Exploring Trading Consequences

In March, a fantastic new resource for studying the history of commodity trade was announced: Trading Consequences.

The project is the product of several years of collaboration between York University, Canada, the University of Edinburgh, UK, the University of St Andrews, UK and the University of Saskatchewan, Canada.

The resource provides multiple interfaces to a rich database of mentions of commodities and locations associated with commodities from the 18th century and up to the mid-20th century. One interface is a Commodity Search tool, which allows you, for example, to search for all documents that refer to jute and display a world map that clusters references by region until you zoom in for individually geolocated entries. You may also search by location to show, for example, all commodities associated in the documents with a location like Hong Kong. In this interface, you can refine the search by collection or by decades. Another powerful way to interact with the material is through an Interlinked Visualization (use the Chrome browser) where, at a glance, you can view the distribution for mention of commodities across the decades. Another Location Cloud Visualization lets you view relative frequency of mentions of a commodity in various places over time.

There is so much that can be said for a monumental effort of this kind, but in browsing through the resource, a few thoughts come to mind:

  1. This project shows what is possible with a broadly interdisciplinary and international collaboration of a team like this.
  2. One of the things that impress me most about this project is the way that they have been open about their progress. See for example some of their blog entries (1, 2, 3) that contain a wealth of insight for others who might want to take on a similar project. They also created a detailed white paper that tells you much more about how the project was carried out. They have also made some of their code available on a repository at GitHub. While there is not much in the way of scripts here, they include CSV files with their lexicon of commodities, and a gazetteer of 1710 ports and cities with ports.
  3. The project is not just a collaboration among universities and across disciplines but shows, like so many of these projects, how much we build upon the open efforts of others. Their blog is on open source WordPress, they use OpenStreetMap for maps, the BSD free licensed Leaflet javascript library for plotting data on these maps and the D3 javascript library for their data visualizations. Their locations link to the Geonames geographical database of placenames and their categories and commodities are tied with Dbpedia structured data which is extracted from Wikipedia.
  4. Just as this project builds on powerful open tools and resources, it also shows how much the world of digitized historical materials are in walled gardens that are sometimes only accessible to institutions that can afford to pay the subscriptions. As a historian or student digs into the Trading Consequences site, they will see a single source may contain hundreds of references that have appeared in the results. The next step will often then be to go into the source in question and explore. Trading Consequences conveniently provides links to the relevant page or source in the various databases. Some sources are open, like this letter in the Royal Botanic Gardens, Kew: Archives collection or this entry in an early Canadian periodical. But many others are not, including entries in the House of Commons Parliamentary Papers – one of the most amazing cases of historic public documents being made accessible through a commercial subscription service (in this case Proquest).

Now that the project is launched, we can look forward to learning more about how historians can use the resource, whether it be as a heuristic tool for discovery or for analysis. One thing I would love to see at Trading Consequences and all projects like it is the development of an open–and well documented–API to the database that would allow outside send queries to the database that returned structured data. This would allow others to continue to build creative ways to interact with this rich data source.

Have you found other great resources of a similar kind? If so, how have you found them useful in your own research?

Return to Top