Brynn Lewis on “Searching for a new Orlando”

A recent graduate of the Computing Science undergraduate program at the University of Alberta, project member Brynn Lewis discusses her contributions to Orlando’s site redesign:

You’ve heard it here; you’ve heard it there; you’ve (hopefully) heard it everywhere: the Orlando Project is undergoing a site redesign. Our ambitious site transformation will deliver a sleek new user interface and improved search functionality, making Orlando both more accessible and more powerful. Our focus is on providing researchers greater access to the core of the project: the data itself.

All of the Orlando Project data—consisting of meticulously researched author entries and associated person, bibliographic, organization, and event documents—is tagged in XML. Marking up documents in this way provides a lot of associated context around the raw text.  This data has powerful potential. Our redesigned site is both flexible and intuitive, encouraging complex and explorative research. Want to find authors who published books in Liverpool? Want to look for women writers with specific family structures? Orlando will easily allow you to do that.

That is where my work, and the work of the site redesign team, comes in. As a research assistant through the Undergraduate Student Research Initiative (USRI) program, I am tasked with helping to revamp and implement a new search interface. The new search includes faceting, allowing users to “drill down” to increasingly specific results. If a user only wants to look at French authors born after 1815 who wrote satirical comedies, they will soon be in luck.  We are also creating an Advanced Search, which will allow users to build complex queries around the XML tag context of a search result. Users will be able to look for particular references to cholera when it is explicitly the cause of an author’s death, or the occupation “sailor” but solely in the context of a romantic relationship. The Advanced Search is designed with enough flexibility to combine and exclude different sets of results, allowing potential researchers to get very specific about their interests.

Our data is primarily stored in BaseX, an open source XML database. XML databases like BaseX allow developers to pull information using XQuery, a language specifically designed to work with XML data. With XQuery, I have written queries used to extract, summarize, and search Orlando information. It allows me to target the specific “path” needed to reach a “node” in a document. A lot of my summer has also been spent integrating my database work with our back-end—our site uses a content management framework known as Drupal, which is based in PHP.

My work has involved decisions about how to best present Orlando’s content, an ongoing discussion around definitions and functionality. How can we tell, based on the data, what an author’s nationality is? What is the best way to model gender while still allowing for the complexity and variety of gender expression?  How can we determine a person’s birth and death date, and what do we do if we cannot extract that information? Sometimes these are higher-level editorial decisions, but on a low level I am also constantly making choices about how to best turn a larger vision of Orlando’s future into a concrete application.

The interface of a site and the way that data is presented to a potential user can have major ramifications to the way that the user understands and interacts with that data. Because of this, work like the Orlando redesign is not always a straightforward process but rather a thoughtful, ongoing conversation about the best way to understand our data. 

This project has not only given me practical experience working with databases and backend development; it has also given the chance to think about wider concepts in the Digital Humanities, especially around how to make data and research tools accessible. It is important to balance the flexibility of a tool with the usability of a tool, and to consider the needs of your audience.  It is also important to consider the base assumptions of your user base, and how that might impact their understanding of the data output. By creating a thoughtful design, we can make sure that researchers are able to both access Orlando and easily interpret and use the data they have accessed.  

So where does this leave us? For me, I am coming out of this project with technical programming and database experience but also (hopefully) a more thoughtful perspective. I was lucky enough to work closely with a talented group of people, including Susan Brown, Mihaela Ilovan, Nia Kathaloni, and Jeffrey Antoniuk who taught me a lot about consciously designing and delivering research tools.  For Orlando, work continues as we bring together the final project pieces. Together we hope to deliver a redesigned site that will excite researchers and allow them to fully take advantage of all that the Orlando Project has to offer. Stay tuned!