New skill sets for metadata management

That was the topic discussed recently by OCLC Research Library Partners metadata managers, initiated by Jennifer Baxmeyer of Princeton, Dawn Hale of Johns Hopkins University and MJ Han of University of Illinois at Urbana-Champaign. Educating and training catalogers has been at the forefront of many discussions in the metadata community. Today’s changing landscape calls for skill sets needed by both new professionals entering the field and seasoned catalogers to successfully transition to the emerging linked data and semantic web environment. Catalogers are learning about and experimenting with BIBFRAME while remaining responsible for traditional bibliographic control of collections. Metadata specialists utilize tools for metadata mapping, remediation, and enhancement. They identify and map semantic relationships among assorted taxonomies to make multiple thesauri intelligible to end users. For the more technical aspects of metadata management, we increasingly see competition for talent from other industries. This may intensify as metadata becomes more central to various areas of government, non-profit, and private enterprise.

Managers want to focus less on specific schema and more on metadata principles that can be applied to a range of different formats and environments. Desired soft skills included problem solving, effective collaboration, willingness—even eagerness—to try new things, understanding researchers’ needs, and advocacy. Although some metadata specialists have always enjoyed experimenting with new approaches, they lack the time to learn new tools or methodologies while keeping up with their routine work assignments. We should promote metadata as an exciting career option to new professionals in venues such as library schools and ALA’s New Members Roundtable. Emphasizing that metadata encompasses much more than library cataloging can increase its appeal, for example: entity identification, descriptive standards used in various academic disciplines, and describing born-digital, archival and research data that can interact with the semantic Web. As one participant noted, “We bring order out of a vacuum.”

Metadata increasingly is being created outside the library by academics and students who receive minimal training, leading to a need for more catalogers with record maintenance skills. Participants noted the need for technical skills such as simple scripting, data remediation, and identity management to reconcile equivalents across multiple registries.  Frequently mentioned sources of instruction include Library Juice Academy, MARCEdit tutorials, Lynda.com, Library of Congress Training Webinars, ALCTS Webinars, Code Academy, Software Carpentry and conferences such as Code4Lib and Mashcat. W3C’s recently published Data on the Web Best Practices and Semantic Web for the Working Ontologist were recommended reading. Crucial to the success of such training is to be able to quickly apply what has been learned. If new skills are not used, people forget what they have learned. Staff feel frustrated when they have invested the time to learn something that they cannot use in their daily work regularly.

We’ve seen a big shift from relying on instructions from the Library of Congress to self-education from multiple sources. Some approaches mentioned by participants:

• Emphasize continuity of metadata principles when introducing an expanded scope of work.
• Take advantage of the Library Workflow Exchange, a site designed to help librarians share workflows and best practices across institutions, including scripts.
• From the recent Electronic Resources & Libraries Conference: “Don’t wait; iterate!” In other words, rather than waiting until staff have all the required skills, let them do tasks iteratively, learning as they go, so they are ready for new tasks when the time comes.
• Have small groups of metadata specialists take programming courses together, after which they can continue to meet and discuss ways to apply their new skills to automate routine tasks.
• Send staff to events such as OCLC’s DEVCONNECT, OCLC Developer Conference being held on 8-9 May 2017 to learn from libraries using OCLC APIs to enhance their library operations and services.
• Create reading and study groups that include cross-campus or cross-divisional staff.
• Expand the scope of current work to enable metadata specialists to apply their skills to new domains or terminology, such as using Dublin Core for digital collections. Involve staff in digital projects from the conceptual stage to developing project specifications, quality assurance practices and tool selection. As an added bonus, this fosters collaborative teamwork relationships.
• Hire graduate students in computer science for short-term tasks such as creating scripts. The students need money and the library needs their skills.

The extent of collaboration with IT or systems staff varies among institutions. Such collaboration is necessary for many reasons, including managing data that is outside the library’s control. Some noted that “cultural differences” exist between the professions: developers tend to be more dynamic and focus on quick prototyping and iteration, while librarians focus first on documenting what is needed and are more “schematic.”  Which is more likely to be successful: teaching metadata specialists IT skills or teaching IT staff metadata principles?  The “holy grail” is to recruit someone with an IT background interested in metadata services. Retaining staff with IT skills is difficult—if they are really good, they can find higher-paying jobs in the private sector. Ideally, metadata managers would like a few staff who have the technical skills to take batch actions on data, or at least know how to use the external tools available to automate as many tasks as possible.