Re: Issue 93

Sounds good to me.

Jim Amsden, Senior Technical Staff Member
OSLC and Linked Lifecycle Data
919-525-6575




From:   Karen Coyle <kcoyle@kcoyle.net>
To:     "public-data-shapes-wg@w3.org" <public-data-shapes-wg@w3.org>
Date:   03/31/2016 11:24 AM
Subject:        Re: Issue 93





On 3/31/16 6:22 AM, Jim Amsden wrote:
> Or:
>
> "SHACL is a language for describing information about, and constraints
> on RDF graphs, and a set
> of rules for assessing the conformance of RDF data to those 
constraints."
>
> This includes:
> 1. that shapes can also be used to capture interesting metadata
> 2. that the conformance of a graph to the shapes is assessed and some
> resulting report is provided
> 3. that shapes can apply to any RDF graph, including OWL or RDFS, not
> just instances.

Excellent point. So we should talk about graphs, not instance data.

Also, the "information about" aspect of SHACL seems secondary to me (not 
saying that it should be, but it seems to be), so I wouldn't want to put 
that first. SHACL does let you annotate the described constraints, but 
to me doesn't serve generally as documentation for anything but 
constraints. So I would put it:

"SHACL is a language for describing constraints on RDF graphs, 
information about those graphs, and a set of rules for assessing the 
conformance of RDF data to those constraints."

Are we there yet?
kc

>
>
> Jim Amsden, Senior Technical Staff Member
> OSLC and Linked Lifecycle Data
> 919-525-6575
>
>
>
>
> From: Karen Coyle <kcoyle@kcoyle.net>
> To: "public-data-shapes-wg@w3.org" <public-data-shapes-wg@w3.org>
> Date: 03/30/2016 04:53 PM
> Subject: Issue 93
> ------------------------------------------------------------------------
>
>
>
> I agree with Harold that the wording in the draft tends to confound the
> validation description language and the rules for implementing the
> validation. Even in the introduction, it says:
>
> "SHACL (Shapes Constraint Language) is a language for describing and
> constraining RDF graphs."
>
> Which I think could be more clearly stated as:
>
> "SHACL is a language for describing constraints on RDF graphs, and a set
> of rules for applying those constraints to RDF instance data."
>
> (Except "applying" isn't right, because it's really a test, the instance
> data isn't modified, so...)
>
> "SHACL is a language for describing constraints on RDF graphs, and a set
> of rules for *comparing* those constraints to RDF instance data."
> or
> "SHACL is a language for describing constraints on RDF graphs, and a set
> of rules for testing the conformance of RDF instance data to those
> constraints."
>
> Or something to that effect. The language itself doesn't do any
> constraining, and the "engine" doesn't do any describing, so we clearly
> have (at least) two different things.
>
> Arnaud suggested [1] that we not talk about engines at all. I need to
> think more about that, but would "rules" meet the definition of "not an
> engine"? If so, I could try to at least identify the areas of the spec
> that need modification, and could suggest wording changes to Dimitris
> and Holger. That assumes that in the investigation of the wording a
> deeper problem isn't found between the concepts.
>
> kc
> [1]
> 
https://lists.w3.org/Archives/Public/public-data-shapes-wg/2015Sep/0228.html

>
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net <http://kcoyle.net/>
> m: 1-510-435-8234
> skype: kcoylenet/+1-510-984-3600
>
>
>
>

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600