RE: [ebXML-dev] UBL versus OAGI BODs?

[Todd Boyle <tboyle@rosehill.net>]

At 11:08 AM 2/6/2003, Marty Burns wrote:
>What do you think the mechanism is that results in congruency? Do they fight
>it out in the market place until one becomes out of fashion? Is there a
>standards forum where it is resolved?

I just wanted to demonstrate somehow, the logical fallacy
of two different models claiming to be accurate models
of common business processes if they were not congruent
with each other.

It is my impression that both UBL and OAGIS are pursuing
both of the goals I mentioned -- (optimum identification
and selection of information elements and business processes
for the whole market, and, accuracy in modeling those scenarios.)

Regarding which vocabulary is better than the other, the
authors and heavy users of OAGIS and UBL will be first to
fully understand any incongruencies.  I have no doubt that
both specifications will improve continuously, at some rate.

Collectively there is a tremendous amount of anecdotal
information and "pieces of the puzzle" among many hundreds
of people, who are not contributing to the improvement.

I was quite impressed by the 38 inhibitors to standards
convergence identified by the work session of 80 standards
professionals, at OMG Interop in 2001.

Somewhere in the future, I'm sure there will be software
platforms and economic mechanisms, that allow all of the
knowledge of hundreds of people to be aggregated.

There is giant economic benefit in standards, but we have
not found a way to pipeline some of those financial gains
to the people who contribute their time and expertise.  Only
the largest users see positive returns on contributing to
standards, resulting in an excessive focus on their needs.

Perhaps, there is some missing enzyme?  DRM technology?
A distributed accounting or micropayments framework?

Standards processes are complicated. Perhaps there needs
to be a dedicated software application for standards processes
(as distinct from modeling, or WebEx, or communications.)

If the goal is a weekly football scrimmage to arbitrate
differences among different schemas perhaps the application
would incorporate a Core Component registry, shared
models, and an accountable long-term message archive.
Perhaps it would be impossible to create a message or
state a position on a message without every word tied to
a common glossary, resulting in massive cross indexing
the different context domains and developers' arguments.
Perhaps a voting infrastructures yeah, not just Yahoo votes

OK you asked. That's my opinion.

You're aware there is a MOU (Memorandum of Understanding)
between our standards bodies sorting out their respective domains.
This includes ISO, IEC, ITU, CEFACT, OASIS and since Dec., OAGIS.

Note the proliferation of MOUs are proliferating in other domains, and I am 
inclined
to be skeptical of the concept of a top-down MOU,
http://www.google.com/search?q=MOU+%22Memorandum+of+Understanding%

oops, 159,000 pages of MOUs...hmmm where is CEFACT...here we go.
http://www.google.com/search?num=30&hl=en&lr=&ie=ISO-8859-1&safe=off&q=MOU+%22Memorandum+of+Understanding%22+cefact

ok here is 2001,  http://www.omg.org/interop/presentations/MoU.pdf
and http://www.itu.int/ITU-T/e-business/mou/

Judging by other domains, MOUs often are mutual arrangements between
governmental i.e. central command models, i.e. pragmatic brokering about
boundaries.   In business contexts, it would be natural to regard these
as division of markets. They might result in eliminating competition for
the privilege of creating standards itself.  Or a brokering among OAGIS
and UBL etc. aligning the existing models resulting in removing
competitive pressures on quality or completeness.  This is only a
theory.  My real concern (as always) is the systematic neglect of
some business processes that are un-economic to incumbent sponsors
of standards.

For example, some basic scenarios between individuals are entirely
unsupported such as the basic "IOU".

Show me a schema where I can simply relinquish ownership and
control of a simple quantity of a resource.  I'll sign it and send it
to my customer by email.    Show me a simple IOU or "You owe Me".
With a URI identifying the reciprocal resource flow.  These atomic
entries are usable, end-to-end by SMEs.

Look at sections 2.2, 2.3, and 2.4 of the MOU.PDF which guarantees
a long term architectural stackism, i.e. a technical outcome, by its
explicit division of domains between ISO, IEC, and ITU.  Stackism
in an ebusiness architecture violates end-to-end.  It has not delivered
the reliability or interoperability among different vendors' components.

Admittedly I need to research the MOUs' activities, to the extent
they are visible on the web, but one thing I can guarantee: these
MOUs are adverse to individuals and will neglect individuals'
scenarios, because there will never be international user groups
representing the individual.  An individual would observe, a
concentrated cost and dispersed benefit pattern.   A platform
vendor will see a mechanism for concentrated benefit and
dispersed cost, and all the economic theories begin to apply,

Respectfully,

Todd Boyle CPA  9745-128th Ave NE  Kirkland WA
http://www.gldialtone.com/end2end.htm
AR/AP everywhere  www.arapxml.net