Fast XML facts!

Frequently when two systems are being designed to communicate, which is increasingly the case in this Internet era, programmers face the sticky issue of how to manage the COM objects, which are created on the separate machines. Obviously, moving the COM objects, classes and ActiveX controls you have in your application around the Internet is not ideal. Moving the data and “state” (the values and content) is a much better approach in terms of maintenance and speed.

With this in mind,how do you structure your data? Do you create your own custom format in which your distributed objects communicate, or do you use XML so that many other systems can in future participate? XML is the perfect medium in which to describe the state of these objects for communications to other systems or objects.

In a disconnected medium such as the Internet, synchronous (where two systems are synchronized in their operations) communications over the web as it is not easy to achieve. Therefore there is a huge need for asynchronous messaging and even for indicating the “state” of that message. Again, XML is perfect for describing this.

Just in case you thought XML is some closed COM object with tricky methods and properties, it isn’t. XML files are simply text files and XML itself is only a text string.

Since they are only structured text strings, this makes them ideal for communications between two components or two systems that were never designed to communicate with each other! Imagine this common scenario. Two products from two different vendors have different custom data formats for their purchase orders - therefore they are unable to exchange their data. But if their developers use the XML approach to structuring their purchase order data, the two systems may be able to exchange data at some time in the future – without overhauling their existing systems!

OK, I hope you are sitting… this one is a biggie. By using XML to structure your data it is possible to achieve backward and even (gasp!) forward compatibility! Future applications will still be able to understand your past data and visa versa.

By now you are wondering just how the structure is being provided. This is achievable because the XML can contain links to other files. This is useful because they can provide a link to an XML file called a “schema” which holds a description of the correct XML structure. Now you may be able to see how versioning of your schemas can result in forward compatibility. If your future document provides a link to a schema, your older program can access that schema and understand the document.

External files that can be used, could include an XSL file (which stands for eXtensible Stylesheet Language), which holds the details on how to appropriately display the XML file. For example, if I sent you an XML file with several hundred lines of numbers in it, the same XML data/message can be displayed as an email, a word document, as a table containing data or even as a graph or simply as text. All I have to do is provide the correct XSL file for each choice!

This is most useful because the XML file itself is only the data, and the “look and feel” is kept separately in the XSL file and only applied according to the context it is in and according to the data it is displaying!

If all of this isn’t exciting enough, XML actually describes data by using a schemas, so it is possible that XML messages and documents can be routed based on the content rather than being routed from one person to another. Subscription to content becomes not only possi ble but also accurate.