Re: Comments about the Concepts in the SHACL Document

Thanks, Peter. I believe before it's worth to address these issues, we 
first need a general agreement on the terminology, how the engine's 
outer algorithms work and how the top-level syntax looks like. Your own 
proposal uses Constraints and Shapes differently than mine, so it makes 
little sense for me to update the current draft if this gets rewritten 
anyway. I also look forward to your suggestion on how to restructure the 
whole document (from a parallel thread). Meanwhile I suggest we continue 
with the raised issues for a few weeks (which could easily turn into 
months).

Holger


On 5/30/2015 6:09, Peter F. Patel-Schneider wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> I read over https://w3c.github.io/data-shapes/shacl/ and noticed a bunch of
> problems.  I have gathered these by concept and added some comments.
>
>
> Shape
> WRONG - "describes a group of constraints with the same focus node"
> WRONG - "is a group of constraints that have the same focus node"
>   - Mentioning focus node at this level is incorrect.
>   - Ignores scope aspect of shapes.
>
> Closed Shape
> ? - closed shape as one of the "other shape-based constraints"
> - - Is closed shape a constraint?
>
> Scope
> AMBIGUOUS - "pre-conditions that must hold before a constraint is applied to
>        a given focus node"
> - - No notion of constraint application defined in document.
> - - Scope is used on both shapes and constraints.
>
> Class Scope -
> AMBIGUOUS - "apply to all instances of these linked classes"
> - - What is an instance of a class?
>
> Constraint
> WRONG - "a condition that can be validated against a graph"
> WRONG - "defines restrictions on the structure of an RDF graph"
> - - Most constraints are validated against a graph and a node in the graph.
> WRONG - "evaluated against a focus node"
> - - Not for global constraints.
>
> Global Constraint
> - - Different from other constraints as they don't have a real scope.
>
> Constraint Violation
> POOR NAME - violations of component constraints may not be an overall
>            violation
> INCORRECT BEHAVIOUR - violations of component constraints are reported, even
>           for OR and NOT
> UNSPECIFIED BEHAVIOUR - constraints with components only specify how to
>              handle errors
> INCOMPLETE - no way of producing different kinds of violations in high-level
>       language
>
> Template
> INCOMPLETE - no connection betweeen query results and violations
>
> Supported Operations
> INCORRECT - validateConstraint needs a focus node for most constraints
> BADLY SPECIFIED - validateNodeAgainstShape ignores shape scope
> BADLY SPECIFIED - validateNode ignores some kinds of scopes
> BADLY SPECIFIED - validateGraph ignores some kinds of scopes
>          - validateGraph does not handle global constraints
>
>
> Missing:
> - - a firm definition of RDF graphs that encode valid SHACL constructs
>    - e.g., optionality of various parts of property constraints
> - - an integrated definition of how SHACL works
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2
>
> iQEcBAEBCAAGBQJVaMdvAAoJECjN6+QThfjzfIsIAJuhdgirBou0nD77I9BpxT6S
> j0AwcU2ZkHJUF01sYDUdlmRrwA3M9AmHmH1PCM2YHTk9mEzrL/f5YvWgZm8IIw/o
> 3AAaS2vJuaI3K5V1a82FjmENUcyZNbhL2/xkR9vA/mDmsRBDTMykcs7KtSUgGZSE
> PJMxDmsuc6+JMm9vboz9EgZVKLw2Nt9zjcgSKQfeYzYkVquPIFa0RN25jWY33JVz
> dATrN3VUXbBloXjP+JHlxM2YAajGdJP08vxXxYtArV6Tq9ZlUrKvRIrPDrRNiKU9
> t0z3y0t62MN+BgC0nLjHHLoxu+CsCYLUOKz+94XUluwemo7tDqRUZDbByIL7yTM=
> =iiu2
> -----END PGP SIGNATURE-----
>

Received on Saturday, 30 May 2015 04:52:15 UTC