Friday, September 22, 2006

ANT, Reseau-Lu, and a wish-list


There is a particularly useful lesson on datastructures posted on this online computer science tutorial. The reason that the lesson is particularly useful to me is that it provided some answers concerning some new technologies being in science studies(made up philosophers, historians, anthropologists, and sociologists- mostly the last group.)

Actually, I think that the wiki on science studies provides a pretty useful tool if you want to find out what it is all about. Since the history of medicine oftentimes converges with the history of science, I have had some experience with this sort of literature.

Of my experience, I have found that one of the most interesting (-and dare I say useful?-) tools to come out of this field is the Actor-Network Theory. Once again, the wiki provides a decent over-view of the theory, but it also leaves out the more recent changes and developments.

This theory is particularly useful as the basis for schematizing the spread of scientific knowledge. Recently, some science studies articles have used certain computer programs to create a two-dimensional representation of these scientific networks. The most current co-citation software being used is Reseau-Lu.

The program was used by Alberto Cambrosio (McGill), Peter Keating (UQAM), and A. Mogoutov (Aguidel) in an analysis of antibody reagent workshops (Social Studies of Science 34/3 (2004):325-64). By using Reseau-Lu, the authors were able to create a two-dimensional map of the cluster-designation workshops that was not limited to just the relationships between people: the map included the relationships between workshop groups, molecules, cell markers, and equipment. Interestingly, the methods employed by the authors to analyze the heterogeneous map were similar to those used by the scientists to develop a cluster designation nomenclature. Here is one of the maps used in the analysis:


Now, back to data structures. The lesson that I mentioned at the beginning of this page helped me to visualize just how this sort of software operates. The nodes seen on the map are basically the nodes described in the lesson. It strikes me that the Graphing Data Structure lesson can help one to appreciate how a program like Reseau-Lu operates. This sort of program could be extremely useful to historians who want a way to represent complex interactions between people, places and objects in the past. With a little tweaking, the program could even be used to represent various periods (years, decades, etc) of the network.

The only thing that I can see preventing me from using this sort of software is price. As I understand it, it happens to be pretty expensive, and it does require some training. Then again, it's not as though my weekends are completely booked these days, so that is not much of an excuse. Money...now that is certainly holding me back...

1 comment:

William J. Turkel said...

Adam, check out the Open Source tools Graphviz and Pajek. No need to spend money at all...