Oct 2000
	Mrs. Saraswathy Ashok, Technical Editor.

Way back in 1960s, IBM created GML (Generalized Markup Language) to address the needs of structuring documents in a standardized form so that reports and other documents could be produced from a single set of source but it could not address the issue on a large scale.

It was later that SGML (Standard Generalized Markup Language) emerged from research done primarily at IBM on text document representation that was expanded and adapted for use in a wide range of industries such as in large aerospace, automotive, and telecommunications companies. It a metalanguage used for defining other languages and each so defined language is called an application of SGML. Although SGML is extremely powerful, its complexity and overhead was the reason behind SGML not being an option for representing hypertext in the early days of the Internet.

In 1989, Tim Berners-Lee designed a tag-based language for marking up technical documents that is shared over the net known as the HTML (HyperText Mark Up Language). Berners-Lee defined HTML in SGML, and is thus an application of SGML.

What do you exactly mean by "markup language"?

Almost all documents have some structure. A markup language is a mechanism to identify structures in a document. Well, this is better understood in the form of an example, say the creation of a file in a word processor. Along with the content that is created, the word processor stores extra information like the instructions to control the layout and the appearance of the words themselves, which is collectively known as the markup.

You have propriety markup languages like the one used by word processors and famous open nonproprietary markup language like the HTML.

In just a few years, the WWW and the HTML had taken the world by storm. Despite the popularity of HTML, it is severely limited to what it can do. But for the ever increasing demand for more flexibility in the Internet systems there was a need for a new markup technology that had the core benefits of SGML and the simplicity of HTML.

It was then XML, the Extensible Markup Language brewed the horizon.

The figure below clearly indicates that XML is a simplified version of SGML; and HTML is an SGML application.

Figure 1. The relationship among SGML, XML, and HTML

Is XML just another HTML? You might wonder why, if XML can be used to generate HTML, you can't write the same HTML directly yourself. Well, HTML being simple is not that flexible, which could be overcome by using XML.

XML documents looks a lot like a HTML document that consists of a mixture of data and markup. With XML you have the freedom to invent your own tags and the structural relationships between them but in HTML, both the tag semantics and the tag set are fixed. For example, in HTML, a h1 is always a first level heading and the tag is meaningless whereas with XML, you could have a Description element that has the as the start tag and ends with an end-tag .

E-Business applications are the new breed of application where you can expect constant changes as the corporations will need to deliver targeted products and services to their customers faster than competition can. XML's extensibility is the key to make this possible and is thus revolutionizing e-business. As it provides flexibility, extensibility and performance benefits in e-commerce environment, the benefits of XML outweigh those of simple HTML.

Example:
Consider an inventory of cars in which each car has a make, model and price. Now suppose you wish to record the mileage and the color of the car also. This is possible with XML. The extensibility features of XML enables you to create a new record for that car with the additional attribute without disrupting the other records as shown below.

In the modern Web applications you have data of very specific types. With HTML it is difficult to represent such data and it is XML that provides means to separate content (data) from presentation (how data is viewed) in web documents. By focussing on the content, let us see how the term metadata relates to XML.

Metadata is nothing but the data that describes data. It defined a common language that allows data to be shared among people and programs resulting in a more effective communication.

Markup languages such as XML provide a means to document metadata in computing.

Defining XML Documents ……… The minimal requirements for an XML based system are illustrated in the diagram below:

Figure 2. The minimum requirements for an XML-based system

Firstly you have the XML document itself that consists of XML character strings collectively known as markup along with the actual information content of the document known as character data. This XML document can be associated with a set of rules that specify what order and occurrence of markup and character data is permitted. For instance you could have a rule stating that all the items must have a unit price.

These rules are housed in the Document Type Definition or DTD. The role that XML processor (XML Parser) plays in XML-based system is that it splits the XML document (with or without the presence of a DTD) into "chunks" of markup and "chunks" of character data and this information is fed through an XML application.

For a document to conform to the XML standards, it must be well formed. A well-formed document is one from which an XML processor can successfully build a tree structure. This can be done much without a DTD. Well-formed XML documents can be further classified as valid if they meet the constraints spelled out in an associated DTD.

What could one accomplish with XML?

There are a number of reasons for application developers to be motivated to use XML in their applications and for generating Web contents. XML provides a simple, robust, scalable, maintainable and reusable programming model to generate static and dynamic contents for web browsers.

As far as browsers are concerned, the capabilities of XML provides a number of benefits which includes:

The content can be manipulated and rearranged. For example, calculations can be performed to generate extra content on the fly. The same content can be made to look different for different users. The contents can be intelligently searched within the browser, based on what it contains.

Searching the web has become so difficult the reason being that search engines rely on full-text searches of information instead of searches of descriptions of information. The problem with such a search is that HTML is weak when it comes to classifying information in a web page. In order to search the web with a reasonable amount of accuracy, we need an effective way to catalog and describe data. Metadata provides a solution to this problem.

The metadata information about any given document could include the following:

Authorship information

Keywords and descriptors

Recent updates and revisions

For example, Netscape has concentrated on using XML vocabularies to describe metadata in a standard way named as the Resource Description Format (RDF). Thus this metadata browser will be able to integrate RDF descriptions of data on the web to create a dynamic navigation tool that allows users to easily navigate the Web.

XML has numerous advantages as it exhibits the following features: Values are marked at the collection level, the object level, and the field level to show what they mean. Validating XML is not a problem because there are two levels of checking available.

Overall structure of an XML file is well defined. Using DTDs.

Nested tags allow different kinds of data to be placed within the same file.

The ordering of the field need not be worried about as the fields are tagged.

If additional data fields need to be added, older implementations will be able to ignore tags that have been added for newer implementation.

The above example mentioned is just one of the areas in which XML proves to be a better option. There is a diversified set of fields where you could see XML in action like:

Online Banking Push Technology with Microsoft Active Channels Concluding thoughts

From above, it is clear that XML is a neat technology that allows to communicate using standard data formats and it can thus contribute to a variety of enterprise application areas like Information Distribution, Searching and Pattern making, Application Integration, Business transactions and Data transformation. Thus XML has survived the initial skepticism and is emerging as an essential tool for the E-Business primarily because of its extensibility. This enables applications to leverage structured and unstructured data thus relieving one from the headache of managing a major MIS initiative.