IRC log of shapes on 2015-08-06

Timestamps are in UTC.

17:58:25 [RRSAgent]
RRSAgent has joined #shapes
17:58:25 [RRSAgent]
logging to http://www.w3.org/2015/08/06-shapes-irc
17:58:27 [trackbot]
RRSAgent, make logs rdf-data-shapes
17:58:27 [Zakim]
Zakim has joined #shapes
17:58:29 [trackbot]
Zakim, this will be SHAPES
17:58:29 [Zakim]
I do not see a conference matching that name scheduled within the next hour, trackbot
17:58:30 [trackbot]
Meeting: RDF Data Shapes Working Group Teleconference
17:58:30 [trackbot]
Date: 06 August 2015
17:58:34 [kcoyle]
kcoyle has joined #shapes
17:59:47 [Arnaud]
present+ Arnaud
17:59:50 [kcoyle]
present+
18:00:00 [pfps]
pfps has joined #shapes
18:00:04 [pfps]
present+ pfps
18:00:09 [kcoyle]
present+ kcoyle
18:01:32 [Arnaud]
regrets: simonstey
18:01:43 [Arnaud]
agenda: https://www.w3.org/2014/data-shapes/wiki/Meetings:Telecon2015.07.06
18:01:47 [Arnaud]
chair: Arnaud
18:02:08 [Dimitris]
present+ dimitris
18:02:18 [Arnaud]
present+ hknublau
18:02:21 [Labra]
Labra has joined #shapes
18:02:28 [Arnaud]
present+ labra
18:04:13 [aryman]
aryman has joined #shapes
18:05:03 [Arnaud]
present +aryman
18:07:42 [Arnaud]
agenda: https://www.w3.org/2014/data-shapes/wiki/Meetings:Telecon2015.08.06
18:08:22 [Arnaud]
present+ TallTed
18:10:00 [Arnaud]
PROPOSED: Approve minutes of the 30 July Telecon: http://www.w3.org/2015/07/30-shapes-minutes.html
18:10:01 [pfps]
minutes look fine to me
18:10:01 [hknublau]
scribenick: hknublau
18:10:15 [Arnaud]
RESOLVED: Approve minutes of the 30 July Telecon: http://www.w3.org/2015/07/30-shapes-minutes.html
18:10:56 [pfps]
What are the goals for the F2F?
18:11:42 [pfps]
I will not be 9 hours behind - I will be on the EDT then.
18:13:52 [aryman]
q+
18:13:54 [kcoyle]
10-6 sounds fine
18:14:02 [Arnaud]
ack aryman
18:14:12 [hknublau]
Arnaud: Discussion about whether we can shift French meeting time to maybe 10 to 6
18:19:08 [hknublau]
Topic: Disposal of raised issues
18:19:12 [pfps]
I have no problem opening either of these two issues
18:19:13 [Arnaud]
PROPOSED: Open ISSUE-77: sh:pattern flags, and ISSUE-78: sh:abstract
18:19:21 [pfps]
+1
18:19:21 [hknublau]
+1
18:19:25 [Dimitris]
+1
18:19:37 [kcoyle]
+1
18:19:46 [TallTed]
+1
18:20:09 [Arnaud]
RESOLVED: Open ISSUE-77: sh:pattern flags, and ISSUE-78: sh:abstract
18:20:11 [hknublau]
(None of them are urgent)
18:20:11 [aryman]
+1
18:20:38 [hknublau]
Topic: ISSUE-3
18:20:44 [pfps]
Eric commented on ISSUE-3
18:21:18 [hknublau]
http://www.w3.org/2014/data-shapes/track/issues/3
18:21:26 [Arnaud]
PROPOSED: Close ISSUE-3, specifying that a SHACL processor is invoked with an RDF dataset (or RDF graph) as its data and a shapes graph for control, per Holger's email
18:21:38 [pfps]
+1
18:21:39 [hknublau]
+1
18:21:43 [TallTed]
+1
18:21:44 [aryman]
+
18:21:47 [aryman]
+1
18:21:48 [Dimitris]
+1
18:21:53 [Labra]
+1
18:22:06 [kcoyle]
+1
18:22:16 [Arnaud]
RESOLVED: Close ISSUE-3, specifying that a SHACL processor is invoked with an RDF dataset (or RDF graph) as its data and a shapes graph for control, per Holger's email
18:24:09 [hknublau]
Topic: http://www.w3.org/2014/data-shapes/track/issues/26
18:27:05 [Arnaud]
issue-26
18:27:05 [trackbot]
issue-26 -- Can extensions invoke the high-level language? -- open
18:27:05 [trackbot]
http://www.w3.org/2014/data-shapes/track/issues/26
18:27:33 [hknublau]
pfps: One way of reading this is would be putting expressions into the middle of SPARQL code. We probably don’t want this.
18:27:52 [hknublau]
… All that is needed in the SPARQL to get back to named shapes.
18:27:59 [aryman]
q+
18:28:50 [Arnaud]
ack aryman
18:28:50 [hknublau]
… sh:hasShape is one implementation mechanism.
18:29:34 [hknublau]
aryman: Is it true that you can wrap any SHACL string into a named shape?
18:30:11 [pfps]
One thing would to allow a, for example, conjunction inside SPARQL code. Another thing would be to allow only names of shapes inside SPARQL code.
18:30:42 [hknublau]
hknublau: I think named shapes are sufficient.
18:30:48 [pfps]
Whether the former is more powerful than the later would require a close examination of the current document - I'm not sure what the answer is.
18:31:42 [hknublau]
Arnaud: Can we limit this to named shapes?
18:32:22 [hknublau]
hknublauc: Any conjunction etc can be wrapped into a named shape if needed.
18:33:06 [Arnaud]
PROPOSED: Close ISSUE-26, stating that extensions can invoke the high-level language referring to named shapes
18:33:16 [hknublau]
+1
18:33:23 [kcoyle]
+1
18:33:28 [TallTed]
+1
18:33:42 [Labra]
+0.5
18:33:44 [ericP]
+0
18:33:49 [Dimitris]
+1
18:33:53 [pfps]
-0.1 I think that this is going to be OK, but I'm not completely sure
18:34:11 [aryman]
+1
18:34:46 [hknublau]
pfps: The design is getting more and more complicating. Harder to find gotchas.
18:34:55 [Arnaud]
RESOLVED: Close ISSUE-26, stating that extensions can invoke the high-level language referring to named shapes
18:35:14 [Arnaud]
issue-51
18:35:14 [trackbot]
issue-51 -- What types of validation results should be returned -- open
18:35:14 [trackbot]
http://www.w3.org/2014/data-shapes/track/issues/51
18:36:32 [hknublau]
Arnaud: Dimitris made a suggestion, iterating with Holger on a proposal, then Peter suggested simplification
18:36:47 [pfps]
q+
18:37:49 [Arnaud]
ack pfps
18:38:38 [hknublau]
pfps: We are getting more and more stuff packed into the results. Would be much better to only thing to get back is a set of violations.
18:39:23 [hknublau]
… handling severity levels etc outside of SHACL
18:39:23 [kcoyle]
q+
18:39:28 [Arnaud]
ack kcoyle
18:40:03 [hknublau]
kcoyle: In that case what do you see SHACL returning? Gradation is needed.
18:40:45 [hknublau]
pfps: Example: Each person must have one name. Result is either OK or a set of violations.
18:42:01 [hknublau]
… this could be a SPARQL result set.
18:42:27 [hknublau]
… no need for Info, Fatal, Error
18:43:16 [hknublau]
… controlling code can find the severity itself.
18:43:20 [hknublau]
q+
18:44:39 [Dimitris]
+q
18:44:42 [hknublau]
… start as simple as possible
18:45:28 [hknublau]
Arnaud: Then there would not be a way to specify levels per shape
18:45:31 [pfps]
The results of trying to validate a shape is just a set of violations - the calling code gets to decide to do what with the violations - the calling code can do this for each shape, or for groups of shapes, or for entire documents of shapes
18:45:45 [Arnaud]
ack hknublau
18:46:32 [pfps]
My proposal is to have none of this in the high-level language, not error levels, not severity levels - so no sh:severity, no sh:Error, no sh:Warning, no [lots of other stuff]
18:48:08 [Arnaud]
ack Dimitris
18:48:36 [hknublau]
hknublau: When you invoke the engine on a graph, there may be multiple severity levels coming back.
18:48:44 [pfps]
q+
18:48:49 [hknublau]
… each violation at least needs a pointer back to the shape.
18:49:56 [Arnaud]
ack pfps
18:50:09 [hknublau]
Arnaud: we should do a straw poll
18:50:36 [hknublau]
pfps: Work-around using two shapes, one for warning, one for error
18:51:38 [ericP]
q+ to ask about non SPARQL impls
18:52:05 [Arnaud]
STRAWPOLL: a) keep super simple, one type of violation, b) define two or three severty levels, c) define an open model providing with several levels and extensions
18:52:27 [ericP]
q-
18:52:40 [hknublau]
a: -1 b: 0.5 c: 0.6
18:52:42 [pfps]
a +1 b 0 c -1
18:52:49 [aryman]
a: 0; b: +1; c: -0.5
18:53:04 [ericP]
a:+1 b:0 c:0
18:53:14 [Dimitris]
a) -0.5 b) +1 c) 0+
18:53:21 [hknublau]
On b) Maybe Info, Warning, Error?
18:53:30 [TallTed]
a -0.75, b +0.5, +1
18:53:42 [kcoyle]
a: 0.5 b: 0.5 c: ?? (depends on the magical user-friendly language)
18:53:45 [Labra]
a: +1, b: 0, c: 0
18:53:55 [pfps]
An "informational" violation seems like a contradiction in terms.
18:54:49 [pfps]
I like pi and e as severity levels.
18:56:00 [ericP]
q+ to ask about the scope of the errors
18:56:07 [Arnaud]
ack ericP
18:56:07 [Zakim]
ericP, you wanted to ask about the scope of the errors
18:56:20 [hknublau]
Arnaud: b) seems to be compromise
18:57:31 [pfps]
q+ to ask how summarization of results could work?
18:57:40 [hknublau]
hknublau: Clarified that each constraint can have severity level
18:58:07 [Arnaud]
ack pfps
18:58:07 [Zakim]
pfps, you wanted to ask how summarization of results could work?
18:58:39 [hknublau]
q+
18:59:08 [aryman]
q+
18:59:19 [hknublau]
q-
18:59:32 [hknublau]
pfps: How would we count results
18:59:33 [Arnaud]
ack aryman
19:00:19 [Arnaud]
https://w3c.github.io/data-shapes/shacl/#violations-types
19:00:45 [hknublau]
aryman: Each engine may stop and return different errors.
19:00:50 [aryman]
q+
19:00:58 [Arnaud]
ack aryman
19:01:21 [hknublau]
Arnaud: Do we really need Info messages? They don’t sound like violations.
19:01:22 [pfps]
I would hope that SHACL processors would not just give up when it encounters the first information violation
19:01:48 [hknublau]
aryman: Info messages are useful, compilers and other tools have them.
19:02:06 [ericP]
q+ to say that i don't know how to combine this with multi-occurance
19:02:11 [TallTed]
optimal/suboptimal/minor error/major error...
19:02:15 [Arnaud]
ack ericP
19:02:15 [Zakim]
ericP, you wanted to say that i don't know how to combine this with multi-occurance
19:03:35 [hknublau]
ericP: Example: I have an Observation with a related Observation and another related property pointing at an Observation
19:04:00 [hknublau]
… same predicate expressed twice
19:04:01 [hknublau]
q+
19:04:11 [Arnaud]
ack hknublau
19:04:49 [hknublau]
hknublau: Results can include subject, predicate, object
19:08:09 [hknublau]
… error level is associated with sh:minCount etc
19:08:27 [pfps]
q+
19:09:27 [hknublau]
ericP: still seeing problems with multi-occurrence
19:09:28 [Arnaud]
ack pfps
19:10:24 [hknublau]
pfps: High level language can only produce Errors
19:11:00 [kcoyle]
q+
19:11:08 [Arnaud]
ack kcoyle
19:11:28 [hknublau]
hknublau: Yes, how else could this be done? Reifying sh:minCount etc?
19:11:39 [aryman]
q+
19:11:54 [pfps]
I'm finding the current document very mixed up with respect to violation reporting. Some parts say one thing and some parts say something contradicting that.
19:12:11 [hknublau]
kcoyle: There should be support to override this per occurrence
19:12:32 [Arnaud]
ack aryman
19:13:07 [kcoyle]
hknublau: not override -- always related to the occurrence
19:13:11 [hknublau]
aryman: I thought we have sh:severity at each constraint
19:13:48 [Arnaud]
https://w3c.github.io/data-shapes/shacl/#sparql-constraints-variables
19:13:52 [Arnaud]
example 37
19:13:58 [aryman]
look at Example 37
19:14:24 [pfps]
But then there is "An sh:Error must be reported if there is no triple that has the focus node as its subject, the sh:predicate as its predicate and the sh:hasValue as its object. Each produced sh:Error must have the focus node as its sh:root and sh:subject, and the sh:predicate as its sh:predicate"
19:14:41 [pfps]
q+
19:15:12 [Arnaud]
ack pfps
19:17:43 [pfps]
I think that we are again running into a vocabulary mismatch problem here - there are two meanings of "constraint" being used
19:18:30 [pfps]
Well, I'm happy only having sh:Error!
19:18:38 [hknublau]
Arnaud: Unless there is a way to override, then the current spec would be too limiting. People want different levels per occurrence
19:19:16 [Dimitris]
We should be able to define severity in shapes and facets
19:19:55 [hknublau]
Arnaud: Holger, could you send clarification email about this?
19:21:22 [hknublau]
… Eric could you clarify your use case in email too?
19:22:31 [pfps]
+1
19:22:49 [ericP]
sure, and here's an example inline:
19:22:51 [ericP]
<BP> a sh:Shape;
19:22:51 [ericP]
sh:property
19:22:51 [ericP]
[ sh:predicate sh:observation; sh:valueShape <SYS>; sh:nodeKind sh:IRI ],
19:22:52 [ericP]
[ sh:predicate sh:observation; sh:valueShape <DIA>; sh:nodeKind sh:IRI ].
19:23:08 [ericP]
that's a good model for multi-occurance
19:23:20 [ericP]
s/model for/example of/
19:23:58 [hknublau]
Arnaud: ISSUE-75 not a big issue
19:24:16 [hknublau]
Topic: ISSUE-23 (Punning)
19:24:39 [hknublau]
http://www.w3.org/2014/data-shapes/track/issues/23
19:25:23 [hknublau]
Arnaud: What is currently in the draft is already a compromise between classes and shapes
19:25:34 [kcoyle]
q+
19:25:35 [hknublau]
… Holger stated punning as a must-have
19:25:43 [pfps]
q+
19:25:54 [Arnaud]
ack kcoyle
19:27:04 [aryman]
q+
19:28:03 [hknublau]
kcoyle: Can we leave this out of the language, e.g. owl:sameAs?
19:28:21 [Arnaud]
ack pfps
19:28:24 [hknublau]
Arnaud: Outside would not be standard
19:28:47 [hknublau]
pfps: The term punning is probably not related.
19:28:59 [hknublau]
… where was this “compromise” discussed?
19:31:22 [pfps]
The only relevant resolution I can find is from 18 February, which is "Start from Eric's Revised LDOM proposal, and explore ways to combine shapes and classes such as punning"
19:31:45 [pfps]
My understanding is that in this proposal shapes are not classes
19:32:33 [hknublau]
pfps: Has it been decided that shapes can be classes?
19:33:02 [hknublau]
Arnaud: The draft was accepted as starting point, and this includes sh:ShapeClass.
19:33:25 [hknublau]
aryman: We cannot prevent people from making a shape also a class.
19:33:50 [hknublau]
… impacts sh:scopeClass
19:34:12 [pfps]
The editor's draft has been accepted as a starting point but doesn't mean agreement to particular bits of it
19:35:09 [hknublau]
Arnaud: Credits to Holger for working on a compromise.
20:16:00 [pfps]
pfps has joined #shapes
20:30:50 [hknublau]
hknublau has joined #shapes
20:34:55 [hknublau]
hknublau has left #shapes
21:07:09 [Arnaud1]
Arnaud1 has joined #shapes
23:27:39 [Zakim]
Zakim has left #shapes