Amy Krane

No abstract was provided for this paper.

The impetus for the creation of the XML specification came from the desire to publish documents on the World Wide Web. HTML was too simplistic for the types of information that needed to be published, and the many optional features of SGML made implementing it too difficult in a web environment, so a simplified version of SGML was created. As the XML specification neared completion, implementors saw a wider range of uses for the new standard besides document publishing including the electronic exchange of non-document data, specifically in the area of electronic commerce.

Exchange of electronic data in commerce is not new; systems such as those based on EDI have been in place for decades. But these have usually been proprietary systems exchanging proprietary data; adding data from other sources has not been possible. In addition, such systems were large and complex, requiring large investments to participate. Many vendors would participate only as required by their buyers. Technically, messages were complex, requiring code books for the interpretation of messages and large amounts of data overhead for the each exchange.

So XML was the right solution at the right time: it was simple, intuitive, and open.

For most of XML's short life time all applications of this new technology have been described using one of two classifications: either authoring, managing, and publishing of documents, the originally intended purpose of XML; or the exchange of non-document data in business or other systems. The distinction between the data used for these categories sometimes relied on the size of the information (large = document; small = exchangeable), or sometimes on the source or destination of the data (published in a book = document; created from a form or extracted from a database = data).

Admittedly large documents may not be suitable for messaging, but that doesn't mean that they have no place in e-commerce. By considering new uses for document data it too can be useful as part of business processes and e-commerce. Adherence to the documents vs. exchange classification prevents an enterprise from getting full benefit of its information resources; restricting or limiting how data can be used will limit its usefulness and its value.

For example, similar information is often found within the enterprise in these following forms:

• end-user documentation
• maintenance manuals
• knowledge management databases
• ERP maintenance schedules
• parts procurement systems

These types of data cover the spectrum from documents to exchange, yet the information content could be the same. Other than size, why is one considered suitable for e-commerce and the other is not? By refusing to categorize these data as one or the other, documents or messages, information managers can use the same content for all of these. If no distinction is made between a repair manual, for example, and the information in a parts database, repair procedures may become as much a part of the parts ordering process as the price list.

The answer to using documents in e-commerce lies in looking beyond what is generally considered as e-commerce. When most people talk about "electronic commerce" they are thinking about the exchange of data relating to a commercial transaction. This model is sometimes called "req to check" because it covers a series of transactions starting with the purchase requisition, proceeding through the ordering process, and ending with payment for the purchase.

I would suggest that electronic commerce begins long before the purchase requisition is created. This is where document content can become part of the e-commerce process.

In e-commerce terms, a catalog is a listing of items that are available for purchase; the buyer will select items from a catalog in order to create a purchase requisition to start the e-commerce process. Catalogs could be from a single vendor or an aggregate of multiple vendors' offerings. But how does the buyer know what items he wants to purchase? By using document data. Without this data, the buyer does not have the information required to make proper selection of items from the catalog.

XML document data can be the entry point for electronic commerce. Rather than the e-commerce process starting at the catalog or purchase requisition, it can start at product information. The buyer could initiate an e-commerce transaction by searching in an equipment maintenance manual, for example, to find the parts required to complete a repair. The maintenance manual is then linked to the parts catalog where items are selected for purchase.

Document data makes the e-commerce process more valuable because the items that are selected more accurately fit the requirements that drove the purchase. Purchasing decisions are more efficient because more complete information is available to the buyer. Granada Research recognized the value that this information gives to the e-commerce process when they stated (1999), "Electronic product information is the new currency in electronic commerce." Furthermore, "Product information is a central requirement to electronic commerce, especially in business-to-business trading relationships." This product information is the document content that lies behind the catalog to give the buyer information required to make the purchasing decision.

Just as document data makes e-commerce more valuable, e-commerce makes the enterprise's document data more valuable. As mentioned above, this data exists in many forms throughout the enterprise. A vendor can take advantage of this product information no matter where it is found in the organization; data shouldn't be discounted just because it's in the wrong format. The promise of intelligent data is that it can be re-used, and reusing data in an e-commerce setting contributes to the company's profitability more than data just sitting on the shelf.