Metadata at IU InfoShare

The Metadata Discussion Group is back! We kicked off our first meeting of the academic year with an infoshare, in which attendees were invited to share a few quick highlights about their metadata work. Attendees discussed project progress, described how their workflows are adapting, and elaborated on how they are coping with new or evolving metadata standards. As the discussion progressed, a number of common metadata challenges emerged.

Couldn’t attend but still wish to share your metadata project? Leave a comment below!

Metadata at IU

Reports & Trends

  • AARC2 description in MARC has expanded far beyond physical books: attendees responsible for providing access to general, special, and archival collections reported record creation for a variety of physical and digital formats
    • Lilly Library’s Bradley Film Collection – initial cataloging outsourced, then cross-walked to MARC for inclusion in IU’s catalog, IUCAT; these films are transferred from film reel to DVD on-demand or if they are in danger (the IU Film Archive folks are arranging the digitization process)
      • Many films have succumbed to vinegar syndrome and have been lost
      • Lilly Library holds manuscripts; see the finding aid for more info
    • Cultural and Linguistic Archive of Mesoamerica (CLAMA) is a collection of digitized and streaming videos; 200+ films are being cataloged for IUCAT access (MARC records)
    • Archives of African American Music and Culture (AAAMC) – has audio and video in their collections, some of it is born digital
    • Media Preservation Initiative – has a database that uses locally-defined metadata elements; maybe useful to help describe AAAMC items?
  • Technical Services Cataloging Division reported expansion into non-MARC data standards: TEI, EAD, and MODS; a few Division staff have cross-trained in DACS; RDA training begins for original monograph catalogers in late September
  • Finding aids are crafted using Describing Archives: A Content Standard (DACS) and then encoded using Encoded Archival Description (EAD); ongoing development of the technical infrastructure supporting Archives Online has helped streamline the workflow and allow for more on-demand digitization in addition to mass digitization
  • Blacklight is being implemented as a discovery layer for IUCAT; data clean-up in progress; different display of records reveals mistakes in metadata; team has had to consider:
    • Browsing by call numbers
    • Other kinds of browse such as faceting
    • Responsive design (example University of Wisconsin-Madison Library catalog)
    • records coded incorrectly (globes example)
    • “Other” facet category – down to 3,000 items, maybe 1,400 items from cataloging fixes
    • what goes in the index – corporate authors were filling up the facets (United States, United States Senate, for example) – but still in searchable index
    • helps with needs in Bloomington, but not sure about other campuses
  • GIS is starting to get big in libraries–larger libraries are hiring GIS librarians
    • UITS has GIS folks that manage the interface for working with digital geo-spatial data here – Indiana Spatial Data Portal; recent release of 15,000 historical maps digitized (made news)
    • State spatial data should receive collection-level cataloging
  • Development of Kuali Open Library Environment (OLE)
    • Holdings metadata had to be completely redone (location and call number type of info):
      • OLE holdings – general info
      • Specific holdings – MARC holdings record
    • Can convert OLE holdings to MARC holdings but not other way
    • Starting from scratch means making very basic decisions, such as how to go about alphabetizing things
    • University of Chicago and Lehigh University expect to implement in mid-2013; IU will be Dec 2014 at earliest
    • OLE doesn’t have an OPAC component; our OPAC is Blacklight; we will have to run Blacklight against OLE data, which is very different than running Blacklight against SirsiDynix;
    • MARCXML is being used for MARC records; Dublin Core is other supported standard (with possibility of support for other standards)

Responding to New or Revised Standards

Common Challenges

  •  Lilly Library has many collections, such as some of their film collections that are available digitally for streaming–how best to deliver these?
  • Many in attendance are finding it hard to adhere to the one-to-one principle, in which one creates a record for each unique item, including any and all
    • TEI encoding of books we’ve scanned results in another record in our catalog for the e-book we produce
    • On-demand publishing could mean analytical cataloging of chapters of books
    • Reel film that is transferred to DVD is like a Xerox copy- do we really catalog all of these DVD transfers?
    • Dissertations – used to order photocopy and catalog it; how many records do we need of the same item in different formats?
    • Content standards, such as DACS often make recommendations–in the case of DACS: pick one format, either the original analog or the digital derivative, and stick with that decision in your institution; however, that decision is impacted by what you’re trying to accomplish and make available–to deliver a streaming file, you have to catalog the streaming file; locally, attendees have responded to this in various ways:
      • Variations/Avalon project takes the one record approach for all versions of a resource; transcript, audio, photograph series are all done separately
  • Granularity of cataloging: for hierarchically arranged resources, such as those held by archives, how granular do you make your descriptions? Series level? Folder level? Item level?
  • When throwing all of your digital resources in one big pot, disparities in content standards and the choice whether or not to use controlled vocabularies is VERY apparent
    • War of 1812 finding aid used DACS to formulate data for various fields in at the collection-level hierarchy; item-level title description conventions (all titles were given as dates due to the complexity of the collection) obscured the fact that some items may not be letters but were, in fact, maps; this was in stark contrast to the maps from the same exhibition available elsewhere that were assigned titles using AACR2
  • When migrating to a different discovery layer, decisions must be made about what to index, etc.; also where is it OK for metadata to be broken? How do these decisions impact the way we discovery resources?
  • The Metadata sub-group of the Image Collections Online Working Group is developing a core set of metadata to be applied across all ICO collections
    • Dates are proving problematic, especially when the one-to-one principle isn’t being strictly obeyed
  • Blacklight is being implemented for IUCAT; a separate instance is being used for the beta Digital Collections Search – can these interfaces combine?

 Topics for Future Meetings

As the infoshare progressed, attendees identified metadata topics they wanted to hear more about:

  • More about audiovisual projects (presentations from different groups)
  • Library of Congress is moving away from MARC – what’s happening?
  • Map cataloging using RDA
  • The One-to-One Principle and its challenges
  • RDA/linked data
  • How data practices and policies affect discovery
  • Libraries as publishers–impact on metadata creation and maintenance?

If you’d like to suggest other topics, please contact the co-moderators.

Author- Julie Hardesty

Julie Hardesty is the Metadata Analyst for the Indiana University Libraries. Her work focuses on the use of metadata standards and best practices to enhance access to and discoverability of academic online resources. She holds a Masters degree in Information Science and a Master of Arts degree in Art History, both from Indiana University.