Executive Summary
2CUL TSI Non-MARC Metadata Working Group
Phase 1 Report
2013 July 30
Team Leads: Jason Kovari (Cornell, chair), Melanie Wacker (Columbia)
Additional Team Members: Robbie Blitz (Columbia), Steven Folsom (Cornell), Wendy Kozlowski (Cornell)
Compile an inventory of non-MARC metadata staffing and expertise at both institutions
The technical services divisions at Columbia University Libraries and Cornell University Library have a combined total of 7 staff members working primarily in non-MARC metadata (Cornell: 4.5 FTEs and Columbia: 2 FTEs). In addition, each institution has at least one additional staff member working with non-MARC metadata, though not as a primary job function.
One of the non-MARC FTE at Columbia is primarily devoted to non-MARC metadata creation and remediation, the other mostly focuses on strategizing, consultations, crosswalking, as well as staff education and training. On rare occasions, other catalogers may work with non-MARC metadata if subject or language expertise are needed. At Cornell, non-MARC metadata work focuses on consultative services and design strategy, with some VRA Core cataloging for the images-for-teaching collection and occasional science or special collections metadata generation. The Columbia non-MARC staff work also with MARC metadata while non-MARC staff at Cornell have little interaction with MARC metadata.
For more information on the staffing and expertise, see the full report: [provide link to the full document once available in Confluence]
Examine and evaluate reporting and decision-making structures at both institutions
Administrative reporting structures and functional reporting structures vary significantly at the 2CUL institutions. Non-MARC metadata staff at each institution report administratively within their technical services units; however, functional reporting and decision-making structures vary on a project-level or initiative-level and rarely occur within technical services for the Metadata Librarians / Coordinator. For non-"Librarian" or non-"Coordinator" staff members, functional reporting and decision-making structures usually align with the Metadata Librarian / Coordinator overseeing the project.
For more information on reporting and decision-making structures, see the full report: [provide link to the full document once available in Confluence]
Identify dependencies and limitations inherent in working with other functional areas at Columbia and Cornell, both within and beyond technical services
Non-MARC metadata projects at both 2CUL institutions have significant dependencies outside of their respective technical services departments, which closely align with the functional reporting structures described above. Limitations often entail technical support and development, including programming (both institutions) and desktop administrative rights (Columbia).
For more information on dependencies and limitations, see the full report: [provide link to the full document once available in Confluence]
Compile an inventory of all non-MARC metadata policies, practices and workflows at both institutions
Columbia has significantly more documentation of non-MARC policies, practices and workflows than does Cornell. At Columbia, metadata documentation is aggregated in various wikis; documentation include data dictionaries, mapping tables, OMEKA documentation, MODS documentation, etc. At Cornell, documentation focuses on VRA Core cataloging and Research Data and Management Service Group projects; most other projects have established workflows and practices but lack formalized documentation.
For more information on policies, practices and workflows, see the full report: [provide link to the full document once available in Confluence]
Establish baseline productivity numbers for activities and projects at each institution to allow for future assessment of potential changes and development associated with 2CUL TSI.
At both 2CUL institutions, the tasks of the Metadata Librarians / Coordinator is difficult to quantify; thus, metrics have not been uniformly captured. This derives from the project-based tasks that dominate non-MARC metadata. The consultative and advisory nature of the Metadata Librarians / Coordinator further limits the statistics that can be uniformly generated. Non-MARC metadata creation is more easily quantifiable and has been captured for OMEKA, Center for Digital Research and Scholarship repository metadata and Libraries Digital Program Division digital projects (Columbia) as well as VRA Core cataloging (Cornell). At Cornell, most digital project metadata is captured outside of technical services with Metadata Librarians performing quality control; statistics are not collected on these projects.
For more information on metrics, see the full report: [provide link to the full document once available in Confluence]
Recommendations regarding a work plan and critical issues to explore in Phase 2
The working group has identified some areas of possible collaboration during Phase 2 of the 2CUL Technical Services Integration; note: the order of the following list does not necessarily reflect prioritization.
Common Development of Single Metadata Capture and Editing Tool
The working group recommends the development of a joint-2CUL web-based metadata capture and editing tool for digital assets. This project's logic seems to closely align with the desire to implement a common 2CUL LMS. By having a common capture and editing tool, the two institutions could work collaboratively on non-MARC projects, facilitate alignment of non-MARC metadata practices and potentially streamline one of the issues plaguing both institutions: enhancing quality of non-MARC metadata and inconsistency of field usage.
Ideally, the capture and editing tool will be able to export metadata in multiple schema and formats (e.g.: Dublin Core, VRA Core, MODS, RDF triples, CSV file, etc.) and include spell-checking and the inclusion of linked data authorities (e.g.: id.loc.gov, GeoNames, VIAF and Getty Vocabularies, when available in Linked Data). For Columbia, this tool would likely replace the need for OMEKA's cataloging interface; for Cornell, this may replace the cataloging interface in SharedShelf.
The working group recognizes that this could require a considerable development effort and buy-in from the departments with which the Metadata Librarians / Coordinator have functional dependencies. Neither institution has a tool with the export functionality described above or that makes use of linked data authorities; if an adequate tool exists on the market, the working group would appreciate consideration into purchasing for both institutions. If locally developed, this tool could yield a shareable product for the metadata community that could be available as an open source solution. The working group envisions a multi-phase project whereby the first phase develops the tool and the second phase develops an API to publish metadata to common platforms at each institution; the creation of automated interfaces with our platforms would work particularly well if both institutions implement a common framework for storage and discovery, such as Hydra.
Research Data Services Collaboration
The working group recommends whenever possible, collaboration in development of research data handling and research data services, particularly as it relates to scientific metadata creation. Because support of research data services is relatively new to libraries, sharing of ideas and experiences will be extremely useful for advancement of our ability to assist patrons in this area. Research metadata support tends to be very domain specific and creating a shared knowledge base between the two institutions would benefit both, and could potentially expand upon 2CUL collaboration outside of technical services.
Formalized Consulting Framework
The working group recommends the development of a more formalized consulting framework between the two institutions. This would allow for being able to transfer consultation requests based on expertise requirements. For instance, a Cornell-based consultation request concerning MODS or OMEKA metadata could be transferred to Columbia whereas a Columbia-based consultation concerning VRA Core or Kaltura metadata could be addressed by Cornell.
Shared Documentation
The working group recommends creating a space for shared documentation related to non-MARC metadata. This would allow metadata practice at both institutions to build upon each other meanwhile providing the information needed if the two institutions align non-MARC metadata practices. Further, if the common metadata capture and editing tool is developed, shared documentation would further aid in 2CUL collaborative metadata projects.
Collaborative Training and Investigations
The working group recommends beginning cross-training for new standards. Further, the two institutions should collaboratively conduct investigations of new tools and standards for metadata creation and management. Co-leading investigations could yield shared tools and practices. Aligning investigations and training could save a net-effort across the two institutions as compared to conducting two separate investigations/trainings.
Metadata Working Group (MWG) Reconfiguration
The working group recommends a reconfiguration of the Metadata Working Groups (MWG) at each institution. Each MWG would engage both institutions in metadata-related forums via Skype, Polycom or Webex. Further, a Columbia representative could join Cornell's MWG Forum Steering Committee and a Cornell representative could join Columbia's MWG. Currently these forums occur at Cornell once-per-month during the academic year and are publicly announced and attended; a steering committee, commonly referred to as Metadata Working Group, plans the forums. To view past forums, see: https://metadata-wg.mannlib.cornell.edu/. At Columbia, the Metadata Interest Group meets in irregular intervals, on average every two months. Meetings are planned by the Metadata Coordinator. Events are only announced to a metadata e-mail list, though anybody is welcome to attend or request to be added to the distribution list. In addition, Columbia's Digital Projects Librarian also hosts "Digital Library Seminars", which are open to the public.
If non-MARC metadata practices at the two 2CUL institutions extend to become fully-integrated, a joint functional-MWG should form. This functional-MWG would be a separate but parallel entity than the current forum-based MWGs; a separate steering group should guide the effort of the forum-based MWG.