Meeting minutes 17th June 2011
- Roll: Martin, Jacques, Robin, Dave, Patrick Regrets: Chet Agenda: Approved as posted
3rd June: http://www.oasis-open.org/apps/org/workgroup/tab/email/archives/201106/msg00010.html Approved w/o Agreed to post to wiki ACTION-20110617-0: Martin to log comments on website feedback from 3rd June to OASIS
Standards Dependency and Citations update:
http://www.oasis-open.org/apps/org/workgroup/tab/email/archives/201105/msg00016.html Patrick presents his email. It includes some HTML that shows some examples of correct and incorrect citations and which oasis specs are guilty. This is a good tool to show the Board that there are consistency issues. It is also good for showing dependencies - if you want to implement a spec what other specs are relied on. Next decision point is what to do with this info and how to present it. Robin: this provides a a good argument and foundation for an OASIS citation database
Public review guidelines:
- Dave summarising Rob Weirs comments, will send on the email. 1)testing might not be done in the same TC, needs to be reflected in the guidelines e.g. odf Robin Cover (OASIS): "Statement of Use", with respect to a Committee Specification, is a written statement by an OASIS Organizational Member stating that it is successfully using or implementing that specification in accordance with the conformance clauses specified in Section 2.18, and tating whether its use included the interoperation of multiple independent implementations. Hence cannot assume its TC members.
- 2) no guidance on whether test artifact should be notes or specs, and not sure whether should require the same review cycle 30/60 days
- There is a concern about what will happen if these are published as guidelines, as they eventually turn into mandates, and may be contrary to spec development in OASIS.
- On interop testing one can observe three types of specs: existing stuff with many vendor impls. Typically lots if compromised shoulds and not many musts brand new - mostly musts and tight conformance - no vested interest third category - small number of close nit vendors which is somewhere in the middle The guidelines seem to be written from the brand new scenario, so need to take into account the other scenarios. Patrick: real difference between testing conformance and testing interop they are not the same thing, and we need to keep this highlighted.
Kavi-5/web site update:
- Martin needs to log feedback
Multi Product Identification:
- Robin goes over his email query. will be making a suggestion soon. can review test suite "parts" as part of our testing guidelines work: are tests specs, or notes?
Summary of new and outstanding action items:
- ACTION-20110617-0: Martin to log comments on website feedback from 3rd June to OASIS