specify domains for MIxS properties

[ramonawalls]

If, as per GenomicsStandardsConsortium/mixs-rdf#9, we are using RDF properties for MIxS attributes, we should decide whether or not to specify domains for some or all of those properties.

[cmungall]

I think the domain would be something like Sample or Study. Do we mean ranges?

I definitely think we want to specify constraints, see https://docs.google.com/presentation/d/10j2dRtHnYZgspiNytaH9eVkQSi4Su0Pojt394uVItlo/edit#slide=id.p

our proposed representation works for rdf, shex, json-schema, and defines a small number of range types

See the NMDC rendering: https://microbiomedata.github.io/nmdc-metadata/docs/AttributeValue

• BooleanValue - A value that is a boolean
• ControlledTermValue - A controlled term or class from an ontology
• GeolocationValue - A normalized value for a location on the earth’s surface
• IntegerValue - A value that is an integer
• PersonValue - An attribute value representing a person
• QuantityValue - A simple quantity, e.g. 2cm
• TextValue - A basic string value
• TimestampValue - A value that is a timestamp. The range should be ISO-8601
• UrlValue - A value that is a string that conforms to URL syntax

[ramonawalls]

This issue was referring to domains, which make mixs much more ontological. Let's save this for a future mixs release. Of course ranges are also important.

[ramonawalls]

Issue no longer relevant.

[turbomam]

MIxS is (intentionally?) vague about domain. Since MimsSoil uses the depth slot, MimsSoil is in the domain of depth

But if a user is filling out the MimsSoil metadata collection template, what is each row? @mslarae13 made the reasonable suggestion that it would be a DNA sample that happens to come from soil.