W3C

- DRAFT -

RDF Data Shapes Working Group Teleconference

08 Feb 2017

See also: IRC log

Attendees

Present
hknublau, Nicky, simonstey, TallTed, ipolikoff, scribe
Regrets
Chair
TallTed
Scribe
simonstey

Contents

<scribe> scribe: simonstey

approval of last week's minutes

<TallTed> PROPOSED: Approve minutes of the 01 Feb 2017 Telecon: https://www.w3.org/2017/02/01-shapes-minutes.html

<hknublau> +1

<Nicky> +1

<ipolikoff> +1

RESOLUTION: Approve minutes of the 01 Feb 2017 Telecon: https://www.w3.org/2017/02/01-shapes-minutes.html

hknublau: we have a new member on the call

tim: Tim Smith, working on semantics since the late 90s
... I've seen many many use cases that could benefit from the results of this wg

TallTed: I don't see your organization listed as w3c member yet

tim: our membership was approved on feb. 2nd

<TallTed> https://www.w3.org/2014/data-shapes/wiki/How_to_Join

TallTed: a few new issues have been raised

<TallTed> https://www.w3.org/2014/data-shapes/track/issues/raised

issue-223

<trackbot> issue-223 -- Should we disallow shapes with mismatching type -- raised

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/223

issue-224

<trackbot> issue-224 -- Can we improve the language around the use of rdf:types for shapes -- raised

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/224

issue-225

<trackbot> issue-225 -- Respond to "Validation results and reports" -- raised

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/225

<TallTed> PROPOSED: OPEN issues issue-223, issue-224, issue-225

<hknublau> +1

<ipolikoff> +1

+1

<tsmith6> +1

<Nicky> +1

<TallTed> +1

RESOLUTION: OPEN issues issue-223, issue-224, issue-225

discussing responses to public comments

hknublau: I opened issues for some of the more substantial comments

issue-223

hknublau: in the example stated in the issue, a nodeshape defines a sh:path -> is also a propertyshape
... currently that's allowed, but might be confusing for others

ipolikoff: I think, currently the spec disallows a construct like that
... but not clear enough

<hknublau> simonstey: (Summary of the email that I sent today)

<ipolikoff> wasn't that about issue-139?

<TallTed> issue-139?

<trackbot> issue-139 -- Can all constraint properties be applied in all scenarios? -- closed

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/139

<TallTed> PROPOSED: CLOSE issue-223 by adding two syntax rules: a) SHACL instances of sh:NodeShape cannot have values for sh:path; b) SHACL instances of sh:PropertyShape must have a value for sh:path

<hknublau> +1

hknublau: we could provide shapes to verify our syntax rules.. but that's a huge undertaking we don't really have time for

<ipolikoff> +1

+1

<TallTed> +1

<tsmith6> +1

<Nicky> +1

RESOLUTION: CLOSE issue-223 by adding two syntax rules: a) SHACL instances of sh:NodeShape cannot have values for sh:path; b) SHACL instances of sh:PropertyShape must have a value for sh:path

hknublau: unfortunately, there are a lot of rules that are very hard to actually check
... but most of them are easy to check

issue-224

<TallTed> issue-224?

<trackbot> issue-224 -- Can we improve the language around the use of rdf:types for shapes -- open

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/224

hknublau: that issue is also about the relation between node-/propertyshapes

ipolikoff: two people said that the current language is weird
... because you can have type declarations for node/propertyshapes
... but they aren't used for determining the actual type
... proposal would be to RECOMMEND the use of type declarations

TallTed: when the type statement isn't the declaration.. what is?

hknublau: I think it would be sufficient to adapt SHOULD for half of the examples

TallTed: but that raises the question of how you select those

simonstey: I propose to remove "However, the presence of any rdf:type triple does not determine whether a node is treated as a node shape or not. "

<TallTed> PROPOSED: CLOSE issue-224 by deleting two instances of "However, the presence of any rdf:type triple does not determine whether a node is treated as a node shape or not."; making SHOULD uppercase in the sh:NodeShape and sh:PropertyShape definition; and by adding rdf:type statements to most if not all examples

simonstey: not sure if we should also change should to SHOULD

TallTed: the difference is whether its defining behaviour or syntax

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119.

https://www.ietf.org/rfc/rfc2119.txt

<TallTed> PROPOSED: CLOSE issue-224 by deleting two instances of "However, the presence of any rdf:type triple does not determine whether a node is treated as a node shape or not."; changing "should" to "is recommended, but not required," in the sh:NodeShape and sh:PropertyShape definitions; and by adding rdf:type statements to most if not all examples

+1

<hknublau_> +1

<ipolikoff> +1

<TallTed> +1

<tsmith6> +1

<Nicky> +1

RESOLUTION: CLOSE issue-224 by deleting two instances of "However, the presence of any rdf:type triple does not determine whether a node is treated as a node shape or not."; changing "should" to "is recommended, but not required," in the sh:NodeShape and sh:PropertyShape definitions; and by adding rdf:type statements to most if not all examples

TallTed: soo.. we have a question on prebinding and one on validation reports

<TallTed> issue;225?

hknublau_: andy & peter are still discussing the prebinding issue.. so better start with the validation reports first

issue-225

<trackbot> issue-225 -- Respond to "Validation results and reports" -- open

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/225

ipolikoff: [giving some background info on the nature of the issue]

+q

<ipolikoff> for example, Each results structure in results(f,V,D,c,s,S) where c has type sh:ClassConstraintComponent and parameter values <sh:class,c> contains a different top-level validation result from f,v,D,c,s,S for each v in V that is not a SHACL instance of c in D and no other top-level validation results.

hknublau_: peter mentions 3 parts explicitely
... we could address 2) fairly easily by putting some boilerplate text in the spec

TallTed: the cleanest way to address the comments would be to raise an issue for all concret ones

-q

scribe: let's discuss/address/resolve those issues next week

ipolikoff: there was also a comment on whether "shapes graph" is defined too narrowly

<TallTed> issue-139?

<trackbot> issue-139 -- Can all constraint properties be applied in all scenarios? -- closed

<trackbot> http://www.w3.org/2014/data-shapes/track/issues/139

TallTed: if this issue was somehow already discussed before, then we should use the results of that discussion as response

issue-139

hknublau_: this issue was discussed for ~1 month.. i.e. whether every constraint component could be used anywhere
... it helped to unify definitions
... I don't think we should have to write testcases and/or spend too much time on that

<TallTed> holger's mail -- https://lists.w3.org/Archives/Public/public-data-shapes-wg/2017Feb/0012.html

hknublau_: I've provided a list of constraint components that can't be used with nodeshapes

tsmith6: as a relative novice reading through the spec, I was very confused about the possibility of defining e.g. mincount for nodeshapes
... and was wondering when this would actually make sense

<TallTed> PROPOSED: reopen ISSUE-139, motivated by surrounding changes in SHACL spec

<hknublau_> +1

+1

<tsmith6> +1

<TallTed> +1

<Nicky> +1

RESOLUTION: reopen ISSUE-139, motivated by surrounding changes in SHACL spec

<TallTed> PROPOSED: CLOSE ISSUE-139 by declaring that NodeShapes are ill-formed if they use any of the properties listed in https://lists.w3.org/Archives/Public/public-data-shapes-wg/2017Feb/0012.html

<hknublau_> +1

<ipolikoff> +1

<tsmith6> +1

<TallTed> +1

+1

RESOLUTION: CLOSE ISSUE-139 by declaring that NodeShapes are ill-formed if they use any of the properties listed in https://lists.w3.org/Archives/Public/public-data-shapes-wg/2017Feb/0012.html

<TallTed> trackbot, close meeting

<trackbot> Sorry, TallTed, I don't understand 'trackbot, close meeting'. Please refer to <http://www.w3.org/2005/06/tracker/irc> for help.

TallTed: I encourage everyone to reread the spec

<TallTed> trackbot, end meeting

<TallTed> trackbot, start meeting

<trackbot> Meeting: RDF Data Shapes Working Group Teleconference

<trackbot> Date: 08 February 2017

<TallTed> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

  1. Approve minutes of the 01 Feb 2017 Telecon: https://www.w3.org/2017/02/01-shapes-minutes.html
  2. OPEN issues issue-223, issue-224, issue-225
  3. CLOSE issue-223 by adding two syntax rules: a) SHACL instances of sh:NodeShape cannot have values for sh:path; b) SHACL instances of sh:PropertyShape must have a value for sh:path
  4. CLOSE issue-224 by deleting two instances of "However, the presence of any rdf:type triple does not determine whether a node is treated as a node shape or not."; changing "should" to "is recommended, but not required," in the sh:NodeShape and sh:PropertyShape definitions; and by adding rdf:type statements to most if not all examples
  5. reopen ISSUE-139, motivated by surrounding changes in SHACL spec
  6. CLOSE ISSUE-139 by declaring that NodeShapes are ill-formed if they use any of the properties listed in https://lists.w3.org/Archives/Public/public-data-shapes-wg/2017Feb/0012.html
[End of minutes]
Minutes formatted by David Booth's scribe.perl version 1.148 (CVS log)
$Date: 2017/02/08 14:20:42 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.148  of Date: 2016/10/11 12:55:14  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: simonstey
Inferring ScribeNick: simonstey
Default Present: hknublau, Nicky, simonstey, TallTed, ipolikoff
Present: hknublau Nicky simonstey TallTed ipolikoff scribe
Found Date: 08 Feb 2017
Guessing minutes URL: http://www.w3.org/2017/02/08-shapes-minutes.html
People with action items:
[End of scribe.perl diagnostic output]