Using DOI's for Component within Data sets - Issues and Results

From Earth Science Information Partners (ESIP)
Revision as of 12:10, November 1, 2011 by Nhoebelheinrich (talk | contribs) (→‎Glacier Photo collection images)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Glacier Photo collection images

  • Pre-assignment of DOIs
    • Needed to find service provider who could help us assign IDs, use existing MD for citation and other purposes
    • We're using EZID Service from California Digital Library which is a member of the DataCite Consortium (1 yr agreement with ESIP Federation D& S cluster
    • We're mapping the existing descriptive metadata associated with the collection to the minimum requirements of the DataCite Metadata Kernel
      • Creator
      • Title (of Photograph)
      • Publisher
      • Publication Year
      • Related Identifiers (DOI of entire collection with IsPartOf relationship declared)
    • Original dump of metadata for this collection was delivered in a .csv.gz file
    • Data was ported to a SQL database
    • Some clean up required to ensure data could be parsed correctly
  • Assignment of DOIs (implementation)
    • DOIs assigned using the EZID API
    • Form of DOI after the prefix used the unique photo-ID assigned to each image, e.g., doi:10.5060/D4RN35SD/{photo_id}. > e.g. doi:10.5060/D4RN35SD/baird1929090101
    • The root identifier doi:10.5060/D4RN35SD has also been registered as a kind of placeholder.
    • Identifier information can be found by going to EZID at [1], and entering the identifier in the "Manage" tab. Or, just append the identifier to http://n2t.net/ezid/id/{identifier}, as in [2]
  • Unresolved issues
    • How to link the assigned identifier to its photo (i.e., an ID is just an ID unless the data object to which it is attached can be linked?
      • Perhaps a splash page needs to be set up for each photo(if not already created as is the case with the Glacier Photo collection) that connects the identifier with the photo, and perhaps, points to the descriptive metadata for the photo? If a splash page has been created, the page for each photo would have to be changed to include the DOI for the photo.
      • The splash page might also include a sample citation of the data as additional, helpful information (using ESIP's data citation guidelines?)
      • In the case of EZID service, can see the descriptive metadata when connecting to the URL form of the ID as well as the owner (coded info), date ID created, etc.
    • How to advertise or make the identifier known once assigned?
      • By assigning DOIs to this collection and by associating the minimum required metadata, DataCite will include these photos in its search service.
      • Additional search services "might" also include the identifier and its metadata in future, e.g., Thompson-Reuters' Web of Knowledge/Web of Science)
    • What if the descriptive metadata changes? Will / how the metadata also be changed in the places where it is included?
    • Metadata can easily be changed using the EZID service, but the responsibility to do so lies not with EZID, but rather the data creator or publisher. A data maintenance policy between data creators / publishers should include this kind of updating responsibility.
    • What happens when a new photo is added to the collection? Do the collection maintainers know what to do? Do they have the appropriate tools? Will they even remember that there is something to do in the first place? This is a matter of incorporating identifier management into the workflow procedures and data management policies of the collection and curating institution, a critical component of the curation of data and identifiers.