OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

ubl message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Transition Statement and ways forward

Dear All,
please find enclosed some thoughts.
Michael Dill

A - UBL as an ISO standard
ISO TC154 has released all the ebXML documents. Consequently it would be the
right addressee for further documents.
re ISO TC154: As far as I know OASIS is a A liaison member and as such OASIS
can submit to ISO, whatever OASIS wants to.

All, what I said was:
1. The votes are votes of P-member countries, and it would be a politically
interesting signal, not to have a positive vote of the country, where OASIS
is incorporated. This simply requires to talk with the U.S. delegation
2. A Head of Delegation of a country has to vote in behalf of the national
responsible body, independent what his personal opinion is. In other words,
he has to follow the decisions of the national body.
3. Therefore I had to make clear, that any statement in the transition
discussion is not an expression of the opinion of the German Delegation.

Your assumption, that the German Delegation oppose, is nothing, what I can
share. There isn't any opposition, because there has not been any

B - ISO and content standards (vs. methodologies). What I pointed out
earlier, is that ISO TC154 approved among others not only the ebXML stuff
the EDIFACT Syntax, but also the TDED and (at least parts of) the UN Layout
Keys. I.e. they are mostly methodology standards, but somewhere also content
oriented stuff. It is worth to explore, what ISO could do under the given
circumstances, where the electronic successors of the 'old' paper oriented
UN Layout Keys have to have a structure, therefore tags/names etc.

C - Increased complexity requires maintenance resources and as I've
mentioned before, that the maintenance of any submission to ISO is
definitively an issue. The work to update the TDED was really huge, and I
still admire the investment of time, which has mostly done by Margaret
Pemperton and Sue Propert.

Apparently, the membership rules require a certain participation to be a
voting member. UBL has probably 13 voting members as of the UBL website,
where 7 are individual ones. Given this experience, I feel it is quite
normal to ask, which resources will be available to do a certain

And maintenance and extension will be necessary. UBL is going to go the same
path, as EDIFACT, X.12, OAGi, RosettaNet and others already went long
before, i.e. both to extend the data and to restrict them by building
subsets (key word: EDI Lite). This will make the stuff more complex and
requires more resources than ever before.

D - schema focus
UBL is schema focussed - the normative stuff are the schemas, whatever
normative means here, and UBL uses the UBL NDR. These NDR meet requirements
as well as other schema languages meet other requirements, e.g. I think that
XBRL has its own merits and target users. Also, there are other payload
syntaxes, RelaxNG, X12 etc.
If we agree that there will be several payload grammars on the market and no
single solution, then we have to look how to organize interoperbility
between them. This is what the core components concept probably has been
developed for. Then the focus of UBL has to move and has to include models
I agree, that UBL has to serve the market, where many people, especially
starters, just want to have schemas in their hand. At the same time, it is
quite clear that a model based approach meets the needs of the more complex
reality, of governments (e.g. U.S. ), of corporates and supply chaines. Thus
models have to become normative and schemas have to become just a derivative
for technical implementation and verification, but nothing, what people
shall edit.

E - customization
Established XML standards like RosettaNet and OAGi have hundreds of
thousands of objects. Even there users ask for extensions!. This means, that
a customization concept cannot exist of subsets and facet restrictions only.
On the other hand, the experiences of extensible XML standards seem to be
that is is very difficult to establish interoperability, where users extend
stuff themselves.
ISO 15000-5 CCTS as a semantic driven methodology looks as well as other
related approaches like ISO 20022 for a refinement mechanism, which allows
to customize by - technically - an extension and - semantically - a
restriction. But again, this is a customization on model level and requires
Customization Guideline documents and standards. Also: among others the
absence of any normative document how to reuse the content of the UBL
spreadsheets as real models makes it impossible to reuse them like an
imported schema, which would allow to develop a migration concept, where
users do not loose a part of their work whenever they upgrade.
As a result, a UBL compliance will be given, if a user reuse the model
library and folloows the rules of the model customization guideline.

F - Neither UBL nor CEFACT has currently any ontology standard document how
to build Core Components and BIE in a consistent manner, which reflects a
greater number of typical business situations, including. People like Anders
Tell noted this quite early. This is another area, where the experiences of
UBL are valuable for global standards.

all the best
Michael Dill

-----Ursprüngliche Nachricht-----
Von: jon.bosak@sun.com [mailto:jon.bosak@sun.com]
Gesendet: Samstag, 13. August 2005 23:54
An: jon.bosak@sun.com; ubl@lists.oasis-open.org
Betreff: Re: AW: [ubl] Transition Statement


| OASIS is US based and thus it would be not favorable, if CEFACT or UBL
| put a proposal forward and the US Delegation for ISO voted NO.

OASIS happens to be incorporated as a nonprofit in the U.S., but it is an
international organization; see the members lists at


OASIS is not a U.S. national standards body and has no closer a relationship
with the U.S. delegation to TC 154 than it has to any other delegation to TC

If what you mean is that it would be futile to attempt to submit UBL for
standardization to TC 154 in the face of opposition from any of the national
delegations -- including yours -- then I agree with you, but this is a
political reality, not an organizational one.

I regard at least this part of your reply as an rhetoric answer, because I
cannot imagine that you assume that I believe in any U.S. control of ISO

OASIS is US based and thus it would be not favorable, if CEFACT or UBL put a
proposal forward and the US Delegation for ISO voted NO.
It is common and reasonable that the national ISO body is contacted first.
The voting members of an ISO TC watch carefully whether a proposal has a
good standing, especially, if it is about content payload, where maintenance
is always an issue.
I remember quite well the big amount of time and resources to maintain the

Michael Dill

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]