Responses to comments received re Committee Specification Draft

I have finally (with sincere thanks to Jim!) put together the spreadsheet containing responses to the specific questions raised at the end of our review session. Please see the attached file. The wording of the responses is essentially drawn from Jim’s responses dated Oct 25.

Since I am new to this process I had to get some guidance from the TC Administrator on the correct approach. One of his recommendations was to firm up the language of the resolution so that an outside party would have the context needed to properly interpret the resolution. I have updated the minutes of Oct 27 to be more precise. They are included below.

Regards,

Dr. Martin Sarabura
Technical Fellow, Office of the CTO

+1 519.502.4819
msarabura@ptc.com

Scribe

Martin Sarabura (PTC)

Attendees

David Honey (IBM)
Harish Krishnaswamy (Software AG)
Jad El-khoury (KTH)
Jean-Luc Johnson (Airbus)
Jim Amsden (IBM)
Martin Sarabura (PTC)
Nick Crossley (IBM)

Regrets

Resolutions

· Request the TC administration to hold a Special Majority Vallot to approve OSLC Core Version 3.0 located at http://docs.oasis-open.org/oslc-core/oslc-core/v3.0/csprd01/ as a Committee Specification.

· Accept the proposal as stated in https://issues.oasis-open.org/browse/OSLCCORE-68

Actions

Martin to submit request for publication
Re-post Martin action items from previous meeting
Jim close issue 78

Chat transcript from room: oslc

[07:04] Martin Sarabura (PTC): Martin will scribe

[07:04] List of attendees: David Honey (Persistent/IBM), Harish K (Software AG), Jad El-khoury (KTH), Jean-Luc Johnson (Airbus), Jim Amsden (IBM), Martin Sarabura (PTC), Nick Crossley (IBM)

[07:05] Martin Sarabura (PTC): https://wiki.oasis-open.org/oslc-core/Meetings/Telecon2016.10.13

[07:05] Martin Sarabura (PTC): Minutes accepted - just a typo

[07:06] Martin Sarabura (PTC): Martin's action items not done - continue to next meeting

[07:08] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-25

[07:08] Martin Sarabura (PTC): Jim to close issue 78

[07:08] Martin Sarabura (PTC): add action item

[07:09] Martin Sarabura (PTC): To main item: Re Mr. Meyer's feedback

[07:10] Jad El-khoury (KTH): https://lists.oasis-open.org/archives/oslc-core-comment/201609/msg00000.html

[07:11] Martin Sarabura (PTC): Jim: OSLC domains are minimally specified in order to make implementation easier

[07:11] Martin Sarabura (PTC): Keep as open. extensible, flexible as possible

[07:12] Martin Sarabura (PTC): Therefore insufficient information in spec to define all constraints in a given implementation

[07:12] Martin Sarabura (PTC): Invented resource shapes as simplification of OWL

[07:12] Martin Sarabura (PTC): Shapes describe resources and support discovery and constrain vocabularies for specific purposes

[07:13] Martin Sarabura (PTC): Servers always free to further extend and constrain the vocabularies

[07:13] Martin Sarabura (PTC): Nick: Recommended to, not just "free to"

[07:14] Martin Sarabura (PTC): No standard set of extended properties, cardinalities and data types. There is a way to describe what those extensions are

[07:15] Martin Sarabura (PTC): We can't specify overly-constrained designs for specific purposes

[07:17] Martin Sarabura (PTC): Martin: Isn't this about how to specify enough so the client can be written more easily?

[07:19] Martin Sarabura (PTC): Martin: Doesn't this make client development more expensive?

[07:19] Martin Sarabura (PTC): Jim: Delegated dialogs, etc capabilities exist

[07:21] Martin Sarabura (PTC): Martin: Not so much for integration as for developing a client in a given domain

[07:22] Martin Sarabura (PTC): Can't force other server to specify or use your constraints

[07:23] Martin Sarabura (PTC): Jim: Even if constraints not speicified, operations could fail. Better approachh is clearly to use reflection but maybe that's not available for some tools

[07:24] Martin Sarabura (PTC): Nick: Historically, initial implementations didn't publish shapes because that was considered a luxury

[07:24] Martin Sarabura (PTC): Over time the importance of shape discovery became more obvious as world became more general

[07:25] Martin Sarabura (PTC): Could still hard code per server type in near term, later on add reflection

[07:26] Martin Sarabura (PTC): Nick: Shapes not 100% solution but surprisingly complete

[07:26] Martin Sarabura (PTC): Jim: Shapes can be described by shapes - possible to extend them in a standard way too

[07:27] Martin Sarabura (PTC): Instance property on the shape makes meta-shapes discoverable

[07:27] Martin Sarabura (PTC): Jim: Puts burden on users to do the discovery, benefit is open world.

[07:29] Martin Sarabura (PTC): Martin: Client developers want a little more constraints than what we are inclined to provide

[07:30] Martin Sarabura (PTC): Agree that we need not bring this into 3.0

[07:30] Martin Sarabura (PTC): Nick: Other standard property we use is owl:sameas for rdf type definitions

[07:30] Martin Sarabura (PTC): and property definitions

[07:31] Martin Sarabura (PTC): Nick: Raise action item to write note to describe ibm use of shapes and owl:sameas and use the shapes to inform dynamic client creation

[07:32] Martin Sarabura (PTC): Jad: Bit confusing as to intentions of shapes - need explanation of how to use shapes

[07:33] Martin Sarabura (PTC): Nick: General idea for shapes in note that Arthur Ryman wrote for w3c

[07:33] Martin Sarabura (PTC): Jim: In the bibliography for the spec

[07:34] Martin Sarabura (PTC): David: Broadly in agreement - system is extensible, way to describe constraints, good for now

[07:35] Martin Sarabura (PTC): Jim: Tim requesting specific extensions - in theory we could consider putting them in on a case-by-case basis. May run into compatibility issues; we also should keep in mind that SHACL may eventually get published

[07:35] Martin Sarabura (PTC): David: Long term not sure what's going on with SHACL

[07:36] Martin Sarabura (PTC): Jim: Not sure what they come up will be usable by our community.

[07:36] Martin Sarabura (PTC): David: Shouldn't be in a hurry given the uncertainties with regards to SHACL

[07:37] Martin Sarabura (PTC): Jim: Many implementations of resource shapes 2.0, seem to be working.

[07:38] Martin Sarabura (PTC): Jim: Plus, controversy and additional time for SHACL indicates risk; leave OSLC shapes as is for now

[07:40] Martin Sarabura (PTC): Jim: Invest in OSLC for JS may provide more value by demonstrating best practices

[07:41] Martin Sarabura (PTC): Martin: This is ongoing work, is fine to proceed

[07:43] Martin Sarabura (PTC): Jad: Should we respond to email? Jim: Have responded by email, not required to respond further unless somebody on TC feels it is required

[07:43] Martin Sarabura (PTC): David: We should encourage Tim to participate on future meetings.

[07:45] Martin Sarabura (PTC): Eg., new type of literal for formatted text seems reasonable

[07:45] Martin Sarabura (PTC): Nick: Response would be to wait to see what happens in SHACL

[07:45] Martin Sarabura (PTC): David: Just not at a point in time where we can assess properly

[07:48] Martin Sarabura (PTC): There is a strong sub-community in SHACL that wants to avoid any UI-related constraints

[07:48] Martin Sarabura (PTC): Nick: We may at some point wish to take up the UI constraints if SHACL chooses to drop it

[07:51] Martin Sarabura (PTC): Martin: Proposes that we request the TC administration to hold a Special Majority Vallot to approve OSLC Core Version 3.0 located at http://docs.oasis-open.org/oslc-core/oslc-core/v3.0/csprd01/ as a Committee Specification.

[07:51] Nick Crossley (IBM): Seconded

[07:51] Martin Sarabura (PTC): Jim: Second

[07:52] David Honey (Persistent/IBM): +1

[07:52] Jim Amsden (IBM): +1

[07:52] Martin Sarabura (PTC): +1

[07:52] Nick Crossley (IBM): +1

[07:52] Jad El-khoury (KTH): +1

[07:52] Jim Amsden (IBM): +1

[07:52] Jean-Luc Johnson (Airbus): +1

[07:53] Martin Sarabura (PTC): Harish votes in favor as voice vote

[07:53] Martin Sarabura (PTC): Carried

[07:53] Martin Sarabura (PTC): Martin action item to submit the appropriate paperwork

[07:55] Martin Sarabura (PTC): Jim: Response to Tim, Martin to respond once more

[07:55] Martin Sarabura (PTC): Open issues on TRS - 70 and 71. Nick: 68 changes made to the proposal

[07:56] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-68

[07:57] Jim Amsden (IBM): Proposal: Accept the proposed resolution

[07:57] Nick Crossley (IBM): Seconded

[07:57] Nick Crossley (IBM): +1

[07:57] Jean-Luc Johnson (Airbus): +1

[07:57] David Honey (Persistent/IBM): +1

[07:58] Jad El-khoury (KTH): +1

[07:58] Jim Amsden (IBM): +1

[07:58] Martin Sarabura (PTC): +1

[07:58] Martin Sarabura (PTC): Harish voice vote in favor

[07:58] Martin Sarabura (PTC): Carried

[07:59] Martin Sarabura (PTC): Jim action item to move the verbiage into the spec

[07:59] Martin Sarabura (PTC): Next meeting to discuss 70 and 71

[07:59] Martin Sarabura (PTC): Meeting adjourned

oslc-core message