OSLC CORE TC Minutes December 22, 2016

["Sarabura, Martin" <msarabura@ptc.com>]

Scribe

• Martin Sarabura (PTC)

Attendees

• Harish Krishnaswamy (Software AG)
• Jad El-khoury (KTH)
• Jim Amsden (IBM)
• Martin Sarabura (PTC)
• Nick Crossley (IBM)

Regrets

Resolutions

Actions

• Jim to finish up the edits
• Martin to check on exact process for resubmitting Committee Specification

Chat transcript from room: oslc

[07:06] Martin Sarabura (PTC): https://wiki.oasis-open.org/oslc-core/Meetings/Telecon2016.12.08

[07:07] Jad El-khoury (KTH): Dec 8 minutes approved

[07:07] List of attendees: Harish K (Software AG), Jad El-khoury (KTH), Jim Amsden (IBM), Martin Sarabura (PTC), Nick Crossley (IBM)

[07:12] Martin Sarabura (PTC): email sent 12/16

[07:12] Martin Sarabura (PTC): Jim&apos;s changes

[07:13] Martin Sarabura (PTC): langString just a one line change, no issues

[07:13] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-25

[07:15] Martin Sarabura (PTC): delete previous line - discussion still about langString

[07:16] Martin Sarabura (PTC): Holger responded that they use the union of xsd:string and rdf:langString

[07:16] Martin Sarabura (PTC): Nick: For OSLC, we are not saying they are equivalent. Saying, interpret as if we said both

[07:17] Martin Sarabura (PTC): They are definitely not equivalent and we don&apos;t want to say that.

[07:19] Martin Sarabura (PTC): https://tools.oasis-open.org/version-control/svn/oslc-core/trunk/specs/resource-shape.html

[07:19] Martin Sarabura (PTC): value type only allows one value

[07:19] Martin Sarabura (PTC): Doesn&apos;t allow literals to have multiple values

[07:19] Martin Sarabura (PTC): Jim knows how to fix this onee

[07:20] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-83

[07:21] Martin Sarabura (PTC): Jim to add list to issue after resolution

[07:22] Martin Sarabura (PTC): allowedValue could be inline

[07:23] Martin Sarabura (PTC): Nick: Typically use allowedValues for enumerated values in which case you need a reference

[07:23] Martin Sarabura (PTC): Jim: Let&apos;s say tool got those allowedValues and iterated, need to be references to get enumerands

[07:24] Martin Sarabura (PTC): Value of the property would be one of the allowedValues - no way to use that value unless it&apos;s a reference

[07:25] Martin Sarabura (PTC): Nick: By having a representation of either can be inline if you so choose

[07:26] Martin Sarabura (PTC): Jim: Why would it ever be reference?

[07:26] Martin Sarabura (PTC): Nick: Most specs in CM changed to either

[07:27] Martin Sarabura (PTC): However list of members may be very large

[07:27] Martin Sarabura (PTC): Therefore justified in many cases.

[07:27] Martin Sarabura (PTC): Jim: Compact resource - doesn&apos;t make sense to inline the icon

[07:28] Martin Sarabura (PTC): Nick: Icon not that big, may make sense

[07:29] Martin Sarabura (PTC): Nick: Definitely don&apos;t want to go through all specs and change - but in future domains we want to think about it more carefully

[07:30] Martin Sarabura (PTC): Jim: Unless somebody raises a defect we should be conservative

[07:31] Martin Sarabura (PTC): Jim: Add note to issue that domains should consider allowing both

[07:31] Martin Sarabura (PTC): https://issues.oasis-open.org/browse/OSLCCORE-84

[07:32] Martin Sarabura (PTC): Jim: Nick needs to review changes - not yet checked into core

[07:32] Martin Sarabura (PTC): Jim: Changes to shapes, not spec

[07:32] Martin Sarabura (PTC): Jim: Will check in, Nick to review

[07:33] Martin Sarabura (PTC): Changes to appendices to documents - maintain change history. Jim has done that

[07:33] Martin Sarabura (PTC): Jim: Ran shape checker and had one problem - Nick has responded: tool is looking for def&apos;n in open-services.net

[07:34] Nick Crossley (IBM): The problem there is the tool is looking for the real definition at open-services.net, and of course that still has the OSLC 2.0 vocab, which does not define those terms. I should look into fixing that some time, so it takes the local input always in preference to the real URI - right now, it takes the local input most but not all of the time of the time, depending on the order in which things are processed or looked up. In the mean time, when it gives that error, you have to check manually to see that you really did define the term correctly in the current vocab. Sorry!

[07:36] Martin Sarabura (PTC): Nick: shape checker sometimes inconsistent in how it does check if defined locally but not found in open-services.net

[07:36] Martin Sarabura (PTC): Will work after everything is published - which is of course too late

[07:38] Martin Sarabura (PTC): Jim will check in changes related to new Error properties - Nick to review as soon as Jim has done so

[07:38] Martin Sarabura (PTC): Martin available to check later next week

[07:39] Martin Sarabura (PTC): Nothing else from Jim today.

[07:40] Martin Sarabura (PTC): Martin to confirm process with Chet before we re-submit documents

[07:42] Martin Sarabura (PTC): Jad will be away Jan 5, Martin to check with David

[07:42] Martin Sarabura (PTC): We should plan to have meeting Jan 5

[07:43] Martin Sarabura (PTC): Meeting adjourned