Minutes TAB call (September 18, 2015)
1) Approval of agenda and minutes from 8/28/15.
2) Status of open action items
3) Update on Board meeting next week
4) Laurent's self-certification feedback request
5) Publishing citation master lists
=== Minutes ===
Chet Ensign (staff)
Jacques Durand (Fujitsu)
Patrick Durusau (individual)
Kevin Mangold (NIST)
Ashok Malhotra (Oracle)
No discussion, no objections. Agenda was unanimously approved.
Approval of minutes:
Draft minutes of call 28 August 2015: https://lists.oasis-open.org/archives/tab/201508/msg00144.html
No discussion, no objections. Minutes unanimously approved.
Action item: Chet - publish approved TAB minutes.
2. Status of open action items.
Status of open action items:
* Patrick to sum up referencing issues in specs, from his reviewer viewpoint. Issue is doing an analysis on problems with references in OASIS specs for reporting to the board.
Patrick report sent to TAB. Close this item.
* Chet - draft a TAB note to the DITA TC expressing our understanding of their need to address some of these comments in next release of DITA and offering to work with TC members on improving these when they start their next round of editorial work.
Chet sent note to TC as part of another email. Close this item.
* Chet - send work plan to Gershon for presentation to board.
Chet sent the work plan to Gershon. It is on the agenda for next Board meeting. Close this item.
* Patrick / Chet - prepare another draft version, approve by email if possible, and if approved forward to Gershon.
Output for IETF RFCs is done but W3C Recommendations still open. Keep this item open.
* Chet - contact Carol Geyer about adding citation lists to Policies & Guidelines page
Chet still needs to follow up with Carol. Keep this item open.
3. Update on Board meeting next week
Chet explained that the Board had a 1/2 hour for the TAB on its call Tuesday 22 September. The topics to be discussed were the proposed changes to the TAB charter and the TAB work plan.
4. Laurent's self-certification feedback request
Ashok volunteered to add a paragraph on conflict resolution.
Group discussed the purpose of the document. Consensus was that we should present a range of options along with their pros and cons and any risks we identify. We should not recommend one or another.
Group discussed idea of allowing companies to post Statements of Use to a TC page similar to the IPR page. We agreed that is neither certification nor conformance and consensus was not to present that idea in the document.
Group discussed other details. Consensus that the TCs should be the ultimate arbiters of test suites, etc.
Kevin suggested that next meeting we start by reviewing changes to the doc so far and accept/reject to get a clean starting point. We also agreed that the doc will need an executive summary section.
5. Publishing citation lists.
Tabled to next meeting.
No other business. Next meeting will be October 2nd at 2:00 PM eastern time.
Minutes respectfully submitted by Chet Ensign on 30 September 2015.
anonymous morphed into Ashok
Kevin Mangold (NIST): meeting is at 2 pm EST, right?
Kevin Mangold (NIST): for some reason i had 4-5PM ET blocked off... i must have put that there a long time ago
Kevin Mangold (NIST): ok, signing out. meeting minutes say 4 PM ET. talk with you all soon.
Chet: Calling in now
anonymous morphed into Patrick
Patrick: Usual number?
Chet: Yes, same as last week
Chet: +1 559 546 1700
Kevin Mangold (NIST): the announcement i got was that starting oct (6?) the new # will be 641-715-3822 ... putting it here so we'll have it written down somwhere
Kevin Mangold (NIST): somewhere*
Chet: Patrick, you dialing in? We'll start as soon as you're here.
Chet: Attending: Ashok, Chet, Kevin
Chet: #1 - Approval of minutes
Chet: ... Agenda
Chet: No disc, no obj, appved
Chet: #2 - approval of minutes
Chet: No disc, no obj, appved
Chet: #3 - status of action items
Chet: Patrick joins
Kevin Mangold (NIST): "Details forthcoming
View your event at https://www.google.com/calendar/event?action="">
Chet: * Patrick to sum up referencing issues in specs, from his reviewer viewpoint. Issue is doing an analysis on problems with references in OASIS specs for reporting to the board.
Chet: Patrick sent email - we can close
Chet: Possibly share with the Board for interest?
Chet: * Chet - draft a TAB note to the DITA TC expressing our understanding of their need to address some of these comments in next release of DITA and offering to work with TC members on improving these when they start their next round of editorial work.
Chet: Chet sent note to DITA TC in email. We'll close this & return later if need be
Chet: * Chet - send work plan to Gershon for presentation to board.
Chet: Done. Can close.
Chet: * Patrick / Chet - prepare another draft version, approve by email if possible, and if approved forward to Gershon.
Chet: RFC list output is done. The W3C part remains. Keep open.
Chet: * Chet - contact Carol Geyer about adding citation lists to Policies & Guidelines page
Chet: Keep open.
Chet: #4 - Update on Board meeting next week
Chet: Chet to send call in details to list
jacques (Fujitsu)1: will join the call in a mn...
Chet: Jacques joins
Chet: #5 - Laurent self-certification feedback request
Chet: Ashok: did I add anything on conflict resolution? Chet: no - Ashok: I can add a para
Chet: Jacques: as far as the role of the TAB is concerned, we don't need to talk about why but rather to put reccommendations - waht are the pluses and minuses of each option
Chet: J: TAB should stay away from saying yes we should or no we shouldn't
Chet: Ashok: agree - I think OASIS is going to do this
Chet: Kevin: agree - that's what we were asked - how to do certification not should
Chet: K: what we could do is emphasize risks if we see any
Chet: Chet: minimalist - just a place for people to post claims of conformance versus the conformance clauses
Chet: Jacques: test suites might not cover all conformance clauses - or might not pass 100% of tests
Chet: Chet: thinking more minimal than that - just someone stating they comply with a spec conformance clause
Chet: J: that's not really conformance then - if there is no test suite there is no conformance
Chet: K: but this is basically the Statement of Use
Chet: J: No
Chet: It is wording but we must be careful with how it is worded
Chet: K: no that is exactly what a SoU claims
Kevin Mangold (NIST): "Statement of Use", with respect to a Committee Specification, is a written statement that a party has successfully used or implemented that specification in accordance with all or some of its conformance clauses specified in Section 2.18, identifying those clauses that apply, and stating whether its use included the interoperation of multiple independent implementations. The Statement of Use must be made to a specific version of the Committee Specification and must include the Specification's approval date. The party may be an OASIS Member or a non-member. In case of a non-member, the Statement of Use must be submitted on the TC comment-list. A TC may require a Statement of Use to include hyperlinks to documents, files or demonstration transcripts that enable TC members to evaluate the implementation or usage. A Statement of Use submitted to the TC must be approved by TC resolution as an acceptable Statement of Use with respect to the Committee Specification. A party can only issue one Statement of Use for a given specification. When issued by an OASIS Organizational Member, a Statement of Use must be endorsed by the Organizational Member's Primary Representative.
Kevin Mangold (NIST): "a party has successfully used or implemented that specification in accordance with all or some of its conformance clauses specified"
Chet: J: but there is no objective way in the SoU of measuring conformance - that is only if we have something to test against
Patrick: J: with no objective criteria, you can't claim conformance - interesting implications for non-machine processable "standards"
Patrick: Kevin: That is a good point about conformance clauses. Perhaps the TC Admin should rule statements of use as invalid in cases where there are vague conformance clauses.
Chet: J: you *can* claim successful usage but *not* conformance
Chet: Ashok: if we end up with a really weak requirement - some ability to make a statement with nothing to back it up - it make hurt OASIS's reputation
Chet: A: we have to do something
Chet: A: have some kind of testing or checklist of validation
Chet: A: I argue that we have to do *something* - maybe not strong 3rd party verification but *something*
Chet: Patrick: just accepting statements is not a good plan. Agree w/ Ashok and Jacques
Chet: P: also, it is hard enough to navigate the site as it is - need to have something well defined that is prominent and drives both OASIS value and the strong validation of its specs. There would be an *incentive* to pass a test suite then. You can say something in a prominent place.
Chet: Kevin: agree completely that we should have a test suite. I was just saying that what was proposed was a publication of statements of use
Chet: K: going back to what Don T. said works for his program - it is up to the vendor to ensure their results
Chet: Chet: what do you think - should we even bring up the idea of a page for publishing SoUs?
Chet: Consensus - no
Chet: K: as discussed in Burlington we could have levels of certification - from checklist at the low end to test suite at the high end
Chet: J: Oasis not likely to be able to enforce a common solution on all TCs
Chet: Chet: does OASIS drive this or does the TC drive it?
Chet: Ashok: what if an outside vendor comes in with a test suite ?
Chet: A: what if the vendor came in with a test suite of their own and said 'here's what we did' - would we publish?
Chet: J: would it be a test for *all* vendors? bcz if someone creates their own test suite, isn't that mostly like a statement of use?
Chet: J: the TC needs to approve any test suite that OASIS hosts
Chet: A: a TC is not exactly a coherent body though - what if people in the TC say 'this isn't what we signed up for?'
Chet: J: the key thing though is that the TC has the authority - and really only the TC has the autority - to really say whether or not a test suite is providing ample coverage
Chet: K: if a company wanted to donate a test-suite to a TC it would be like any other contribution
Chet: K: TC could decide what to do with it.
Chet: P: think it is ok to require that a party providing a test suite be an oasis member
Chet: P: what if you had a TC formed specifically to create a test suite for a spec
Chet: P: would help drive membership in oasis
Chet: Next meeting Oct. 2nd at 2:00 eastern
Chet: K: one thing we should do next meeting is go through all suggested changes and accept / reject to reduce possibility of error
Chet: J: needs to prep something more of an executive summary level for it before putting it out
Chet: A: we could just list out the options there
Sent transcript to: email@example.com