Re: [poe] IM wording for classes and their relationships

@riannella thanks for all the changes you have applied so far.
Re your list above:
1. What does "**more** than just the RDF model" mean? Class and Property are defined by the RDF Schema and this is part of RDF (see https://www.w3.org/standards/techs/rdf). (Note: the Web Annotation Rec tells "The examples throughout the document are serialized as [JSON-LD] using the Context given in Appendix A of the Annotation Vocabulary"  and the Vocabulary makes heavy use of RDF(S) and OWL - I guess nobody will assume the Web Annotation does not build on RDF.)
2. thanks
3. In practice such a "single point of information" does not work. If one looks into [2.5.1 Permission](http://w3c.github.io/poe/model/#permission) and can't find there that this is a subclass it would be hard to argue "you should have looked at the list of classed below the IM figure". If an overview makes a basic definition (like: "this is a subclass") then this should be reused in a detailed definition.
4. I would appreciate having "Property" in the headline. 
like 2.1 Asset Class --> 2.2.1 Relation Property --> 2.2.2 Part Of Property - to override the assumption all sub-items of a section about a Class are subclasses.
5. Sorry, the lists of properties for Atomic and Compound Constraint are quite different and defining different sets of properties and a different use of them for instances of the same class does not fit well into the ODRL design. Ok, that's not an IM wording issue, should be discussed by a specific issue.
6. welcome
7. To be precise: I suggest to use such wording templates for defining properties:
a/ cardinality 0..1: A {domain class name} may have one or none {property name}  for {range class name}
b/ cardinality 1: A {domain class name} must have one {property name}  for {range class name}
c/ cardinality 0..unbounded : A {domain class name} may have none, one or many {property name}  for {range class name}
d/ cardinality 1..unbounded : A {domain class name} must have one and may have many {property name}  for {range class name}
(note: hm, 2.7 doesn't show a "non-normative" statement, can't get your thought)
8. thanks for a good start in 2.7. But there are still such "contains", e.g. 2. Information Model -> list of classes, Permission: "... The Permission may also **include** the Duty  ..." or 2.1.1 Set "... An ODRL Policy of type Set MUST **include** at least one Rule. ..." (my browser shows 59 occurrences of "include" in the IM document, thanks for sorting them out)
9. sorry, the related to does not help much in this case. The list of properties of the Policy Class (still) does not show properties for a target Asset, an assigner Party etc. But that's more an issue as discussed in #202 





-- 
GitHub Notification of comment by nitmws
Please view or discuss this issue at https://github.com/w3c/poe/issues/200#issuecomment-311277916 using your GitHub account

Received on Tuesday, 27 June 2017 07:35:44 UTC