cgspace-notes/content/posts/cgcore-types-harmonization.md

2.8 KiB

+++ title = "Harmonization of CG Core Output Types" date = 2021-02-21T13:27:35+02:00 description = "Proposed changes to CG Core types after review of several CGIAR repositories." categories = ["Notes"] tags = ["Migration"] url = "cgcore-types-harmonization" draft = true

+++

Proposed changes to the CG Core controlled vocabulary for output types after review of actual usage by several CGIAR open access repositories.

With reference to CG Core v2 draft standard by Marie-Angélique as well as DCMI DCTERMS.

Proposed Changes

As of 2021-01-18 the scope of the changes includes the following fields:

  • cg.creator.id→cg.creator.identifier
    • ORCID identifiers
  • dc.format.extent→dcterms.extent
  • dc.date.issued→dcterms.issued
  • dc.description.abstract→dcterms.abstract
  • dc.description→dcterms.description
  • dc.description.sponsorship→cg.contributor.donor
    • values from CrossRef or Grid.ac if possible
  • dc.description.version→cg.reviewStatus
  • cg.fulltextstatus→cg.howPublished
    • CGSpace uses values like "Formally Published" or "Grey Literature"
  • dc.identifier.citation→dcterms.bibliographicCitation
  • cg.identifier.status→dcterms.accessRights
    • current values are "Open Access" and "Limited Access"
    • future values are possibly "Open" and "Restricted"?
  • dc.language.iso→dcterms.language
    • current values are ISO 639-1 (aka Alpha 2)
    • future values are possibly ISO 639-3 (aka Alpha 3)?
  • cg.link.reference→dcterms.relation
  • dc.publisher→dcterms.publisher
  • dc.relation.ispartofseries will be split into:
    • series name: dcterms.isPartOf
    • series number: cg.number
  • dc.rights→dcterms.license
  • dc.source→cg.journal
  • dc.subject→dcterms.subject
  • dc.type→dcterms.type
  • dc.identifier.isbn→cg.isbn
  • dc.identifier.issn→cg.issn
  • cg.targetaudience→dcterms.audience

Out of Scope

The following fields are currently out of the scope of this migration because they are used internally by DSpace 5.x/6.x and would be difficult to change without significant modifications to the core of the code:

  • dc.title (IncludePageMeta.java only considers DC when building pageMeta, which we rely on in XMLUI because of XSLT from DRI)
  • dc.title.alternative
  • dc.date.available
  • dc.date.accessioned
  • dc.identifier.uri (hard coded for Handle assignment upon item submission)
  • dc.description.provenance
  • dc.contributor.author (IncludePageMeta.java only considers DC when building pageMeta, which we rely on in XMLUI because of XSLT from DRI)