Meetings/Telecon2015.12.10

Event details

Agenda

  1. Scribe nomination
  2. Roll Call
  3. Approval of November 12 minutes

  4. Next meeting January 7 2016 (note this is 4 weeks later)
  5. Actions from the previous meeting
  6. Topics
  7. (Last 15 minutes) Road map
  8. Other business

How to prepare for this meeting

  1. Read the specs that are currently available for TC review. (* = recently available)
  2. Read the above issues and comment if you wish

Voting Rights

Held by:

Minutes

Chair

Scribe

Attendees

Regrets

Resolutions

Actions

Chat transcript from room: oslc

[07:36] ian green (ibm): ian to scribe

[07:37] ian green (ibm): approve minutes from 12th nov.

[07:38] ian green (ibm): Next meeting Jan 7th 2016 (24th Dec cancelled).

[07:40] ian green (ibm): Propose to have an owner ("volunteer") to push through and make progress on a particular document

[07:41] ian green (ibm): Proposing primary reviewer for Attachments - no volunteer

[07:41] ian green (ibm): Proposing primary reviewer Delegated dialogue - Ian & Martin P

[07:42] ian green (ibm): Proposing primary reviewer Resource Preview - Jean-Luc & David Honey

[07:43] ian green (ibm): Proposing primary reviewer core vocabulary - David H

[07:43] ian green (ibm): Proposing primary reviewer (de facto) Discovery - Martin P

[07:43] ian green (ibm): Proposing primary reviewer OSLC Overview - Martin S.


[07:45] ian green (ibm): Agenda item: Issues

[07:46] ian green (ibm): https://issues.oasis-open.org/browse/OSLCCORE-50

[07:47] Martin Sarabura (PTC): https://tools.oasis-open.org/version-control/svn/oslc-core/trunk/specs/discovery.html#resourceConstraints

[07:47] ian green (ibm): Unnecessary constraints on discovery resources - not necessary to specify that they are read-only.

[07:49] ian green (ibm): Use cases include an authenticated user to rename via a serviceprovider resource.

[07:50] ian green (ibm): Another is a service provider catalogue acting as a central registry. Providers can register there.

[07:53] ian green (ibm): Ian: Why do we specify read-only at all?

[07:53] ian green (ibm): (ie in general)

[07:54] ian green (ibm): Jim: review uses and decide whether we need such a property.

[07:55] ian green (ibm): David: does this mean that clients must always fetch RS to determine writability?

[07:56] ian green (ibm): David: what happens on PUT when a property is changed? Is such a change ignored, forbidded, etc.?

[07:56] ian green (ibm): Jim: there is a clause to define the behaviour here

[07:58] Jim Amsden (IBM)1: Proposal: Examine all ResourceShape Properties and unless there is a compelling reason for readOnly to be specified true, remove the property and leave it unspecified in order to allow servers more flexibility in supporting dynamic client needs.

[07:58] ian green (ibm): David: some properties must be read-only

[08:05] ian green (ibm): Discussion about backwards compatibility of changing read-onlyness or leaving it unspecified

[08:07] ian green (ibm): Concern over potential impact since readOnly property would need to be revised.

[08:10] ian green (ibm): MartinP: discuss this in the context of "overriding" published RS

[08:12] ian green (ibm): Jim: in specific case of issue 50, more dynamic discovery is possible without changing meaning of RS. Server can allow that values of such properties to change in other ways

[08:15] ian green (ibm): MartinP: value of use cases in issue 50 insufficient to struggle with dealing with implications of read-onlyness.

[08:17] ian green (ibm): Jim: LDPCs offer enough flexibility in the discovery use cases without needing a spec. change.

[08:18] ian green (ibm): call dropped. re-dialling into conf.

[08:19] ian green (ibm): Proposal: do not change the meaning of read-only

[08:20] ian green (ibm): Votes:

[08:20] David Honey (IBM): +1

[08:20] ian green (ibm): +1

[08:20] Martin Pain (IBM): +1

[08:20] Martin Sarabura (PTC): +1

[08:20] Jim Amsden (IBM)1: +1

[08:20] Jean-Luc Johnson (Airbus): +1

[08:20] ian green (ibm): Proposal accepted

[08:21] ian green (ibm): MartinS: should we include in non-normative section how to deal with such use cases

[08:21] ian green (ibm): Jim: takes action to write this up

[08:22] Harish (SoftwareAG): +1

[08:22] Harish (SoftwareAG): for the earlier proposal not to change the meaning of read-only

[08:23] Harish (SoftwareAG): got bumped off the call - rejoining


[08:24] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-43

[08:26] Jim Amsden (IBM)1: <#dcterms-title> a oslc:Property ; oslc:name "title" ; oslc:propertyDefinition dcterms:title ; oslc:occurs oslc:Zero-or-one ; dcterms:description "The summary of this shape."rdf:XMLLiteral ; oslc:valueType rdf:XMLLiteral ; oslc:readOnly true .

[08:27] Jim Amsden (IBM)1: <#ResourceShape> a oslc:ResourceShape ; dcterms:title "A shape resource describes the contents of and constraints on some set of described resources." ; oslc:describes oslc:ResourceShape ; oslc:property <#dcterms-title>, <#oslc-describes>, <#oslc-property>, <#oslc-hidden> .

[08:31] ian green (ibm): Discussion about how to avoid introducing a new Class

[08:51] ian green (ibm): Lots of discussion on the meaning of "shared" or "common" oslc:Properties and whether we need an aggregating RS

[08:55] ian green (ibm): David: use a human-written table rather than using the tool to communicate the "common" properties

[09:00] Martin Pain (IBM): dcterms:subject: http://dublincore.org/documents/dcmi-terms/#terms-subject

[09:02] ian green (ibm): Meeting closes.

Meetings/Telecon2015.12.10 (last edited 2015-12-17 18:26:05 by sarabura)