The ALCTS Preconference “Cataloging and Metadata for the Web: Meeting the User Where They Are” was held on Friday, June 23, 2017. Of the 63 attendees, approximately half were from academic libraries, while the rest were from public libraries, special libraries, and a few other types of organizations. The preconference included seven speakers with questions and answers throughout the day.
Ted Fons, Principal Consultant, Third Chapter Partners, set the tone for the day with his presentation “Into the Hands of Readers: The State of Visibility of Libraries on the Web.” Fons began by recommending that we be mindful how we talk about web visibility for libraries. Librarians should not get caught up with the technical issues and vocabulary but should be more focused on goals and desired outcomes. When we talk about linked data it should be couched in terms that reflect what we want to achieve with it. Our projects are less focused on outcomes and more about tactics and details, often losing sight of the actual goal. If the goal is to have readers find library content from wherever they are then we need to follow the rules of the web. The framework for the web consists of traditional ranked results, sponsored links and the knowledge card. Fons recommends that librarians get their content into the knowledge card (a block of information inset on a web search results screen) as it’s more effective than simple words. Take ownership of it and do not depend on others to create your data. It is important to engage with outcome-oriented projects that involve our discovery layers and third party vendors.
Kenning Arlitsch, Dean of the Library, Montana State University, began his presentation, “Improving Visibility in Search Engines: How collections and organizations benefit from Search Engine Optimization and Semantic Web Identify,” by discussing the importance of having a formal Search Engine Optimization (SEO) program to support linked data projects. He explained that having a SEO program will allow search engines to discover library websites and institutional repositories, to more accurately crawl the websites and to better understand the metadata obtained from them. The Semantic Web Identity (SWI) improves the search engines’ ability to recognize organizations. He described the University of Utah digital library team’s projects that included the more than a million newspaper pages and 500,00 objects of different formats, the Mountain West Digital Library, Utah Digital Newspapers, Western Waters Digital Library, Western Soundscape Archive and the USpace institutional repository. He discussed the challenges for institutional repositories including SEO deficiencies in the library, and technical and organizational SEO deficiency themes. He provided a brief background on the SWI, SWI research activities and data collected by ARL institutions, the basic SWI process and a summary of the research conducted. He concluded by describing the challenges for librarians and actions that could be undertaken for greater impact.
Three of the speakers discussed their experiences as early adopters partnering with Zepheria to increase the visibility of their library collections on the web.
Sharon Day, Director, Branch Services & Collections, Edmonton Public Library, “Meeting Users Online: The Edmonton Experience”, discussed their experience as an early adopter partnering with Zepheria. She stated their intended goals to increase the number of catalog records indexed and searchable by search engines, increase in referral traffic to their OPAC, and increase new users visiting the OPAC. She described the Library.Link Visibility Lifecycle, the transformation from MARC to BIBFRAME to BIBFRAME/Link.epl.ca and the SEO Tools Used (Google My Business and CMS SEO Descriptions). She discussed the various data points analyzed including date and frequency of data refreshes/publication to link.epl.ca, number of MARC records harvested, number of BIBFRAME resources created, number of link.epl.ca URLs created, dates sitemaps were submitted to and processed by Google, and number of link.edpl.ca URLs submitted and indexed by Google. She also discussed the challenges they encountered which included loss of control of data, the data not being refreshed quickly enough, and the harvesting of inappropriate records.
Susan Allen, Director of Technology Services, Worthington Libraries, discussed their experience partnering with Zepheria. She discussed their LibHub/Library.Link timeline, the reasons that they chose LibHub/Library.Link, training exercises, their readiness assessment, and the details from their data pilot and the post-pilot results. Challenges to discoverability include that it took several months to see results in Google searches, the community name must be included in searches as the geolocation is not working as well as they would have liked and they would prefer that users would be directly brought to the library rather than the portal page as it creates confusion. Their plans going forward include staying aware of BIBFRAME and linked data development, assess and measure the impact of their project and future experimentation and field testing with Zepheira.
The presentation “Library.Link at Multnomah County Library Year 2” was given by Erica Findley, Cataloging/Metadata Librarian, Multnomah County Library. In her presentation she gave a brief background on their Library.Link Assessment Team activities, their motivation for becoming an early adopter, their goal of making resources available on the internet and goals for year three. She stated that the goal of using linked data was “making connections and transforming the way data work.” Going forward they are making a change to the redirect and are hoping for better raw data to compare libraries as a result of their new data dashboard.
The sixth presentation, “Tools for Enhanced Discovery on the Web”, was given by Jim Hahn, Orientation Services & Environments Librarian, University of Illinois at Urbana-Champaign, and Qiang Jin, Cataloging Librarian, University of Illinois at Urbana-Champaign. In their presentation, the speakers discussed a project to transform approximately 300,000 e-book records in their collections into linked data base using the BIBFRAME model 1.0. The process included extracting MARC XML, transforming to BIBFRAME XML, enriching with open linked data, and indexing by Google Custom Search Engine. They generated queries to gather unique bibliographic identifiers for e-books, programmatically extracted MARC XML and programmatically transformed MARC XML to BIBFRAME XML. They described the enrichment process including the tools and resources used. Their analysis of their outcomes showed that a limited number of unique titles in the top search results and the average search results were approximately half way down the page demonstrating that there was more work to be done. The next phase of their project is to transform the records into BIBFRAME 2.0.
The day wrapped up with a detailed summary that included highlights from each presentation provided by Ted Fons.