RE: Re: [ebxml-mktg] RE: Gartner and ebXML - distribution of thearticle

[James Bryce Clark <jbc@lawyer.com>]

At 01:21 PM 9/11/02, Hickman, Michael (GXS) wrote:
> > "I
> > agree completely with David that the ebXML set of specifications is
> > modular enough and open enough to be vastly more accessible to SMEs and
> > small-scale EDI users..."
>
> Who says that "openness" and "modularity" make
> *anything* more accessible to anyone, let alone SMEs?

The original ebXML specs do. 
(www.ebxml.org/specs/ebREQ.pdf) 
Do you think we were wrong?  

    (1) Openness is essential to SME accessibility. 
It permits easy migration of user data (as opposed to locked-in
proprietary formats), which in turn permits lower cost integration with
existing legacy systems and legacy data structures. 
    By way of example, XML as a representation of data is
more broadly useful than an Oracle or MS SQL7 file.  This is *not* a
critique of the use of proprietary software:  rather, any database
program that exports into universally transportable *formats* (like XML)
is better than one than cannot.

    (2) Modularity is essential to SME
accessibility.  E-commerce and EDI architecture require multiple
interlocking functions.  Any "solution" that requires an
all-or-nothing preemptive stack change is more expensive, more
disruptive, and farther out of reach than one which permits gradual
migration layer by layer.  
    By way of example, when migrating from 1980's EDI over
VANs, some people first normalized to RNIF or OAG BODs, over the same
VANs, and then started sending them over SOAP.  Some kept their X12 or
EDIFACT docs, first migrated the transport from VAN to AS1 or SOAP, are
still transporting EDI docs, and will upgrade their business document
payload later.  This flexibility of upgrade paths is good, and
requires modularity.
    Also, having a neutral taxonomy of the necessary
layers is essential for buyers (SMEs) to cut though overbroad function
claims from sellers.  Users need to be able to see clearly in this
smoky and mirrored space.  Modularity helps.   

> Nothing I've seen about ebXML makes
> it more accessible than any other technology. What will make this
> technology accessible is when vendors package the technology into an easy
> to use product that doesn't require a huge staff or specialized technical
> skills to run, and allows companies to quickly get integrated with their
> trading partners.

    Yes it will help when mass market vendors ship. 
Usually this takes a while, for any new technology.  In my view, the
time needed to make a system working, open and functional is worthwhile,
because it promotes accessibility of e-commerce.  In contrast, 
strongly proprietary, insufficiently functional, or dysfunctional systems
do not.  Rather, they suck up all the user's purchasing power,
pre-empt other solutions, lock in the buyer, and then may fail to
deliver, leaving necessary tasks unfulfilled.   
    After two years we are just evolving a bunch of big
name betas.  How much faster did you expect this to occur?  The
only way to move faster is to pump out nonfunctional vaporware or
premature & incomplete "standards", to try keep the market
to ourselves, and push all the implementation problems on the early
adopters.  I know some people who do this well;  there aren't
many of them at ebXML.  Sorry.

Regards   Jamie Clark

~ James Bryce Clark 
~ American Bar Association Business Law Subcommittee on E-Commerce 
~ 
www.abanet.org/buslaw/cyber/ecommerce/ecommerce.html
~ 1 310 293 6739  jbc@lawyer.com