|

Recommendations for repository managers on personal names

May 6, 2023

44 Leave a comment on paragraph 44 0 Write personal name/s as displayed in the deposited document and provide a persistent identifier enabling unambiguous identification, such as ORCID.

45 Leave a comment on paragraph 45 0 There are two main approaches to handling personal names in repositories:

46 Leave a comment on paragraph 46 0

  • using a unified preferred form, as defined in an authority file;

47 Leave a comment on paragraph 47 1

  • capturing the names as they are rendered in the deposited document.

48 Leave a comment on paragraph 48 0 The former approach is typical of library catalogs, where the unified form is used as a catalog heading. Depending on the country, names that are originally written in a non-Roman alphabet will be Romanized, or, conversely, transcribed/transliterated according to the rules used in a particular country. If a repository offers embedded metadata that can be imported into reference managers and preformatted recommended citations, this approach may not be optimal because the format of the name in the repository will differ from that in the publication.

49 Leave a comment on paragraph 49 0 If names are captured as they are displayed on deposited publications, the name of the same person will appear in the repository in various formats. In this case, it is important to use persistent identifiers, such as ORCID, to ensure proper identification and connect various name versions.

50 Leave a comment on paragraph 50 0 While in previous version of DSpace a workaround was required to display various name versions in a user friendly way (e.g. by means of an additional in-house application), DSpace CRIS and DSpace 7 not only support bidirectional integration with ORCID, but also treat persons as entities (CRIS entities and configurable entities, respectively) – e.g. https://scholars.lib.ntu.edu.tw/cris/rp/rp00095 (DSpace CRIS).

51 Leave a comment on paragraph 51 0 It is also important to ensure that persistent identifiers are exposed via OAI-PMH. PIDs in Dublin Core™ Working Group has developed recommendations to make it possible to expose persistent identifiers including ORCID, via OAI-PMH. Two solutions are proposed and both cover several use cases.

Option 1: Using an ‘id’ attribute with Dublin Core properties


Both PID and label are known
   <dc:creator id="https://orcid.org/0000-0003-1541-5631">Walk, Paul</dc:creator>

Label is known, but PID is not
<dc:creator id="">Walk, Paul</dc:creator>

PID is known, but label is not
<dc:creator id="https://orcid.org/0000-0003-1541-5631"></dc:creator> or
<dc:creator id="https://orcid.org/0000-0003-1541-5631"/>

(This option is not suitable if it is necessary to include more than one PID.)

Option 2: Using nested properties for identifiers


PID and label are known
<dc:creator>
	<dc:identifier>https://orcid.org/0000-0003-1541-5631</dc:identifier>
	<foaf:name>Walk, Paul</foaf:name>

Label is known, but PID is not
<dc:creator>
<foaf:name>Walk, Paul</foaf:name>
</dc:creator>
or:
<dc:creator>Walk, Paul</dc:creator>

PID is known, but label is not
<dc:creator>
	<dc:identifier>https://orcid.org/0000-0003-1541-5631</dc:identifier>
</dc:creator>

In this option, it is possible to provide multiple PIDs for the same property
<dc:creator>
	<dc:identifier>https://orcid.org/0000-0003-1541-5631</dc:identifier>
  <dc:identifier>http://paulwalk.net</dc:identifier>
	<foaf:name>Walk, Paul</foaf:name>
</dc:creator>

JPCOAR metadata schema also includes an element for researcher identifier, jpcoar:nameIdentifier (https://schema.irdb.nii.ac.jp/en/schema/3-1). It is exposed for Institutional Repositories Database (IRDB:https://irdb.nii.ac.jp/ ) via OAI-PMH, however, different types of IDs (KAKEN ID, ORCID, researcher ID, and others) are not integrated.


Source: https://comments.coar-repositories.org/personal-names/