Robert Sutor

1. The need
2. The initiative
3. The status

The need

EDI (Electronic Data Interchange) has been, and continues to be, a very successful technology for company-to-company e-commerce. The adoption of EDI in the Fortune 1000 is very large - some estimates place it as 98%. However, the costs and complexity of implementing EDI and doing the transactions are often too much for smaller and medium-sized companies. Outside the Fortune 1000 (most of the companies in the world), the adoption rate is approximately 5%. Smaller companies need to get into the electronic supply chain.

Many new companies have sprung up to handle auctions, exchanges, and marketplaces on the Internet and XML is proving to be a useful technology for them. If you look across all these activities, there are many different "standards" in use, from the messages exchanged, right down to the "envelopes" used to enclose them.

If we could develop common ways of wrapping the messages and model re-usable business objects that are instantiated via XML, we could give EDI an infrastructure for moving forward, we would allow vertical industries to build on this infrastructure and concentrate on their domain-specific e-business, and we would reduce the number of specifications that needed to be supported by middleware for intra- and inter-enterprise business integration. Forrester Research has estimated that by 2003, fifty times the number of transactions will take place than took place in 1999, and so the need for rapid standardization of this infrastructure is critical.

Previous

Table of Contents

The initiative

In September, 1999, the United Nations/CEFACT group announced that it was joining forces with OASIS, the Organization for the Advancement of Structured Information Standards, to create the Electronic Business XML initiative (ebXML). Like other EDI groups such as ASC X12, the UN/CEFACT group had been exploring ways to use XML and the Internet to extend the benefits of EDI beyond the large companies and the large countries, to small companies and developing nations. They had many years of experience with business-to-business e-commerce via their development and support of the EDIFACT standard. UN/CEFACT partnered with OASIS because of its mission to make standards such as XML easier to use in industrial and commercial settings. OASIS is a consortium of more than 120 companies and is frequently asked to advise industry and government groups on the status of and best ways to use XML. In 1999, for example, OASIS participated in panel discussions in a United States White House session on e-commerce and in a conference held by NATO. OASIS is also the host for the XML.org project, an effort to create a global linked web of repositories for user and application access to XML resources such as DTDs and schemas.

The purpose of the ebXML initiative is to provide a coherent solution for the problems outlined in the first paragraph above, that is, to create a single global electronic market. The timetable needed to be aggressive: a 15 to 18 month schedule was proposed with specifications being released at most every 6 months. Note that these specifications are meant to be used by for people developing other XML e-business standards and so the active participation by many interested organizations is needed.

It was clear from the beginning that the aggressive schedule would not allow us to create a new solution from the bottom up. Therefore, ebXML has actively solicited and received assistance from participants in the Object Management Group, Open Applications Group, Open Travel Alliance, Health Level Seven, ACORD, and several other standards groups.

The first meeting of the ebXML initiative was held in San Jose, California, in November, 1999. Over 120 people attended and more than fifty organizations were represented. I must say that these people were all very anxious to start working on a solution and some were uneasy having to listen to presentations rather than immediately forming working groups. Of course, with so many perspectives represented, it was important to bring everyone up to speed on the EDI/XML work and the work done by companies such as Ariba, Commerce One, Oracle, Microsoft, and IBM related to e-commerce on the Internet.

By the end of the meeting, eight project teams had been formed and charged with defining the full scope of the ebXML work. The teams are

Each of these project teams has a web page accessible from the ebXML web site at http://www.ebxml.org.

In January, 2000, the second plenary session was held (in a not-very-warm Orlando, Florida). This meeting was different from the first in that the project teams spent three full days in working sessions. These team meetings continued the work that had been carried on by teleconferences since November. Additionally, liaisons from the teams met with each other to coordinate their work. The Requirements group and the Transport/Routing and Packaging group were the first to release draft specifications after this meeting. These were made available on the ebXML web site in March, 2000.

Previous

Table of Contents

The status

In May, 2000, the ebXML initiative met in Brussels for its third plenary meeting. The project teams presented their draft specifications. In this presentation I will review their results (which were not available at the time this talk description was developed) and discuss the remaining work to be done by the consortium.

Previous

Table of Contents