Editor’s note: The Isis Current Bibliography is a constant work-in-progress, and for this installment of the column, HSS Bibliographer Stephen Weldon invited two of his graduate assistants to explain the workings of a cool new feature of the online bibliography: the addition of maps for visualizing the geographical areas related to the subject matter of the citations in IsisCB Explore. This addition will expand the scholarly value of the Isis-CB, not only as a growing resource of secondary scholarship but also as a dynamic research tool.
by Paul Vieth & Kraig Bartel
In this age of networked data, the biggest asset of the IsisCB is perhaps not the bibliographic records it contains, but the metadata relationships we have carefully constructed between those records. In order to make sense of those relationships, we maintain a parallel database of authorities: concepts, events, people, time periods, institutions, publications, and places. In order to make these authority relationships more accessible and actionable to researchers, we have now added a map on each authority page. This visualization will assist researchers to make sense of the geographic distribution of the history of science, technology and medicine.
This map serves two purposes. First, as the IsisCB is a paratextual companion to secondary literature in the history of science technology and medicine, the map allows users to interactively explore records organized geographically, “diving down” through the map to the citations it visualizes. Second, as the IsisCB also functions as a primary text, the map shows the geographic contours of the scholarly landscape of every authority contained in the database, In this way, the maps reveal for each topic both the well-trodden ground and the terra incognita, and generate questions about scholarly trends and lacuna in need of attention.
Designing and implementing this geographic data visualization involved theoretical, practical, methodological, and technological hurdles. Given the nature of cartographic demarcation and the fact that this is a database of historical information, we had to contend with the tensions and incommensurabilities between the ever-in-flux boundaries of political divisions and the rigid fixity of static two-dimensional representations of Earth. For example, how do we plot scholarship that takes place in the Roman Republic versus the Roman Empire of 40 AD versus the Roman Empire of the 4th century AD? What was Italy before 1861? By what boundary was the Korean peninsula circumscribed at the turn of the century? Under Japanese occupation? During the Cold War? In the end, we decided to elide these distinctions and present the map not as an ontologically authentic representation of the data in all its incommensurate messiness, but as a research heuristic meant to facilitate the playful interactions of researchers, texts, and metadata. Like those scholars who identify as historians of Warring States “China” or Medieval “Italy,” this map operates under the same presentist logics of significatory convenience.
For any given institution, publication, person, place, or concept we generate a map of related places connected to all of the citations connected to it. In creating the map, we have been constrained by the types of mapping software currently in use and by existing geotagging standards. The software we chose to use was a choropleth map, i.e., a map that colors areas based on the number of “hits”: the more hits, the greater the saturation of the color. Current open-source choropleth software requires the use of standard two-letter country codes (ISO 639-1), so all mappable geographic entities are linked to this code. In addition, we have associated all of our geographic entities—even those that cannot be given a country code—to a corresponding authority in the comprehensive GeoNames database. What this means in practice is that cities are placed in their current country, as well as geographic regions such as “South Asia,” associated with several countries, in this case: India, Pakistan, Bhutan, Nepal, Bangladesh, and Sri Lanka.
We faced many accessibility and usability challenges in implementing this map, and discussed the theoretical and practical concerns involved for several months prior to production. These maps appear on the Authority pages and add to the wealth of other data about each authority and its relationships to other authorities. We designed the map, and redesigned the page structure, to accommodate space constraints. We also considered aspects of cartographic justice—and accessibility. We chose a Winkel tripel projection that minimized the relative spatial distortions across the globe but still provides a familiar equatorial orientation. We’ve also worked on the color scale and how to portray the discrete color quanta so as to optimize its legibility for the most users.
We hope that this map enriches the IsisCB as both a paratextual accompaniment to its bibliographic data and as a text itself. We also hope that this map helps users navigate our database more efficiently and creatively and enables users to find and to create meaning in the data.
Paul Vieth is a PhD student & Kraig Bartel is a PhD candidate in the Department of the History of Science, Technology, and Medicine at the University of Oklahoma.