emory-libraries / OpenEmory

open-access repository for scholarly articles, based on Fedora Commons
open.library.emory.edu
2 stars 2 forks source link

Spike: Assess gaps between prototype data and Hyrax/Fedora 4 data #117

Closed eporter23 closed 1 year ago

eporter23 commented 1 year ago

In Sprint 5 we added basic PCDM properties as well as FileSet properties. As a next step we need to assess what other containers and relationships may be needed for Hyrax compatibility.

abelemlih commented 1 year ago

Additional requirements include:

eporter23 commented 1 year ago

@abelemlih I think Hyrax also requires all works to be part of an Admin set. We can figure out how to create one manually or once we get a Hyrax running, it should do that for us. Example from fedora 4 test object:

ns002: isPartOf
http://fedora-cor-test.library.emory.edu/fcrepo/rest/prod/ad/mi/n_/se/admin_set/default
eporter23 commented 1 year ago

@abelemlih I also wonder if we need to populate at least Core Hyrax metadata noted here? https://samvera.github.io/metadata_application_profile.html

abelemlih commented 1 year ago

@eporter23 @bwatson78 I added some findings to the following file regarding some of the questions above.

I am still researching ordered members and access control, and will add my findings to the same file. Let me know if you have any questions.

eporter23 commented 1 year ago

@abelemlih thanks for these notes! it is interesting about Admin sets not being required. When we ingest items to Curate, for example, both bulk import and UI work creation enforce a work being added to an admin set. In both cases they are always deposited as members of the Default Admin Set (even if no explicit selection is made).

abelemlih commented 1 year ago

@eporter23 @bwatson78 I added my last comment concerning ordered members to the document above. Please let me know if there are any additional questions/topics I should look into.

abelemlih commented 1 year ago

Both Emily and Brad reviewed my findings and recommended I close this ticket.