Re: ERP standards

[Todd Boyle]

>>please feel free to enlighten me if a standard exists that is practical.

Neil, your comments are quite rational.  I agree, without reservations,
there has not been any standard for the exchange of transactions among
the internal applications of a company that had sufficient following to
provide any payoffs.  OAGIS, could have been a candidate.  But it has
some fundamental weaknesses.  ERP is not even a candidate inasmuch as
it ignores the 50 million SMEs in the world.

Now in 2002 you have three broad choices.

1. General Ledger standards - By this I mean, standard conceptual
elements and standard names for things like transaction dates, times,
parties, accounts, etc. necessary for exchange of information to/from
accounting or business systems.  There are three groups on the planet,
today, who give a damn about GL standards.

- Eric Cohen and his group at the XBRL Consortium,
- Robert Lemense and the D14 domain committee of EDIFACT, and,
- our group at ArapXML who produced the OMG GL and OMG AR/AP models,
and are members of the OMG and the Core Components workgroup of
UN/CEFACT.

Inevitably our three GL groups will zero in on the accurate picture and
combine the models. Meanwhile they don't talk to me so, I can't tell you
what the h*ll they are up to.  XBRL does not listen to anybody or share
their work in progress, or allow your vote on it, unless you're either
a target for their XBRL Framework, or, paying the $10,000 annual dues.
Thats $800/month, for the privilege of then, contributing even more money
and time to build the standards.

I actually was stupid enough to fly to Orlando in December to meet with
the XBRL GL group.  For a full 8-hour day, they did maintain an
astonishing wall of confidentiality, just as CPAs do in commercial
negotiations, never disclosing anything of their positions in design of
a general ledger schema or economic ontology.

The UN/CEFACT bodies only conduct their dialog in private discussions,
and in physical meetings every 6 months in international locations,
usually outside the US.  In that sense they are like the Davos group. I
have asked many times for any drafts or even discussions of principle
design, but the invariable result from this group is some assertions of
political process, releases of whatever new regulatory body they have
created.  Supposedly, the D14 of the UN/CEFACT will publish some kind of
GL model soon, perhaps at the Barcelona meeting in Spain, in March.

Here is a typical encounter with Robert Lemense who never participates
on technical or design discussions.  The guy is 65, he is part of
the French EDI establishment.  He was a champion of ENTREC.

http://lists.ebtwg.org/archives/ebtwg-ccs/200111/msg00084.html
http://lists.ebtwg.org/archives/ebtwg-ccs/200111/msg00092.html

Bear in mind, the world is not beating down the doors looking for
a GL specification or even a family of EAI integration schemas like
OAGIS, SMBXML or QBXML.  They happen to work pretty well. But
what difference does that make if *none* of the commercial software
vendors is utilizing them?  other than perhaps, their own proprietary
interface (if you're lucky)

It is only the individual and SME who really needs a GL standard...

2. e-business integration standards.

Obviously, the number of industry specific semantic models has grown,
and have gotten much more detailed and accurate in every industry.  Look
at all these diverse standards! --new and old, continuing to evolve and
develop.

http://www.diffuse.org/0201-ec.html
http://www.diffuse.org/0112-ec.html

There's also the nearly daily news on Robin Cover pages, but
that is just within the universe of XML (technology-specific),
http://xml.coverpages.org/

These are not bad news and these, are the real battleground where e-
business semantics are being forged. Not the centeralized standard
bodies.   So, the question is, similar to General Ledger interface
standards:  how can horizontal interop. be achieved in a world of
excellent vertical schemas being used in every industry? There are two
answers really.  Bigtime mapping infrastructures like Biztalk Server or
EAI platforms, or, hopefully, some future metadata registry and
open source code, that enables developers toachieve mapping more easily.

3.  The Core Components framework.

Core Components is the common metadata architecture that applies the
principles of ISO 11179 to the business domain.  This is a very large
subject and the place to start is perhaps reading some easy warmups,
from the magazines on the web.

Core components technical specification provides the rules, for designing
semantic elements. Users can combine them anyway they like.  This is not
about prescribing anything, it is about nailing down the most obvious
and wellknown entities like dates, parties, locations, products, contacts,
and the vocabulary for commitments and fulfillments.  These conceptual
entities are already well established in contract law. There is no doubt,
their brief definitions can be stacked up like a dictionary, with unique
identifiers, and we can all get down the road with a single language.

The Core Components framework removes the infighting over the naming of
the element, or the syntax of expressing it as EDI, XML etc. or national
biases or *any other objection.*   Since it is fundamentally a
dictionary of atomic elements, you can assemble them into any document
you desire. There is no doubt, this is the way forward. Core components
can describe all of those excellent vertical XML schemas.  They don't
have to cooperate and they can wish it wasn't true.  Nothing can stop
you from creating a core component version of AnythingXML, which is
therby, interoperable to some degree, with your own component model.
Nothing can stop me from interoperating with Robert Lemenses' thing
if he ever publishes it, or with the XBRL GL which is occasionally
released to the public after it's been decided by their members. And,
nothing will stop the users of ARAP GLIE's from abandoning it and
adopting the XBRL or the UN/CEFACT GL.

This is where my fingers get tired.  You really should install Poseidon
and join with Arne and I to continue the work on the version 2 of ARAP
Submission to the OMG with its associated set of Core Component
semantics. Let's make it just better.  The registry is a meritocracy.
Regardless of whoever discovers, and correctly defines, the atomic
entities, or the correctly designed aggregate entities, they will be
there for 100 years.  Picture yourself during the renaissance, when
scholars argued over the definitions in the Oxford dictionary.  That's
what's happening here, except that it will not take long.  A couple more
years.

Thanks for listening if you're still there,
Todd
Todd Boyle CPA  9745-128th Ave NE  Kirkland WA
International Accounting Services, LLC  www.gldialtone.com
address@hidden  425-827-3107  project www.arapxml.net



> At 04:04 PM 2/23/02, Neil Tiffin wrote:
> At 3:18 PM -0800 2/23/02, Todd Boyle wrote:
>> GNUE project is certainly not unique in ignoring various
>> standards of course.  We should count our blessings and
>> salute Neil, Derek, and other key developers for their generosity
>> in offering this open source project to the community.  They're
>
> Having worked with GNUE for almost 2 years I think the issue is NOT the
> lack of desire to use standards.  I for one would much rather use
> someone else's prior work in the form of standards instead of trying to
> create a beast from scratch.
>
> My problem is that I am not an accountant and don't have the time to
> sort through all of the noise (standards that are being proposed, but
> will never be implemented or represent an accepted standard).
>
> I have not found an accounting standard that applies to GNUe.  There are
> all sorts of standard that are vying for control of how accounting is
> done.  But I have not found one that is geared for internal systems.
> Most of the ones mentioned, so far, have been for data interchange and
> they are not currently practical for high volume transactions internal
> to a company.
>
> Of course, my look at accounting standard has only been cursory, so
> please feel free to enlighten me if a standard exists that is practical.
>
> Neil