See also: IRC log
<Arnaud> scribe: TallTed
<Arnaud> PROPOSED: Approve minutes of the 16 Nov 2016 Telecon: http://www.w3.org/2016/11/16-shapes-minutes.html
<ericP> +1
<hknublau> +1
RESOLUTION: Approve minutes of the 16 Nov 2016 Telecon: http://www.w3.org/2016/11/16-shapes-minutes.html
kcoyle: questions on 2.1
[discussion of various wordings]
<Arnaud> kcoyle: "in a shape" doesn't make sense to me when a shape is a node - it is merely a URI
<ericP> +1 to "probability field"
[discussion of approval of edits]
Arnaud: will be done when we publish the next version
... know it's not easy to be perfect but we need to be careful not to miss changes from the history
Holger: added the two Peter pointed out
<Arnaud> PROPOSED: Open ISSUE-197, ISSUE-198, ISSUE-201, ISSUE-202, ISSUE-203, ISSUE-204, ISSUE-205, ISSUE-207, ISSUE-208, ISSUE-209, ISSUE-210, ISSUE-211, ISSUE-212, ISSUE-213
<hknublau> +1
<kcoyle> +1
+1
RESOLUTION: Open ISSUE-197, ISSUE-198, ISSUE-201, ISSUE-202, ISSUE-203, ISSUE-204, ISSUE-205, ISSUE-207, ISSUE-208, ISSUE-209, ISSUE-210, ISSUE-211, ISSUE-212, ISSUE-213
kcoyle: have been doing a deep-review of whole document, currently have a number of notes on Section 3, and expect to have more on 4.
<kcoyle> https://docs.google.com/document/d/1O24vnnZuTWQgi2-U_-lnY-IjV9EbRyxNtrYC3zq2pnM/edit
kcoyle: how can we handle these, other than an ISSUE for each?
ericP: found and suggested removal of a lot of "chattiness" / redundancies
<ericP> PROPOSED: Remove the "sht:data-format sht:TURTLE" triple.
<hknublau> +1
<ericP> +1
<kcoyle> 0
+1
RESOLUTION: Remove the "sht:data-format sht:TURTLE" triple.
<ericP> PROPOSED: Rename sht:node to sht:focus (or initial focus)
<ericP> PROPOSED: Rename sht:node to sht:focus
<hknublau> 0
<kcoyle> 0
+0
<ericP> +ε
<ericP> +1
RESOLUTION: Rename sht:node to sht:focus
<ericP> PROPOSED: Separate the target* parsing from validation with a structure like the one suggested in https://lists.w3.org/Archives/Public/public-rdf-shapes/2016Nov/0030
<ericP> <S123> a sht:SelectionTest
<ericP> mf:action <example-data.ttl> ;
<ericP> mf:result
<ericP> [ sht:focus <F1> ; sht:shape <S1> ],
<ericP> [ sht:focus <F2> ; sht:shape <S2> ],
<ericP> [ sht:focus <F3> ; sht:shape <S2> ].
<ericP> <S123> a sht:SelectionTest
-1
<ericP> mf:action [
<ericP> sht:data <Is1_Ip1_DEC1_234.ttl> ;
<ericP> sht:control <someFocii.ttl>
<ericP> ];
<ericP> mf:result
<ericP> [ sht:focus <F1> ; sht:shape <S1> ],
<ericP> [ sht:focus <F2> ; sht:shape <S2> ],
<ericP> [ sht:focus <F3> ; sht:shape <S2> ].
PROPOSED: Separate the target* parsing from validation with a structure like the revision suggested above
<hknublau> 0
+0
<kcoyle> +1
<ericP> +1
RESOLUTION: Separate the target* parsing from validation with a structure like the revision suggested above
<Arnaud> trackbot, close action-47
<trackbot> Closed action-47.
[TQ tests being added, some other related side topics...]
ISSUE-196?
<trackbot> ISSUE-196 -- Should we delete filter shapes? -- closed
<trackbot> http://www.w3.org/2014/data-shapes/track/issues/196
<Arnaud> https://lists.w3.org/Archives/Public/public-rdf-shapes/2016Nov/0028.html
Arnaud: we combined the two proposals into one because holger would only agree to remove filters if we add the flag to disable a shape/constraint
... let's address Peter's point about the resolution: any node is considered conforming/valid != ignored
<Arnaud> PROPOSED: Amend resolution on ISSUE-196 changing "ignored" to "valid"
<hknublau> +1
kcoyle: sh:disabled seems to add a lot of complication to satisfy a singular use case
... in a large complex shape, effect seems hard to determine
old -- "Close ISSUE-196, deleting filter shape and instead, adding a boolean flag sh:disabled which (if true) means that a shape or constraint is ignored (i.e., any node is considered conforming/valid)."
<ericP> +1
-0.5
PROPOSED: Amend resolution on ISSUE-196 to "Close ISSUE-196, (1) deleting filter shape and (2) adding a boolean flag sh:disabled which (if true) means that a shape or constraint is always passed (i.e., all focus nodes are considered conforming/valid)."
... Amend resolution on ISSUE-196 to "Close ISSUE-196, (1) deleting filter shape and (2) adding a boolean flag sh:disabled which (if true) means that all focus nodes are considered conforming/valid."
<kcoyle> +1
<hknublau> +1
+0.5
<ericP> +1
RESOLUTION: Amend resolution on ISSUE-196 to "Close ISSUE-196, (1) deleting filter shape and (2) adding a boolean flag sh:disabled which (if true) means that all focus nodes are considered conforming/valid."
ISSUE-194?
<trackbot> ISSUE-194 -- stems in value sets -- open
<trackbot> http://www.w3.org/2014/data-shapes/track/issues/194
hknublau: discussion following resolution remembered by Dimitris & Holger was not captured in minutes, but was applied when spec was edited
Arnaud: records are not perfect, at this point it doesn't really matter how we got where we are
... important question is: what we do now?
... two different questions: (1) do we remove stemming from spec as currently specified, (2) do we add what ericP was looking for originally
kcoyle: still hoping for examples, comparing the options
... can't agree to drop sh:stem without seeing what one would have to do with sh:pattern
<Arnaud> http://w3c.github.io/data-shapes/shacl/#StemConstraintComponent
4.5.4 sh:stem
<hknublau> sh:pattern "^https://www.w3.org/People/" ;
<hknublau> sh:or ( [ sh:pattern "..." ] [ sh:pattern "...b..." ] )
<hknublau> + sh:nodeKind sh:IRI ;
https://www.w3.org/TR/xpath-functions/#regex-syntax
<ericP> http://piratepad.net/hFXe67Dl1l
<ericP> compare =without stems= and
<ericP> =with stems as Parameters=
hknublau: want to know whether if we need to publish a spec before going to CR we can publish now
Arnaud: not sure the spec is stable enough, still have a lot of open issues
hknublau: concerned we're not making enough progress and won't be able to go to CR before Xmas
Arnaud: that's possible
... it doesn't do us any good to ignore public comments because we will have to deal with them no matter what
... Best way to make progress is to make concrete proposals on how to resolve issues
<Arnaud> trackbot, end meeting