ebXML at Five - What’s a heaven for?

Tim McGrath

When Mike Rawlins suggested an update on his ebXML review I looked back on the experience of implementing some of the ebXML framework. I also have had the privilege of being exposed to other ebXML initiatives through various committees and conferences.

My recent involvement with the work of ebXML has been in creating a common set of XML documents. As such I will not comment too much on the work of the ebXML "infrastructure" specifications except to say that as the gloss fades from the Web Services technologies, ebXML seems to be holding its own very nicely, especially in the Asian market.

With regard to the documents themselves, ebXML clearly identified that delivering XML schemas for business documents was outside its scope. What ebXML did provide was a conceptual framework for building such documents known as the Core Components Technical Specification. I like to think of this as an evolutionary step rather than a final solution. The next link in this evolution required someone to build XML document schemas based on the CCTS.

As Jon Bosak, who organized and led the working group that created XML (and now chairs the OASIS UBL Technical Committee) says

The Universal Business Language (UBL) specification was developed as an international, royalty-free standard for business documents in XML syntax specifically to avert a crisis in electronic business caused by competing XML syntaxes.

UBL was one of the first implementations of the ebXML Core Component Technical Specification. And as an early adopter of the ebXML CCTS, UBL's development work contributed to CCTS version 2.0. Following the release of UBL 1.0 in 2004, we now have significant implementation experience that we intend to use for verification of the next CCTS release.

As a community driven initiative, UBL (like open source) has no real monitor of its adoption. In fact, we would be concerned if we could monitor usage. However, we do know of several significant implementations because our membership includes many people from organizations using UBL in their core business.

The level of adoption of UBL has been heartening to those involved. With UBL 2.0 being released later this year we suspect UBL will cross the chasm of technology adoption to a stage where it becomes a pragmatic tool of choice.

One of the keys lessons here is that the development of ebXML is continuing. It is evolving based on the experiences of implementation.

Maybe we had overly ambitious expectations for ebXML. But as Robert Browning (who I don't think was involved with the project) said,

Without stretching beyond our grasp we cannot make innovation.

It would be naive to think that back in 2002 we all knew how the concepts of Core Components would pan out. Although Mike's original comments at the time do seem prophetic. UN/CEFACT is still grappling with the harmonization of Core Components. It is not a simple problem to solve.

The question we should be asking is "Are we making progress?"

Well, the concepts ebXML identified for the specification of the components themselves has enabled projects like UBL to develop a practical library of re-usable components. It is not the cure for cancer Mike was hoping for, but it is proving to be a viable alternative to the anarchy that existed prior to ebXML.

So I suggest that at least one implementation of ebXML CCTS has delivered benefit to the B2B community.

Would UBL have existed without ebXML? Personally I think not. At least not in the form it has taken.

Whilst possibly not the route that some of it proponents thought it would take, ebXML provided the framework that enabled UBL to create the XML schemas that enable enterprises of any size and in any geographical location to conduct business over the Internet.

And I think that's what ebXML set out to achieve.

Tim's Bio:

Tim McGrath led the Quality Review Team for the ebXML initiative between 2000 and 2002 and was a member of the ebXML Steering Committee. He is also a co-author of "Professional ebXML Foundations" and more recently "Document Engineering: Designing Documents for Business Informatics and Web Services" with Bob Glushko.

Since 2002 he has been a member (and is currently vice-chair) of the OASIS Universal Business Language (UBL) Technical Committee.

About  |  Services  |  Resources  |  Contact

©Rawlins EC Consulting 2007