massively complex data sets or because it’s cool. Both are super valid reasons, but the approaches that evolve almost always focus on visualizing data for some outside third-party, unrelated to the data. Of course, all of us who do visualize are also obsessed with visualiing our own data, but we don’t count. Although we weren’t able to deploy our visualizations to too many people, Fernanda/David and myself/Jeff were interested in what it means to visualized data for the people visualized. What would people do with these artifacts? How could ethnography be done with such tools?
The folks at Info@Vis! emphasize four basic elements of social network representations: presence, identity/affiliation, interaction, and communication. Personally, i find it peculiar that there is no discussion about the audience of the visualizations. Perhaps they are presuming that they are only visualizing for social scientists. But, even for social scientsits, another purpose of visualizing data is to engage those being visualized. It operates as a mirror and having that artifact provides you with such a grounding artifact upon which stories can be told.
Although i was unable to go to Hawaii to present out paper, Fernand tells me that it was quite well received there which makes me super happy.
Odd coincidence. Came across another of Freeman’s papers on another of the blogs I read. Same time frame, I think: http://moreno.ss.uci.edu/groups.pdf
Alternatively, here’s the link to his pubs page (http://moreno.ss.uci.edu/pubs.html) which does contain some more recent work…
cgb
Social Circles: Visualizing email list traffic
Danah Boyd links to social circles, a tool from marcos weskamp that can subscribe itself to a mailing list and…
Most humans are much more capable of seeing patterns in graphically visualized data and remenbering them, than in any other form of representation. I think this is the best argument for visualizing data. We literally SEE more in visualizations.
Visualization is the prequel to many kinds of interactions (navigation, exploration, etc.).
I’ve been playing with the same tools you have for a few years, as you know, and I’ve kind of decided that networks just aren’t often useful for the users. I wrote a review of literature searching for uses of social networks for end users, but didn’t find much that I found persuasive.
http://www.ics.uci.edu/~danyelf/publications/SocialNetworksforEndUsers.pdf
(That doesn’t mean they aren’t interesting, cool, and sometimes insightful–but I don’t see people wanting to frequently explore their own networks. No one introspects all that much, in any context)
In the end, I think, where they win out is as guidance to the general shape. Knowing what we do about a network, we can use it to design other things. I claim you should build the search interface on email to take advantage of the cliques and clusters in your social network, for example.
(And, yes, I have a paper about that, too.)
I just posted a collection of graphs reviewing the activity present in the MemeStreams community durring 2003. More information can be found at the URL’s thread.
In our case, we are creating visuals for both third-parties (anyone who visits MemeStreams) and the users who’s data is being mapped. It seems to us thus far that users spend more time looking at visualizations of their own data, then that of others..
In physics visualization has rather old history and its main purpose is to help understanding the cause/mechanics of the solution found. Visualization is not the answer itself and it needs some expertize to make most out of it. I don’t know much about social science but if the visualization is a new tool to it, I am strongly interested in how people use visualization.
Noel – that may be the approach in physics, but i disagree that that is the only way that it has to be. I really enjoyed creating visualizations as a tool for ethnography. But i think that you have to construc them differently to do so.