Dxwg: Consider the addition of a conformanceTarget property

[nicholascar]

Following the OGC's Modular Specification [[modspec]], consider adding a property conformanceTarget (domain: prof:Profile, range: ?) that allows for the linking of Profiles to the sorts of thing or instances of the sorts of thing that are expected to conform to it.

What might the inverse property of this be?

[kcoyle]

Nick, can you give a functional link to the OGC spec? Thanks.

[smrgeoinfo]

In answer to @kcoyle question The Specification Model - A Standard for Modular specifications (08-131r3)

[smrgeoinfo]

An example spec using this approach:
https://docs.opengeospatial.org/is/16-008/16-008.html .
Requirements for three standardization target types are considered:
• Logical Model
• Encoding
• Data instance.

[smrgeoinfo]

The Energistics Energy Industry Profile of ISO 19115-1:2014 (v1.1) (https://www.energistics.org/wp-content/uploads/2017/03/EIP_v1.1.pdf) Defines a URI to identify instance documents conforming to the profile in section 4.2.9 Metadata Standard:
http://w3.energistics.org/energyml/profiles/EIP/v1.1/metadataStandard_citation.xml

[dr-shorthair]

Also see this paper: Configuration management of a system of interdependent standards as mentioned in https://github.com/w3c/dxwg/issues/72#issuecomment-382207466

[dr-shorthair]

_Figure 1. Conceptual model of normative dependencies in ISO/TC 211 standards_

_Figure 2. Conceptual model of a modular standard (from OGC, 2009)_

[nicholascar]

It looks like we might have bypassed this Issue by 1. detailing the competency questions used in the definition of prof:isProfileOf and by 2. dealing with resources via roles.

For 1.:
IF A isProfileOf B
AND X conformsTo A
THEN X conformsTo B

For 2.:
If a profile wishes to indicate a resource that can be used to assess conformance to it, such as a SHACL file, then that could be indicated as:

:A 
    a profi:Profile ;
    prof:hasResource :R .

:R
    a prof:ResourceDescriptor ;
    prof:hasRole role:conformanceTest .  # an imagined role, yet to be fully defined

[nicholascar]

No further comment on this issue in a week and it appear to me to be bypassed (see comment above) and we are running out of time for vocabulary editions so I'm marking it "due for closing".

[rob-metalinkage]

I think we can leave this - the OGC model has a target per requirement - not per "conformanceClass" (aka Profile) - so its not a direct evidence of the usefulness of such a pattern. I think this can be considered as a future work item, because the typology of conformance targets will require some thought and experience.

[kcoyle]

To close we need to poll the participants.

All - please upvote if you agree to leave this as a future work item.

[nicholascar]

Closing after WG notification period of due-for-closing and the vote above to leave this for Future Work. We'll review all Future Work-tagged Issues, close or open, in a final Issue search before final draft of this WG Note.