RLG Programs 2008 Annual Partners Meeting
Renovating Description and Practice Breakout
Session Recap Sent to Participants

June 2-3, 2008

The following is a transcription of the eight flipcharts participants voted on in the breakout session, in order of the ones that garnered the most votes:

  1. develop systems, methods and tools for integrating and sharing (to the network) metadata contributions from curators, subject librarians, experts, users, etc. (includes capture at multiple levels, annotation, etc. as well as descriptive terms) [51 votes]
    • current practices and levels » use-case scenarios
  2. create systems and tools to manage and expose semantic, hierarchical, granular relationships (across pillars) [39 votes]
  3. develop tools, including computational analysis, for automated metadata extraction and creation (including descriptive, administrative, technical, etc.) [39 votes]
    • carve out which sources
    • what would have the most impact
  4. share (with goal of shared/wider implementation) ideas, best practices and tools to streamline metadata creation workflows in and from diverse environments [38 votes]
  5. develop systems, methods and tools for capturing, managing and continually enriching rivers of metadata (includes external source capture and batch process management) [37 votes]
  6. gather and map terminologies and identities from multiple domains, languages and scripts and link to workflows [30 votes]
  7. use FRBR to create work- and expression-level descriptions for use across "pillars" [25 votes]
    • FRBR manifestations can help us manage our colletive collection
  8. create discovery middleware (including links from terminologies and identities
Transcription of the original set of flip charts noting the "key issues" raised by the breakout session participants were then synthesized into the above items by Carol Mandel and Bob Wolven. Numbers in brackets indicate the project area the key issue was merged into.

  • vision/conceptual model of where to be in 3-5 years for current investment
  • data mining and computational analysis [5]
  • use of terminologies and vocabularies, including multi-script and multi-lingual, harmonization [4]
  • names, identities, organizations [4]
  • manifestation identification at the network level [4]
  • systems for capturing and managing "rivers" of metadata, including external/commercial sources ("tributaries") [1]
    • cross-pillar metadata searching nontext, visual and audio
    • digital and licensed
    • work relationships
    • user access regardless of source
  • integrate metadata contributions from curators, librarians, experts and end-users [2]
  • automatic metadata creation [5]
  • link tools to real workflows and environments and have impact
  • non-Latin scripts [4]
  • FRBR for (use) work descriptions across communities and for relationships [6]
  • discovery tools that leverage user and expert metadata [3]
  • social networks to capture user metadata (á là Wikipedia) and expertise [2]
    • capturing metadata at both local and network level
  • more sharing of workflows, best practices, and tools for technical and administrative metadata
  • creation, capture and use connecting and integrating specialized work in technical and administrative metadata to larger community [5]
  • linking and semantic techniques and metadata to make sense of relationships [6]
    • metadata for relationships
  • managing relationships, granular and hierarchical data [6]
    • building blocks for the "community catalog"
  • leverage discovery tools – for terminologies, names and places developed outside the community [3]
    • "middleware" between end-user and the metadata
  • methods and tools for cross-domain, L-A-M [8]
  • embryonic metadata; batch tools for metadata creation and enrichment [1]
  • special collections at scale
  • sharing ideas and tools across different environments [7]
    • best practices for metadata creation and conceptualization
    • streamline workflow in diverse environments

Other updates:

The May 8, 2008 RLG Programs Metadata Tools Forum that Cheryl Gowing mentioned attending during our breakout brought together nine tool developers and their intended user communities together.  These tools were developed to expedite metadata creation or editing and could be considered one way of "sharing best practices and tools for streamlining metadata creation workflows in and from diverse environments". They were selected for the forum because they could be used by multiple institutions and relied on commonly available software most institutions would already have in-house or based on open source components. Almost all attendees saw at least one tool that could be used in their own environments. The summary sheets describing each of the nine tools demonstrated with contact information for those who want to know more and a recap of the discussion following the demonstrations have been added to the RLG Programs Metadata Tools Forum Web page.

The Prototype a Cooperative "Identities Hub" project could help move us to the area the group thought would have the most impact: share the network level metadata contributions from curators, subject librarians, experts, users, etc. The RLG partner working group to identify the components that would have the most benefits and to develop use case scenarios for potential target audiences is still being formed. We would like the working group to represent a range of RLG partners and staff representing "diverse environments". We welcome your recommendations for RLG partner staff for this working group. Since this area received the most votes, it is the one we would like to follow up on first.

For more information

Karen Smith-Yoshimura

Program Officer