Re: URe: ERP standards

[Todd Boyle]

I've been a bit busy, but it's time to address LEMENSE Robert's points.

At 01:22 AM 2/25/02, LEMENSE Robert wrote:
> Thank you Zack.
> I appreciate forwarded "behind the back" comments from Mr Boyle.
>
> Robert Lemense

My comments were posted to a public mailing list.  I have said
all of these points to you directly.  You and Zack Coffin seem
to characterize my comments as deceitful or a breach of some trust you
have placed in me.

Why not address the architecture points I have made?  This is a
software project. My points are substantial and worthy of
discussion.  There is nothing deceitful about posting them.
Do you expect that your absence should inhibit me from
posting them?

You are occupying a position as chairman of the D14 in a standards
institution.  I am not.  It is cowardly and unprofessional for D14 to
formulate standards and circulate drafts and communications
without publishing work in progress, on semantic entities for accounting
systems.   I want those drafts.

The joint work by the EDIFICAS group and the private vendor
consortium, XBRL was also improper. While the general public was
not provided drafts of the XEGL during its period of drafting,
members of the D14 and EDIFICAS, public standards groups,
were reading them, and influencing their design as well.

The extremely slow process of D14 causes economic waste.  That
process can be speeded up by a vigorous, public discussion
of accounting vocabulary.

I didn't ask for the D14 to be my representative in deciding interface
semantics for my software. In fact, I protest D14 involvement in
accounting semantics because your semantic definitions to date,
(ENTREC etc. and the XEGL schema) are not well designed, and
are not platform independent.  The semantic entities of your
work products have baked into them, assumptions that create
advantages for large enterprises at the expense of individuals and
small business.  They are not sufficiently documented by MIGs.

They also do not adequately support recent advances
in business process models.  ENTREC is bound to the EDI
syntax which is useless to small business and has been completely
rejected by small business software vendors.  XBRL GL is bound to
XML, which many businesses reject.  We want syntax-neutral
representations.

For these reasons I objected to the D14's automatic entitlement
to the role as registration authority for core components for the
accounting domain.   That's all.  Nothing personal.

> Todd Boyle wrote:
>
> .../...
> > 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.
> .../...
>
> > 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.
>
> If Mr Boyle is a member of the Core Components workgroup of UN/CEFACT (But 
> is he really ???) he should know that the UN/CEFACT bodies do not "conduct 
> their dialog in private discussions, ...."
> The general rotation schedule is 2 meetings a year: 2 x 2 meetings 
> alternatively Europe and US, with in between 1 in ASIA, or 1 in NZ-AU.


There you go again, with this fixation on membership and hierarchy. In the 
world of large organizations, I guess you would like to   ostracize me from 
the "in group".  But there is no group.  I am an individual and I don't 
give a f*ck whether I'm a member of anything.

Robert may not be aware that the ebXML and ebTWG Core Components workgroups 
are open processes.  They have leaders, and editors, and 
contributors.  They have active mailing lists and it is possible to 
participate, as a private individual.  I contributed in some ways from 
2000-2002.  I went to a couple meetings.  I reviewed the spec, implemented 
parts of it in software, provided corrections, etc. and I came here to the 
GNUE list to evangelize it.

I'm listed as a contributor on page 3 of the current draft of the Core 
Components Technical Specification v1.8.  The project leader decides 
whoever he wants to list.  Who cares?

> > In that sense they are like the Davos group.
>
> United Nations like Davos Group ????

Yes.  Davos. The domain groups of the UN/EDIFACT are like the Davos 
group.  They do whatever they want, and the general public hasn't got a 
clue how far reaching their actions are.

Perhaps you should explain why we need government involvement to facilitate 
semantic standards.  Did governments produce languages or compile the 
Oxford English Dictionary?  I think not.  Is the United Nations a place 
where technology or R&D or higher learning is produced?   The involvement 
of governments with industry usually results in transfers of wealth from 
the individual, to the state and the selected industries.

The reader should be advised that in France, all aspects of accounting are 
tightly controlled by law.  They are even required to use a chart of 
accounts imposed by the government.

When the D14 turns up as the Registration Authority for all of the 
accounting elements in the global core components registry, that is where 
it becomes a threat to me, and likely, makes the core components less 
useful to individuals and small business.  At that point, one can only go 
back to proprietary accounting applications.    But who knows what the D14 
is planning.  It's not published.

> .../...
> >
> > 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.
> >
>
> a) Robert Lemense "never participates..." on egotistical and unilateral 
> thoughts for UN dealing is "consensus".
>
> b) The guy is a forerunner in accounting standardisation search; therefore 
> he is possibly a respectable ancestor. However, he is not (yet) 65 ...

I will concede this point.  You are a forerunner in accounting 
standardization search, and I have benefited significantly from studying 
the ENTREC creation.  Your initials points in the XEGL schema are also 
worth reading, for any GL researcher. I am also indebted for your 
generosity in sending me the French Plan Comptable book.

That's why I posted your name to begin with. You are a public figure.

But that is as far as I would go.   You're one of the guys who works on GL 
schemas.  If you were not injuring my interests with your closed process 
and liaisons with the XBRL consortium, I would be more sympathetic.

> c) .../... "is part of the French EDI establishment..." ???     Pure 
> disinformation !
>
> The guy is mandated by, and representing CSOEC (- Conseil Supérieur de 
> l'Ordre des Experts Comptables) that is the French Council of Certified 
> Accountants Institute:
> The guy is also a member of EDIFICAS Europe, organisation founded by 
> European accountants professional institutes, individual accountants, 
> individual auditors and the European Federation of Accountants. In France, 
> EDIFICAS is a member of EDIFRANCE, e-Business and EDI forum.
>
> d) "He was a champion...." WAS is not right:
> should read "He IS a champion...
>
> The guy is not only a champion of ENTREC; he is a champion of all 
> UN-Standard Messages developed by D14 since 1997 that are:
> - BALANC
> - BUSCRD
> - CHACCO
> - CHAMAP
> - ENTREC
> - INFENT
> - LEDGER
> - REGENT
>
> Robert Lemense

Oh who cares.  This is a software list.  Your post gives no useful
information.  Meanwhile, the work of D14 is unfinished, and
sequestered in some offline discussion.

- ENTREC/CHACCO have been a market failure.  Even in their home
  market (Large enterprises, EDIFACT EDI users, mostly European),
  the EDIFACT messages for accounting have been apparently not
  adopted by any significant users.   In a world of 6.5 billion people,
  a successful global standard is one which serves millions of users.

- ENTREC/CHACCO are *totally* irrelevant to GL integration.
  Companies use ORBs, middleware and more recently web services
  when they have an EAI  requirement.  Business developers would
  hardly think of EDI adapters and EDI message designs for
  passing data among systems.

- EDIFACT and EDI themselves, are also a market failure.  It works well
  for its intended use.  EDI reportedly has 400,000 sites.  EDI has
  plateaued. EDI is never going to provide the slightest usefulness to
  even 1% of 6.5 billion people, as individuals.  It is an organization
  technology.

- KEY POINT:  The Core Components workgroup has run several years,
  to create the beginnings of a syntax-neutral, platform-neutral way
  of representing metadata.  Now, that work is fairly complete and
  may be released soon for public review.  You can get a copy of that,
  and the thousands of email messages during the years of development,
  on the public internet (Unlike the ENTREC.)
  The Core Components Technical Specification (CCTS) is at
  http//lists.ebtwg.org/archives/ebtwg-ccs/200202/msg00089.html

- The CCTS provides a foundation for a large scale technical and
  political process of creating definitions of everything in the
  business marketplace. Articulating semantic entities for accounting
  use cases, turns out to be a very challenging task that can't be done
  by some guys in private, who meet every 6 months in worldwide
  locations and don't have a mailing list.  It also can't happen when
  everybody is so concerned with their image and position that
  they can't talk in an open dialog.

There has never been any technical, academic or even usenet or list
posts, visible on the internet about the ENTREC, CHACCO etc. messages.
Even their authors website do not provide an EDI MIG (Message
Implementation Guideline).  These ENTREC, CHACCO etc are perfectly
irrelevant to the ongoing work by many people, for a meaningful standard
for the integration of business processes, of which general ledgers are
a small part. Somebody please tell me, how to use these to integrate my
Quickbooks with other local applications?  Or with a web service?

Under the current leadership in D14, from 1997 to the present their
total contribution was ENTREC messages.  Meanwhile, think of the
incredible advances since then!  Here we are in 2002 and there is
still no semantic draft from D14, and yet, they want to be the
Registration Authority for the global vocabulary for accounting.

Now I am going to tell you the weakness of my position, one that makes
me very uncomfortable, and makes every accountant uncomfortable,
including the D14 and XBRL communities.

There is no authentic "accounting domain" in the whole entire domain of
computer technology, or telecommunications.  There is financial
reporting, which is a legal and political arena.  There are certainly,
user interface requirements such as double entry representations,
journal metaphors, etc. that are necessary for accountants to do their
jobs.  These are trivial things to provide by software.

There is an internal integration "domain" perhaps.  That is EAI, and
more recently, BP standards.

But there is not any authentic accounting integration domain for the
XBRL, or the D14 or anybody else, I believe.

There is a need for accountants to cooperate with business process
architects and with lawyers, defining the semantics of business
transactions and financial reporting, and only as a secondary task, to
define the user interface requirements such as double entry
representations necessary for accountants to do their jobs.

Accounting and settlement are just *aspects* of a business exchange.

If you promise to pay me $100 for my car, and I give you my car, does it
matter if you don't make an accounting entry?  Or if the bank does not
post the $100 from your account to my account? All of these problems are
ridiculously easy for computers. You put a $3 signing device in the
hands of the public, so that they can sign contracts (including
releasing money) without a military grade firewall.

Accounting is just like banking.  You have two people, fighting for
control over a single lever or switch, namely, what entry to post.
That's a political problem not a standards problem.

Let me ask another way.  Is making a payment, a business process???
The posting of an information entry in a ledger? And one which was
in every case, preceded by and required under the terms, of a contract?

Payments is NOT a business process.  Then, why do we need a D6 domain
group? Why are Payments the first sets of business processes that get
submitted whenever theres a new BP standard??

Similarly, why do we need a D14?

People will be debating this for decades.  There are over 4000 messages
in the XYZ forum on YahooGroups, for example, in less than 1 year,
for and against AICPA XYZ proposal.

This is a historic thing which will affect payments, too.   All of
accounting and payments come after the fact of business transactions.
They are deterministic.  There is no need for today's labor force of at
least 20 million people working in accounting and payments.  In the
networked world, all that manual manipulation is actually harmful.

Accounting, in the small business arena, is a system of social
organization.   That is why, in an Orwellian way, the Registration
Authority issue is so important, and the D14 is so anathema to me. Under
their control, the semantics will be controlled by accountants and
banks, in a way that perpetuates the role (and fee) of human
intermediaries instead of encryption technologies, signatures, and
objective transaction attributes between sovereign parties themselves,

TOdd