Undefined Future Uses

Last week I attended a lecture titled “Computing and the Practice of History” by Dan Cohen, Director of the Center for History and New Media  at George Mason University.   He focused on three things to explore how the digital world is changing the way historians conduct their work:  1) Archives/Collections – the foundation for all work; 2) Methods; and 3) Scholarly communication.  While many specialized collections continue to attract scholars to explore unknown territory, there are also new opportunities coming from mega collections such as HathiTrust and meta-mega collections such as Europeana, Open Context, NINES and others.  These collections benefit from new tools that can reveal the texts from different perspectives, in many cases beginning with a quantitative analysis that can lead to new questions.

He went on to say that these collections create a platform that supports not only the collection itself but also other connections.  He quoted Roy Rosenzweig, founder of the Center for History and New Media, who believed in creating ” a generative platform for undefined future uses”.  Such platforms must be open and support APIs. They must also be able to disclose their metadata as another means of exploring the collection that can “enable or disable” forms of inquiry.

These observations certainly ring true with CDL’s experience in aggregating collections and supporting platforms for their use, such as for eScholarship or Calisphere.  While these services may have once been focused on becoming portals, now they are more aligned with being platforms supporting a range of uses.  We know that most users arrive not through the front door but from a referral in another source or from a web search engine.  Objects need to be able to stand on their own rather than relying on an organized pathway to their place in the collection.  They must provide context to reveal the larger collection or their related associates within it.  But there is more to be done to enable those “undefined future uses” if we think about how the text (in the case of eScholarship) and the metadata could be mined, or how to take advantage of commentary and corrections from users.  We should focus on other ways to make these services function “at the network level” or to think about the web first, something Cohen noted as a principle when designing new services.  The problem to solve these days is not discovery but how to provide context and the means to select and filter, either within the service itself or to allow others to do so.  Search facets have been used for this purpose, but there are other methods to reveal the peaks and valleys within a collection and to enable deeper exploration.  Part of our curatorial role is to analyze collections and be open to having others analyze them, to help shape them for future scholarship.