OSLC Core Meeting October 20, 2010

Last week's meeting

Meeting logistics

See the OslcCoreMeetings for more information, more dial-in numbers and on-line meeting information.

• Conference Access
  • Toll free: 1-866-423-8350
  • Toll: 1-719-387-8273
• Participant passcode: 558663

Agenda

Review actions taken since last meeting:

• AI Dave J: review guidance, review core - make changes to make the RDF/XML and XML distinction clear
• AI Dave J to clarify Atom usage, content type of Atom entry content
• AI Dave J: add link annotation to representation guidance for XML
• AI Dave J to clarify that union is the correct interpretation having both for allowedValues and allowedValue
• AI Dave J to clarify oslc:range using Ian's language

Minutes

Attendees and notes from the meeting

Attendees

• DaveJohnson
• PaulMcMahan
• SteveSpeicher
• NickCrossley
• ScottBosworth
• JimConallen
• MikeLoefler?

Topics discussed

We reviewed changes to representations guidance, found some small issues which Dave will fix:

• Lingering RDF/XML reference in XML section
• Use content-types in the section titles for the sake of clarity
• In the Atom rule 1.4, don't say MUST

Next we discussed how to finalize representation guidance and Arthur pointed out that it really is part of the Core spec, since we reference it from the Core spec and show examples of it from the Core spec. Arthur and Scott proposed adding the representation to the Core spec as appendix B and then linking to the examples from there.

That led to a discussion of which parts of the spec are normative vs. informative. Dave will look at other spec efforts and proposed how to mark sections as normative and informative.

Dave lamented that we only have long full-form examples in the appendices and no small simple examples in the spec to show how things work and Arthur responded that perhaps we need a primer, which can be talky, show examples and explain how things work in a more conversational way. Scott and Dave agreed that this could solve some problems that newcomers have in making sense of OSLC.

Finally, we discussed how a provider should respond when it does not support part of the query syntax. Consensus is that the spec should say something like this: "providers MUST respond with an error when they do not fully support incoming query syntax expression and SHOULD respond with a 400 and an explanation.