Descriptive Metadata Call 2015-06-24

Owned by Bill Branan
Jun 16, 2017
4 min read

Time: 1pm EDT / 10am PDT

Call-In Info: Google Hangout: https://plus.google.com/hangouts/_/g2jey2y5cjcnggkmymxziudmw4a 

Moderator: Carolyn Hansen (U of Cincinnati)

Notetaker: Carolyn Hansen

Attendees:

 

Agenda:

  1. Metadata Survey
    1. Discussion about survey questions 
    2. Contact User Interface Group?
    3. Timeline
  2. Next Steps
  3. MODS To RDF discussion (Added by Steven Anderson - BPL)
    1. Current Committed / Confirmed Institutions For This Effort in Immediate Future:
      1. Boston Public Library
      2. New York Public Library (non-Hydra, non-Islandora)
      3. University of Maryland
      4. Amherst College

Link to survey: https://docs.google.com/forms/d/1azuJ_zDZN8JomL5ObgqltRLyRR5FS94bIl0h67t2e9k/edit?usp=sharing 

 

Notes:

1. Metadata Survey

  •  Carolyn: Daniel Enright from Symplectic will be joining the group. I've shared the survey with him and we are waiting on feedback from him and Corey Harper. Also, Corey contacted the User Interface Group regarding the survey; they expressed interest and we are waiting on more feedback from them too. Hoping to to deploy survey in early July and get feedback by end of July. Some questions still need to be refined

2. Next Steps

  • Carolyn: We will continue to refine questions. Waiting on feedback from Daniel, Corey, and User Interface Group. 

3. MODS to RDF discussion

  •  Steven: Would like to create official subgroup of Descriptive Metadata Group to focus on MODS mapping to RDF; separate subgroup is needed because this deals mostly with legacy MODS data, and not all folks on main group will be interested in this or it won't apply to their institution; looking to get institutions together (BPL, NYU, Amherst, and others) to create mapping and get feedback from broader community
  • Julie: How broad of feedback are you looking for? Julie says this is larger issue, are their other groups beyond us that would be interested in this since what LC has done with MODS RDF is not working for many 
  • Steven: Initial steps are to set up mapping; this mapping could work for other platforms using MODS RDF 
  • Arwen: Concern with splintering off a subgroup. We will have two parallel development processes and then have to reconcile them. There is no target to be mapping to yet
  • Chrissy: Thought the scope of the Descriptive Metadata Subgroup was to develop recommendations. Concern about having official subgroup to develop different recommendations from the larger groups. We don't want to have conflicting recommendations 
  • Steven: Target mappings may not be as granular; even by doing this work, it starts discussion of targets
  • Arwen: At UCSD, we represent everything in MODS, some level of detail that we don't see the usefulness for, also in the process of developing a new data model, Dublin Core with DPLA layer on top of it, w/ local idiosyncratic data as well. have had conversations with DAMS group, seriously reduced representations of title, so much in MODS that representing all of it is a big deal and a challenge. That's why the MODS RDF work has become very complex. What's the goal? If it's to map MODS in a good-enough way for use in Hydra, that's one thing. Representing all of MODS is another.
  • Chrissy: This is similar to what we're doing. Taking our best pass at it now and revisit it later after community recommendations have been written, difficulty converting complex subjects and dates (handling them similar to UCSD), UCSB not attempting to retain complex subjects (cataloging community as a whole is moving away from this). What fields are posing similar challenges?
  • Arwen: With catalogers, we are actually stopping the distinction between personal and corporate names (in the DAM, not the catalog). This should be in the authority record itself. Some things can fall away because of the functionality we will get from the linked data.
  • Chrissy: for ETDs, setting up harvest of catalog, not attempting to ingest a second record in the repository. decision to identify what is the master database for types of records. don't necessarily worry about losing data when we crosswalk,
  • Arwen: We are hoping to archive master metadata from ingest (not happening yet)
  • Julie: Are you getting URIs in place before transformations to RDF?
  • Arwen: no, ingesting with internal URIs, everything has a local authority record, looking at how to reconcile this with external vocabularies. focus first to get in RDF with local identifiers, next step is how to connect that with the greater world
  • Chrissy: we're in active development of image head, have URIs of metadata we're starting with, moving forward, hoping to do something similar to what UCSD is doing, for MODS records coming in, we added URIs before ingest using tools like OpenRefine. 
  • Steven: it is a big issue, part of the reason to have a dedicated group 
  • Arwen: might be a challenge of communication to the community, the relationships between the work the groups are going to do.
  • Steven: even when we do move to linked data, we want to move records back to MODS as well (have systems that are reliant on MODS)
  • Steven: inform people about the MODS subgroup (add questions or info into survey), feedback with mappings, giving feedback as needed, 
  • Julie: Steven are you still needing people to participate in the group? Get in touch with Steven. Trying to get started next week?
  • Kelcy: Should there be a wider call than Hydra? It's not a Hydra-specific issue
  • Steven: Wants to avoid very large calls because it will slow things down. If you know of institutions that are not Hydra, not Islandora, get in touch with them.
  • Chrissy: Interesting discussions with catalogers, since many MODS fields are MARC based and MARC is moving to BIBFRAME. What fields are legacies from MARC?
  • Carolyn: What will MODS subgroup need from larger group?
  • Steven: Need members to work on this; feedback from larger group. Group will begin forming in early July