Archive for September, 2013

Sep 27 2013

Classics, “Digital Classics” and Issues for Data Curation | DH Curation

Published by under Metadata

Some required reading for anyone interested in the Perseus Project at Tufts: Classics, “Digital Classics” and Issues for Data Curation | DH Curation.

Comments Off

Sep 26 2013



In the last post I mentioned that my job is not all about provenance. But sometimes it is, and doing the research to both identify the owner, and provide the ability to track that information provides a great deal of value to our records.  In this case, the owner of this book on the plague was Silvester Gardiner, a Boston physician from the 18th century, and a bit of a loyalist.  In 1778, his name appeared on the proscription and banishment act, and his property was confiscated and sold at auction, including his personal collection of rare books.

Provenance is an area rich for discovery in the Digital Humanist toolkit, and recently a group of students from Swathmore created an app that visualizes the provenance of rare manuscripts in the Schoenberg Database of Manuscripts.  Of course it is only possible to create that app with the metadata that had been added consistently to the collection.  In other words, if you don’t track it, you can’t find it.

Also, notice that the Wikipedia article uses a VIAF entry for authority control? If my record uses the authorized name then the possibility for linked data and “discoverability” increases, which is exactly what metadata librarians are trying to accomplish.

Comments Off

Sep 12 2013

Data Dictionaries

Published by under Metadata

Screen shot of the Hydra Admin Tool Data Dictionary

Screen shot of the Hydra Admin Tool Data Dictionary


Think being a rare book metadata librarian is all about identifying formats, answering questions of provenance and determining whether the binding is sheep or goat?  That happens, but the truth is that a large portion of my job is building Data Dictionaries, which keep track of the elements we want to describe in an online linked environment.  The Data Dictionary above is for the Hydra Administrative tool we built in collaboration with DCA and Digital Curration Experts for associating Dublin Core Metadata in the Fedora repository.  It is a collaborative work in progress that lives on Google Drive, allowing different authors to contribute their expertise as we seek to define a minimum common set of core terms.  It will ensure that records entered through the administrative hydra head provide a predictable level of description, and that the metadata is entered as consistently as possible across our different collections.  It is not as exciting as identifying a heretofore unaccounted autograph, but it is one of the most important ways we will be able to ensure future interoperability of objects as we begin digitization at Tisch Library.

Comments Off

Sep 05 2013

Not Dead On Arrival: Using Library Metadata to Ensure Future Use

My CAMWS panel discussion:

As teaching and learning becomes an increasingly digital activity, it is important to consider how future scholars will access, comment on, and even reuse the original scholarship produced by students and other researchers in this medium.  One way to ensure longevity of the end product, as well as future interoperability, is to use librarians and their specific knowledge of metadata standards in the online environment.  This is important because the tools of digital scholarship are always evolving, and the way our digital world looks and feels now, may not be the way it functions in 20 years.  The last thing students and their teachers want is for their work to be effectively dead on arrival, in a format that can’t be sustained, or scaled for future use.

That the humanities are beginning to take advantage of our connected online world is hardly surprising.  The Perseus Digital Library and its offshoots are examples of this at Tufts, but are by no means singular.  Consider the output from the Scholar’s Lab at the University Virginia, or even the Women Writer Project, and you can get a sense of the shift occurring in higher education. Texts are now mined, parsed and visualized in ways that that take advantage of new tools that seem to emerge every few weeks.  One of the challenges created by this shift is the long-term preservation of the intellectual output of our scholars.  Consider this: walk into any major research library, search the catalog, fill out the appropriate paperwork and you can probably view an original 16th century commentary on a Latin classic.  Do you think the same will be said of the work that you, or your students, are doing in the digital realm 20 years from now?  To put it another way, more work is being done in the online environment, yet the digital objects that we are creating are frequently far more ephemeral than their physical counterparts.  So how do we ensure that the same protections we have afforded books are also given to our digital objects?

For over a century librarians have been particularly effective at both preserving the traditional carriers of intellectual content (i.e. books) and providing the tools to discover them. Librarians are now evolving the principles they developed for bibliographic control in order to assist these digital projects.  Where librarians are particularly helpful is with the application of metadata.  Metadata is simply: “structured information that describes, explains, locates, or otherwise makes it easier to retrieve, use, or manage an information resource.” Metadata can assist both the longevity of a project and its discoverability–it contextualizes, and future-proofs your work.  By bringing librarians into projects early, they can start thinking about the ways to describe the output and the systems that will be necessary to support it.

Partnering with the Classics Department at Tufts University, Tisch Library assisted with two projects that highlight the issues facing digital scholarship, and were precursors to the Perseids platform.  In 2011, students in a Medieval Latin course began working with a digital collection known as the Miscellany Collection.  This collection highlights 32 manuscripts and printed leaves. By associating the images with descriptive elements from the Dublin Core metadata standard, the students were able to visually verify information about the leaves, as well as produce their own translations.  While the Miscellany was a beta project, designed to encourage discussion within the library about supporting digital humanities, Dublin Core is an established standard used by most libraries.  This means that the scholarly work now associated with the digital objects will be preserved for future use.  That is, while the front facing web presence might change, the resulting intellectual product will be preserved by a backend system designed to maintain digital content for the long-term.

In 2012, Tisch Library provided assistance with a Text Encoding Initiative (TEI) project pursued by a Master’s candidate in the Classics.  TEI is a robust metadata markup language used in the digital humanities, which libraries are just beginning to support.  By reaching out to the library early, the student established the guidelines for how to describe the key linguistic elements, and felt confident that the preservation pieces were in place for the project.  While long-term preservation and description of a complex digital project like this one is an admirable outcome in itself, projects like these also force librarians to understand the sort of tools used by scholars to work with the final product.

For Tisch Library this is truly exciting because it moves us into the realm of collaborative creation. By once again bringing the library into discussions about Perseids, we hope to collaborate with our colleagues by not only providing the support and infrastructure to ensure the long-term discoverability of this new environment for student and professor research output, but also contributing to the tool itself.


Comments Off