IRC log of shapes on 2016-11-09

Timestamps are in UTC.

13:33:03 [RRSAgent]

RRSAgent has joined #shapes

13:33:03 [RRSAgent]

logging to http://www.w3.org/2016/11/09-shapes-irc

13:33:05 [trackbot]

RRSAgent, make logs rdf-data-shapes

13:33:05 [Zakim]

Zakim has joined #shapes

13:33:07 [trackbot]

Zakim, this will be SHAPES

13:33:07 [Zakim]

ok, trackbot

13:33:08 [trackbot]

Meeting: RDF Data Shapes Working Group Teleconference

13:33:08 [trackbot]

Date: 09 November 2016

13:33:25 [ericP]

scribenick: ericP

13:33:34 [ericP]

marqh: happy to scribe

13:33:39 [ericP]

... how do i do so?

13:33:48 [ericP]

marqh: happy to marqh

13:34:03 [ericP]

scribenick: marqh

13:34:12 [marqh]

present+

13:34:15 [hknublau]

present+

13:34:19 [Labra]

present+

13:34:20 [ericP]

present+

13:34:27 [simonstey]

present+

13:34:32 [ericP]

: Approve minutes of the 2 Nov 2016 Telecon: http://www.w3.org/2016/11/02-shapes-minutes

13:34:37 [ericP]

PROPOSED: Approve minutes of the 2 Nov 2016 Telecon: http://www.w3.org/2016/11/02-shapes-minutes

13:34:40 [marqh]

ericP: first topic, approve previous minutes

13:34:43 [ericP]

topic: admin

13:34:45 [ericP]

PROPOSED: Approve minutes of the 2 Nov 2016 Telecon: http://www.w3.org/2016/11/02-shapes-minutes

13:35:00 [marqh]

ericP: exciting bits: end of minutes

13:35:08 [Labra]

** I will leave in 10 mins and be back after another 20mins

13:35:22 [marqh]

... opened issue 193 targets can be defined

13:35:49 [marqh]

... closed issue 191, about valueTypes et al.

13:36:10 [simonstey]

+1

13:36:11 [ericP]

+1

13:36:11 [Labra]

+1

13:36:12 [hknublau]

+1

13:36:17 [ericP]

RESOLVED: Approve minutes of the 2 Nov 2016 Telecon: http://www.w3.org/2016/11/02-shapes-minutes

13:36:33 [ericP]

PROPOSED: Change time of regular weekly call to 8:30am US Eastern

13:36:59 [ericP]

+1

13:37:03 [hknublau]

+1

13:37:03 [marqh]

ericP: quorum is marginal, but best we can do

13:37:12 [simonstey]

+1

13:37:16 [Labra]

0

13:37:21 [ericP]

RESOLVED: Change time of regular weekly call to 8:30am US Eastern

13:37:23 [marqh]

+1

13:37:48 [ericP]

PROPOSED: Virtual F2F scheduled on 16 Nov 08:00-12:00 EST

13:38:01 [ericP]

-> http://doodle.com/poll/nzfev6sgnu38695p doodle poll

13:38:14 [hknublau]

+1

13:38:20 [ericP]

+1

13:38:21 [marqh]

ericP: virtual face to face meeting proposal

13:38:29 [simonstey]

+1

13:38:43 [Labra]

0

13:39:06 [ericP]

RESOLVED: Virtual F2F scheduled on 16 Nov 08:00-12:00 EST

13:39:12 [ericP]

topic: Disposal of Raised Issues

13:39:24 [ericP]

issue-194

13:39:24 [trackbot]

issue-194 -- stems in value sets -- raised

13:39:24 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/194

13:39:57 [marqh]

ericP: when we had a discussion of what we call stems in sheks, we weren't specific about how they are used.

13:40:12 [marqh]

... any value set is an enumeration of a long or

13:40:34 [marqh]

... this can include URIs that a certain substring, and excluding other substrings

13:40:45 [marqh]

... this is about URIs starting with a substring only

13:40:56 [ericP]

PROPOSED: Open ISSUE-194

13:41:00 [ericP]

+1

13:41:01 [simonstey]

+1

13:41:05 [Labra]

+1

13:41:06 [hknublau]

+1

13:41:09 [marqh]

... the proposal is to raise ISSUE-194 to address this

13:41:10 [marqh]

+1

13:41:16 [ericP]

RESOLVED: Open ISSUE-194

13:41:30 [ericP]

issue-183

13:41:30 [trackbot]

issue-183 -- Eliminating the term "Undefined" -- open

13:41:30 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/183

13:42:13 [marqh]

ericP: this came for dimitris, but he is not here

13:42:19 [ericP]

issue-184

13:42:19 [trackbot]

issue-184 -- Property paths and value nodes -- open

13:42:19 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/184

13:42:32 [marqh]

... we may leave this issue unresolved

13:43:09 [marqh]

hknublau: i can pick this up. In have avoided the term and teh change is now dealt with. the issue can be closed, in my view

13:43:28 [marqh]

... the background is about the term 'undefined'

13:43:35 [marqh]

... what terms to return a node or nothing

13:43:49 [marqh]

... closest in SPARQL is a SPARQL error

13:43:53 [marqh]

... so I have just used this

13:44:11 [marqh]

ericP: I understand, i would like dimitris to agree to closure

13:44:29 [TallTed]

TallTed has joined #shapes

13:44:41 [marqh]

hknublau: dimitirs may just be late, let's defer in case he arrives

13:44:45 [marqh]

ericP: agreed

13:45:10 [marqh]

ericP: lets move to issue-191

13:45:15 [ericP]

ISSUE-191

13:45:15 [trackbot]

ISSUE-191 -- Should the value types of parameters be constraints -- closed

13:45:15 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/191

13:45:33 [marqh]

hknublau: this issue is already closed, resolved last week

13:46:17 [ericP]

ISSUE-140

13:46:17 [trackbot]

ISSUE-140 -- SHACL needs to support validation of individual nodes -- open

13:46:17 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/140

13:46:25 [TallTed]

present+

13:47:38 [marqh]

ericP: one way to look at this is that a this can be handled with distinct mechanism

13:47:52 [marqh]

.. the spec can describe no shape pairs

13:47:59 [marqh]

... i have made a proposal

13:48:19 [marqh]

... one approach i would propose is to move the section on how we sift through graphs and compose node shape pairs

13:48:30 [marqh]

... then have a seperate operation that mkaes it clear it is done one

13:48:33 [marqh]

... once

13:49:09 [marqh]

hknublau: it sounds like we need a compromise here. sheks are not fond of targets, in my view targets are useful

13:49:14 [Labra]

** I have to leave now...be back in 20 mins...

13:49:27 [marqh]

... why don't we meet in th3e middle, have some targets in examples and some without targets

13:49:43 [Dimitris]

Dimitris has joined #shapes

13:49:49 [marqh]

.. look under focesNodes. this describes how focusNodes can be determined

13:49:57 [marqh]

... I woudl like input from sheks on

13:50:10 [marqh]

... it is also possible to pass a node into a processor by some means

13:50:25 [marqh]

... a sentence to put in here to mention this explicitly would be useful

13:50:40 [marqh]

... add som examples with no target, with text to clarify node origins

13:50:43 [Dimitris]

present+

13:50:45 [marqh]

... this could be a good compromise

13:50:59 [marqh]

ericP: i am content with that

13:51:39 [marqh]

ericP: it does make it harder to separate the 2 processes, but that may not be a problem

13:52:12 [ericP]

PROPOSED: resolve issue-140 by having half of the examples include target* and the other half not.

13:53:02 [ericP]

+.5

13:53:20 [marqh]

TallTed: that puts significant weight on having examples of both kinds in each case

13:53:28 [hknublau]

+0.5

13:53:39 [marqh]

ericP: can readers extrapolate and understand

13:54:03 [simonstey]

+q

13:54:25 [marqh]

some examples in one formulation and some in the other, we set it up so that is how it is implemented

13:54:34 [ericP]

ack next

13:54:58 [marqh]

simonstey: to support what TallTed has said, some example with or without target, people will read into this

13:55:14 [marqh]

... so examples consistency is helpful

13:55:25 [Dimitris]

q+

13:55:45 [ericP]

ack next

13:55:50 [marqh]

... some statements saying 'target omitted in this example' will get missed by readers

13:56:35 [TallTed]

maybe include a comment-line next to each target triple, saying "that triple is optional"

13:56:41 [marqh]

Dimitris: we can have examples with targets, then ones showing how targets may be ommitted in some cases

13:57:04 [marqh]

ericP: could have targets in a seperate block, illustrating that it provides functionality

13:57:17 [marqh]

Dimitris: this is a differnet type o fvalidation

13:57:33 [marqh]

Dimitris: just need to show how to omit them in specific csaes

13:57:52 [marqh]

ericP: having targets in there is problematic for useability

13:58:06 [marqh]

.. inherent cost of having things written into the shape graph or data graph

13:58:11 [marqh]

... inherent cost of having things written into the shape graph or data graph

13:58:22 [marqh]

... making the shape graph predestined for some data

13:58:32 [hknublau]

strongly disagree, except for sh:targetNode

13:58:47 [marqh]

... my goal is to show how to add that, but you don't have to have a target for validation

13:59:01 [simonstey]

+q

13:59:14 [ericP]

ack next

13:59:16 [TallTed]

or a comment line saying "targets could be set here, as ..."

13:59:37 [marqh]

Dimitris: having both ways, showing targets can be ignored is useful

13:59:58 [hknublau]

different CSS would be ok

14:00:02 [marqh]

ericP: duplicating all the examples is one way, could get busy

14:00:18 [marqh]

... could have a section showing the alternative approach

14:00:18 [simonstey]

+q

14:00:45 [ericP]

ack next

14:01:10 [marqh]

TallTed: consistent style as a link

14:01:16 [marqh]

hknublau: i like this approach

14:01:44 [marqh]

simonstey: can do some fancy javascript to enable switching things on and off in examples

14:02:29 [marqh]

ericP: editorial question; i suspect flat styling is sensible

14:03:32 [ericP]

PROPOSED: close issue-140 by adding some consistent style and a link to say that target* is not needed for invocations by API

14:03:40 [hknublau]

+1

14:03:43 [simonstey]

+1

14:03:43 [Dimitris]

+1

14:03:52 [ericP]

+1

14:03:56 [TallTed]

+1

14:04:10 [marqh]

hknublau: do you want to include text into bullet 9.1

14:04:20 [marqh]

ericP: i'll come up with something for this

14:04:28 [ericP]

RESOLVED: close issue-140 by adding some consistent style and a link to say that target* is not needed for invocations by API

14:04:45 [ericP]

ISSUE-183

14:04:45 [trackbot]

ISSUE-183 -- Eliminating the term "Undefined" -- open

14:04:45 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/183

14:04:45 [hknublau]

(bullet 2.1 Focus Nodes)

14:05:11 [ericP]

-> https://github.com/w3c/data-shapes/commit/b367dcd030c652bcb7f7818bb01177dbb7f1d50b holger's edits addressing issue-183

14:05:35 [marqh]

Dimitris: this is an old one

14:05:46 [marqh]

hknublau: this is about sparql functions which return nothing

14:06:03 [marqh]

... term is SPARQL error, i have used 'undefined' not an official term

14:06:32 [marqh]

Dimitris: happy to close this

14:06:32 [ericP]

PROPOSED: Close ISSUE-183, accepting the edits made by the editors to eliminate the term undefined which alters the definition of the sh:hasShape SPARQL function

14:06:38 [hknublau]

+1

14:06:40 [ericP]

+0

14:06:45 [marqh]

+1

14:06:47 [simonstey]

+1

14:07:00 [Dimitris]

+1

14:07:04 [TallTed]

+1

14:07:07 [ericP]

RESOLVED: Close ISSUE-183, accepting the edits made by the editors to eliminate the term undefined which alters the definition of the sh:hasShape SPARQL function

14:07:15 [ericP]

ISSUE-184

14:07:15 [trackbot]

ISSUE-184 -- Property paths and value nodes -- open

14:07:15 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/184

14:07:54 [marqh]

Dimitris: use sparql definitions for property paths

14:08:12 [marqh]

... when on property paths don't take duplicate nodes

14:08:21 [marqh]

... issues including cardinality

14:08:42 [marqh]

... proposal is to work with sets

14:08:54 [marqh]

hknublau: this is already implemented in the spec

14:09:06 [marqh]

... the question is whether people are against this change

14:09:14 [ericP]

PROPOSED: Close ISSUE-184, agreeing that on property constraints with sh:path the value nodes are a set with no duplicate value nodes.

14:09:20 [hknublau]

+1

14:09:22 [Dimitris]

+1

14:09:27 [marqh]

+1

14:09:28 [ericP]

+1

14:09:31 [TallTed]

+1

14:09:35 [simonstey]

+1

14:09:56 [ericP]

RESOLVED: Close ISSUE-184, agreeing that on property constraints with sh:path the value nodes are a set with no duplicate value nodes.

14:10:00 [ericP]

ISSUE-185

14:10:00 [trackbot]

ISSUE-185 -- Processing order for filters and constraints -- open

14:10:00 [trackbot]

http://www.w3.org/2014/data-shapes/track/issues/185

14:10:52 [marqh]

Dimitris: after some of peter's comments, introduced ordering, proposl: there is an ordering

14:11:19 [simonstey]

+1

14:11:21 [simonstey]

+q

14:11:30 [ericP]

ack next

14:11:53 [marqh]

simonstey: are there any implication that arbitray order allows you to do, rather than this strict order

14:12:02 [marqh]

... is there something you can't do

14:12:17 [marqh]

... is there minimal implication

14:12:51 [marqh]

Dimitris: can make single query for whole constraint, can choose to operate on fewer nodes

14:13:11 [marqh]

... when a failure occurs, not easy to define what happens

14:13:30 [marqh]

simonstey: if validation fails, based on node not to be included, but filter not applied yet?

14:14:05 [marqh]

... if first step fails, then could end up with different results

14:14:11 [marqh]

ericP: ambiguous

14:14:47 [marqh]

Dimitris: might not complete result back when failure occurs

14:15:10 [marqh]

simonstey: if filter wants only to check constraint on 'male' persons

14:15:19 [marqh]

... now, do you do filtering afterwards?

14:15:53 [marqh]

... failure might not occur if filtering done first

14:16:10 [marqh]

hknublau: i think this can be avoided if we always evaluate both: enforce this rule

14:16:20 [marqh]

... in this case, these are not returned as failures

14:16:30 [marqh]

... even if there is a failure, it woudl not propagate up

14:17:13 [marqh]

simonstey: do we need to consider a crashdue to the failure

14:17:19 [marqh]

hknublau: i don't think so

14:18:12 [marqh]

ericP: this is a form of negation and failure; sometimes you need to avoid ssituation where you can infer p and not p

14:19:18 [marqh]

ericP: Proposal: SHACL does not force whether filters or constraints are evaluated first for validating a node against a shape.

14:19:37 [ericP]

PROPOSED: Close ISSUE-185, agreeing that SHACL does not define whether filters or constraints are evaluated first when validating a node against a shape, but both must be evaluated

14:19:44 [marqh]

simonstey: if constraint is evaluated before filter, both must be evaluated

14:19:49 [simonstey]

+1

14:19:49 [hknublau]

+0.5

14:19:53 [Dimitris]

+1

14:19:55 [ericP]

0

14:19:57 [marqh]

q+

14:20:00 [Labra]

0

14:20:17 [TallTed]

q+

14:20:28 [ericP]

marqh: re readability: should we just say "but all must be evaluated"?

14:20:34 [ericP]

ack next

14:21:01 [marqh]

TallTed: as i recall, the point on filters was to eliminate from constraint validation things which may be problematic

14:21:34 [marqh]

simonstey: maybe not to include focus nodes, it may be useful

14:21:36 [marqh]

q+

14:21:52 [marqh]

simonstey: may want to filter out certain nodes

14:22:04 [ericP]

ack next

14:22:08 [marqh]

TallTed: may have a million zeros and six ones, why invest time on things which don't matter

14:22:18 [marqh]

hknublau: similar with sh:and and sh:or

14:22:37 [marqh]

hknublau: every filter is an and or an or, can be turned around

14:23:01 [marqh]

TallTed: if can't hint at order, every job may last for ever

14:24:02 [ericP]

marqh: i feel like we don't need to mention ordering. it's defined either way

14:24:14 [marqh]

TallTed: then they are not filters, they are constraints

14:24:22 [ericP]

s/mention ordering/mention ordering in the standard/

14:24:33 [marqh]

simonstey: if validation of the filter fails, are you returning any results

14:25:06 [marqh]

... if constraint fails, you woudl return a validation results

14:25:07 [Dimitris]

q+

14:25:23 [ericP]

ack next

14:25:29 [marqh]

TallTed: i see not value in not saying: this is teh order that this should be dealt in

14:26:32 [hknublau]

q+

14:26:39 [ericP]

ack next

14:26:39 [marqh]

marqh: interested in the reason to maintain the felixibilty nod not have order

14:27:00 [simonstey]

+q

14:27:22 [ericP]

ack next

14:27:23 [marqh]

Dimitris: have and and or where order does not matter, this is consistent

14:28:04 [marqh]

hknublau: unfortunately, had long discussion with karen, on what is considerd a focusNode. we agreed that a focusNode is a node which passes all teh filters

14:28:13 [marqh]

... target - filters are focusNodes

14:28:33 [marqh]

... need order that things are defined

14:28:43 [simonstey]

q?

14:29:06 [marqh]

TallTed: filter should be evaluated before constraints by the spec

14:29:38 [marqh]

TallTed: i believe hinting is a good thing, ordering helps

14:30:02 [marqh]

... harder in RDF world to have ordering, but having a class of things to deal with

14:30:03 [ericP]

ack next

14:30:29 [marqh]

simonstey: and and or are completely different things. ordering with regard to constraints is needed

14:30:43 [marqh]

... checking the filter afterwards is not good

14:31:22 [Dimitris]

q+

14:31:35 [marqh]

... if either of the ands fail, you could do a lot of work not required

14:31:50 [ericP]

ack next

14:32:27 [marqh]

Dimitris: two and statements, order matters. engine evaluates

14:33:04 [marqh]

simonstey: if you have an and and 2 shapes, if one fails, can return faster result

14:33:49 [marqh]

... with filter and constraint, if constraint first, fails, then filter passes, return result, if ordered differently, diferent results returned

14:34:03 [marqh]

q+

14:34:31 [marqh]

Dimitris: i am not sure, i think the same thing happens

14:35:01 [ericP]

ack next

14:36:03 [ericP]

marqh: it sounds to me like we should change the issue-185 proposal "SHACL *does* not care about order" to "SHACL *does* care about the order"

14:36:22 [marqh]

marqh in favour of changing ISSUE-185 to propose that filters are evaluated before constraints

14:37:06 [marqh]

Dimitris: i mind, but i am willing to consider

14:38:20 [marqh]

simonstey: i think that Dimitris has a valid point why cases are not that different. in both cases if you evaluate wrong side first, then the very last would fail, the outcome is the same; i get this

14:38:58 [marqh]

... unless a strong reason exists for flexibilty, it is useful to require that filters are evaluated first

14:39:40 [marqh]

... could make statement that it doesn't matter to the result, but the filters shall be evaluated first

14:39:56 [marqh]

TallTed: this is a useful detail for the reader

14:40:34 [marqh]

... if shape2 does not pass filter, then the work is trivially satisfied

14:41:24 [marqh]

Dimitris: if we have order, we cannot model the shape as sh:and and sh:or

14:42:39 [marqh]

... if there is no ordering, then we can model teh filter like an sh:or

14:43:11 [marqh]

TallTed: the point of filter is to say don't evaluate

14:44:07 [marqh]

... we had a graphic showing that the filter pre-processess before constraints

14:46:15 [marqh]

TallTed: is filter about results or about focusNodes?

14:46:36 [marqh]

ericP: it is about removing results that would otherwise fail

14:47:01 [marqh]

ericP: i think this is about result reporting: take 5 errors and return zero

14:47:19 [marqh]

TallTed: then it should be about results filtering; it should be moved to the results section

14:47:42 [marqh]

... int that case, it would be not be about choosing focusNodes

14:48:07 [marqh]

ericP: has the effect of filtering out results

14:48:17 [marqh]

TallTed: question is what is being applied to

14:48:34 [marqh]

... if about trimming results, it comes at a different point, doing a different job

14:48:51 [marqh]

ericP: not only about trimming results

14:48:56 [marqh]

TallTed: it is a focusNode trimmer

14:49:34 [marqh]

ericP: in C pragmas are written in as par tof the compiler

14:49:49 [marqh]

... Dimitris raised the point that it would be nice to let the system be smart

14:51:10 [marqh]

q+

14:51:36 [ericP]

ack next

14:53:12 [marqh]

marqh: i am quite taken with the point that oreder should be enforced; i don't think the counter argument is strong

14:53:37 [marqh]

ericP: Dimitris point is a good one, where the filter is very complex and may fail a lot

14:54:38 [Dimitris]

this is more ordering

14:55:48 [marqh]

TallTed: do we need 2 types of filter

14:55:53 [marqh]

ericP: pre and post constrasin

14:56:01 [marqh]

marqh: i'm not convinced 2 filter are needed

14:56:36 [marqh]

simonstey: or and not are not intended to be used as filter

14:57:58 [marqh]

ericP: how can a test be set up to test the order of processing

14:58:06 [marqh]

TallTed: can write long test

14:59:33 [marqh]

... can make a test with a div0

14:59:36 [marqh]

q+

14:59:51 [ericP]

ack next

15:00:21 [ericP]

marqh: if you have an impl, you can test pieces, even if it's not an end-to-end integration test

15:00:59 [ericP]

ericP: so write it in English, not in standard test case format

15:01:03 [Dimitris]

q+

15:01:54 [marqh]

Dimitris: can provide a new definition int he spe to close this issue

15:02:00 [marqh]

... definition of validation

15:02:02 [Dimitris]

A node validates against a shape if and only if either it does not validate against some filter of the shape or none of the constraints in the shape produce a validation result or a failure for the node.

15:02:37 [marqh]

ericP: who will write a proposal for change

15:03:38 [ericP]

ACTION: marqh to propose changes to the definition of validation to close issue-185

15:03:38 [trackbot]

Created ACTION-46 - Propose changes to the definition of validation to close issue-185 [on Mark Hedley - due 2016-11-16].

15:03:53 [ericP]

ack next

15:04:06 [marqh]

Dimitris: it's fine, i'm done

15:04:21 [ericP]

trackbot, end meeting

15:04:21 [trackbot]

Zakim, list attendees

15:04:21 [Zakim]

As of this point the attendees have been marqh, hknublau, Labra, ericP, simonstey, TallTed, Dimitris, .5

15:04:29 [trackbot]

RRSAgent, please draft minutes

15:04:29 [RRSAgent]

I have made the request to generate http://www.w3.org/2016/11/09-shapes-minutes.html trackbot

15:04:30 [trackbot]

RRSAgent, bye

15:04:30 [RRSAgent]

I see 1 open action item saved in http://www.w3.org/2016/11/09-shapes-actions.rdf :

15:04:30 [RRSAgent]

ACTION: marqh to propose changes to the definition of validation to close issue-185 [1]

15:04:30 [RRSAgent]

recorded in http://www.w3.org/2016/11/09-shapes-irc#T15-03-38