Meetings/Telecon2016.07.21

Event details

Agenda

  1. Scribe nomination
  2. Roll Call
  3. Approval of June 23 minutes (July 7 meeting was cancelled)

  4. Next meeting August 4, 2016

  5. Actions from the previous meeting
    • Nick to update ReSpec template and send email pointing to it

    • Nick to compare his local copies of TRS and ILDP documents to those found online (see references in minutes)

    • David to raise issues for known gaps such as what to do if there are two create events for the same resource, or support for selective indexing.

    • Nick to raise issue around support for delegated access control

    • Jim to raise issue regarding support for notifications via MQTT or other such publish/subscribe method

  6. Topics
  7. Other business

Voting Rights

Held by:

Minutes

Chair

Scribe

Attendees

Regrets

Resolutions

Actions

Chat transcript from room: oslc [07:06] Martin Sarabura (PTC): Martin volunteers to be scribe

[07:07] Martin Sarabura (PTC): https://wiki.oasis-open.org/oslc-core/Meetings/Telecon2016.06.23

[07:07] Martin Sarabura (PTC): June 23 minutes accepted

[07:08] Martin Sarabura (PTC): https://wiki.oasis-open.org/oslc-core/Meetings/Telecon2016.07.21


[07:08] Martin Sarabura (PTC): First topic: Add CCM to core tc?

[07:09] Martin Sarabura (PTC): Jim: Possibility of diluting the core focus - fragmentation of resources vs coupling of independent deliverables

[07:10] Martin Sarabura (PTC): Martin: Readers would consider all TC outputs to be part of spec anyway regardless of which TC published it

[07:10] Martin Sarabura (PTC): Jim: New TC domain committee being formed

[07:11] Martin Sarabura (PTC): Mandate to migrate documents to OASIS structure without making substantive changes

[07:12] Martin Sarabura (PTC): What about two OSLC TCs? One for core technologies, one for domains

[07:12] Martin Sarabura (PTC): Thus Change Mgmt would naturally belong in domains, config mgmt in core

[07:14] Martin Sarabura (PTC): David: A good model, you can choose which to attend and hopefully the one you choose to attend is aligned with your interests

[07:14] Martin Sarabura (PTC): Jim: Core was a fair amount of work, similarly config mgmt. Putting both into same TC could make meetings quite long...

[07:15] Martin Sarabura (PTC): Action: Martin to check with Chet on our current publication status

[07:15] Martin Sarabura (PTC): Jim: Maybe not as much to do in these areas now

[07:17] Martin Sarabura (PTC): Jean-Luc: Config mgmt has impact on domains and on core - no way to express how service provider is configuration enabled

[07:18] Martin Sarabura (PTC): That could be a short-term solution, then we can talk about reorg later

[07:20] Martin Sarabura (PTC): Jim has had this conversation with Nick previously - domains including config mgmt are largely independent of core

[07:21] Martin Sarabura (PTC): Note all core capabilities are optional, including discovery

[07:23] Martin Sarabura (PTC): Of course a service provider that is less capable in the area of discovery has a more limited applicability

[07:24] Martin Sarabura (PTC): Jim: There is a method to determine whether config mgmt is supported

[07:25] Martin Sarabura (PTC): David: Hybrid model also possible - same service provider can serve some resources that are configuration-aware, others not

[07:26] Martin Sarabura (PTC): Perhaps best for ccm tc to address

[07:27] Martin Sarabura (PTC): Jim: Support for configuration management is a property of container, not server


[07:28] Martin Sarabura (PTC): Action items

[07:29] Martin Sarabura (PTC): Nick has not yet modified and republished respec - Jim was still able to use it

[07:29] Martin Sarabura (PTC): Leave on the list

[07:29] Martin Sarabura (PTC): Jim noticed a few other things that may need to be updated

[07:30] Martin Sarabura (PTC): Based on feedback from Chet

[07:30] Martin Sarabura (PTC): Jim to tell Nick about these other possible changes

[07:30] Martin Sarabura (PTC): Jim will post as issue

[07:31] Martin Sarabura (PTC): Nick's local copies same as online

[07:31] Martin Sarabura (PTC): David has raised issues - action item complete

[07:32] Martin Sarabura (PTC): 70 - two creates

[07:32] Martin Sarabura (PTC): also 71, 72

[07:32] Martin Sarabura (PTC): Nick posted issue re delegated access control - 68

[07:33] Martin Sarabura (PTC): Jim's action item covered by 72


[07:34] Martin Sarabura (PTC): Next topic: TRS document first pass

[07:34] Martin Sarabura (PTC): Jim has a few issues: 69 - missing references RDF-VAL etc

[07:34] Martin Sarabura (PTC): 69 should be closed

[07:35] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-74

[07:36] Martin Sarabura (PTC): Do we need acc namespace? Already published, we can't get rid of it

[07:37] Martin Sarabura (PTC): David: Does access control have wider applicability than TRS?

[07:37] Martin Sarabura (PTC): Likely it does, therefore we should keep it separate

[07:38] Martin Sarabura (PTC): Close it, don't merge

[07:38] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-73

[07:40] Martin Sarabura (PTC): Jim proposes publishing as separate document(s). Reason is practicality - we're not ready to add the document to the current core lineup

[07:40] Martin Sarabura (PTC): David: Agree, especially since many fewer implementations of TRS

[07:42] Martin Sarabura (PTC): Doesn't require a vote since we would not be changing anything in our process

[07:42] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-75

[07:43] Martin Sarabura (PTC): Jim: Problem is TRS CUD in current spec are individuals, not classes and yet shapes table has properties for them

[07:43] Martin Sarabura (PTC): eg., type, change and order

[07:44] Martin Sarabura (PTC): New class called change event having 3 three properties and having 3 instances

[07:44] Martin Sarabura (PTC): David: Ontology would align better

[07:46] Martin Sarabura (PTC): Jim: Change CUD to 3 new classes with new properties, create change event shape that constrains those classes, not subclasses of anything

[07:48] Martin Sarabura (PTC): David: Set of data is same, with same meaning - why 3 classes?

[07:48] Martin Sarabura (PTC): David: Propose abstract base class, with 3 concrete instances

[07:48] Martin Sarabura (PTC): Jim: Inheritance doesn't work in resource shapes so you're not really gaining anything

[07:49] Martin Sarabura (PTC): So it's OK to have one class

[07:50] Martin Sarabura (PTC): Comments on the document?

[07:50] Martin Sarabura (PTC): David: Sections 3.1 through 3.4 are empty.

[07:52] Jim Amsden (IBM): https://tools.oasis-open.org/version-control/svn/oslc-core/trunk/specs/tracked-resource-set.html#resources

[07:54] Martin Sarabura (PTC): Link in the agenda had fragments in it which somehow resulted in those sections being trimmed out

[07:54] Martin Sarabura (PTC): Jim: ReSpec problem

[07:55] Martin Sarabura (PTC): Jim: Some mistakes in the shapes. Investigating...

[07:57] Martin Sarabura (PTC): ChangeEvent doesn't have name. Currently ReSpec says it has a name trs#Deletion

[07:58] Martin Sarabura (PTC): Jim will have to figure out what's going on.

[07:58] Martin Sarabura (PTC): Meeting adjourned

Meetings/Telecon2016.07.21 (last edited 2016-08-01 01:54:19 by sarabura)