Minutes of Oct 13, 2009

["Jacques R. Durand" <JDurand@us.fujitsu.com>]

For review,

 

Jacques

Test Assertions Guidelines (TAG) TC meeting

Event Description:
------------------

Tuesday 2pm (California time), 13 October 2009

Host confcall: Fujitsu
US Toll Free: 877-995-3314
US Toll/International: 210-339-1806
Passcode: 9589308

Agenda: 
-------

1. Admin:
- approval of past minutes.

2. Guidelines document: Follow-up on latst version.
- review of latest draft (1.0.8.2) from Stephen:
http://www.oasis-open.org/committees/document.php?document_id=34492
- discuss and/or review:
. Guidelines "normative" new style.
. TA outcomes: where to describe (Currently end of 4.2) . 
conformance statements in Predicate? (see Jacques/Kevin) . reference to EARL lge (W3C) 

3. TA markup deliverable:
- latest schema draft:
http://www.oasis-open.org/committees/download.php/34466/testAssertionMarkupLanguage-0-8-2.xsd
- UBL invoice example:
http://www.oasis-open.org/committees/document.php?document_id=34598
- latest markup spec document draft, and XML notation inside
http://www.oasis-open.org/committees/document.php?document_id=34626


Participants:
------------

Stephen Green, (Document Engineering Services)

Jacques Durand, (Fujitsu)

Kevin Looney  (Sun Microsystems)

Dennis Hamilton (individual)

Tim Boland (NIST)


Excused:

Paul Rank (Sun)



Action Items:
-------------



Minutes: 
-------


'''1. Admin:'''

- approval of past minutes:
- September 29: approved.


'''2. Guidelines document: Follow-up on latst version.'''

Roadmap:

JD: PR scenarios: if the markup is wrapped up soon, we could package both docs 
(guidelines + markup) for a joint PR which could be 60 days again, since the markup
would need be 60 days as its first PR period. Advantage: people have the guideline
handy when reviewing the markup, and can comment on the guidelines as well.
Could impact the guidelines.
DH: If the package is intended to become parts of a same standard, should be a joined PR.
DH: although they still need be kept separate. To keep in  mind: the OASIS format of
the Committee spec determines the ISO standard format. If we submit these to ISO, 
the need be in same packaging mode already in OASIS.
JD: should we shoot for Oct 27 for completing guidelines?
SG: Oct 27 seems realistic.
SG: we could have a smaller PR for guidelines?
JD: we could but would complicate things, if we package both together.
DH: in a pPR package, some parts can be accessory, provided just as context.
SG: can we shoot for 4 weeks to get to a stable mark-up ?
JD: yes.

Current Guidelines review:

- Conf Clause: needs to be more significant. Needs to identify
"conformance target(s)", here typically a conforming "representation of the TAG TA model".
(the mark-up being just one of them). Also, need to be more assertive of what MUST
be found in a TA. The Advanced features are in fact normative as well - just optional.
So teh Conf Clause can shoot for a "minimal" TA.
- Prescription level: should refer to ISO keywords in Glossary (in addition to RFC2119?)
- Diagrams: JD thinks the large one thatintroduce concepts like "shared" not described
in spec, is too complete. Belongs to markup. But the one that focus on TA only could work,
extended with more detailed Norm Source element.
- TA outcomes: moved in section 3, but now what we had at the end of  4.2 needs be removed.
- insert of EARL references needed (from Kevin).
- We can't say that some sections are normative, some are not. By deafult all is
normative (except Appendices). Editorial style should make use of keyword when possible
(SHALL / MUST, SHOULD, etc)
Even advanced features are normative as long as they can be complied with, even if optional.
- new 4.8 section on "Conformance Statements", after the "Case of Multiple Specs".

SG: could remove large diagram, or put it in appendix?
JD: if we have the small TA diagram, may not need the large, which still requires us to explain
a lot more about "shared", etc. 



'''3. TA markup deliverable:'''

JD: simplified schema notation looks good. Makes it possible to put schema in Appendix.
DH: or even as a separate external doc referred to by URL.
SG: "shared" TA set element:
we may need more implementation experience to understand the real requirements there.
SG: the focus of the mark-up could be on the TA, as opposed to the TA set.
SG: Conf Clause: the minimal conf profile would not require support for:
- recursive TA set,
- complex semantics for "sharing" in a TA set.