Re: Clarifying word

That doesn't work for me, because once again we have no definition of 
"shape" other than saying that it is an instance of sh:Shape, which 
makes it an instance of a class, and, AFAIK, an instance of a class 
cannot "point to" anything. I'm also pondering how a graph can contain 
instances of classes. It seems to me that a graph may have sub-graphs, 
and the subjects of those may be instances of classes, but it's gotta be 
triples all the way down. This seems to be a mixing of structure and 
semantics. For now, let's get the triples well-defined, and then we can 
worry about the semantics of classes.


On 4/16/16 7:35 PM, Irene Polikoff wrote:
> I think it is quite simple:
>   * A SHACL shape is an instance of sh:Shape. A shape points to
>     constraints (or conditions) an RDF node is compared against to
>     determine if it conforms to the shape. For example, a shape
>     example:Issue may point to two constraints: one that says that the
>     property example:submitter must have exactly one value and the value
>     must be a string and one that says that the property
>     example:submissionDate must have exactly one value and the value
>     must be a date.
>   * When an RDF node conforms to conditions specified by a shape it is
>     said to be valid against a shape.
>   * A shape can also define what RDF nodes are to be validated (checked
>     for conformance) against it. This is called a scope of a shape. When
>     a shape doesn’t specify a scope, its scope is any RDF node.
> The specification also uses the following terminology:
>   * RDF graph containing shapes is called “shapes graph”.
>   * RDF graph containing data to be checked for conformance (validated)
>     is called “data graph”.
>   * When examples talk about specific nodes that are being checked
>     against a shape, they use a term “focus node”.
>   * A report produced as a result of checking RDF data against the
>     relevant shapes is called a "validation report".
> Irene Polikoff, CEO
> TopQuadrant, Inc. <>
> Technology providers making enterprise information meaningful
> Blogs —,
> LinkedIn —
> Twitter -
> On 4/16/16, 4:49 PM, "Karen Coyle" <
> <>> wrote:
>     I decided to take Peter's request that more people read through the
>     document, figuring that I would only be able to do a portion of it
>     before it got over my head. However, I haven't gotten very far due to
>     what I presume is some of that lack of consistency that Peter has
>     mentioned.
>     The introduction (1.) has these sentences:
>     "SHACL groups descriptive information and constraints that apply to a
>     given data node into shapes. This document defines what it means for an
>     RDF graph, referred to as the "data graph", to conform to a graph
>     containing SHACL shapes, referred to as the "shapes graph"."
>     "A shape may include a scope which defines which nodes in the data
>     graph
>     must conform to it. When a data node is checked for conformance to a
>     shape, that node is referred to as the focus node. The output of the
>     validation process is a validation report which indicates whether or
>     not
>     the data graph conforms to the shapes graph."
>     In these we have "shapes", "SHACL shapes", "shapes graph", "nodes",
>     "data nodes" "focus nodes".
>     Shortly thereafter we have "shape definitions", and a "shapes graph
>     that
>     defines these constraints has two shapes."
>     The main problem is the use of "shape/shapes" some times and "shapes
>     graph" at others, with the implication (but not stated) that a "shapes
>     graph" can consist of one or more "shapes." However, I'm not sure
>     what a
>     shape is in this context, since it is by definition in the form of a
>     graph.
>     Note also that the examples in that section consist of multiple graphs,
>     that is there is no subject that holds them together. I believe they
>     should have a symbolic "top node" that shows that they belong to a
>     single graph even though there are subgraphs.
>     I'm happy to write alternate text for some of this, but in this case
>     I'm
>     not clear on what is intended.
>     There are other areas where I can suggest better wording. I'd rather do
>     edits in a copy than try to explain them. Would that be ok?
>     kc
>     --
>     Karen Coyle
> <>
>     m: 1-510-435-8234
>     skype: kcoylenet/+1-510-984-3600

Karen Coyle
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600

Received on Sunday, 17 April 2016 05:29:27 UTC