Skip to content

Latest commit

 

History

History
164 lines (91 loc) · 7.15 KB

2022-05-03.srap_meeting_14.md

File metadata and controls

164 lines (91 loc) · 7.15 KB

DC SRAP meeting 14

1. Opening of the meeting

Juha opened the meeting at 14:05 CEST

2. Appointment of the minutes taker (Juha will act as the chair)

Osma selected as the minutes taker.

3. Approval of the agenda

Agenda approved.

4. Minutes of the previous meeting (2022-04-19)

Minutes were published at https://github.com/dcmi/dc-srap/blob/main/meetings/2022-04-19.srap_meeting_13.md

5. Review of proposed SRAP Publication statuses

Approved the following list of statuses:

  • public draft
  • submitted manuscript
  • preprint
  • postprint
  • publication
  • updated publication

See #11

Tom: What is an update?

Juha: It's an updated publication - something has been publicly available but later changed.

Tom: Could it be called "updated publication" then?

Juha: Yes, we can change that.

Juha: Can we approve this as final (for now) so we can move into other parts of the profile?

No objections, status list approved. This will be a part of the SRAP specification that will eventually be sent to the Usage Board.

Juha: These should not be set in stone - if there are good reasons to change, the discussion can be reopened.

6. Review of proposed SRAP Date subproperties

See #20

Date subproperties and their definitions in:

https://github.com/dcmi/dc-srap/blob/main/terms/date_subproperties.md

were approved by the attendees of the 13th meeting of the WG. Is this good enough for public review or do we still need to change something? (all)

Osma: What about dateAccessioned? This seems to be automatically generated by DSpace repositories, meaning the time when the document was uploaded. A bit lower level, technical timestamp than these.

Juha: We can return to this later; I am not fundamentally opposed if there is a good case for it.

Jan: dateAccessioned seems more like administrative metadata of the repository, more than about the resource itself.

Alasdair: A date in the repository more than relevant for resource itself.

Juha: Technical, administrative. If we want to add it, need to think how it relates to the statuses.

Osma: Maybe better not to include.

Juha: Do we need to change dateUpdated as well?

Tom: Could it be dateUpdatedAsPublication? This would follow the same pattern as the others. dateUpdated is too general.

Juha: We can provide guidance. For instance, dateMissing could relate to any status whereas dateRetracted only involves publications or updated publications. Rename to dateRetractedAsPublication.

Jan: Dates are already very specific. Do we want them to be more generalizable for DCMI? These are now specific to libraries and academic publication processes in journals.

Alasdair: For dateUpdated, if we don't make it specific, we could run the risk that people don't read the scope notes and use it improperly.

Juha: Should we provide essential information in the name of the property or just in the description? We are toeing a fine line here. dateUpdatedAsPublication is a useful clarification - we don't want people to use it for when a preprint is turned into a postprint.

Tom: Resources will accumulate many dates, which gives a record of status changes. If we used less specific, more generic properties, that would need more data modelling.

Juha: It's important to keep a record so we can determine who came up with an idea first. Other formats are not doing this yet.

Alasdair: Would we allow multiple similar dates for a series of updated?

Juha: That might require creating a new resource. Minor updates can be done in the same resource/record, but major updates would require a new one.

Juha: I think this is a reasonable starting point. Are we ready to approve this set of date properties?

Tom: Has Karen given any comments?

Juha: She has commented in the past, I've considered them in the current proposal.

Approved the following date subproperties:

  • dateAvailableAsPublicDraft
  • dateReceivedAsManuscript
  • dateSubmittedAsPreprint
  • dateSubmittedAsPostprint
  • dateAccepted
  • dateAheadOfPrint
  • datePublished
  • dateUpdatedAsPublication
  • dateRetractedAsPublication
  • dateMissing
  • dateLost

7. Affiliation information

See #3

Description of the new class and properties should be written in a Markdown snippet (Osma).

There is an example (by Osma) of expressing affiliation information in XML: https://github.com/dcmi/dc-srap/blob/main/profile/affiliation-example.xml

Osma: Have not written up the Markdown snippets yet - next meeting. (Shows XML example: paper 20 years ago with three authors and affiliations at the time - very simple record with data and DC terms - title and three authors expressed with dct:creator. In all cases, value of creator element is not just a name, but element "Affiliated Agent". PID attribute used to express identifier - could be ORCIDs, Wikidata IDs - any sort of semi-persistent identifier. Structured way to express information given on the first page of the paper. Question whether it is appropriate to use the attribute pid= in this way. Agent and Affiliation as properties of creator. Attribute id= is reserved in XML - so cannot use - this was a bug in what we discussed in Porto - Opened Issue 17 in 'dcmi/pids_in_dc' repo. With id= can only have one value - and value must be unique within the context of the XML document.)

Juha: Problem with PID: "persistent identifier" should not be just an ordinary URL. Would use that only for an ISNI or ORCID.

Osma: I can edit the example to put in real ORCIDs or other PIDs.

Alasdair: We could take quick jump up to my agenda point - been contacting people who were interested. Alasdair, Osma, Scott, John on a call? My question: If we do not use id= (good reasons not to).

Tom: Could publish as a DCMI Note.

Juha: Can't be part of SRAP, it's more generic. SRAP can use it if it's published.

Alasdair: There are different use cases for PIDs, including also subjects.

Juha: Similar function as MARC subfield zero.

Alasdair: If we had published best practice guidelines, we could explain how to use this attribute.

Tom: If you introduce a new XML attribute, is it good practice to introduce it with a namespace? If it required a namespace, I guess it wouldn't be that much extra work. Important to follow established conventions in the XML world.

Juha: So we will have a recommendation on PIDs in DC and we can offload this to a group of other people.

8. Any other business

8.1 Drafting the SRAP specification in Markdown

There is already a crude Markdown version of the original SRAP draft specification (2021-01-08): https://github.com/dcmi/dc-srap/blob/main/profile/srap_proposal_20210108.md

Snippets created after the 13th meeting (if any).

8.2 Describing creator-related PIDs (ORCID, ISNI, ROR) in Dublin Core – status of the WG (Alasdair)

8.3 Next work items

Juha: We could take up new element such as grants. Will discuss with Osma and Tom.

9. Next meeting

Scheduled to take place 2022-05-17, 13:00-14:00 UTC

10. Closure of the meeting

Juha closed the meeting.