Snagging list - Githubissues

pesc-org / json-ld-transcript

Creating a machine readable, semantic PESC transcript

https://www.pesc.org/college-transcript.html

1 stars 0 forks source link

Snagging list #16

Open philbarker opened 11 months ago

philbarker commented 11 months ago

Queries arising from other issues that need wider discussion before decision can be made:

In Students Name:

[ ] compositeName is a legacy field, not widely used. Probably doesn't matter much. displayName would be more useful (optional). Could use full name from CEDS.
[ ] Probably ought to expand enumeration to include things like preferredName.
[ ] GenerationCodeOrSuffix should be valueDataType xsd:token (it currently isn't because of limitation in limitation in tap2shacl

4pins commented 10 months ago

Sounds right. Our approach is similar but diffrent.

http://specification.sifassociation.org/Implementation/NA/4.3/CommonTypes.html#obj:BaseNameType

philbarker commented 9 months ago

In organization

[ ] What more specific types of educational organization do we need.
[ ] What levels do we need.

4pins commented 9 months ago

@philbarker Is this where you meant to make the most recent comment. From my point of view, it doesn't fit in this (closed) ticket.

philbarker commented 9 months ago

Yeah, this is just a tracking issue for miscellanea we need to pick up on later. It's not closed.

philbarker commented 8 months ago

In Organization Identification

[ ] Need to figure out how best to relax constraints so that implementers can use their own Ord ID scheme, hopefully while still warning if no recognized scheme is present.
[ ] See whether CEDS will treat OPEID as identifier scheme

Currently must have one of the recognized schemes.

philbarker commented 8 months ago

Names (mostly organization)

[ ] Org names in CEDS are tokens, we need langStrings.

Requires change in CEDS or us to do something different to CEDS.

philbarker commented 7 months ago

Residency (and address??)

[ ] Country code should come from a (long) picklist
[ ] State code should come from a (long) picklist

philbarker commented 1 month ago

In LanguageShape and language of instruction

[ ] the only test is that the language code is an alpha-3 string; no test that it is a valid ISO 639-2 code.
[ ] are extended codes allowed?

philbarker commented 1 month ago

Need to make sure that each individual tap sheet has enough of the shapes to check the type of objects that are in other tap sheets. For example, no use saying that targetContactPoint has value shape ContactPointShape if nothing of that shape is defined. It's enough to define the expected type as rest can be checked in the relevant TAP sheet.