[INSPIRE Harvester] Open questions around mapping

[jrcastro2]

1. Authors Affiliations (authors.affiliations / record_affiliations)

Current fields:

• authors.affiliations.value
• record_affiliations.value

Other identifiers:

• authors.affiliations_identifiers.schema
• authors.affiliations_identifiers.value

Questions:

• What is the plan for exposing or storing the ROR identifier? What structure will the field have?

2. Editions Field

In Inspire, theses typically do not have an editions field, but other record types might.

Potential mapping options:

1. Add to imprint
2. Use version field
3. Add to description with type other
4. Add to title

Question:

• Which approach is better for mapping edition information? we propose to add to imprint

3. Funding Information (funding_info)

Inspire includes detailed funding information.

Questions:

• Should we import extensive funding data? (f.e. matching the vocabularies used in Zenodo)
• Is it important for researchers to have all funding/grant info imported into CDS?
• if we don't import all the grants, shall we raise errors and fail to import (harvest)?

4. Number of Pages

Inspire provides number_of_pages.

In CDS-RDM, the preview feature already shows the number of pages for uploaded files. The information is automatically counted for the uploaded file, so it does not require curation.

Question:

• Can we safely drop this field or is there a use case for keeping it?

5. Publication Type

Inspire provides publication_type.

Questions:

• How is this field currently used in INSPIRE?

6. Texkeys

Inspire includes texkeys.

Questions:

• How is this being used?

7. Withdrawn Records

Inspire provides a withdrawn flag.

Questions:

• Should we restrict access to files of the importer record if withdrawn = true?
• Any other special handling for withdrawn records?

8. DOIs

Inspire records may contain multiple DOIs.

Observations:

• If a DOI has a CDS prefix, we could use it as the main DOI.
• Other DOIs could be added as related identifiers or alternative identifiers.

Questions:

• If no CDS DOI exists, how should we select the main DOI?
• Should we import all other (non-CDS) DOIs as related works? Will you be able to curate the cases where the external DOIs should be an alternative identifier of the record?

9. Deleted

Questions:

• Could this appear, that we have deleted records marked for CDS?

10. publication_info.hidden

Questions:

• What is this hidden field used for?