XML for Not Yet Techies 2

XML is a system for defining other languages. XML is for creating documents that are so structured that they are smarter than the average web site.

For example. If you know HTML or have seen it, you know that the way to tell a web browser to begin a new paragraph is with a "<p>" symbol placed inside the document right where you want the new paragraph to start. That "<p>" tells the browser to start a new paragraph. It does not tell the browser or the reader what the paragraph is about.

What if you could put tags into a sentence such as this: <moviestar>Drew Barrymore</moviestar> waved to the cameras. Or: <author>Drew Barrymore</> autographed her latest book.

Anybody with the least familiarity with HTML can recognize the tag structure. But movie star and author are not standard HTML tags. Today’s browsers would ignore them, tomorrow's XML ready browser will understand them.

The second sentence tells the browser of the not too distant future that it is not referring to the Drew Barrymore who is a movie star, but the one who has written books. They are the same person of course, but they are two different aspects of her. The <moviestar> tags could be further subdivided into different parts of her career, say as a child in ET or as an adult, or even into individual movies.

So what, you say? You knew that already. Yes, you as a human being could read those two sentences without the XML tags and, knowing who Drew Barrymore is, understand. But your browser can’t, yet. Soon, it will.

And what are the practical on the XML computer job advantages to this? What if you want to search the web for information on her first book. If you put "Drew Barrymore" in a search engine now you will get hundreds of thousands if not millions of entries. Many will concern her movie career. Some links might be about the Drew Barrymore who works at the HI-Tech Co’s research lab. I made that up, but when you search even for famous people it’s not unusual to get hits for many different people.

But if you tell the search engine to look only for the "<author> tagged "Drew Barrymore"s, you will be much more successful.

That sounds like anarchy, I know. How can we get away with it? You structure how your document’s information is classified, as well as how it’s presented. (HTML is not going to go away.)

That is what the "extensible" part means. You can extend it as far as you want to take it.

Now, if you think about this for a minute, you’ll realize that it’s additional work, and most of us don’t like doing additional work unless it’s necessary. My own opinion is that as it becomes more common across the web, most of us will come to adopt a commonly recognized style sheet set of tags that will be very popular. We will not try anything new or innovative unless it’s necessary. In fact, there will probably be software that will automatically tag our routine documents for us.

XML - Page 3