Home | Business News | Browse by Publication | X | XML Journal

The myths of "standard" data semantics: faulty assumptions must be rooted out.

Article Excerpt
Much of the literature heralding the benefits of XML has focused on its application as a medium for application interoperability. With (a) the Internet as a platform, (b) Web services as the functional building block components of an orchestrated application, and (c) XML as a common data format, applications will be able to communicate and collaborate seamlessly and transparently, without human intervention. All that's needed to make a reality is (d) for everyone to agree on and use XML tags the same way so that when an application sees a tag such as it will know what it means.

This intuitive understanding makes a lot of sense, which is why so many organizations have sprung into existence to create their own vocabularies (sets of tags} to serve as the "lingua franca for data exchange in ." This intuitive understanding is so pervasive that it's even a key part of the U.S. GAO recommendations to Senator Joseph Leiberman (chairman of the Committee on Governmental Affairs, U.S. Senate) on the application of XML in the federal government. This report warns of the risk that:

... markup languages, data definitions, and data structures will proliferate. If organizations develop their systems using unique, nonstandard data definitions and structures, they will be unable to share their data externally without providing additional instructions to translate data structures from one organization and system to another, thus defeating one of XML's major benefits.

The perspective of these efforts is that the standardization and promotion of the data element definitions and standard data vocabularies (SDV) will solve the application interoperability problem. Unfortunately, this intuitive understanding--like many intuitive understandings--doesn't survive the trials of real-life application because important (and seemingly trivial) assumptions are poorly conceived. This article will examine some of these assumptions and articulate several myths of "standard" data semantics.

For the purpose of this article, the term "namespace" will be used to denote a domain in which a vocabulary (i.e., a set of data element names or tag names) is used with a particular meaning and purpose.

Context of Position

"Context" plays an important and largely unrecognized role in data semantics. Similarly, context plays an important role in understanding the position expressed in this article. To set the context for the comments expressed herein, I'll summarize the perspective from which they originate.

My background is in industrial and systems engineering rather than computer science. I'm more concerned about the overall system behavior of integration and interoperability, and about the human element within the system. I also have 20 years of experience in the development of schemas and the standardization of data exchange schemas within the International Standard Organization (ISO). This latter experience, in particular, has made me acutely...

View this article FREE - Now for a Limited Time, try Goliath Business News
Free for 3 Days!