For years we have been advising that the technical protocols such as SOAP, WSDL etc are in reality relatively easy to establish, whereas the really difficult problem is to establish common semantics for the contents of the documents. This week we have an announcement from the OASIS UBL Technical Committee which is a stimulus to look again at this interesting area.
The introduction of XML has been incredibly successful; in fact it has been too successful because it has led to the development of many different versions of XML documents many covering very similar content. The question therefore is how many versions of common business documents such as Purchase Order and Invoice, do we really need?
UBL ANNOUNCEMENT
The OASIS Universal Business Language TC this week announced that UBL 1.0 has been approved as an OASIS Committee Draft and is now available for general use.
UBL adds semantic context to an extensible framework such as Web Services or ebXML. Web Services defines transport concepts such as Service, Message, Policy, Transaction, etc, whilst UBL defines the business payload that would be carried by a Web Service with a library of XML schemas for reusable data components such as Address or Item, and schemas for documents such as Order or Invoice that would be composed from those components.
The approval of such an initiative ought to be good news. Standardized business documents and processes will be vital to the success of many Web Service scenarios, especially in a B2B context where there is a many to many relationship between participants and it makes no sense for them to waste time each defining their own schemas, yet alone the problems of transforming them to each participant’s individual schema.
And to emphasize the aims of the initiative, UBL is designed to provide a universally understood and recognized commercial syntax for legally binding business documents and to operate within a standard business framework such as ISO 15000 to provide a complete, standards-based infrastructure that can extend the benefits of existing EDI systems to businesses of all sizes. UBL is freely available to everyone without legal encumbrance or licensing fees.
ISSUES TO CONSIDER?
Sounds great. But wait a minute - there are a few issues here that we need to consider.
1. The UBL Specification dated 1st May 2004 says it is "Designed as an implementation of ebXML Core Components Technical Specification 2.01". No mention anywhere of Web Services. Apart from the use of SOAP for messaging, ebXML continues to use its own protocols in contrast to Web Service throughout the rest of its stack. We noted the intent by the ebXML community to progressively migrate to Web Services in a recent NewsWire Item, but in this specification, there is no mention of Web Services, so is this a problem? Actually it appears not. The UBL is independent of transport and is a document and schema level standard; ebXML uses XSD, so the UBL may be considered as implementation neutral.
2. The sponsors of UBL include ACCORD, CommerceOne, Hitachi, NEC, Oracle, SAP, SeeBeyond, Sterling Commerce, Boeing, Sun Microsystems and US Dept of the Navy. But Microsoft, IBM and BEA are notable by their absence! Now looking back over the archive materials we note that IBM officials have made various non committal statements over time that may be summarized as "we are watching what happens".
Let's try to interpret what's going on. There are two primary conclusions you might come to.
a) standardization of documents is very different to technical protocol standardization because there will always be diversity in content. IBM and other platform vendors might prefer to stand back from this and allow market forces to decide. Does a platform vendor really care about the content carried on the platform? Well the answer to that is YES, if the conflicting content standards militate against the success of the platform.
Or b) IBM and Microsoft are adopting their customary standardization strategy, which has been demonstrated for WS-Reliable Messaging. Viz: if an existing committee has already been established, and there is not an opportunity to submit a developed standard, then IBM and Microsoft prefer to continue developing their standard outside of a formal committee structure.
In this case, in the absence of any comment from IBM in particular, we prefer the first alternative. This is not just CBDI being charitable, rather we assess that the platform vendors are quite happy to see others take on this really difficult challenge, and the progress to date suggests their bet will prove correct.
3. But UBL is not the only game in town! While UBL has UN/CEFACT and OASIS backing, there is an existing standardization process running over at the OAG, which has considerable success over many years developing Business Object Documents (BODS) harmonizing the business document interoperability between the major packages. There is actually some overlap between the OAG and the OASIS UBL TC - NEC, Oracle, Sun and Sterling Commerce are supporting both. IBM is a member of the OAG. We note there are moves to collaborate on business document specification through cross standard body initiatives, but there is clearly some considerable overlap and specification conflict that needs to be resolved.
Even more interesting is of course the vast number of vertical sector industry organizations that have already developed their own versions of core concepts, which have been the real early movers in this area. THEY are streets ahead, and will see UBL as disrupting their activity. They will be asking the question - to what extent does an insurance or automotive sector company benefit by having cross industry standardization of core concepts and documents? Useful perhaps, but probably not a high priority.
UN/CEFACT PROVIDES BROADER VISION
Perhaps a more compelling story is made if the standardized business documents are seen in context with a collaboration methodology and framework. Last year we reported on the UN/CEFACT efforts to develop a Business Collaboration Framework (BCF). This provides a layered architecture for collaborative initiatives, and a technology neutral implementation approach including both ebXML and Web Services. However last August UN/CEFACT stated that they "will support only one document-centric approach to XML content and its desire is that UBL will be the foundation for that approach."
SUMMARY
The UBL announcement is interesting and provides foundations for wider adoption, development and rationalization. However it is going to take a long time before standards are universally adopted in this area.
We suspect many potential users will be looking for broader cross industry support, not necessarily platform vendors, but key influencers in consistent document deployment including application vendors and BPO specialists. This list should of course include Microsoft as vendor of their MBS applications.
Broader cross industry support would be a useful step to reinforce the separation of the document standards from the technical infrastructure. Standards need loose coupling also!
Finally it would be useful to hear from the platform vendors to confirm they are not adopting Plan B as outlined above. |
Home