Browsing by Subject "Digital collections"
Item Session 2C | Wrangling Serial Titles and Place Names in the UNT Libraries’ Digital Collections(2022-05-24) Phillips, MarkThe UNT Libraries’ Digital Collections has grown to include over 3 million unique digital resources including maps, newspapers, photographs, audio, and video records. These digital collections use the UNTL metadata format, that is based on Dublin Core and includes qualifiers that allow for more specificity about a field to be represented. While the UNTL metadata format works well in describing a wide range of digital resources held in our collections, one thing that has not been modeled well historically is the concept of a “Title” such as a serial title for a newspaper, like the Austin American-Statesman or a “Place” such as Denton, Texas. This past year we have taken the first steps to manage titles and place names in a more robust way in the UNT Libraries’ Digital Collections. This involved the creation of a system to model the concept of a Title and the concept of a Place that could be populated with information that provides descriptive and specificity to adequately represent these concepts. Trying not to reinvent the wheel, this approach leveraged data from the Library of Congress databases to link title records with existing LCCN and OCLC numbers. Likewise places are linked with Geonames and Wikidata to provide equivalences between systems. Finally appropriate user interface elements were integrated into the system to expose this information to the end user so that they are able to make use of this effort in identification and disambiguation of these concepts. This presentation will present the problem we were facing, explain the approach, and provide examples of next steps in this space.Item Session 3B | Bias in Online Library Collections Searches(Texas Digital Library, 2022-05-25) Hetrick, EmmaThe main issue of this one hour workshop will be biases in library digital collection catalog searches. I will begin by analyzing the Harry Ransom Center’s digital collections within the context of critical archival studies, critical Internet studies, critical race studies, and search engine studies. My aim is to identify how the "search" function of these collections operates, and determine how well it provides relevant results for searches about identity. This study is modeled on Safiya Noble’s Google searches in her book Algorithms of Oppression. I conclude that the “search feature” is limited in yielding relevant results, and furthermore, that the digital images in the database are not labeled in a manner to yield more desirable results. This research contributes to several of the efforts described in the HRC's Diversity Action Plan, and I hope will provide actionable recommendations to increase the accessibility, inclusivity, and representation of the HRC's collections and the way they are described online. To that end, I want to open the conversation to workshop participants and their experiences with searching library catalogs in an effort to brainstorm changes that can be made to these catalogs. No prior knowledge or experience is required, but attendees will be encouraged to try some searches of their own either before or during the workshop.