Page tree

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  1. Open Actions
    1. No updates
  2. Administrivia
    1. Reminder: Schema Board nominations open 1 March 2021. Board members at the end of their term: Alan, Keith, Benn, Mario, Terry
  3. schacGender
    1. Heather reached out to the federations that have registered items in the URN Registry, and of the five that answered, none use this attribute
    2. Three options:
      1. Change the name
      2. Change the definition
      3. Deprecate (with community consultation) - this has full consensus of the schema board.
        •  Heather Flanaganwill send out a poll to determine what SCHAC attributes are commonly used in case any others need to be deprecated, and whether the currently experimental attributes need to be moved to fully supported
  4. Status of splitting the eduPerson spec into core and supplemental
    1. Splitting Protocols From Schema Documents
    2. Suggest we start with the attribute dictionary as originated from the I2 TAP work. That dictionary can then be represented in several forms (e.g., LDIF). It's still an abstract concept; not for developers to tell them how to implement their client. The attribute dictionary would then be broken down further into profiles (e.g., SCHAC, eduPerson, voPerson). This is a big change, but if done well, you can transform attributes in a deterministic way across protocols.
    3. If we do this, we have to go through the exercise of taking the eduPerson and other schema attributes into something that will fit in the attribute dictionary. For eduPerson, about half easily map into the TAP (potentially eventually the REFEDS) attribute dictionary, and the other half require a little more work. If we want to do this for SCHAC and voPerson, will need to do the same
    4. Nothing actually changes on the wire today; this is all behind the scenes abstraction work.
    5. The parent doc would be narrative; the JSON docs would be sub-docs within specific protocol space.
    6. This effort could offer easier, consistent attribute mapping.
    7. Next immediate step: review all schema for alignment with the attribute dictionary.
      •  Keith Hazelton will continue to work on the REFEDS/TAP attribute dictionary based on additional schema (SCHAC, voPerson, remainder of eduPerson)
      •  Benjamin Oshrin will send out an update to the schema-discuss list re: the REFEDS/TAP attribute dictionary
  5. Entity Categories - status
    1. R&S 2.0 - group has been making good progress, but we may be getting stuck on OIDC mappings. The R&S 2.0 working group is not really the right place to discuss attribute mapping, so the next call will discuss whether we need to not include OIDC attributes in this entity category (yet)
    2. Authentication Only, Anonymous Authorization, Pseudonymous Authorization - REFEDS SC is working through the voting process