1. Roll call Will be in Barcelona for F2F #2?
Bill Burcham no ?
Mavis Cournane no YES
Mark Crawford YES YES
Fabrice Desr� no YES
John Dumay no ?
Matt Gertner no ?
Arofan Gregory YES YES
Phil Griffin no YES
Eduardo Gutentag YES (left y:00) YES (pending travel approval)
Eve Maler YES YES
Dale McKay YES (left x:55) no
Joe Moeller no ?
Sue Probert no ?
Ron Schuldt YES no
Lisa Seaburg no ?
Gunther Stuhec YES YES
Paul Thorpe no ?
Marion Royal yes (observer)
Quorum not reached as of x:09. We proceeded informally.
1'. Next week's call
Eve can't make it next week. Mark will chair.
2. Acceptance of minutes of previous meetings
20 February 2002 telecon:
http://lists.oasis-open.org/archives/ubl-ndrsc/200202/msg00072.html
Deferred.
3. Adoption of agenda
Today we are scheduled to resolve tag structure/data dictionary issues
and code lists.
Adopted informally.
4. Action item review (till x:20)
ACTION: Arofan to join Mark in writing the tag structure material.
It needs to be written in a "normative" way (i.e., as instructions
ready to be put into the NDR document).
A draft has been prepared, but it needs work before being released.
Arofan will send to Mark, who will take over ownership.
ACTION: Eve to update the code list position paper to reflect the
ideas brought up in the 20 Feb discussion, and disseminate it to the
LC SC before their meeting on 21 Feb.
DONE.
ACTION: Bill and Mavis to champion the URI/URN issue and determine
an approach. Mavis has talked to Kelly Schwartzhoff.
In progress.
ACTION: Arofan, Tim, Gunther, and Lisa to develop example and code
with LC SC. This example should grow to illustrate the modnamver
proposal.
DONE except to put the namespacing in and to create the head file.
New ACTION: Eve and Dale (and everyone!) to comment on the schema
code produced by Gunther et al. by COB 28 February.
ACTION: Arofan to liaise with the CMSC on the issue of code list
extensibility and subsetting.
DONE.
ACTION: Ron to write up a description of the "partial document"
use case by next Monday for review by the NDR SC and LC SC. Sue
to forward Ron's NDR mail on this to the LC list.
DONE.
ACTION: Matt to do a writeup on the "document design time" (fixed
vs. varying) assignment of roles.
In progress.
5. Code lists
Position paper:
http://lists.oasis-open.org/archives/ubl-ndrsc/200202/msg00073.html
- What news from the CM SC on code lists? No news.
- Has the LC SC had any reaction to the position paper? None so far.
- Identifiers for internal and external code lists: need "namespaces"?
. Would this be usable in the context methodology?
. What are the XML processing ramifications?
- Style of allowing a custom code: keyword or flag?
Eduardo proposed a hybrid solution to both these issues: a single
code field that contains a QName, where a custom code is merely
the use of a "namespace" that is not the regular one for that
field:
<!-- with xmlns:baskinrobbins and xmlns:my defined above -->
<IceCream IceCreamFlavorCode="baskinrobbins:Chocolate"/>
<!-- vs. -->
<IceCream IceCreamFlavorCode="my:DragonflyRipple"/>
The UBL documentation would note the expected (minimally accepted)
code list "namespace" support for each field. We can't, however,
require that all values from these baseline lists be supported.
The UBL documentation will have to define "namespace" URIs for
some external code lists that don't have obvious URIs of their own.
This would mean that even UBL-internal code lists would not be
schema-validatable. (Unless XSD gets revised at some time in the
distant future to allow separate validation of enumerated values'
namespace URI part and local part...)
We would need to encourage the various owners of external code
lists to define the codes as QNames in order for us not to lose
functionality when we're able to incorporate their code list
schema modules into UBL. It would be bad for them to hardcode
non-prefixed values *or* to hardcode prefixes.
This approach means that subsetting is easy, because it could
amount merely to defining a custom code list that provides only
selected values from some other (one or more!) code lists.
Extension is similarly easy. What the context methodology would
then do is change the recognized namespaces for that code field.
All this hangs on proper documentation/definition of code lists.
New ACTION: Eve to change the code list paper by COB 27 February
to reflect that day's discussion.
6. Top-level tag naming
Position paper:
http://lists.oasis-open.org/archives/ubl-ndrsc/200202/msg00074.html
- What are UBL/ebXML implications of this use case?
- Does/should the top-level tag naming affect the usage?
Ron speaks on the potential use of XML for A2A integration. He
wants to ensure that the "leaf-level" information can be understood
in the context of the various higher-level constructs in which it's
used. Eve says that in order to give context to an instance of an
element in the core library, one would need to define some document
type framework; you could either use UBL's high-level documents, or
you could define your own.
When you pull information (in XML form) out of a database, it loses
its native context. If we are going to take the message level and
establish it as its own context, we also want pointers to external
(universal) taxonomies so that the native meaning isn't lost. A
way to do this is to use a UDEF attribute on these extracted pieces
of information.
Mark clarified UBL's relationship to ebXML. He suggested that
Ron also talk to the EBTWG folks if he wants this vision to become
part of ebXML.
Regarding the specific issue of top-level element naming, it appears
that the tag name is not the definitive description of the meaning
of the tag; the dictionary entry (or the UDEF value or whatever) is.
Because of this, we think that those who want to exchange novel
packets of UBL information should be defining their own "top-level
elements" in an extension, so that they have something to hang the
semantic definition off of.
The dictionary entry would indicate which constructs are top-level,
so the tag name doesn't necessarily need to do this. Arofan's
suggestion is to name the "message level" after the process that
it facilitates. Mark suggests to look at the EDIFACT list of
messages and pick names from there (though sometimes they're not
as process-specific as they should be).
UBL may need to have a common set of attributes on all top-level
elements. These may need special naming rules because they, like
the top-level element itself, aren't "properties" of a higher-level
object class.
New ACTION: Mark and Arofan to propose a method for naming top-level
elements and attribute and put it into the tag structure paper.
New ISSUE: We need to discuss (with the LC SC) the question of
putting UDEF identifiers on UBL elements. We will put this on the
agenda for Barcelona.
New ACTION: Ron to make a specific proposal on retaining "native
context" of data elements when they are assembled into something
new for the purpose of A2A or B2B integration. To be done in time
for Barcelona attendees to read and digest it.
7. Roles and abstract types
Position paper:
http://lists.oasis-open.org/archives/ubl-ndrsc/200202/msg00076.html
Is Bill's notion of "role" the same as the "role" context driver?
We think that the LC and CD SCs are the ones that really need to
examine the proposal.
New ACTION: Eve to check with Bill on whether he will be in Barcelona
and whether he's willing to "sell" the proposal to the other SCs.
8. Other tag structure and data dictionary issues (led by Arofan and Mark)
New ACTION: Mark to ensure that the notes on other tag structure
issues in the various February meeting minutes get incorporated into
the new draft of the tag structure paper.
9. Planning for March 11 deadline
- Feb 28-Mar 6: tag structure and element vs. attribute papers
- Mar 7-11: all paper owners make necessary edits
Current status:
- modnamver: in useful draft form, but not done
- code lists: will soon be in nearly final form
- elements vs. attributes: will soon be in useful draft form
. will cover empty elements
- tag structure: will soon be in useful draft form
. will eventually cover the role model paper
- role model: this has not reached group consensus, but is good input
New ACTION: All paper owner to make edits March 7-11.
9. Adjourn
Adjourned y:56.
--
Eve Maler +1 781 442 3190
Sun Microsystems XML Technology Center eve.maler @ sun.com