New Alliance Associates Opinions

ne of the major problems of industry-wide acceptance of XML is a lack of standards

Over the past year and a half, I have attended seminars, subscribed to XML list groups, bought books and in general anxiously awaited the flowering of this next great technology break through called XML (extensible markup language). This subset of HTML is supposed to let business seamlessly communicate with each other using the Internet quickly and cheaply.

In spite of all the rhetoric and high expectations we're not there yet.

I love standards because there are so many of them. Oh, there are a number of groups who are furiously working on developing standards. Groups like RosettaNet, Microsoft's Biztalk, DISA, OASIS, and Commerce ONE. Various trade associations including electronic manufacturers, real estate, and automotive are also pushing their versions of e-commerce using XML. Too many groups with competing objectives are all publishing DDTs (data definition types) or schemas and instead of bringing different business together, just the opposite is happening.

It's dejavu all over again

Back in the dark ages of the 70's and 80's when EDI was struggling for acceptance, the event that led to wide spread adoption of EDI was the acceptance of published standards from groups like DISA, with its X12 standards and EDIFACT sponsored by the UN. Exactly how wide spread EDI became is questionable. After more than 20 years only 5% of business in this country are EDI enabled even though 95% of the FORTUNE 500 are. The reason most of the SME (small to medium enterprise) business never adopted EDI was there simply was not enough benefit to justify the expense. It was never just a matter of technical issues to be resolved. In order for a business to cost justify EDI two factors had to be present. First, was an established list of trading partners with whom information needed to be exchanged in sufficient volume and over a long enough time frame to justify the expense.
Second, the trading partners also had to be EDI capable and willing to expend the internal resources necessary to maintain their side of the transactions.

There are a two widely held positions within the EDI trading community, One is that most of the SME companies that have adopted EDI technology did so at the instance of a large dominant trading partner. There are many examples of large companies in retail and automotive who have forced their vendors to do EDI with them. Two, in many cases even after a company installs an EDI system, they do not integrate it into their back office systems and as a result loose any benefit that might be gained from reducing data entry and its associated errors. I have heard many times "they used to fax us their orders now we get an EDI document which we print out and manually enter into our system."

So what is XML going to do that EDI couldn't? Will it be as easy to use as a fax? After all you need to send a fax is someone's phone number. Will virtually any business be able to use it with out special equipment and training? Is it going to be something you can explain how to do over the phone in a sixty second conversation?

The answer to all these questions is no! At least not anytime soon. The tools and the standards just aren’t there yet. I believe that in the near future commercially available tools will handle most of the technical issues and allow users to concentrate on content. After all about five years ago when I was trying to develop our first web site the tools and technology were scarce and very basic. Today, ten year olds are putting up web sites in a few hours that professionals couldn't do in a month back then.

I predict that in the future we will be able to buy a small box that handles translation, encryption, repudiation as well as security and communications that will just plug into a network as a dedicated server. Until we get there it appears that only the big guys will be able to justify XML or ebXML to solve the problems that EDI could not address and expand the use of e-commerce into small and medium enterprises.