On 9/7/16 5:48 AM, Péter Király wrote:
Dear Karen,
some more comments
The examples sometime a bit mixed, e.g.
- the issue identifiers sometime don't match at the graph and at the
validation result (in 5.2.1 there are two issue2, and zero issue1, in
the sh:or's and sh:not examples there are three issue2.)
- in 5.2.1 there are wrong language labels
- it is not quite clear why some shapes and graphs are boxed and
others not (in the examples).
Yes, thanks, this is an area we are cleaning up this week. I'll let you
know when we think we've got all of the formatting consistent.
Regarding to the content: I am just curious if there are LessThan and
LessThanEquals why there is no GreaterThan and GreaterThanEquals? Yes,
we can use a reversed statement or combine Less.. with Not, but I
would prefer their availability.
The argument is that these are pair-wise constraints, so you would
adjust the direction of the values rather than use GT. If you can think
of an example of when that would not be convenient, I can take it to the
W3C group. (This may make more sense if you look at section 4.6 of the
main document, here[1].)
kc
[1] http://w3c.github.io/data-shapes/shacl/#constraints-property-pairs
Cheers,
Péter
2016-09-07 11:47 GMT+02:00 Péter Király <[email protected]>:
Hi Karen,
I started to reading it, and I find it quite helpful.
I have a suggestion: for me the formal definitions (such as "Shape :=
label:IRI|BNode, targets:Set[Target], filters:Set[Shape],
constraints:Set[Constraint]") would be more readable if they would be
in monospace characterset - similarly than the examples.
"This signifies that a Shape has four components called label,
targets, filters, constraints. The label is either a IRI or BNode, the
targets are a set of Targets, the filters are a set of Shapes, and the
constraintsis a set of Constraints."
Here I would expect a bit more explanations something like "targets
are a set of Targets (the elements which are selected as the subject
of validation)".
I am not sure whether the result in the example for 5.1.3 Datatype
section is right. I would expect issue2 is right because it is a
xsd:dateTime, and issue1 is wrong because it is a xsd:date, and not
the other way around.
Do you know any existing implementation or is there a project working
on the implementation?
Best regards,
Péter
2016-09-05 17:21 GMT+02:00 Karen Coyle <[email protected]>:
Folks,
There is a W3C standard (SHACL)[1] in development that would address the
issue of validation of RDF graphs. The standard itself is, as standards tend
to be, long and not an easy read. Eric Prud'hommeaux and I (both committee
members) have created a first draft of a brief reference document, in the
form of an Abstract Syntax of the core vocabulary of the SHACL standard. We
welcome any comments or corrections to this document, and any suggestions
for making it better. The draft is at:
https://www.w3.org/TR/2016/WD-shacl-abstract-syntax-20160825/
Comments should be sent to the mail list at:
[email protected]
However, I will also entertain any discussion that takes place here, which
feels less formal than posting to a W3C list. Our goal is to make SHACL Core
as clear as possible for first time users. If this becomes a W3C standard,
it will probably eventually become available in various RDF-related tools.
Thanks,
kc
[1] https://www.w3.org/TR/shacl/
-------- Forwarded Message --------
Subject: FIRST PUBLIC WORKING DRAFT: SHACL CORE ABSTRACT SYNTAX AND
SEMANTICS
Resent-Date: Wed, 31 Aug 2016 16:46:10 +0000
Resent-From: [email protected]
Date: Wed, 31 Aug 2016 09:45:36 -0700
From: Karen Coyle <[email protected]>
Reply-To: [email protected]
To: [email protected] <[email protected]>
**Please forward to interested lists**
As announced on the W3C blog[1], the first public working draft of the SHACL
Core Abstract Syntax[2] has been published by the RDF Data Shapes Web
Working Group.[3]
"This document defines an abstract syntax for the core SHACL (SHApes
Constraint Language). It is derived from the SHACL specification and is a
non-normative version of the content of that specification."
We are soliciting comments (and questions) on this first draft. Please
comment at [email protected].
---------
[1] https://www.w3.org/blog/news/archives/5749
[2] https://www.w3.org/TR/2016/WD-shacl-abstract-syntax-20160825/
[3] https:////www.w3.org/2014/data-shapes/
--
Karen Coyle
[email protected] http://kcoyle.net
m: +1-510-435-8234
skype: kcoylenet/+1-510-984-3600
--
Péter Király
software developer
GWDG, Göttingen - Europeana - eXtensible Catalog - The Code4Lib Journal
http://linkedin.com/in/peterkiraly
--
Karen Coyle
[email protected] http://kcoyle.net
m: +1-510-435-8234
skype: kcoylenet/+1-510-984-3600
