Separation of Data from Process

[Previous] [Next]

XML offers different cost-reducing or revenue-enhancing opportunities to a lot of people. For technical writers, XML provides a syntax that allows them to capture the meaning of their documents. For the enterprise programmer, XML provides a syntax for getting data between objects. For the programmer building systems to provide e-commerce transactions between business partners, XML provides a syntax to capture the richness of the transactions, making the transactions accurate and timely.

To all of these people, however, XML provides at least one common benefit: XML allows them to separate their data from the processes that act on that data. This is an important point, and you will see it repeated and illustrated throughout the book.

Separating data from the processes that act on the data is nothing new. A purchase order number that exists as a field in a relational database has a particular meaning. The number can appear in many different places: on a purchase order, on an invoice register, on an invoice, and so on. No database designer would ever think of storing that number with formatting instructions indicating how it is to be rendered on a particular form.

However, when it comes to other information types, we include formatting details quite often. Word processors are notorious for combining data and processes like well-marbled beef. Consider a chapter title that is displayed as a 24-point-Helvetica-bold-centered piece of text. A human reading a line formatted this way on a piece of paper or a computer screen can infer that the line is a chapter title. But to get a computer to look for an actual "chapter title," we need to tell the computer how a title is rendered, because a computer doesn't have a human's cognitive ability to recognize patterns. Humans' ability to deal with ambiguity is something that computers don't (yet) have. For example, we might see something in the middle of our text that we don't recognize as being written in the language we are accustomed to because the phrase is rendered in italic. We know why it is rendered in italic because of our ability to recognize patterns and figure out why distinctions in typography are used. This ability is a two-sided sword, however, because sometimes we get it wrong. And in order for a computer to distinguish between the foreign phrase and something else rendered in italic, we need to explicitly mark each phrase differently.

XML provides a syntax that allows you to define each information object in an unambiguous way. By doing so, you can capture the information as one object and then process it with many different applications depending upon the requirements at the time.